the impacts of jamie™s ministry of food...
TRANSCRIPT
The Impacts of Jamie’s Ministry of
Food Australia
Jessica Kate Herbert, BHS (Hons)
A thesis submitted in fulfilment of the requirements for the
degree of Doctor of Philosophy
Deakin University
March 2015
iii
Acknowledgements This thesis forms part of a commissioned evaluation of Jamie’s Ministry of Food
Australia. In doing this research, I have had the privilege to work with an amazing
team. I would like to express my sincere thanks to my supervisors who supported
and encouraged me throughout this PhD process. Firstly, my primary supervisor
Professor Marj Moodie has provided constant support and encouragement and
never doubted my abilities. Thanks to both associate supervisors, Lisa Gibbs and
Boyd Swinburn. Associate Professor Lisa Gibbs has provided invaluable
knowledge, guidance and patience while I figured out the world of qualitative
analysis. Professor Boyd Swinburn has an inspiring mind, and I feel very
privileged to have had the opportunity to work with him.
other members of the Jamie’s Ministry of Food evaluation team. Anna Flego wa
iv
v
Declaration I declare this thesis does not exceed 100,000 words.
Abbreviations EPPI-centre the Evidence for Policy and Practice Information and
Co-ordinating Centre
GBP English pounds
GEE Generalised estimating equations
HILDA the Household, Income and Labour Dynamics in Australia
JMoF Jamie’s Ministry of Food
NCD Non-communicable disease
PRISMA the Preferred Reporting Items for Systematic Reviews and Meta-Analyses
SD Standard Deviation
SE Standard Error
SEP Socio-economic position
TGF The Good Foundation
UK the United Kingdom
US the United States of America
USD United States of America Dollars ($)
vi
Papers and presentations arising from this thesis Peer-reviewed journal articles
Herbert J, Flego A, Gibbs L, Waters E, Swinburn B, Reynolds J, Moodie M. Wider impacts of a 10-week community cooking skills program - Jamie’s Ministry of Food, Australia. BMC Public Health. 2014; 14 (1):1161. Flego A, Herbert J, Waters E, Gibbs L, Swinburn B, Reynolds J, Moodie M. Jamie’s Ministry of Food: Quasi-experimental evaluation of immediate and sustained impacts of a cooking skills program in Australia. PLoS One. 2014; 9(12). Flego A, Herbert J, Gibbs L, Swinburn B, Keating C, Waters E, Moodie M. Methods for the evaluation of the Jamie Oliver Ministry of Food program, Australia. BMC Public Health. 2013; 13(1):411.
Flego A, Herbert J, Gibbs L, Waters E, Keating C, Swinburn B, Moodie M. The evaluation of Jamie’s Ministry of Food, Ipswich: Final evaluation report. Melbourne Deakin University, 2014. Available from: http://www.jamieoliver.com/jamies-ministry-of-food-australia/ Conference presentations
Herbert J, Flego A, Gibbs L, Waters E, Swinburn B, Reynolds J, Moodie M. Jamie’s Ministry of Food reduces take away/fast food expenditure [Poster]. Australian New Zealand Obesity Society; October; Sydney, Australia. 2014. Herbert J, Gibbs L, Moodie M. “More than just a meal in a box”. Early findings from the evaluation of Jamie’s Ministry of Food Australia programme [Poster]. International Congress of Obesity; March; Kuala Lumpur, Malaysia. 2014.
Herbert J, Flego A, Gibbs L, Waters E, Swinburn B, M. M. A longitudinal qualitative study of Jamie’s Ministry of Food Australia programme [Poster]. European Congress of Obesity; May; Liverpool, UK. 2013.
vii
Abstract Australians are facing a number of challenges that discourage home cooking.
Healthy eating is vital for healthy living and meals prepared at home from fresh
ingredients or “scratch” are invariably more nutritious than meals purchased
away from home. There are a number of factors that influence people’s abilities
and opportunities to prepare meals from scratch at home. As a result, there has
been a decline in the consumption of home-cooked meals from scratch. There are
a number of public health initiatives in Australia and internationally aimed at
promoting cooking skills in the population as a means of improving the
population’s diet. The benefits of community-based cooking skills interventions
for individuals have been reported. However, the evidence surrounding the
impacts of these programmes is insufficient and further research is required to
understand their effectiveness. Celebrity chef Jamie Oliver auspices one such
cooking skills intervention named Jamie’s Ministry of Food (JMoF), which is
currently being implemented in the UK, US and Australia. Participants attending
JMoF Australia participate in weekly 90-minute classes over a ten week period.
The programme teaches basic skills to participants to enable them to prepare
simple, fresh, healthy food quickly and cheaply. This PhD sets out to determine
what were the impacts of JMoF Ipswich Australia as a community-based cooking
skills intervention?
increased participants’ skills,
viii
A longitudinal qualitative study sought to explore participants’ expectations
participants’ skills, knowledge, attitudes, enjoyment and satisfaction of cooking
ix
participants’ cooking self
participants’ enjoyment of preparing meals. Findings suggested the
environment. Improvements in participants’ confidence and enjoyment of cooking
’s
x
Table of Contents
Acknowledgements ................................................................................................ iiiDeclaration vAbbreviations vPapers and presentations arising from this thesis ................................................... viAbstract viiChapter 1. Overview ......................................................................................... 1
1.1 Background ............................................................................................. 11.2 Aim and scope ......................................................................................... 21.3 PhD candidate’s role in the evaluation .................................................... 21.4 Outline of this thesis ................................................................................ 3
Chapter 2. Literature review ............................................................................. 62.1 Chapter overview .................................................................................... 62.2 The importance of cooking skills for health ............................................ 62.3 The impacts of community-based cooking skills interventions ............ 152.4 Jamie’s Ministry of Food (JMoF) ......................................................... 532.5 Summary ............................................................................................... 59
Chapter 3. Mixed methods evaluation of Jamie’s Ministry of Food Australia programme 60
3.1 Chapter overview .................................................................................. 603.2 Evaluation and rationale ........................................................................ 603.3 Theoretical framework to guide evaluation ........................................... 613.4 Programme logic model ........................................................................ 653.5 Evaluation research questions ............................................................... 663.6 Mixed methods evaluation design ......................................................... 683.7 Summary ............................................................................................... 72
Chapter 4. Quantitative methods ..................................................................... 734.1 Chapter overview .................................................................................. 734.2 Quantitative paradigm ........................................................................... 734.3 Research aims and objectives ................................................................ 754.4 Quantitative study parameters ............................................................... 764.5 Ethical considerations ............................................................................ 924.6 Summary ............................................................................................... 93
Chapter 5. Qualitative methods ....................................................................... 945.1 Chapter overview .................................................................................. 945.2 Qualitative paradigm ............................................................................. 945.3 Research aims and objectives ................................................................ 985.4 Qualitative study parameters ................................................................. 985.5 Ethical considerations .......................................................................... 1095.6 Summary ............................................................................................. 110
Chapter 6. Quantitative results ...................................................................... 1116.1 Chapter overview ................................................................................ 1116.2 Participants: quantitative study ........................................................... 112
xi
6.3 Jamie’s Ministry of Food sample compared to the Ipswich population 1166.4 Programme outcomes .......................................................................... 1206.5 Adjusted analyses of outcomes ........................................................... 1316.6 Supplementary analysis ....................................................................... 1356.7 Summary ............................................................................................. 138
Chapter 7. Qualitative results ........................................................................ 1397.1 Chapter overview ................................................................................ 1397.2 Participants: qualitative study ............................................................. 1407.3 Participants’ motivations and intentions ............................................. 1447.4 Cooking social norms and cooking identity ........................................ 1467.5 The celebrity factor, Jamie Oliver’s influence .................................... 1517.6 Social benefits of the programme ........................................................ 1547.7 Food purchasing behaviours ................................................................ 1587.8 Barriers and facilitators to cooking ..................................................... 1597.9 Summary ............................................................................................. 165
Chapter 8. Discussion of findings ................................................................. 1668.1 Chapter overview ................................................................................ 1668.2 Research questions and programme logic model ................................ 1668.3 Major findings ..................................................................................... 1688.4 Evaluation strengths and limitations ................................................... 1878.5 Summary ............................................................................................. 192
Chapter 9. Implications and conclusions ...................................................... 1949.1 Research implications .......................................................................... 1949.2 Significance of this PhD ...................................................................... 2029.3 Conclusions ......................................................................................... 203
References ..................................................................................................... 204Appendices ..................................................................................................... 225
Appendix 1. Quality Assessment Tool for Quantitative Studies…………… 226 Appendix 2. Jamie’s Ministry of Food Australia manifesto……………....... 231 Appendix 3. JMoF protocol journal article ……………...…………...……. 235 Appendix 4. Jamie’s Ministry of Food Australia questionnaire…...………. 244 Appendix 5. Ethics documents……………………………………….......... 252 Appendix 6. Interview guide……………………………………………...... 261 Appendix 7. JMoF Australia primary outcomes journal article…………….. 263 Appendix 8. JMoF Australia secondary outcomes journal article…….……. 282 Appendix 9. Adjusted results for age, gender, employment and combined... 297
xii
List of Tables
Table 1: Cooking skills interventions – intervention and control groups ............. 21
Table 2: Cooking skills interventions – pre-post design, no control group .......... 26
Table 3: Cooking skills interventions – measured post-intervention only, no control group ......................................................................................................... 36
Table 4: Interventions with a follow-up period ..................................................... 43
Table 5: Cooking skills interventions reporting significant change over time by outcome measure ................................................................................................... 45
Table 6: Introduction of the researcher ................................................................. 95
Table 7: Demographic characteristics of participants by time point ................... 117
Table 8: Prevalence of health indicators, population 18 years and over ............. 119
Table 9: Predicted mean scores in cooking confidence, knowledge, attitudes beliefs, skills, enjoyment and satisfaction outcomes by group (T1 and T2) ....... 124
Table 10: Predicted mean scores in cooking confidence, knowledge, attitudes beliefs, skills, enjoyment and satisfaction outcomes for the intervention group (T1, T2, T3) ......................................................................................................... 125
Table 11: Predicted mean scores in behaviours and health and well-being by group (T1 and T2) ............................................................................................... 133
Table 12: Predicted mean scores in behaviours and health and well-being for the intervention group (T1, T2, T3) .......................................................................... 134
Table 13: Supplementary analysis: ordinal logistic regression and multilevel mixed model analysis .......................................................................................... 136
Table 14: Supplementary analysis: secondary logistic model (generalised estimated equation) for cooking confidence ....................................................... 137
Table 15: Summary of interview themes and domains ....................................... 139
Table 16: Qualitative participant profile at time of first interview ..................... 144
Table 17: Summary of complementary quantitative and qualitative findings ... 169
xiii
List of Figures
Figure 1 Ecological framework for cooking influences ........................................ 10
Figure 2: Literature search .................................................................................... 20
Figure 3: Jamie’s Ministry of Food Centre, Ipswich, Queensland, Australia. ...... 58
Figure 4: Jamie’s Ministry of Food programme logic model ............................... 67
Figure 5: Quantitative and qualitative paradigms ................................................. 69
Figure 6: Mixed methods design ........................................................................... 71
Figure 7: Quantitative study design ....................................................................... 77
Figure 8: Qualitative evaluation time frame and study design ............................ 100
Figure 9: Qualitative data analysis process ......................................................... 106
Figure 10: Questionnaire response rates ............................................................. 113
Figure 11: Correctly answered nutrition knowledge by group and over time ..... 123
Figure 12: Mean daily fruit and vegetable consumption - comparison of JMoF intervention participants and Queensland (QLD) state-wide 2012 data ............. 127
Figure 13: Take-away food consumption comparison between intervention participants and Queensland ................................................................................ 128
Figure 14: Jamie’s Ministry of Food evaluation - qualitative participation, August 2012 to July 2013 ................................................................................................ 142
Figure 15: Improved programme logic model .................................................... 167
1
Chapter 1. Overview
1.1 Background
In today’s society, Australians are facing a number of challenges that discourage
home cooking. Changes in the western world’s dietary patterns and lifestyles have
resulted in an excess energy intake that contributes to dietary-related disease and
conditions such as obesity and diabetes (1). In a world of increasing demands and
opportunities, time is a precious commodity and supermarket shelves are stocked
with quick and easy meal solutions in the form of convenience foods (2).
However, these meals are often high in energy and contain high levels of salt,
saturated fat and sugar. The decline in the population’s cooking skills and cooking
confidence may prevent the preparation of food at home (3, 4). This is
concerning, because preparing food from fresh ingredients generally produces
more nutritious and healthier food (5).
e the population’s cooking
cal interest is celebrity chef Jamie Oliver’s Jamie’s Ministry of Food
2
1.2 Aim and scope
This PhD aims to determine the impacts and effectiveness of JMoF Australia as a
community-based cooking skills intervention. This thesis specifically evaluated
the first Australian fixed-site, ten-week JMoF programme in Ipswich,
Queensland. The evaluation adopted a broad socio-ecological approach and
employed mixed methods to determine the impacts and sustained effects of the
programme.
1.3 PhD candidate’s role in the evaluation
In 2010, The Good Foundation and the Queensland Department of Health
commissioned Deakin University to evaluate the JMoF Ipswich programme. All
data from the evaluation of JMoF Ipswich are presented in this thesis and the
contribution of the PhD candidate to each component of the evaluation is clearly
specified below. The candidate was instrumental in the initial planning and design
of the evaluation together with the evaluation coordinator and the evaluation
investigators (three of whom are also the candidate’s PhD supervisors).
The evaluation team included the following people:
Principal Investigator:
Professor Marj Moodie (Principal PhD supervisor) Deakin Health Economics,Deakin University
Investigators: Associate Professor Lisa Gibbs (Associate PhD supervisor) Jack Brockhoff Child Health and Wellbeing Programme, Melbourne School of Population and Global Health The University of Melbourne
Professor Elizabeth Waters Jack Brockhoff Child Health and Wellbeing Programme Melbourne School of Population and Global Health, The University of Melbourne
Professor Boyd Swinburn (Associate PhD supervisor) WHO Collaborating Centre for Obesity Prevention, Deakin University
Project Coordinator:
Anna Flego Deakin Health Economics, Deakin University
PhD Candidate:
Jessica Herbert Deakin Health Economics, Deakin University
3
1.4 Outline of this thesis
This section provides an outline of the structure of the thesis. Chapter 2, Literature
review, describes the importance of home-cooked meals and their role in healthy
diets. The barriers and influences on home-cooked meals that have contributed to
a decline in cooking skills are described, followed by a systematic examination of
cooking skills interventions. The findings from this systematic review indicated a
lack of conclusive evidence of the sustained effectiveness of such programmes
and the consequent need for stronger evaluations. The evaluation of JMoF
contributes to the evidence base in this area. The final section of the literature
review describes the JMoF Ipswich programme.
Chapter 3, Mixed methods evaluation of Jamie’s Ministry of Food
programme, provides a rationale for the mixed methods evaluation. This chapter
4
outlines the programme logic, the research questions that were used to guide both
the quantitative and qualitative studies, and how both studies will be brought
together.
Chapter 4, Quantitative methods, sets out the methodology used in the
quantitative evaluation of JMoF Ipswich. The quantitative study used a pre- and
post-assessment of the intervention with a six-month follow-up and a ten-week
wait-list control. The programme’s effectiveness was measured against two
primary outcome measures, namely cooking confidence and vegetable intake. A
range of secondary outcomes relating to healthy diets, personal development,
social connectedness and affordability of healthy meals were also measured.
Chapter 5, Qualitative methods, details the methodology used in the
qualitative evaluation. Repeat interviews over three time points were conducted
with a small sample of participants to understand their expectations and
experiences of the programme, the facilitators and barriers to cooking and any
unanticipated programme outcomes.
Chapter 6, Quantitative results, presents the findings from the analysis of
the pre- and post-questionnaires to measure changes in outcomes over time as a
result of the JMoF Ipswich intervention. Six-month follow-up data are also
presented to explore sustainability of effects over time.
Chapter 7, Qualitative results, presents the findings from the thematic
analysis of the longitudinal qualitative interview data from JMoF participants to
answer the research questions around programme expectations and experiences,
and participant barriers and facilitators to cooking. In addition, emergent findings
around the celebrity impact of Jamie Oliver are highlighted as well as the social
norms and identities that participants place around cooking.
Chapter 8, Discussion, brings together the findings from the quantitative
and qualitative studies. Results are compared and contrasted to paint a broader
understanding of the outcomes of the JMoF Ipswich programme in the context of
5
the literature and social theory. The strengths and limitations of the research are
addressed.
Chapter 9, Implications and conclusions, the implications and significance
of the findings are addressed followed by the drawing of a short conclusion.
6
Chapter 2. Literature review 2.1 Chapter overview
This chapter begins by describing the importance of cooking in a broad context. It
details the significance of a healthy diet for disease prevention. It explores the role
of home-cooked meals in the promotion of healthy lifestyles and how lifestyles
have changed to incorporate fewer home-cooked meals made from raw
ingredients. The term made “from scratch” has been used going forward to
represent foods prepared from raw ingredients and is a term frequently used by
celebrity chef Jamie Oliver. This chapter describes the influences that impact on
the preparation of home-cooked meals. The literature around cooking skills
interventions is systematically reviewed, and the impact and effectiveness of such
programmes appraised. The chapter closes with a description of JMoF Ipswich,
the cooking intervention which is the focus of this PhD research.
2.2 The importance of cooking skills for health
2.2.1 Non-communicable disease and diet
Adequate nutrition and healthy diets are important determinants of health. They
assist in the prevention of non-communicable diseases (NCD) such as Type-2
diabetes (11), cardiovascular disease (12), stroke (13) and some cancers (14, 15).
Obesity is associated with a higher risk of NCDs and it is forecast that the burden
of obesity and its associated health costs will only continue to rise (1, 16, 17). In
Australia, three in five adults are either overweight or obese (18, 19). In 2005, the
direct annual cost of overweight and obesity was estimated at AUD 21 billion,
with an additional average annual cost in terms of government subsidies of AUD
35.6 billion (20).
2.2.2 Changes in dietary patterns and diet quality
7
Changes in the global food system are reflected in shifting dietary patterns and
lifestyle choices (1). There has been a dramatic increase in the consumption of
energy-dense foods that are high in saturated fat, salt and refined carbohydrates
and a decrease in the consumption of fruits, vegetables, complex carbohydrates
and dietary fibre (1). These changes in consumption patterns have led to excess
energy intake and over nutrition in today’s society (2). Changing lifestyles have
endorsed a high demand for fast-food, ready-made meals and convenience foods
which are high in energy and low in fruit and vegetables (17). The food industry
has strategically responded by offering quick and easy meal solutions for
consumers; the promotion and intense marketing of these products have
accelerated their increased consumption (2, 21). The use of “pre-prepared” foods
at home as a full meal or as part of a meal has become normal and acceptable
behaviour (22). As a result eating meals at home may not always be the healthy
option, due to a greater reliance on convenient, part or fully prepared meals and
take-away foods (23).
2.2.3 Food preparation and consumption today
There are a number of barriers that discourage or prevent people from cooking
meals at home, and there is a discourse about the lack of home-cooking and food
skills in today’s society. Lack of time is a commonly cited barrier to meal
preparation at home (25-27). Women are predominantly the food preparers and
8
are responsible for cooking in the household (28, 29). Working mothers report
time constraints in terms of both preparing meals and teaching cooking at home,
resulting in the greater use of convenient and fast foods to feed their families (30).
As previously mentioned, such pre-prepared foods are readily available in today’s
food environment (31). Changing lifestyle patterns have created a shift in the way
people prepare and eat meals; for example, it is acceptable to eat meals either
away from home in restaurants or at home using take-away or purchased pre-
prepared meals (22).
based on food prepared from “ ”
“ ”
one’s
9
“ ”
2.2.4 Environmental and personal influences on cooking
There are a number of different influences in today’s environment that impact on
individual choices. A review by Bisogni et al (2012) explained that healthy eating
is complex and diverse, and carries different meanings for different people
particularly at different life stages and through various life experiences (44). In
addition to life experience, an ecological perspective suggests that there is no
single factor to account for a particular behaviour such as cooking. Instead,
multiple factors – social, cultural and economic – contribute to behavioural
choices and actions (45). The use of an ecological model aids understanding of
the dynamic interplay between human behaviour and social and physical
environments (46, 47). Behaviour through an ecological perspective is presented
in terms of five levels, namely interpersonal, intrapersonal, organisational and
institutional, community and public policy, or societal (47-50). A social-
ecological model can be used to map the various variables at play at each level,
thereby facilitating the visual depiction of the multiple influences impacting on
individual behaviour (50). Figure 1 (adapted from Fitzgerald et al (2009) and
Robinson et al (2008) (51, 52)) illustrates the social-ecological framework used to
explain cooking choices. There are often multiple variables operating at each level
that influence behaviour. It is important to note that these interact cumulatively
and broadly across the different levels influencing an individual’s physical,
emotional and social well-being (47, 50). The next section describes some of
these influences that impact on an individual’s cooking behaviours.
10
Figure 1 Ecological framework for cooking influences*
*Adapted from Fitzgerald et al (2009) and Robinson et al (2008) (51, 52)
Cooking and consumption patterns have been shown to vary according to
demographic factors such as gender, socio-economic position, race and place of
residence (29, 53, 54). Cooking knowledge and cooking confidence are negatively
associated with education level and income (21, 29). For example, increased
income is linked to increased confidence to prepare certain foods (29). In one
Australian study, results showed that more vegetables were bought by households
where the main cook had the confidence to prepare them (55). Confidence to cook
and prepare vegetables was influenced in turn by the main cook’s level of
education and income (55).
11
One’s attitude and beliefs may also influence food choice
Self-efficacy, or confidence in one’s ability, is a good predictor of behaviour
change and shapes the outcomes that people expect to achieve (59, 60). Low
cooking confidence not only affects cooking skills but also people’s attitudes
towards meal preparation (61). In contrast, confident cooks have been found to
like the experience of cooking and tend to have a larger repertoire of cooking
techniques and dishes (61). In a cross-sectional study in Canada, children with
greater cooking self-efficacy and who were involved in the dinner preparation
increased the odds of increased family dinner frequency (62). Cooking confidence
was seen to rise with tacit skills such as skills of judgement, timing and planning,
12
allowing people to decrease the effort associated with cooking and encouraging
them to cook more (22).
Short showed an inter-relationship between cooking, cooking skills and people’s
cooking practices and food choices (22). Food choices are influenced by many
different factors such as personal tastes, cultural beliefs, public and private sector
policies and financial pressures (63, 64). Socio-cultural influences have been
shown to influence people’s food choice and consumption. For example in US,
African Americans eating healthy has meant their conformance with the dominant
culture and relinquishment of their cultural heritage (65). In addition to socio-
cultural influences, personal eating and food preferences of other family members
may influence household food purchasing and preparation of food in the home
(61). This suggests that it may not be the cook who ultimately controls a family’s
eating habits (61).
In a qualitative study conducted by Spence and van Teijlingen (2005), mothers
reported that they put their children’s diet and well-being before their own, and
this was found to be a motivator towards learning new skills in order to improve
their children’s diet (66). This links in closely with the next section on
provisioning, which is another key societal influence on food preparation
attitudes.
Domestic work has been conceptualised using the term "provisioning".
Provisioning refers to "the work of securing resources and providing the
necessities of life to those for whom one has a relationship of responsibility" (67,
p.383). This work is never-ending given that activities must regularly be
performed to ensure a family’s health and well-being. Food provisioning is one
sub-category of provisioning and refers to activities associated with securing and
providing food resources including the choices and actions involved in feeding
13
oneself and others (68). Women are typically the key caregivers and the
responsible food providers within the home (22, 28, 37). Women are central to
food choices and preparation (38). Whilst this gender difference may change over
time, as more men contribute to domestic food preparation, for the most part there
is still a consistent division of labour in terms of household food preparation. The
current literature shows that women are more likely to carry this responsibility
compared to men (37, 38, 67, 69).
a “
” for those “ ” was coined by
. The notion of a “ ” conveys complex inter
instead of their preferred idea of a “ ”
to prepare healthy meals due to work and children’s extracurricular activities
14
There is limited evidence in the academic literature on the role of celebrity chefs
and their influence on cooking and behaviour change. In Australia, there is
currently an abundance of cooking demonstration and cooking reality
programmes which often air during prime time television viewing. These are, for
example, Master Chef, My Kitchen Rules, Save with Jamie just to name a few.
Caraher (2000) states that celebrity cooks and television (TV) cooking
programmes are not likely to have a major impact on cooking behaviour, but may
influence and broaden cultural trends (70).
years old in an effort to promote children’s appreciation of healthy foods
’
change, Jamie Oliver’s television Jamie’s School Dinners
. Oliver’s second social cause –
–
15
argues that Jamie Oliver as an “ ”
what is acceptable “
”, foster individual self
2.3 The impacts of community-based cooking skills
interventions
With the decline in cooking skills, the response by public health practitioners has
been to try to improve the population’s ability to prepare food from “scratch”. In
recent years, there has been a proliferation of community-based cooking skills
interventions which aim to improve people’s cooking and change attitudes and
behaviours towards healthy eating patterns. The use of a hands-on practical
learning approach through cooking classes has been shown to be an excellent
starting point towards behaviour change, particularly in the short term (29, 77).
Incorporating cooking activities within nutrition education interventions can
increase programme participation and show gains in terms of food-related
preferences, attitudes, awareness, knowledge, self-efficacy and consumption
behaviours in both adults (3) and children (78).
16
2.3.1 Methods
A systematic review of community-based cooking skills interventions targeted at
adults was conducted with the aim of identifying positive impacts in terms of
cooking skills, behaviours and dietary outcomes comparative to a control group.
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses
(PRISMA) (79) and Cochrane guidelines for the systematic review of health
promotion and public health interventions (80) were used as guides for this
review.
17
For an intervention to be included in this review, it needed to meet specific
inclusion criteria. The intervention needed to target adults over the age of 18
years. Secondly, it had to entail a hands-on practical cooking element
incorporated as the primary activity of the intervention or as a key and sizable
component within a broader intervention. Interventions without a practical hands-
on element, for example, nutrition education only or cooking demonstrations
without a practical element, were excluded. Multifaceted programmes
incorporating more than just a cooking intervention were also excluded. For
example, interventions including physical activity as a component were excluded
because of the difficulty in determining the extent to which the outcomes were
attributable to the cooking component. Thirdly, the intervention had to be set in a
community-based setting, such as community health, hospitals, schools
communities and faith-based settings. This community-based criterion excluded
interventions that provided for one-on-one programmes or commercially-driven
programmes, such as an Italian cuisine cooking school.
18
The following databases were searched to identify literature published in peer-
reviewed journals: Academic Search Complete, CINAHL Complete, Education
Research Complete, Global Health, Health Source: Nursing/Academic Edition,
MEDLINE Complete, Psychology and Behavioural Sciences Collection,
PsycINFO, SocINDEX with Full Text, and PubMed.
The following terms were used to search the databases: (Cook* OR “food
skills”) AND (Health OR diet OR behaviour OR behavior OR skill OR
esteem OR “healthy eating” OR social) AND (
based OR “community level” OR “population level” OR kitchen).
All interventions with a control group and reporting effectiveness were assessed
using a quality assessment checklist (82, 83) called the Quality Assessment Tool
for Quantitative Studies (84). The strength of each study with a control group was
ranked as strong, moderate or weak. The ranking criteria are found in Appendix 1.
For all qualitative studies, appraisal was based on the Cochrane Collaboration’s
Guidelines for systematic reviews of health promotion and public health
interventions (80). The Cochrane Collaboration states that whilst there is no
formal consensus on the guidelines for reviewing qualitative studies, it is
important to include qualitative studies within reviews of public health
interventions. The tools used incorporate three quality assessment
recommendations detailed in the literature: (I) Quality in Qualitative evaluation: a
19
framework for assessing research evidence (85); (II) How to read a paper:
Papers that go beyond numbers (qualitative research) (86); and (III) The critical
appraisal skills programme (CASP) ten questions to help make sense of
qualitative research (87).
2.3.2 Findings from community-based cooking skills interventions reviewed
Figure 2 shows the number of studies screened and included for review at each
stage of the process. A total of 7,973 studies were located during the database
search, with an additional 17 sourced from the grey literature. After accounting
for duplication and performing a title scan, 358 abstracts were reviewed and 69
full-text studies were examined to determine eligibility. A total of 47 full-text
studies were included and reviewed, ten included a control group (Table 1) whilst
the other 37 did not include a control group (Tables 2 and 3).
The ten studies including a control group are documented in Table 1. Five
randomly allocated participants to the control or intervention groups (88-92),
whilst five studies (four interventions) used non-randomised methods of
allocation to the control and intervention groups (77, 93-96). Twenty-six studies
entailed pre- and post-assessment of the intervention without a control group
(Table 2) (57, 97-121). The remaining eleven studies provided post-assessment
measurements of the intervention effect (Table 3); most of these eleven used
qualitative methods (56, 66, 122-129), except for one (130). All 47 studies were
reviewed and findings are presented; however, particular attention is paid to those
studies with a control group, especially when summarising the impacts and
outcomes of the intervention.
20
Figure 2: Literature search
*large numbers of duplicates were found because multiple databases were searched.
Duplicates removed (n=3,889*)
Records identified through database searching
(n=7,973) Id
entif
icat
ion
Additional records identified through other sources (such as non-peer reviewed evaluation
reports) (n=17)
Total studies found (n=7,990)
Titles screened (n=4,101)
Records excluded, (n=3,743) Titles did not relate to nutrition
or cooking intervention
Scre
enin
g In
clud
ed
Elig
ibili
ty
Full-text articles assessed for eligibility
(n=69)
Total articles included (n=47)
Studies including a control group
(n=10)
Abstracts screened (n=358)
Records excluded, (n=289)Abstract did not match
eligibility criteria
Full-text articles excluded (n=22)
Reasons included; population group <18 years old, did not
included “hands-on” practical cooking element, included physical activity or other
intervention which would not help indicated cooking impact
Studies without a control group
(n=37)
Table 1: Cooking skills interventions – intervention and control groups
Author, Year Country
Study design Appraisal score
Population and sample
Intervention constructs and duration
Measurement tools Outcomes
Randomised controlled trials UK studies Curtis et al 2012, UK (88)
Randomised parallel-design, comparing three interventions. Pre and post- assessment of intervention, with 6 and 18month follow-up.
Strong* Some confounders in terms of SES, and no blinding.
Socially deprived families with poor diets. Equal sample among interventions A, B, C. Baseline n=589, 3months post-intervention n=444, 6-month follow-up n=369, 18-month follow-up n=198.
Family Food and Health Project Three interventions: A) 2h education only session B) 2h x 6weeks ‘cook and eat’ intervention C) personalised goal setting + ‘cook and eat’.
3-day food diary at baseline, 3months post intervention, 6months follow-up and 18months follow-up.
Three months post-intervention intervention C consumed less fat ((P=0.02), starch (P=0.04) and more total carbohydrates (P=0.001) compared to intervention A, carbohydrates were significantly sustained at 18months follow-up (P<0.05). Significant reduction in energy density of diets at 18months follow-up in both intervention B (P<0.001) and C (P<0.001) compared to A. Intervention B participants consumed more carbohydrates than intervention A (P=0.04).
US Studies Gold et al 2014, Dakota, US (90)
Randomised parallel-design, comparing three interventions. Pre and post- assessment (2weeks) of intervention.
Strong* While sample size small, only 15 participants in each group were required.
Refugees and immigrants (<3years in the US); part of the English language learning community. Food map (n=27), Cooking class (n=21), control (n=25).
Three interventions: A. 2h food safety map discussion. B. 2h x 2 cooking classes C. Control – no intervention. Both A and B received a food safety kit.
Self-report survey; paper based pre-intervention survey and telephone interview post-intervention measuring food safety knowledge and behaviours and behavioural intentions.
Food safety knowledge significantly increased in the food map (A) and cooking (B) interventions compared to the control (C) for knowledge of cooking meat (P<0.001) and refrigeration of food (P<0.001). All food safety behaviours significantly increased in food map and cooking class compared to control group (P<0.05) including washing hands, cross contamination, refrigeration of food and cooking of food. Significant differences found between food map (mean=3.7) and cooking intervention
22
Author, Year Country
Study design Appraisal score
Population and sample
Intervention constructs and duration
Measurement tools Outcomes
(mean=4.7) and the intended use of a food thermometer (P0.002).
Wenrich et al 2012, Pennsylvania, US (92)
Random allocation to control and intervention group. Pre and post- assessment of intervention, with 3month follow-up.
Moderate* Outcome assessors were not blinded. Limited valid and reliable data collection tools available.
Low income families living in rural area, n=25 couples in both intervention controls (couples included one food preparer and one partner).
8 x 2h weekly session. The intervention group were taught skills through taste testing and hands-on cooking; the control group received written materials only.
Self-reported survey using some validated items and measuring: readiness to eat vegetables, self-efficacy, and vegetable intake, and dietary history. Meal diaries were kept over the course of the intervention.
No significant differences were found between intervention and control groups. In both intervention and control groups, consumption of deep-orange vegetables and total carrots, broccoli and lettuce, plus total vegetables in a recipe significantly increased. Amongst partners, significant improvements in all vegetable categories in both intervention and control groups over time, but not between groups. Intervention food preparers with a high Meal Diary score (meaning greater improvement in target vegetables) had significant improvements in total target vegetables (P=0.009), total carrots, broccoli and lettuce (P=0.007) and total vegetables in recipes (P=0.008) compared to food prepares with a low Meal Diary score. Amongst control participants, the number of family meals per week significantly decreased by 1 meal/week compared to the intervention group which stayed the same.
Fulkerson et al 2010, US, Minnesota (89)
Random allocation to control and intervention group. Pre and post-
Strong* Low sample size, not powered to detect behavioural outcomes. No
Parent/child pairs of healthy children aged 8-10 years old and their parents who prepared the meals at home (n=44 families).
Healthy Home Offerings via the Mealtime Environment 5 x 90 min sessions, over 3 months. Focus on
Child anthropometry, 24 hour dietary recall of children, Self-reported validated and reliable psychosocial surveys measuring family dinner frequency
While not significant, an increase in parental self-efficacy (P=0.55) compared to control. No significant change in dietary outcomes for children (P=0.08). Intervention children reported significantly higher food skills development (P<0.01) and frequency of helping to make the family dinner (P<0.001).
23
Author, Year Country
Study design Appraisal score
Population and sample
Intervention constructs and duration
Measurement tools Outcomes
assessment of the intervention, with 6 month follow-up.
mention of 6-month follow-up results, except in abstract.
improving meals prepared at home for children.
and source of food, parent cooking self-efficacy, child food preparation skills.
Levy and Auld 2004, Colorado, US (91)
Random allocation to control and intervention group. Pre and post- assessment of intervention with a 3month follow-up.
Strong* Self-selected students. Tested questionnaire for face validity and reliability using test-retest methods
College students (n=33 intervention and n=32 control).
2h x 4 cooking sessions, plus a 45 minute supermarket tour. Control group attended a 1h cooking demonstration.
Valid and reliable self-reported questionnaires; eating habits survey measuring baseline eating history and experience, cooking habits survey measuring cooking attitudes, behaviour and knowledge and post-intervention 72 hour food preparation recall.
Significant increase in attitudes around liking to cook in intervention group (0.4 mean gain) compared to control (0.1 mean gain) (P<0.01). Significant increased confidence to use different cooking techniques in intervention (0.7 mean gain) compared to control group (0.3) (P<0.01). While not significant, results showed the intervention group was more likely to prepare dinner than eat out and ate out less frequently than control group.
Non- randomised control trials UK studies McKellar et al 2007, Scotland, UK (95)
Pre and post (3months) assessment of the intervention and 6-month follow-up.
Moderate* Confounders between intervention and control groups (more deprived in the intervention group).
Females aged 30-70 years old with rheumatoid arthritis living in socially deprived areas. (Sample at 6-month follow-up n=130, intervention n=75 and control n=55)
2h session for 6 weeks, cooking skills classes emphasised the use of a Mediterranean-type diet.
Dietary assessment measured with food frequency questionnaires; cardiovascular risk measured using anthropometry. Clinical and laboratory assessment (such as blood test, blood pressure and joint assessments) and clinical
Significant increase in weekly portions of: fruit, vegetable and legume consumption (P=0.016) and decreased consumption of monounsaturated and saturated fats (P=0.022) post intervention compared to control. Cardiovascular risk factors reduction in intervention group with a significant drop in systolic blood pressure (P=0.016) compared to no change in the control group. No change in other measures over time and between groups. Significant reduction in clinical global assessment (P=0.002) and pain scores
24
Author, Year Country
Study design Appraisal score
Population and sample
Intervention constructs and duration
Measurement tools Outcomes
assessment measured through global pain measurements and laboratory assessment. Measured baseline, 3 month and 6-month.
(P=0.049) at 6 months compared to control group.
Wrieden et al 2007, Scotland, UK (77, 96)
Pre and post- assessment of intervention and6-month s follow-up, with a wait-list control.
Moderate* High dropout rate at follow-up (44%). Low sample size, not sufficiently powered to detect behavioural outcomes.
Participants were from areas of social deprivation (T1 n= 113, T3 n=63)
Cook Well programme 2h session for 10 weeks (7 sessions were practical)
Self-reported general, cooking skills and food frequency questionnaire and 7-day food and shopping diaries. Measured at baseline, post intervention and follow-up. Qualitative interviews with small sample at T3.
No change in food frequency at baseline and six month follow-up. Increase in fruit and vegetable consumption, however fruit consumption was the only food to show a significant change of one portion per week (P>0.05) between baseline and post intervention. Increased food preparation from cooking with basic ingredients by the intervention group (68% at baseline to 90% at follow-up) compared to no change in control group. Significant increase in confidence to follow a recipe (P>0.05) between baseline and follow-up, compared to control group. Small significant difference in weight gain between intervention and control group between baseline and 6-month follow-up (P=0.049). The intervention group observed a slight decrease in weight (mean -0.6kg) from T1-T3, which differed to a slight increase in the control group (mean 1.8kg).
Kennedy et al 1998, UK (94)
Pre and post- assessment of the intervention
Moderate§ The researchers’ perspective
Low income mothers from socially deprived urban neighbourhood (n=23
Friends with Food 2h sessions for 10 weeks.
12 x programme development baseline focus groups. Session observations.
Intervention participants reported lifestyle barriers such as financial constraints, influences from their partner and family and nutrition education as barriers impacting on
25
Author, Year Country
Study design Appraisal score
Population and sample
Intervention constructs and duration
Measurement tools Outcomes
with 3month follow-up; control was assessed post- intervention (no baseline data for control group).
and ethical issues or approval was not reported.
intervention, n=13 matched controls)
Qualitative 1h semi-structured interviews to measure and quantitatively score nutrition knowledge and allow for emerging themes (post-intervention and 3 month follow-up between intervention and matched control group).
their cooking. Participants also indicated successful cooking was more likely with family support. Intervention participants significantly increased nutrition knowledge in 2 out of the 4 intervention groups (P<0.05) compared to no change in control group.
US studies Birkhead et al 2014, Louisiana, US (93)
Interrupted time series. Post- assessment of intervention with retrospective control.
Weak* Limited details of the data collection and analysis methods used.
First and second year medical students attending cooking learning module, n=125
Medial students received 2h training then delivered 6 cooking classes to underserved urban communities in New Orleans.
Cross-sectional annual school wide self-report surveys measuring competencies, attitudes, health habits and participation in cooking learning module. Retrospective annual reports, prior to leaning module implementation, were used as comparison, n=352
Compared to the prior year when the curriculum was not available, students who completed the cooking learning module were more likely to perceive proficiency in nutrition knowledge for weight loss (P=0.021), antioxidants in health (P=0.012), aerobic exercise (P=0.001) and hydration (P0.008).
* Quality Assessment Tool for Quantitative Studies (84) § Qualitative critical appraisal skills programme (CASP) 10 questions were used to appraise qualitative research (87) (44). BMI = Body Mass Index
26
Table 2: Cooking skills interventions – pre-post design, no control group
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Australian studies Lee et al 2010, Victoria, Australia (114)
Pre and post- assessment of intervention.
Community Kitchens in Frankston on the Mornington Peninsula. Participants (n=63), facilitators (n=17) and project partners (n=10).
Duration unknown. Community Kitchen where participants are led by a facilitator to plan, cook and share a nutrition meal.
Written surveys (to determine reach of programme), 11 x focus groups before and after cooking session and telephone interviews with facilitators.
Qualitative findings reported perceived benefits from participants after attending including the development of cooking, food and kitchen skills, new recipe modification, ability to make quick and easy meals and increased confidence to try new recipes. Social support was also common with participants reporting a positive experience to establish friendships, socialised and enjoy with others.
Porter et al 2000, Queensland, Australia (117)
Pre and post- assessment of intervention.
Participants with chronic mental illness returning to the community after hospitalization (n=3).
16 x 1h individual life-skills programme sessions, 2 times per week for 8 weeks.
Qualitative case study approach and The Functional Needs Assessment-Programme for Chronic Psychiatric Patients (FNA) to assess food skills.
Improvements in FNA levels for all three participants. This includes improvements in skills relating to dining, kitchen, food preparation, money management, shopping and purchasing and social etiquette.
Foley and Pollard 1998, Western Australia, Australia (57)
Pre and post- assessment of intervention, 4year follow-up.
Food Cent$ targets low income earners. Evaluation measured community advisors training to be Food advisors (n=150) and participants attending budgeting and cooking session (n=373). N=83 paired pre-post and
Food Cent$ 1.5h x 4 classes (food advisor training, budgeting session, and cooking and supermarket tour).
Pre-post diet check measuring dietary change, post-session questionnaire measuring demographics, intention to change, 6-week follow-up and 4year follow-up questionnaire measuring confidence to run session, dietary and spending change. Focus group with
Significant dietary change was reported after attending the budgeting session with a significant increase in persons who spread margarine thinly (P=0.004), which rarely ate sweets (P=0.013) or bought cakes (P=0.001).
27
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
follow-up attending cooking and budgeting session.
advisors who did not conduct a session.
UK studies Garcia et al 2013, Scotland, UK (110)
Pre and post- assessment of intervention, 12month follow-up
Parents of young children. (baseline n=102 and follow-up n=44)
2h session for 4-8 weeks Self-reported questionnaires measuring food consumption patterns, confidence, knowledge, enjoyment and any perceived benefits to participants’ families (which were only measured at follow-up)
12 months after the intervention there were significant increases in all confidence measures, a significant reduction in the number of times ready-made meals were consumed from 2-4 times/week to 1 time/week (P=0.001). Significant sustained increases in fruit (P<0.05) and vegetable (P=0.001) consumption measured in times/week at follow-up. Majority of participants at follow-up agreed there were benefits to their family and their families’ nutrition.
Caraher and Lloyd 2013, South Liverpool UK (99)
Pre and post- assessment of intervention, 3 month follow-up
Socially disadvantaged families (n=11 families)
Can Cook Families Programme 2 day training course, where chefs lead the programme.
Pre/post assessment using: semi-structured interviews, cooking skills questionnaire, 24hour food recall, food storage photographs and 7day shopping records.
Significant mean increase in fruit by 0.8 serves/day (P=0.01) and vegetable by 1.5 serves/day (P=0.009) after the intervention. Increased spending on fruit and vegetables and significant increase in skills and confidence of using basic ingredients (P=0.001), following a recipe (P=0.01), tasting new foods (P=0.025) and preparing new recipes (P=0.038).
Lawe 2013 Nottingham, UK (113)
Pre-post assessment of intervention.
College students at Nottingham Trend University. (n=unknown).
Student cookery clubs including 2h x 4 weekly sessions.
Mixed methods evaluation: Self-report pre/post questionnaire about eating habits and knowledge of
Results did not report change over time nor did they indicate analysis methods or sample size. Post-intervention participants reported “good-excellent” levels of confidence in
28
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
healthy eating. Qualitative class observation in first and last classes, with informal participant discussion. Focus group with first year university students
trying different recipes and improved likelihood that participants would eat healthy after attending the course.
Children's Food Trust 2011, Sheffield, UK (102)
Pre and post- assessment of intervention, 6-month follow-up of some participants.
Areas of high deprivation targeting families, children and adults. The number of adults and children unclear (n=1,754 in total).
Let’s Get Cooking 8-6 week activity cycle. National English cooking skills programme targeting children and their parents implemented across the UK.
Self-report pre-post questionnaire measuring change in eating habits.
Methods indicated that adults completed the questionnaire however all results presented appear to refer to children sampled. 92% of participants indicated they were using skills learnt at home. On average, participants had also shared that knowledge and skills with at least one other person. 58% of participants reported eating healthier after the programme, with results being sustained 6months after the intervention.
Davis et al 2009, Southampton UK (107)
Pre and post- assessment of the intervention, 12month follow-up.
Mostly women (81%) of South Asian decent (n=46).
Breaking the Cycle Project Duration unknown, however during implementation 330 people participated in 10 taster sessions and 28 cookery club sessions providing practical skills to improve diets.
Self-reported questionnaires, reflection sheets, focus groups and telephone interviews.
Positive dietary behaviours were shown: 42% of participants were using low fat milk at baseline which increase to 86%, there was a decrease in consumption of sugary foods from 23% to 5% and an increase in the consumption of fruit and vegetable from 85% to 100%, however significant of results were not reported. Other outcome measures were not reported such as attitude to healthy eating, changes in eating patterns, changes in shopping and cooking practices, barriers to changes and maintenance of changes.
29
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Participants indicated that their motivation to become food trainers was led by a need in the community and a family history of nutrition-related ill health.
US Studies Kokkonen-May et al 2014, Minnesota, US (112)
Pre and post- assessment of intervention.
Supplemental Nutrition Assistance Programme (SNAP) eligible participants in three regions of Minnesota (n=45).
Simple Good Cooking 2h x 6 sessions (with a choice of 20 hands-on activity lessons with options ranging from general, fruits, vegetables, protein, grains, and dairy.
Self-report survey measuring self-efficacy, confidence, food behaviours and food and health knowledge.
After the intervention there were significant increases in confidence to plan a health meal (P<0.000), prepare a healthy meal (P<0.000) and make food from scratch (P<0.03). After the intervention there were significant increases in fruit and vegetable consumption (P<0.001). Fruit increase from 0.8cups/day to 1.3cups/day. Vegetables increase from 1.1cups most days to 1.4cups most days. Behaviour changes after the intervention included shopping from a list (P<0.001), making a dish with beans (P<0.001) and doing physical activity at least 30minutes/day (P<0.001).
Bielamowicz et al 2013, US (97)
Pre, mid and post-assessment of intervention.
Participants attending Do Well, Be Well with Diabetes (DWBW) Programme in 86 counties. DWBW was a 9 week diabetes education programme Pre-test n= 2,853, Post-test n=1026
Cooking Well with Diabetes 4 lessons. The Cooking Well with Diabetes was an extension of the DWBW intervention.
Mixed methods. Quantitative questionnaire measured at baseline, after third class, and post-intervention during last class. Measuring self-report blood glucose, haemoglobin A1Ctest, cooking behaviours, knowledge of healthy cooking practices for diabetes.
Knowledge scores increased significantly (P<0.05) between baseline and after the intervention. Behaviours measuring healthy cooking for diabetes practices increased in 8 out of 10 measures (P<0.01) after the intervention. Qualitative findings indicated that participants were able to put knowledge into practice after attending the intervention.
30
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Qualitative open-ended questions to provide insights into the programme.
Condrasky et al 2013, South Carolina, US (106)
Pre-post assessment of intervention, 3month follow-up with monthly phone calls every three months for 15months.
Cooks within African America churches (n=100).
Cooking with a Chef One 8 hour workshop, plus monthly mailed information on cooking tips and recipes for 15 months post intervention and a half day booster session around Easter and Christmas (for a five year period).
Mixed methods. Pre-intervention Cooking with a Chef validated questionnaire measuring cooking attitudes, self-efficacy, behaviours, cooking techniques and fruit and vegetable use. Pre and post cooking skills and cooking confidence measure. Feedback over the phone via ‘technical assistance phone calls to participants three month post-intervention and every third month thereafter for 15months.
Significant increases were found post intervention in cooking skills (P<0.0001) and confidence (P<0.0001). Qualitative findings indicated that participants were using less salt and incorporating more fruit and vegetables into their menu planning.
Condrasky, Hegler and Sharp 2011, South Carolina, US (105)
Pre and post- assessment of the intervention.
School food services workers (n=111).
Cooking with a Chef 1 x 8 hour workshop Hands-on nutrition education programme adapted from the original Cooking with a Chef aimed school nutrition operators.
Cooking with a Chef validated questionnaire measuring cooking attitude, produce consumption self-efficacy, cooking self-efficacy, self-efficacy for basic cooking and self-efficacy for using fruit and vegetables.
Significant change in attitudes (P=0.03) and cooking skills (P<0.0005). Significant correlation between produce consumption self-efficacy with cooking technique self-efficacy and self-efficacy for using fruit, vegetables and seasonings.
Condrasky et al Pre and post- Parents and caregivers of Cooking with a Chef Mixed methods No change in fruit and vegetable
31
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
2006 South Carolina, US (103, 104)
assessment of the intervention.
preschool aged children, n=29 (intervention n=15, control n=14).
2h session x 6 lessons. The intervention included interactive sessions facilitated by a chef and nutrition educator who taught culinary skills and hands-on learning.
24h Food recall, food behavioural checklist and pre and post questionnaire measuring mealtime practices, use of flavours in cooking at home, fruit and vegetable intake and parental support. Post-intervention focus groups were also conducted.
consumption, however small increases were reported in children’s fruit intake (P<0.096). Parents and caregivers increased confidence, skills and awareness of the use of flavours and confidence to prepare healthy meals at home.
Rustad and Smith 2013, Minneapolis, US (118)
Pre and post- assessment of the intervention.
Ethnically diverse, low income women aged 23-45 years of age (n=118).
6 x 75-90 min session. 3 sessions included hands-on interactive nutrition education via cooking.
Food frequency questionnaire and self-reported questionnaire measuring nutrition and health knowledge and behaviours.
Significant increase in 7 out of 11 knowledge questions (P<0.01) in statements around sodium in food, diet and cancer, nutrients in foods, nutrition labels, use of herbs and spices and physical activity. Significant increase in nine out of 11 behavioural questions (P<0.01) with increased vegetable intake, use of herbs and spices in cooking, use of nutrition labels, doing physical activity, preparing healthy meals for their families. There was also a significant correlation between knowledge and behaviour.
Francis S 2012, Wyoming, US (109)
Pre and post- assessment of the intervention, 16week follow-up.
Adults Wyoming communities (86% females). n=19 (n=14 at follow-up).
Eat to Your Heart’s Content 2h x 4 sessions. Each session consisted of cooking practicum and nutrition education.
Self-report questionnaire measuring heart-healthy lifestyle practices.
Significant improvements were found for all measures between baseline/post intervention and baseline/follow-up (P<0.02) including improvements in knowledge, cardiovascular disease risk factors, dietary improvements in fibre, fruit and vegetables, health food choices, cooking methods and behaviours.
32
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Chavent et al 2012, Connecticut, US (101)
Pre and post- assessment of the intervention.
Child care professionals (n=15).
Share Our Strengths Cooking Matters Two weekends between 9am and 2pm. Modules included: creating positive time, setting a healthy example, recipe ingredients, food safety, nurturing a healthy eating, food labels, healthy food, portion size and food allergies.
Pre and post surveys. Significant improvements were found after the programme in confidence to add healthy food to a recipe (P=0.0003), compare food prices (P=0.005), and plan meals within a budget (P=0.04).
Shankar et al 2007, Washington DC, US (119)
Pre and post-assessment of the intervention, 4month follow-up.
African American women living in 11 public housing communities (n=212).
7 x 90 minute sessions in total (2 sessions a week for 3 weeks, followed by 1 “booster” session 6weeks post intervention).
Self-reported change in dietary knowledge, attitudes and behaviours.
There were no significant changes in fruit and vegetables consumption after the intervention and at follow-up. Participants who attended 5 or more sessions experienced a significant reduction in % calories from fat between baseline and post intervention (3.08% less P<0.001) and between baseline and follow-up (2.97% less P<0.001). For participants who attended 1-4 session2, there was a significant 3.86% reduction in calories from fat (P≤0.05), however this was not sustained.
Swindle et al., 2007, Denver, US (120)
Pre and retrospective pre-post assessment of the intervention, 3 and 6months
Limited income families (>90% women) in the Denver metropolitan area. Baseline, n=19, retrospective pre-post n=53 (n=12 completed
Operation Frontline’s Eating Right 6 week community targeted nutrition education programme taught in English and
Two forms of ‘pre-test’ were measured, one at baseline before the intervention and a retrospective pre-test. Self-reported retrospective
Retrospective pre/post test indicated that all shopping, general and eating behaviours significantly changed (P<0.001) post- intervention. Behaviours were not sustained between post- test and 3 (n=27) or 6 (n=12) month follow-up.
33
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
follow-up.
initial pre-test), n=27 completed 3month follow-up, n=14 completed 6month follow-up.
Spanish. Curriculum includes meal planning, cooking, shopping smartly and food safety.
pre-post standard Operation Frontline survey was used plus the pre-intervention survey measuring eating, shopping and general behaviours.
Comparing the traditional pre-test with the retrospective pre-post-test (n=12) results showed that the only significant difference between was in shopping behaviours which had a higher mean (2.5, SEM 0.2) compared to the retrospective pre-test (1.9, SEM 0.1), however both were significantly different post intervention (2.9, SEM 0.1).
Newman et al 2005, US (116)
Baseline and 12months follow-up.
Women who had been previously treated for breast cancer (n=739).
12 x monthly cooking classes, newsletters and 15-23 telephone counselling session (median number of classes attended was 5). The primary intervention was whether telephone counselling improved dietary targets. This study reported on whether the use of cooking classes (and newsletters) over a 12 month period improved adherence to study dietary targets.
3-day 24 hour dietary recall via the phone.
Mean dietary intake were reported: vegetable intake increased from 4 to 8.1 (including juice) serves per day (P<0.001), fruit intake increase from 3.6 to 4.2 (P<0.001), fibre increased from 22.1 to 30.3 grams per day and fat decreased from 27.7 to 21.7 % energy. Dietary adherence was aided by cooking class attendance (P for trend <0.01).
Woodson et al 2005, Nevada, US (121)
Pre and post- assessment of the intervention.
African American faith communities through church facilities (n=326).
Food for Health and Soul 6 x 60 min classes aiming to retain cultural food and reduce dietary fat, sodium, sugar and
Pre and post self-report Eating Style Questionnaire measuring behaviours plus stages of change and perceived benefits from the programme.
Significant improvements in fat (P<0.001) and sodium intake (P<0.001) after the intervention. There was advancement throughout the stages of change indicators; however there was no significant difference post-
34
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
increase fibre. intervention.
Brown and Hermann 2005, Oklahoma, US (98)
Pre and post- assessment of the intervention.
Members of 29 Oklahoma Counties (n=229 youths with mean age of 12 and n=373 adults with a mean age of 57).
Duration of intervention unclear, trainers provided an average of 8 classes over 2 months.
Self-report questionnaire reflecting class content around fruit and vegetable consumption and food safety behaviours.
Fruit significantly increased in adults from 1.5 to 2.1 serves per day (P<0.0001), Vegetable significantly increased from 2.1 to 2.7 serves per day (P=0.0001). 11% significant increase in washing hands before preparing food in adults (P<0.0001), 8% of adults significantly increased washing produce before using them (P<0.0008).
Chapman-Novakofski et al 2005, West Virginia, US (100)
Pre and post- assessment of the intervention.
Adults with diabetes or their care givers. (N=239).
Dining with Diabetes Programme 2h session x 3 classes.
Self-report measures of knowledge based on Stages of Change Theory and Social Cognitive Theory.
Significant change in healthy eating practices (P<0.05), confidence to change their meals (P<0.05) and knowledge around diabetes and nutrition (P<0.05) post-intervention.
Canadian Studies
Keller et al 2004, Ontario, Canada, (111)
Pre and post- assessment of the intervention during the evaluation year.
Senior men aged 65 years and over and retired n=19 pre (1 year after intervention began) and post measurements during evaluation year, n=10 key informant interviews.
Men Can Cook 2h classes run once a month over 8 months. Classes involved preparing and consuming a meal.
Mixed methods. Pre and post self-reported questionnaires assessing change in cooking habits, nutritional knowledge and eating habits. Semi-structured interviews mid-way through intervention relating to men’s experience and outcomes of the cooking group.
No statistical significant change over time, however improvements were shown in pleasure for cooking and cooking confidence, improved appreciation and attitude towards healthy eating and new foods. Qualitative results found programme had a positive effect on sense of self-worth and connection with others. The men reported a positive experience of the intervention.
35
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Dewolfe and Greaves 2003, Canada (108)
Pre and post- assessment of the intervention, 3month follow-up.
Low income (majority female, some low education) Baseline n=42, post intervention n= 20, 3 month follow-up n=17.
The Basic Shelf Experience. 6 week programme where participants would plan, prepare and eat the meal.
Mixed methods. Self-reported questionnaire measuring food resource utilisation, food insecurity and self-esteem. Focus groups measured at follow-up to understand participants’ perspective and programme experience.
No significant change in food security or change in self-esteem. Qualitative results found the programme could not change the amount of money that people had but participants felt they were doing something positive to stretch their food dollars further. Participants received support from other group members. Participants felt they built confidence in themselves and in others, around shopping, planning, and preparing meals and using a cookbook.
Marquis et al 2001, British Columbia, Canada (115)
Pre and post- assessment of the intervention.
Participants using Tillicum Haus Friendship Centre’s Community Kitchen in Nanimo, British Columbia. N= 14.
Q’wlut tu cicut (My Parents are cooking). Total intervention of 30 weeks. 20 weeks education and practice training to run a cooking group, followed by 10weeks of running their own cooking clubs, recruiting friends and families and cooked together on average 3.3 time (range 1-12 times).
Two qualitative focus groups. The first focus group determined participants’ goals and expectations, subsequent focus groups determined if goals were met.
Perceived benefits of attending the intervention included socialisation, budgeting skills, food shopping knowledge, obtaining a food certificate (which led to employment for some), making new meals, enjoyment of cooking and the programme and sharing knowledge. There were also reports on improvements on food security.
36
Table 3: Cooking skills interventions – measured post-intervention only, no control group
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Australian Studies Abbot et al Sydney, Australia (56, 122)
Post-assessment of intervention.
Aboriginal community in western Sydney, Australia (n=23).
4h classes over 18 sessions. Participants ranged from attending 2-9 classes. The intervention focused on home cooking on a budget, and health eating for weight and diabetes management.
Semi-structured in-depth interviews exploring participant experience, benefits after the course in healthy eating behaviours and cooking skills.
Participants reported strong motivating factors for change that included a recent health diagnosis and desire to influence family dietary habits. Participants reported improvements in cooking skills and knowledge as well as positive health changes such as weight loss, increased well-being and diabetes self-management.
Foley et al 2011, Queensland, Australia (126)
Post-assessment of intervention.
Intervention targeted Aboriginal and Torres Strait Islander (young mothers group and men’s groups) (n = not reported).
3x 3h workshops or 4x 2h workshops.
Focus group conducted post-intervention.
Intervention was successful from perspective of nutrition professionals, students and community leaders; participants were motivated and increased cooking healthy meals at home and were provided with an opportunity to cook and taste new foods. Additional benefits to the facilitators to engage with participants.
UK studies Spence et al 2005, Scotland, UK (66)
Post-assessment of intervention.
Lower class areas targeting low income families (n=6).
Now You’re Cooking 8 week programme teaching basic cookery skills and inexpensive health and tasty dishes
Semi-structured interviews conducted six month post-intervention
Key motivation to attending cooking classes was to improve their children's diet. Participants learnt cooking skills, budgeting skills, boosted self-esteem leading to employment opportunities. Participants reported improvements in their children’s diet, yet not their own, while participants without children were more likely to change their own diet.
37
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Snowdon, W 1999, Bedfordshire, UK (128)
Post-assessment of intervention.
Local Southern Asian community (n=60).
3 x 2h sessions. Post-intervention evaluation form n=60. Qualitative end of club evaluation session discussion n=20 Qualitative follow-up evaluation session discussion (face-to-face n=5 and phone interviews n=2).
In the short term, participants commented that they enjoyed the course and felt they had learnt how to make cooking healthier. Many also commented that a motivating factor for change was a family member’s illness (for example, diabetes). In the long term, participants maintained nutrition knowledge and perceived changes in their diet by improving the nutritional quality of their cooking. Some also passed on nutrition messages to others.
US Studies Archuleta et al 2012, New Mexico, US (130)
Post-assessment of intervention during last class and one month follow-up.
People with diabetes and their family members in New Mexico baseline n=614
Kitchen Creations 3h x 4 session consisting of diabetes nutrition education followed by meal preparation and consumption. Each participant received materials and cookbooks.
Pre-survey measured demographics only. Post and follow-up surveys measured knowledge and behaviours.
Improvements were shown in knowledge around food and cooking for diabetes (significance testing was not reported). Overall knowledge was high post intervention, with significantly higher knowledge score between Hispanics compared whites (P=0.0004) and females compared to males (P=0.0199). On average, participants’ intention to change behaviours improved.
38
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Canadian Studies Engler-Stringer and Berenbaum 2006, Canada (123-125)
Post-assessment of intervention.
Collective Kitchen attendees including: low income single mothers, men, seniors, and disabled, homeless or under-housed people. n=9 informant interviews for example staff members who support community kitchens, n=10 group leaders, n=20 participants
Observation of ongoing collective kitchen period for 6 weeks-6 months.
Participant observation, in-depth individual interviews about their experience with the community kitchens.
Participants described; learning about the nutritional value of food, new methods of cooking, changing grocery shopping habits, label reading and bulk buying. Programme facilitated formal and informal peer-assisted learning around cooking and nutrition, with opportunities for participant and leaders to share information and skills. Qualitative findings included decreased social isolation, increased social support, participation and sharing resources in their community as well as knowledge of where to find help. Participants perceived an increase in short-term food security, decreased psychological distress associated with food insecurity, food produced were favourable in terms of quality and acceptability in comparison to charitable handouts and gained the ability to cook greater quantities of food.
Tarasuk and Reynolds 1999, Toronto, Canada (129)
Post-assessment of intervention.
Low-income household participants and community kitchen facilitators (participants interviewed n=14, facilitators interviewed n=6, no control group)
Programme duration not stated. Community Kitchens were ongoing facilities for the community to regularly come together to cook and prepare food for their families.
Observational methods and semi-structured interviews of kitchen users with 4 months to 5 years use of kitchen.
Community Kitchens aided participants with coping skills and provided social support. Could not resolve food insecurity because the programme could not change household economic circumstances.
39
Author, Year Country
Study design Population and sample Intervention constructs and duration
Measurement tools Outcomes
Moldofsky Z 2000, Toronto, Canada (127)
Post-assessment of the intervention.
Large low-income immigrant population from a Food Bank (n=10).
Meals Made Easy Ten series of cooking classes. 2h classes once per week in blocks of 3-6 classes. Interactive cooking programme to help Food Bank clients.
Self-assessments at the end of each class and focus group.
The programme challenged participants’ negative thoughts around food and cooking to improve cooking attitudes and baking skills. The programme encouraged self-esteem, social acceptance and a reduction in social isolation. Participants perceived improvements in their cooking self-efficacy, cooking more economically and more nutritiously and incorporating more variety into their diet.
The studies included were from Australia (n=6), UK (n=12), US (n=21) and
Canada (n=8). Whilst the study settings varied they were all conducted in
community-based settings. These included: community centres (n=29),
community kitchens or food banks (n=7), hospitals (n=2), schools or other
education settings (n=7) such as language centres, and religious settings such as
churches (n=2).
Overall, there were few studies with a large sample size. The majority of studies
had a small sample size with only 14 out of 47 studies reporting a sample size of
over n=100 participants. The Family Food and Health Project was the only
controlled intervention to claim to have achieved the sample size required to
41
detect a change in percentage food energy from changes to dietary fat and
carbohydrate intake; it had 55 participants in each group (88). Compared to other
interventions, McKellar et al (2007) also included a larger sample size with 130
participants (n=75 intervention and n=55 control); however, sample size
calculations were not discussed (95). The Cook Well programme addressed issues
of maintenance of sample size over the course of the evaluation and reported a
44% dropout rate at six-month follow-up. Difficulties were reported in retaining
participants in socially deprived population groups, and in potential research
fatigue given the multiple measurement tools used (77).
All interventions included in this review had an element of hands-on cooking.
Cooking skills were used as a mechanism to improve participants’ diets, target
cooking confidence, knowledge and attitudes to increase food preparation at
home, and reduce risk factors associated with lifestyle diseases. In addition to
teaching cooking skills, many interventions also included elements of nutrition
education, food knowledge and goal setting (88). Some programmes worked with
chefs to teach culinary skills to the population. For example, the Cooking with a
Chef programme in the US entailed an interactive chef facilitated hands-on
learning intervention with a number of different community groups who then
influenced other people (103-106). The Can Cook Families Programme in the UK
also worked with chefs to pass on cooking skills to others (99).
42
Intervention duration ranged from one day workshops (105, 106) to interventions
running for over 12 months (116). Most interventions were in four and 10 or 12-
week blocks.
The studies variously used quantitative and qualitative methods in the collection
of data. The majority of intervention studies used self-report measures to
quantitatively measure changes in outcomes such as skills, knowledge,
confidence, attitudes and behaviours. Self-report measures included: food diaries
of shopping, preparation and consumption, food frequency questionnaires or
validated questionnaires. The Barton et al (2011) Short Cooking Skills
Intervention Questionnaire (131) was used in a number of studies (77, 99, 110).
Other quantitative measures included anthropometric measures (weight, height
and waist circumference) and clinical and laboratory assessments such as blood
tests, pain measures, glucose levels and blood pressure. Qualitative measures,
including semi-structured interviews, class observation and focus groups, were
used to gain an understanding of participant experiences, behaviours and attitudes.
43
Table 4: Interventions with a follow-up period
Follow-up period* Number of interventions (n=47)
3 to 4 months post-intervention Interventions with a control (91, 92, 94) 3 Interventions without a control (99, 106, 108, 109, 119, 120) 4 6 months post-intervention Interventions with a control (77, 89, 95, 96) 4 Interventions without a control (102, 120) 2 12 months post-intervention Interventions with a control 0 Interventions without a control (107, 110) 2 More than 12 months post-intervention Interventions with a control (88) 1 Interventions without a control (57, 106) 2 No follow-up period Interventions with a control (90, 93) 2 Interventions without a control (56, 66, 97, 98, 100, 101, 103-105, 111-118, 121-130)
27
*Follow-up period taken from the longest measured time point.
As part of this thesis, the nine studies (Table 1) that reported results and contained
a control group were critically appraised. Four were classified as “strong” studies
(88-91), four as “moderate” (77, 92, 94, 95) and one “weak” (93). The difference
between the ranking of strong, moderate and weak was the presence of bias. Bias
included: selection bias, study design, confounders, blinding, data collection
methods, withdrawals and dropouts (84). No studies reported being blinded or
double blinded, however this would be difficult in a community-based setting.
Sample size was low in most of the studies, and they often lacked sufficient power
to detect behavioural outcomes. In the majority of studies the data were based on
self-reported responses and therefore were open to bias.
The cooking skills interventions reported a variety of outcomes, as outlined in
Tables 1, 2 and 3. The outcomes information is summarised in Table 5. Cooking
44
skills interventions have been shown to make significant improvements in dietary
outcomes (77, 88, 95, 96), cooking behaviours (77, 90, 96), knowledge (90, 93,
94), skills (89), confidence (77, 91, 96), attitudes (91) and health outcomes (95,
96), as shown in Table 5. The most frequently measured outcomes were related to
diet and cooking behaviours. Multiple outcomes were reported in four of the 10
studies with a control group. Three studies reported changes in two different
outcomes (90, 91, 95) and Wrieden et al (77, 96) reported changes in four
different outcomes. Eleven of the 26 studies with a pre-post design and no control
group reported multiple outcomes. Two studies reported changes in four outcomes
(99, 109), four studies reported changes in three different outcomes (100, 105,
112, 118) and five studies reported changes in two different outcomes (97, 98,
101, 106, 110). Each outcome will be discussed separately. Not all found
significant improvements post-intervention.
Three out of the nine cooking skills interventions (with a control group) showed a
significant positive impact in terms of dietary outcomes (77, 88, 95). The Cook
Well Programme showed significant increases in fruit consumption post-
intervention (P>0.05), however improvements in fruit intake were not sustained
six months after the intervention (77, 96). Sustained dietary outcomes were shown
by McKellar et al (2007) who found significant changes in healthy eating
behaviours three months post-intervention, with improvements in fruit, vegetable
and legume consumption and a reduction in monounsaturated and saturated fat
(95). Eighteen months after the Family Food and Health Project, which compared
three interventions (education only, cooking classes and goal setting with cooking
classes), persons attending the cooking classes reported significant reductions in
energy-dense diets and increases in carbohydrates and starchy foods compared to
the education-only group (88).
45
Table 5: Cooking skills interventions reporting significant change over time by outcome measure1 Outcome measured Studies measuring
outcome (n= 37), (study citation)
Studies indicating significant change in outcomes over time (n=24) (study citation)
Interventions with a control Dietary outcomes 4
(77, 88, 92, 95, 96)b
3 (77, 88, 95, 96)b
Cooking behaviour 5 (77, 89-91, 93, 96)b
2 (77, 90, 96)b
Confidence 4 (77, 89, 91, 92, 96)b
2 (77, 91, 96)b
Knowledge 3 (90, 93, 94)
3 (90, 93, 94)
Skills 2 (77, 89, 96) b
1 (89)
Attitudes 2 (91, 93)
1 (91)
Health outcomes 3 (89, 95, 96)
2 (95, 96)
Interventions without a control Dietary outcomes 17
(57, 98, 99, 102-104, 107, 109-113, 116, 118-121)
10 (98, 99, 109, 110, 112, 116, 118, 119, 121, 126)
Cooking behaviour 19 (57, 97-101, 103-109, 111, 112, 118-121)
10 (57, 97-101, 109, 112, 118, 120)
Confidence 7 (99-101, 105, 106, 110, 112)
7 (99-101, 105, 106, 110, 112)
Knowledge 10 (97, 100, 101, 109-113, 118, 119)
5 (97, 100, 101, 109, 118)
Skills 4 (99, 105, 106, 117)
3 (99, 105, 106)
Attitudes 8 (105-107, 110, 111, 115, 119, 121)
1 (105)
Health outcomes 3 (97, 108, 109)
1 (109)
1Total number of studies included n=37a, some intervention effects measured in multiple studies, significant outcomes P<0.05. aStudies in Table 3 not included, because outcomes did not report change over time. bWrieden et al’s Cook Well intervention was reported twice via a report (96) and publication (77).
46
A total of 12 interventions (two with a control group and 10 without) reported
significant improvements in cooking and food-related behaviours. In the
interventions with a control group, the 10-week Cook Well programme influenced
participants’ food preparation behaviours six months after the programme (77,
96). Participants increased their frequency of cooking with basic ingredients at
home from 68% at baseline to 90% at follow-up, with no changes found in the
control group’s behaviours (77, 96). A food safety intervention targeting refugees
and immigrants in the US showed significant improvements in food safety
behaviour after attending a cooking class compared to the control group (90).
There were significant improvements in behaviours around washing hands, food
cross contamination, refrigeration and safe cooking of foods (90).
improvements in participants’ behaviours aro
47
Nine interventions reported significant improvements in participants’ confidence
and self-efficacy after attending a cooking skills intervention. Two interventions
had a control group (77, 91) and seven interventions used pre-post designs
without a control group (99-101, 105, 106, 110, 112). College students in the US
were randomly assigned to cooking classes or a cooking demonstration to
improve knowledge, attitudes and behaviours towards cooking (91). Participants’
confidence to use different cooking techniques significantly increased in those
attending the cooking classes compared to those who attended cooking
demonstrations (P<0.01). The Cook Well Programme measured cooking
confidence using five validated cooking confidence questions (77, 96). Results
showed significant improvements in confidence to follow a recipe (P>0.05)
between baseline and six-month follow-up in the intervention group only (77).
The same validated cooking confidence questions were used to by Garcia et al
(2013) (110) and the Can Cook Families Programme (2013) (99). The Can Cook
Families Programme reported significant increases in confidence in using basic
ingredients (P=0.001), following a recipe (P=0.01), tasting new foods (P=0.025)
and preparing new recipes (P=0.038) post-intervention (99). In addition, the
Cooking with a Chef programme in the US reported significant improvements in
participants’ cooking confidence after attending this hands-on culinary nutrition
intervention. The Cooking with a Chef model has been used with a number of
different population groups and significant increases in participants’ confidence
were shown when the programme targeted school (105) and church (106) food
services providers.
A total of eight interventions reported improvements in cooking and food
knowledge after attending a cooking skills intervention. Three interventions with
48
a control group reported significant improvements in participants’ cooking and
food knowledge around food safety (90) or improvements in nutrition awareness
(93, 94). Five interventions using pre-post designs and no control group also
showed significant improvements in cooking and food knowledge post-
intervention. Knowledge around cooking for diabetes prevention was targeted in
Cooking Well with Diabetes, a cooking extension from a diabetes education
programme implemented in 86 counties (97). This study reported increased
knowledge of healthy cooking practices for people with diabetes and nutrition
knowledge around recipe modifications in order to incorporate less salt, fat and
sugar, and add more fibre (97).
Cooking skills and techniques are relatively hard to measure because of the wide
differences in the definition and interpretation of the term “cooking” (22). Four
interventions out of six that attempted to measure cooking skills reported
significant improvements. A randomised controlled trial, Healthy Home Offerings
via the Mealtime Environment, measured child-parent pairs after completing the
interactive nutrition education and meal planning intervention together as a child
parent dyad. Children are reported on in this instance as the intervention targeted
both children and their parents. They found that children in the intervention group
reported significantly higher food skills development (P<0.01) and increased
frequency of helping to make the family dinner (P<0.001) compared to control
children, whilst parental mean self-efficacy for making healthful changes was
higher, although not statistically significant in the intervention group compared to
the control group(P=0.55) (89). Amongst interventions without a control group,
the Cooking with a Chef programme improved participants’ cooking skills over
time (105, 106). In addition, the Can Cook Families Programme increased
participant skills and confidence (99).
49
A number of studies attempted to change cooking and food attitudes but only two
out of 10 interventions reported significant changes in attitudes. Levy et al (2004)
reported significant increases in cooking enjoyment in the intervention group but
not in the control group (P<0.01) (91). Secondly, the Cooking with a Chef
programme reported improved attitudes in the school food services workers after
attending the programme. However, this intervention did not have a control group
with which to make a comparison (105).
Three studies reported significant changes in health-related outcomes. The Cook
Well evaluation reported small, yet significant differences in weight gain between
baseline and the six-month follow-up (P=0.049). The intervention group observed
a slight decrease in weight (mean -0.6 kg) from time 1 to time 3 compared to a
slight increase in the control group (mean 1.8 kg) (96). McKellar et al (2007)
reported improvements in the health status of participants with rheumatoid
arthritis after attending a cooking course that taught them the preparation of a
Mediterranean-type diet (95). Results showed significant reductions in blood
pressure and pain scores in the intervention group at six months follow-up (95).
Lastly, the Eat to Your Heart's Content intervention found significant
improvements in cardiovascular disease risk factors after attending the four-
session intervention which consisted of cooking and nutrition education (109).
Results could not be compared to a control group (109).
Qualitative findings offered some additional insights into outcomes that were not
reported within the quantitative studies. Findings related to food security, social
support and flow-on outcomes to family members after participating in an
intervention. Engler-Stringer and Berenbaum (2006) conducted an evaluation of
the Collective Kitchens programme (123-125) based on participant observations
50
and interviews with programme participants, facilitators and stakeholders. The
benefits reported included enhanced opportunities to develop cooking skills,
cooking and food knowledge, social support and decreased stress relating to food
security. Tarasuk and Reynolds (1999) evaluated community kitchens using
observation and concluded that they enabled social support but were not a means
to resolving food insecurity in participants (129). In Australia, an evaluation of a
Victorian Community Kitchen, which involved a facilitator-led programme where
participants planned, cooked and shared a nutritious meal, found that enhanced
social support was common; participants reported that the programme provided a
positive opportunity to establish friendships, socialise and enjoy cooking and
eating with others (114).
resulted in a positive effect on participants’ sense of self
Now You’re Cooking
Participants reported improvements in their children’s diet
51
Six of the 47 interventions reported on programme costs (90, 91, 95, 96, 126,
129). The cost per person varied widely depending on the nature of the
intervention. Gold et al (2014) indicated that the cost to run their intervention was
approximately USD 5 per person for food supplies (90); whilst Levy et al (2004)
reported the average the food cost per participant in their programme as USD 22
for the programme (91). A practical cooking workshop in Australia for Aboriginal
and Torres Strait Islanders reported an approximate cost of AUD 2.20 to AUD
5.20 per person or a total of AUD 68 per workshop (126). In a six-week
intervention in the UK the cost per person was GBP 84 (95).
52
2.3.3 Review discussion and conclusion
Findings showed that cooking skills interventions can improve reported dietary
outcomes, cooking and eating behaviours, cooking knowledge, skills, confidence
and attitudes. They may also have an impact on some health outcomes, such as
reducing cardiovascular risk factors which was shown in one study. In addition, a
number of qualitative evaluations of cooking skill programmes reported increased
opportunities for social support and decreased stress relating to food security
issues, outcomes that were not present in the quantitative assessments.
53
2.4 Jamie’s Ministry of Food (JMoF)
2.4.1 Programme history
JMoF is a community-based programme that teaches basic cooking skills to non-
cooks. Celebrity chef Jamie Oliver introduced the programme in the UK in 2008.
A British initiative during World War II inspired JMoF. During the war, the
government introduced a national network of food advisors and cooking teachers
to assist the public in using available food rations (133). The programme in its
current form is about “keeping cooking skills alive” by passing cooking skills on
and teaching people how to cook from fresh ingredients (133). The programme
originated in the UK and is currently operating in Australia, the UK and the US.
54
Jamie’s Food Revolution. This television series showed Jamie working on a
continues to operate and is called Huntington’s Kitchen
used a “ ” model to explore participants’ actions and
“ ” to “ ” “ ” to “ ”
was positively impacting participants’
55
2.4.2 Jamie’s Ministry of Food (JMoF), Australia
The JMoF programme was brought to Australia by The Good Foundation (TGF),
a philanthropic independent charity established by the electrical and white goods
retailer The Good Guys. The Foundations aim is to “inspire all Australians to
embrace a healthy lifestyle” (139). At the commencement of the Australian
programme in 2010, the programme’s mission statement was to “provide an
engaging community focused programme that teaches basic cooking skills and
good nutrition to non-cooks all over Australia, regardless of age, demographic or
ethnicity, to improve their quality of life and health” (140). A manifesto to the
programme is also available, see Appendix 2. In 2014, this mission statement was
changed to read, “Jamie’s Ministry of Food aims to educate, empower and inspire
people to love and enjoy good food, learning how to cook, understanding where it
comes from and recognising the power it can have on health, happiness and
finances” (140).
56
Jamie’s Ministry of Food
The first of the JMoF Australia programmes was launched in Ipswich,
Queensland in April 2011. The Ipswich site was jointly funded by philanthropist
Mr Andrew Muir (owner of The Good Guys) and the Queensland State
Government, as well as local partners. Ipswich was intentionally chosen because
it is an area of significant social disadvantage (141) and has high prevalence
levels of overweight and obesity (142).
57
Ipswich is located 40 km from the Queensland capital city of Brisbane. It is a
mid-sized regional city with an estimated population of 183,105 persons; it is
currently experiencing modest but sustained population growth of 4.6% per year
(143). In 2011, the majority of the Ipswich population (81.2%) spoke only English
at home; amongst the remaining 18.8%, the most common languages were
Vietnamese (3.6%), Samoan (1.6%), Spanish (0.6%), Hindi (0.4%) and Mandarin
(0.4%) (144). Only 3.5% of the population was of Aboriginal or Torres Strait
Islander descent, which is a similar proportion to that in the total Queensland
population (3.6%) and marginally higher than in the Australian population (2.5%)
(144).
58
Figure 3: Jamie’s Ministry of Food Centre, Ipswich, Queensland, Australia.
59
2.5 Summary
This chapter has provided an overview of the importance of cooking skills for
healthy diets and the barriers around preparing home-cooked meals. The chapter
demonstrates that a number of factors influence the healthiness of the people’s
diets and cooking skills and home-cooked meals are but several pieces of the
puzzle to improving the population’s diets. The chapter has highlighted that there
are a number of influences on the preparation of home-cooked meals via various
socio-ecological factors pertaining to an individual's personal capabilities,
demographics, beliefs and the environment. The chapter critically reviewed
cooking skills interventions in the community that aimed to improve the
population’s cooking skills; however, these interventions are generally poorly
evaluated, meaning more rigorous research is required in this field. The chapter
concluded by describing JMoF Australia, and highlighted the programme’s lack
of evaluation to date. It should be noted that there is limited evidence around
cooking practices and interventions within the Australian context as most of the
research has been conducted in the UK, US and Canada. The next three chapters
highlight the methods used in this evaluation of JMoF Australia.
60
Chapter 3. Mixed methods evaluation of Jamie’s Ministry of Food Australia
programme 3.1 Chapter overview
In Chapter 2, the literature around cooking was identified and the background to
Jamie’s Ministry of Food (JMoF) Australia discussed. Chapter 3 provides a
rationale for the evaluation of JMoF and positions the research in social theory.
This chapter outlines the overarching research questions used to guide the
research and the mixed methods study design. Chapter 3 is the first of three
chapters to describe the methodology for this evaluation.
3.2 Evaluation and rationale
In 2011, TGF and Queensland Health commissioned Deakin University, in
collaboration with colleagues at The University of Melbourne, to conduct an
evaluation of the JMoF Ipswich programme as operating in Ipswich, Queensland,
Australia. The purpose of the JMoF evaluation was to inform the practice of
programme developers, implementers, evaluators and the wider research
community interested in cooking skills programmes, and to inform policy around
investment in the programme. Given the limited quantity and quality of evidence
on cooking skills (as outlined in Chapter 2) this research offered an opportunity to
build the evidence on the effectiveness of practical cooking skills programmes.
3.2.1 Jamie’s’ Ministry of Food (JMoF) programme objectives
The programme initially did not have predefined programme objectives.
Consultation occurred between TGF, Queensland Health and the evaluation team
to determine clear programme objectives. The following three programme
objectives were identified:
61
1. To provide opportunities, to people of different ages and demographic
background, to experience and learn how to cook healthy meals quickly
and cheaply;
2. To increase programme participants’ cooking skills, knowledge and self-
efficacy;
3. To increase programme participants’ enjoyment of food and social
connectedness.
3.3 Theoretical framework to guide evaluation
In Chapter 2, a broad range of inter-related factors that influence an individual’s
cooking behaviours were described. This section describes the theoretical
framework, which is the foundation underpinning and guiding this research; it
consists of inter-related theories that allow for reflection on specific factors that
influence behaviours. It is important to ground research in theory because theories
provide a justification for an individual’s actions. Theories are used to explain and
predict events or situations in order to make findings meaningful and
generalisable (145). Theories are considered a set of inter-related concepts,
definitions and propositions which describe and identify why a problem may exist
(146). They also help us to discover various influences on behaviour, for example,
knowledge, attitudes, self-efficacy, social support and available resources (146).
–
62
participants’ behaviours.
63
a person’s environment . An individual’s environment includes both the
rnal to the person as well as the person’s social
individual’s perception of their environment or situation, for example, their time
“
”
belief in one’s
belief influences “
”
’
person’s self
shape an individual’s capacity through mastered skills enabling them to succeed
64
person’s capacity through and developing skills may influence one’s
mood. It is important to control one’s somatic reaction
’
that a behaviour will “ ”
65
’
3.4 Programme logic model
The purpose of the JMoF evaluation is to determine the immediate impacts of the
programme on participants as well as any sustained outcomes that may result over
a longer period. A programme logic model, Figure 3, was developed by the
candidate in collaboration with the evaluation team in order to focus the
evaluation inquiry and show potential causal pathways and influencing factors for
behaviour change. This model was used to help define the evaluation, the
objectives and research questions.
66
3.5 Evaluation research questions
As outlined in Chapter 1, this study aims to determine what were the impacts of
JMoF Ipswich Australia as a community-based cooking skills intervention?
More specifically, the evaluation aims to answer the following research questions
surrounding participation in the JMoF programme:
1. Did JMoF increase participants’ skills, knowledge, attitudes, enjoyment,
and satisfaction of cooking and cooking self-efficacy (confidence to
cook)?
2. Did JMoF result in broader positive outcomes for participants in terms of
behaviour change to a healthier diet, more affordable healthy meals, and
improve self-esteem and social connectedness?
3. What were participants’ expectations and experiences of the programme?
4. What were the moderators, barriers and facilitators related to participants’
cooking skills behaviour change?
5. Did participants experience any unanticipated outcomes after attending
JMoF?
understanding participants’ experience
Figure 4: Jamie’s Ministry of Food programme logic model
68
3.6 Mixed methods evaluation design
A mixed methods approach was adopted because the utilisation of both
quantitative and qualitative methods would yield a better understanding of the
JMoF programme compared to a single study design approach. Mixed methods
research acknowledges that there are multiple legitimate ways of making sense of
the social world (132). There are a number of different definitions of what
constitutes mixed methods research. Johnson et al (2007) collated the ideas and
definitions used by leading mixed methods researchers and concluded that mixed
methods research is defined by the following; “the type of research in which a
researcher or team of researchers combines elements of qualitative and
quantitative research approaches (e.g., use of qualitative and quantitative
viewpoints, data collection, inference techniques) for the broad purposes of
breadth and depth of understanding and corroboration” (154, p.123).
different methods to “
”
69
Figure 5: Quantitative and qualitative paradigms*
*Diagram adapted from Crotty (1998) (156)
Epistemology
Theoretical perspective
Methodology
Methods
Objectivism
Post-positivism
Deductive inquiry
Questionnaires
Subjective
Social-ecological
Inductive/ Abductive
Interviews
Study design Quantitative research Qualitative research
70
71
Figure 6: Mixed methods design
Mixed methods study What were the impacts of JMoF Ipswich Australia as a
community-based cooking skills intervention?
Quantitative evaluation methods (Chapter 4) and results
(Chapter 6)
Quantitative research questions
1. Did JMoF increase participants’ skills, knowledge, attitudes, enjoyment, and satisfaction of cooking and cooking self-efficacy (confidence to cook)?
2. Did JMoF result in broader positive outcomes for participants in terms of behaviour change to a healthier diet, more affordable healthy meals, and improve self-esteem and social connectedness?
Qualitative research questions
3. What were participants’ expectations and experiences of the programme? 4. What were the moderators, barriers and facilitators related to participants’ cooking skills behaviour change? 5. Did participants experience any unanticipated outcomes after attending JMoF?
Qualitative evaluation methods (Chapter 5) and results
(Chapter 7)
Mixed methods integration Interpretation of results
(Chapter 8)
d i t tiMi d h
72
3.7 Summary
Chapter 3 has described the rationale for this mixed methods evaluation and
positioned the research in social theory. Two theoretical frameworks –
experiential learning theory and Social Cognitive Theory – will guide this work.
A programme logic model was also developed to test causal pathways to
behaviour and to develop research questions. This chapter justifies reasons for a
mixed methods approach and explains how the quantitative and qualitative
methods were combined. The next two chapters describe the methods for the
quantitative and qualitative studies, separately.
73
Chapter 4. Quantitative methods
4.1 Chapter overview
In Chapter 3, the overarching mixed methodology of the JMoF evaluation was
discussed along with the rationale for using a mixed methods design. This
chapter, Chapter 4, details the quantitative methods. Quantitative studies are vital
for providing evidence to demonstrate programme effectiveness (159). The
methods described in the quantitative study have been published by Flego,
Herbert et al (2013); the candidate provided significant input to this paper in her
role as second author (8). A copy of this methods paper is provided in Appendix 3
This chapter goes into more detail than is outlined in the journal article and
provides justification for the methods used.
4.2 Quantitative paradigm
The approach taken to this quantitative study is a post-positivist approach, which
is a modified version of a positivist approach.
Positivism was popularised in the 19th century by philosopher Auguste Comte
and was founded on a "given" that reality is certain; "Objects in the world have
meaning prior to, and independently of, any consciousness of them." (156, p.27).
This means that reality can be understood by attempts to search for universal laws
through objective observation, measurement, recording and verification (156,
160). Positivism relies on an objective approach which puts aside bias and beliefs
creating distance between the researched and the researcher (156). A positivist
approach searches for and develops universal and scientific laws, which can be
deduced from hypothesis testing and standardised statistical analyses (160).
74
This version of positivism has evolved over time and its limitations have been
recognised. One limitation of such an approach is the view that anything can be
objectively measured and that all findings are certain and constitute an accurate,
unambiguous account of the world (156, 161). Another limitation is that a
positivist believes in a scientific, systematic and organised world, which is
removed from the “lived world” or reality that people experience (156). This
focus on facts does not acknowledge underlying mechanisms of observations
(156). These restrictive limitations led to the evolution of post-positivist
approaches.
Post-positivism is a moderate version of positivism. It both entails the belief that
reality exists and values objectivity; but it recognises uncertainties and critiques
them (156). This approach acknowledges probability rather than certainties, while
reality exists it is not unchangeable and absolute, thereby recognising the
possibilities of bias (156). It also accepts that observation can alter those being
observed, and challenges the notion of absolute objectivity (156). With these
uncertainties in mind, this quantitative research takes a post-positivist approach.
75
This study adopts a deductive approach to the research. Quantitative studies
commonly denote from an objective view of knowledge and often use deductive
methods to test a general theory (159). Deductive research aims to test hypotheses
by collecting data through measurement and observation (162). Deductive inquiry
is driven by the research question and involves the use of questions that have
predetermined responses seeking specific answers based on theoretical concepts
(161). Deductive approaches must have a hypothesis to test; this bring us to the
next section that sets out the research aims and objectives.
4.3 Research aims and objectives
The overarching research question for this mixed methods evaluation is
what were the impacts of JMoF Ipswich Australia as a community-based cooking
skills intervention?
In order to explore this overall research aim, the quantitative study specifically
aims to address the following two research questions:
1. Did JMoF increase participants’ skills, knowledge, attitudes, enjoyment, and
satisfaction of cooking and cooking self-efficacy (confidence to cook)?
2. Did JMoF result in broader positive outcomes for participants in terms of
behaviour change to a healthier diet, more affordable healthy meals, and
improve self-esteem and social connectedness?
76
4.4 Quantitative study parameters
4.4.1 Study design
The quantitative study used a longitudinal quasi-experimental design with a wait-
list control group. “Quasi-experimental” refers to the use of non-randomised
methods of allocating participants, often in natural or real-world settings (163).
Random assignment of participants to an intervention or control group is
considered the gold standard of research methods and is an important feature in
experimental studies because it helps protect against threats to internal validity
(164). However, randomisation is often difficult to achieve in community-based
interventions. The use of a non-randomised design was an appropriate choice for
the JMoF programme as randomisation was neither suitable nor feasible. Many
participants attending JMoF wanted to go with family and friends; randomisation
would have made this impossible and may have potentially jeopardised their
recruitment to the programme.
77
Figure 7: Quantitative study design
related to “ ”
ng and highlighted that if the “ ”
78
4.4.2 Sample size
Sample size calculations were conducted to determine the likely numbers required
in each group to ensure that the study was able to detect a statistically significant
change. The available literature around cooking skill interventions is limited, and
there were no precise sample size calculations available related to measuring an
effect of cooking confidence. Therefore, the sample size calculations were based
on the second primary outcome of daily vegetable intake, for which there were
data available to predict an effect size. A cooking skills programme in the UK
found that an effect size of one serve of vegetables per day was unlikely to be
achieved by programmes of this nature (77).
79
4.4.3 Recruitment
During the two-year data collection period participants were continuously
recruited to the evaluation. The programme was open to all members of the
general public. Once a participant registered for JMoF they were automatically
assigned to a control or intervention group based on their allocated programme
start date. The invitation to participate in the evaluation was controlled by the
programme implementers, who set up computer generated rules in the JMoF
participant database operating system Salesforce (170). Salesforce automatically
emailed members an invitation to be involved in the evaluation and a unique web
link to a Plain Language Statement and a questionnaire at pre-programmed times
based on their programme start date. For all non-responders or participants
without an email address, a paper version of the questionnaire was sent.
participant’s date of birth was manually checked against the date
80
20 “ ” store charge card, redeemable at any
4.4.4 Data collection
At each time point, participants completed a 15-minute self-report questionnaire
consisting of 40 multiple-choice questions (Appendix 4). The questionnaire was
designed in collaboration with the evaluation team and key stakeholders to elicit
data to cover all four research domains as described in the programme logic
model (Figure 4).
’
81
Cooking and food knowledge questions were based on content taught within the
JMoF programme. The specific nutrition knowledge questions, which aimed to
align with the nutrition messages embedded within the programme and to
individually test knowledge around salt, fat and sugar, were sourced from a
nutrient knowledge questionnaire (171). Questions about cooking and eating
attitudes and beliefs were adapted from questions used in a food choice
questionnaire administered to French and English populations (174).
Cooking confidence (a primary outcome) was used as a proxy to measure cooking
self-efficacy. Cooking confidence questions were adapted from a two-page
questionnaire developed to assess cooking skills interventions (131), which also
incorporated a question on the willingness to try new foods. One additional
question was added – confidence that what they cooked would turn out well; this
addressed self-efficacy and was taken from the questionnaire used in the
evaluation of the Men Can Cook! Programme in Ontario, Canada (111). The
cooking confidence questions were adapted to a five-point Likert scale ranging
from “not at all confident” to “extremely confident”.
82
Questions about shared enjoyment of cooking and meal satisfaction were adapted
from questions used in the evaluations of The Stephanie Alexander Kitchen
Garden (71) and Men Can Cook!, a men’s cooking group (111).
Measures to investigate cooking behaviours were adapted from the cooking skills
questions developed by Barton et al (2011) (131). Items included measuring
changes in weekly frequency of cooking the main meal from basic ingredients and
the inclusion of vegetables with the main meal.
The primary outcome used to underpin the sample size calculations was self-
reported daily vegetable intake. Questions about daily and weekly consumption of
vegetables, fruit and take-away food were sourced from the Queensland Health
self-report health status survey (169).
Questions about household food expenditure and food affordability and budgeting
were also asked. Household expenditure questions were adapted from The
Household, Income and Labour Dynamics in Australia (HILDA) Survey (173).
Questions about weekly social eating and normative eating behaviours, such as
whom the participant ate with and where they ate their dinner meal, were adapted
from The Stephanie Alexander Kitchen Garden Evaluation questionnaire (71).
83
Rosenberg’s self-esteem scale was used to test global self-esteem (175). The
Rosenberg self-esteem scale is a well validated and reliable measure of self-
esteem based on 10 questions on a Likert scale; the scores are combined to
produce an overall self-esteem score. Scores ranged from 0-30, for ease of
interpretation scores were were categorised as low self-esteem (less than 15)
normal self-esteem (15-25) and high self-esteem (over 25). The measure of
perceived general health status was taken from the Queensland Health self-report
health status survey (169). Self-reported height and weight were used to calculate
participants’ BMI using the BMI formula: kg/m2 (176).
Demographic questions were taken from a range of sources to facilitate
comparisons with the Australian and Queensland populations, where possible.
The demographic questions related to education level, employment and income
levels, cultural ethnicity (18, 169, 173, 177) as well as household characteristics
(131).
The first stage of questionnaire development involved discussion on whether the
questionnaire content matched the domains that it aimed to measure (163). The
first phase of the content validity process involved a workshop with the JMoF
evaluation reference group, in July 2011, to determine the evaluation direction
and design and to agree on the key domains to be tested. Present at this workshop
were two programme implementers, a government representative, five chief
investigators of the evaluation study, the evaluation co-ordinator (a research
fellow) and the PhD candidate. The chief investigators, academics with expertise
in research, shared their thoughts about different questions to be used. The
programme implementers were able to comment on the content relating to the
programme and relevance to programme participants. The government
84
representative provided insight into other programmes and areas of research in
Queensland and questions of particular relevance to government as an investor in
the programme.
informal focus group discussion at the end of the participants’ class.
There were two modes of data collection; a web-based questionnaire was the
primary method of data collection, supplemented by a paper-based mailed version
sent to non-responders. The data collection occurred over approximately two
years, between 25 November 2011 and 6 December 2013. Due to the rolling
nature of the programme, questionnaire distribution occurred weekly. During the
piloting of the questionnaire it became apparent that in-class survey completion
85
was not an appropriate delivery method because this period was too rushed and
took time away from the class and clean-up time. Therefore, it was decided that a
web-based version would be the most appropriate method of survey distribution.
implementer’s participant database system ‘Salesforce’, web
86
4.4.5 Data management
All data were entered and stored in Microsoft Excel, checked for errors then
imported into STATA (version 12.0) (181) for analysis preparation. The web-
based questionnaires were downloaded weekly from Qualtrics into an Excel
format. Online data were combined into a master Excel file and checked for
duplicates and computer errors to ensure the online survey system was
functioning properly. Paper-based questionnaires responses were entered
manually into Excel by the researcher and two casual research assistants. Data
from the web-based questionnaires and paper-based questionnaires were initially
stored separately because the paper-based questionnaires required rigorous
checking for errors.
’
“ ” category were checked to see if the “ ” responses fitted one of the
87
=INDEX(A2:A30,RAND()*100+1)
The sample determined was double entered and compared to the original data for
any discrepancies. The overall number of errors found was 0.8%.
4.4.6 Data analysis
All analysis was conducted using STATA (version 12.0) (181). The purpose of
the analysis was to explore changes both within and between the intervention and
control groups and over time (T1, T2 and T3). Statistical analyses were based on
the set of individuals who registered for the programme, responded to an
invitation to participate in the evaluation, and subsequently completed the
baseline questionnaire. Significance testing was performed for all outcome
variables to generate a P-value which shows the probability of observing a
difference between the intervention and the control group; the smaller this P-value
the stronger the difference between groups is (183). Results were deemed
88
significant if P<0.05. Follow-up T2 or T3 data were only used if the participant’s
T1 baseline data were available.
Demographic and baseline characteristics of the participant sample were
summarised using standard summary statistics (means and standard deviations)
and non-parametric statistics (medians and inter-quartile ranges). Demographic
variables included: gender, age, Aboriginal or Torre Strait Islander status,
language spoken at home, postcode, employment status, education level,
household yearly income, household characteristics, and who the participant
attended the programme with. Descriptive statistics were compiled and chi-square
analysis was performed to test for any differences between the intervention and
control groups at baseline.
89
‘ ’
‘ ’
90
Food expenditure data were collected in AUD using seven multiple choice
categories to indicate a weekly price range. Total food and drink ranged from $0
to $200 or more, and fruit and vegetable and take-away food expenditure ranged
from $0 to $100 or more. The data were then re-coded by the mid-points and
analysed on a continuous scale.
Rosenberg's global self-esteem consisted of a series of 10 questions, which were
combined to generate an overall self-esteem score (175). Scores ranged from 0-
30: low self-esteem (less than 15), normal self-esteem (15-<25) and high self-
esteem (25 and over) (187). This variable was analysed as a continuous variable.
Perceived general health was measured on a 5-point scale with poor = 1, fair = 2,
good = 3, very good = 4, excellent = 5. General health was analysed as a
continuous variable.
91
Two analyses of each outcome variable were conducted: (I) comparisons between
groups of their changes over time from T1 to T2 (equivalent to testing for a group
by time interaction), and (II) comparisons of the three time points (T1, T2 and T3)
within the intervention group. The latter tested for the sustainability of any
programme effect at six-month follow-up.
Each model-based analysis was re-run adjusting for covariates (age, gender,
employment status and a combination of the three) when the covariates exhibited
baseline differences (that is., differences at T1) between the control and
intervention groups.
92
Cohen’s d statistical test
4.5 Ethical considerations
Ethics approval was obtained from Deakin University Human Research Ethics
Committee (reference number: HEAG-H 117_11) in October 2011. Consent was
required for all participants in the evaluation. Each participant with an email
address was sent an email inviting them to complete the web-based questionnaire.
The email indicated that consent was granted by completion of the web-based
questionnaire. Once participants had clicked on the link to the questionnaire,
information about the evaluation was included in a Plain Language Statement.
93
Postal questionnaires were sent with a paper copy of the Plain Language
Statement, consent form and an invitation letter, which was adapted from the
initial email content. The invitation email, Plain Language Statements and consent
forms are found in Appendix 5. All data were de-identified. The data collected
were initially in an identified form, however each participant was assigned a
unique identifier to facilitate the linking of data collected at different time points.
All identifiable data (excluding their ID codes) were removed once entered into
the database. Identifiable data were stored separately to the central database.
4.6 Summary
This chapter has described the methods used in this quantitative study. In
summary, the evaluation was open to all persons registered for the JMoF
programme who were 18 years old or over. The design was a longitudinal quasi-
experimental design with a wait-list control. Intervention participants were
measured at three time points; before programme commencement, after the
completion of their programme, and at six months follow-up. The wait-list control
group were measured twice; 10 weeks prior to commencing their JMoF class and
at the commencement of the programme. There was no six-month follow-up
control; instead, results were compared to Queensland state-wide data. An online
questionnaire was used followed by a paper postal questionnaire to maximise
recruitment. All repeated measures were analysed using multilevel mixed model
analysis or generalised estimating equations (GEE). The next chapter reports the
methods for the qualitative study.
94
Chapter 5. Qualitative methods
5.1 Chapter overview
In Chapter 4, the quantitative methods were presented. This chapter presents the
qualitative methods. The qualitative study aims to provide a deeper understanding
of participants’ experiences of the programme and to explore the barriers and
facilitators to cooking. In Chapters 2 and 3, relevant literature and theoretical
concepts were reviewed to seek explanations to explore factors that may impact
on cooking behaviours. The same literature has been taken into account in the
development of this qualitative evaluation. This chapter outlines the qualitative
methods used in this evaluation including the qualitative paradigm, research aims
and objectives, study design parameters related to participant sampling and
recruitment, and data collection and analysis.
5.2 Qualitative paradigm
Qualitative methods are useful for understanding human behaviour. They provide
opportunities to explore the reasoning behind particular behaviours, their
interpretation, and the meaning that they give to the person’s own and other
people’s actions (190). In other words, they facilitate understanding of the ‘why’
and ‘how’ people do the things they do (191). The use of a qualitative component
in the evaluation of the JMoF programme provides for an exploration of
individual experiences of the programme. It provides a voice for participants and
offers the capacity to explain both significant and insignificant quantitative
findings, thereby providing the ‘why’ to differences that may be of consequence
(191). A qualitative approach is an appropriate fit to explore the perspectives of
JMoF participants around food and cooking in order to understand the broader
socio-cultural and environmental factors influencing their choices and actions.
95
Table 6: Introduction of the researcher
Reflectivity – about the researcher I, Jessica Herbert the researcher and writer of this thesis, grew up in a way that was somewhat different to many other children. I am the daughter of a retired Australian Army Officer and my family and I moved from place to place, sometimes internationally, throughout most of my childhood and adolescence. This constant moving made me appreciate new people, with interest and curiosity. I believe this childhood has also made me resilient, adaptable and flexible. My middle-income family upbringing gave me a fortunate start in life. I lived in a stable family home where I was always fed; family holidays were a regular experience and education and new experiences were highly valued.
equity for all and my beliefs in people’s rights of safety and good health. I am
5.2.1 A social constructionist approach to the research
96
The social world plays key roles in influencing people in different ways. With this
in mind, the approach to the qualitative evaluation assumes a social
constructionist approach. This implies that people construct meaning in various
ways, as people engage with the world they begin to interpret it differently (156,
161, 193). Reality can be constructed differently by others and it is ultimately
influenced by a number of everyday factors. Through language, the construction
of reality is created and transmitted to others (194). Crotty (1998) uses an analogy
of a tree to explain the concept of social constructionism; “What the
‘commonsense’ view commends to us is that the tree standing before us is a tree,
it has all the meaning we ascribe to a tree. It would be a tree with that same
meaning whether anybody knew of its existence or not. We need to remind
ourselves here that it is human beings who have constructed it is as a tree, given it
the name, and attribute to it the association we make with trees. It may help if we
recall the extent to which those associations differ even within the same overall
culture. “Tree” is likely to bear quite different connotations in a logging town, an
artists’ settlement and a treeless slum” (156, p.43).
factors will influence participants’ personal values, beliefs and cultural influence
5.2.2 An abductive approach to the research process
97
The aim of conducting a qualitative study was to provide insight into and
understanding of JMoF participants’ attitudes, beliefs and practices around
domestic cooking. In order to do this, an abductive approach to the research was
taken. Abductive research uses a combination of both inductive and deductive
techniques informed by theoretical and empirical literature (161). Deductive
inquiry, as the method of inquiry described in Chapter 4, aims to test hypotheses
(162) through the use of methods such as multiple choice questions (161).
Inductive inquiry, on the other hand, aims to understand and establish patterns
from the data collected; it is more of an open, creative and exploratory process
collected through an accumulative process to confirm patterns, themes and
interactions as they emerge in order to produce generalisations (157, 161). Patton
(2002) argues that “the strategy of inductive designs allow the important analysis
dimensions to emerge from patterns found in the cases under study without
presupposing in advance what the important dimensions will be” (161, p.56).
There is an element of deduction used within the qualitative study design, drawn
from pre-existing literature and theoretical concepts that have been used to guide
the methods and data collection. The exploratory nature of the qualitative work
also allows for new ideas and patterns to emerge and to not be restricted by pre-
existing ideas found in the literature. Morse and Mitcham (2002) believe that
research must have a focus of inquiry and make use of some inductive techniques.
They suggest considering the literature to make the analysis smarter and to
develop a skeleton which is guided by the literature and can be used to focus and
facilitate the inquiry (195). The inductive component of the research facilitates the
filling out of the skeleton. An abductive approach moves back and forth between
inductive and deductive inquiry and facilitates the discovery and understanding of
the motives underpinning the behaviour of different people.
5.2.3 Conceptual frameworks
The conceptual frameworks that guide this qualitative study are taken from a
social-ecological model, with constructs of Social Cognitive Theory and
experiential learning theory applied at the interpersonal and intrapersonal level to
understand behaviour. An ecological theory recognises that multiple factors can
98
influence health behaviours, as described in Chapter 2. To complement this
perspective, constructs of Social Cognitive Theory (45, 196) and experiential
learning theory can be applied to the conceptual framework of the qualitative
study, as described in Chapter 3. Constructs of Social Cognitive Theory were used
to guide the analysis of the qualitative data obtained. The next section outlines
how the data were obtained before describing how the data were then analysed.
5.3 Research aims and objectives
The overarching research question for this thesis was to investigate what were the
benefits of community-based cooking skills programme using Jamie’s Ministry of
Food Australia programme as a case study? As pointed out previously, healthy
eating is complicated and individuals place complex meanings around this topic,
which warrants further investigation as to why people eat what they eat. Cooking
skills impact at one level of this complex issue, and it is important to understand
what drives people to act in ways that are healthy or unhealthy in terms of the
food that they prepare and eat. As part of the broader mixed methods study, the
qualitative study specifically aims to address the following three research
question:
participants’ expectations and experiences of the programme?
What were the moderators, barriers and facilitators related to participants’
5.4 Qualitative study parameters
99
5.4.1 Study design
A longitudinal qualitative design was selected to follow programme participants
over the course of their JMoF journey. Programme participants were interviewed
three times over a period of approximately eight months. Repeated semi-
structured interviews were used to build an understanding of programme
participants’ food behaviours and cooking habits over time and their expectations
and experiences of the programme.
prospective and retrospective accounts of participants’ experience
nature allowed the researcher to use the participants’ responses from previous
if any reported changes in participants’ described behaviour had been mainta
100
Figure 8: Qualitative evaluation time frame and study design
101
5.4.2 Sampling
Participants were selected using purposive sampling, utilising maximum variation
(161, 204) to ensure that a diverse group of participants was recruited. Specific
factors considered when sampling included: socio-economic status, age, gender,
family structure and cooking confidence level. By including these different
elements, the selected sample would allow for the opportunity to capture different
expectations and experiences of the programme.
Qualitative studies typically have a smaller sample size compared to quantitative
studies. The emphasis in qualitative studies is to select participants who provide
in-depth, meaningful data, rather than relying on a large sample size to make
statistical comparisons (161, 205). An important factor with a qualitative sample
is whether the data can answer the research questions posed, and if the sampling
processes were flexible enough in order to do so (205). It was believed that the
102
sampling methods used in this study met these requirements because a diverse and
purposively selected sample was used.
5.4.3 Recruitment
Recruitment of participants for the qualitative evaluation occurred in two ways.
Firstly, the quantitative baseline survey included a question asking if the
participant was willing to be contacted for a future interview. One hundred and
ninety-nine people indicated they were happy to be contacted and provided
contact information. Initial recruitment was potentially confined to those
participants who were beginning the JMoF programme during the data collection
period, timetabled for August to September 2012. This was the period the
researcher was in Ipswich, because funding was limited for the researcher to
travel and stay in Ipswich. This meant only 24 people were eligible. All 24
participants were contacted via phone to schedule an interview time; however
only 12 were recruited. The remaining 12 were unable to participate due to time
commitments or interview scheduling constraints, such as not being available
during the researcher’s scheduled time in Ipswich.
during the researcher’s class observations of gro
103
5.4.4 Data collection
Semi-structured interviews were deemed the most appropriate method of data
collection to answer the research questions and gain insight into participants’
perspectives, such as their thoughts, feelings and intentions (161). The interviews
used a conversational style, with the researcher remaining neutral and non-
judgemental, to allow participants to share their perceptions. Interviews were kept
focused with the use of an interview guide. The interview guide was used as just
that, a guide, which allowed the interview to have some focus specifically relating
to the study’s objectives and, at the same time, was flexible enough to not restrict
104
the freedom of the participants to raise issues and the researcher to probe and
explore issues as they emerged in the interview (161). Questions were framed to
be open-ended, which allowed participants the chance to respond in their own
words (161). Patton (2002) suggests that one weakness in using an interview
guide may be that topics are unknowingly omitted. However, this was not thought
to be a concern because the researcher remained flexible and allowed enough time
during each interview to explore new topics if they arose. The full guides for each
interview are found in Appendix 6. Brief versions of the topics for each interview
are listed below.
Interview one
Participant motivations for joining the programme
Expectations about participating in the programme
Current cooking and food behaviours and attitudes
Interview two
Feedback about the programme and programme environment
Current food and cooking behaviours and attitudes
Perceived changes made after the programme
Unexpected outcomes
Reflections from interview one
Interview three
Current food and cooking behaviours and attitudes
Perceived changes made after the programme, and if change was or was not
sustained
Unexpected outcomes
Reflections from interview two, and over the last six months
105
The first few interviews were conducted with the researcher and a senior
researcher present. During this time, the interview structure was discussed by the
two researchers and refined.
interview during the researcher’s time in Ipswich.
the researcher’s
’
106
5.4.5 Data analysis
Qualitative data analysis was a challenge. While the methodological literature
offered guidance and direction, there is no formula of exactly how to transform
qualitative data into insightful findings (161). There were instances when the
researcher felt completely out of her depth with the challenge of: (I) managing the
vast amount of data presented, and (II) trying to creatively explore the data and to
synthesise a story which both met the research questions and provided an accurate
and conceptual account of participants’ experiences of the programme and the
motives, intentions and reasons behind their behaviours. Due to the longitudinal
nature of the study, data analysis occurred concurrently with data collection.
Essentially data analysis began with the first instance of data collected.
Figure 9: Qualitative data analysis process*
107
*adapted figure from Green et al (2007) (208)
The data immersion process first began during conduct of the interviews. The
second process of immersion occurred during the transcription phase. All
interviews and memos were transcribed verbatim. The researcher transcribed 34
of the 42 transcripts. Nine transcripts obtained during the third round of
interviews were transcribed by a professional transcription agency (A1
Transcriptions, http://www.a1transcriptions.com.au/). This service was utilised
due to the availability of funding from a small internal grant. The process of
immersion for these nine interviews began from the error checking stage. The
next process of immersion was to check each transcript for errors by listening to
the interview and reading the transcript. The transcript was then printed and re-
read to begin the coding process. This process of immersion started to stimulate
ideas and provide a clearer picture of the issues being researched (208).
The coding process involved applying descriptive labels to the transcripts by way
of examining and organising the data (208). Coding was first conducted via
printouts of each transcript; highlighting and making notes in the margins as the
transcripts were read and re-read, linking back to the immersion process. Notes
were made on topics and issues present, and any other points of interest. The
second stage of coding was done electronically. All interview transcripts and
memos were uploaded into a qualitative software package, NVivo (NVivo 9
[programme]: QSR International Pty Ltd 2011), to assist with data management.
108
The linking of codes was the next step in the analysis process, to create coherent
categories (208). Codes that related to one another were grouped together,
keeping in mind codes that did not fit or link with other codes. This process began
the generation of categories and starting to form explanations of the data,
providing descriptive details of what was found (208). The longitudinal nature of
this study added another element to the analysis process. Firstly, interviews at
each time point underwent the immersion and coding process to create categories.
Once all three interview points had undergone this process, the categories were
compared over time and looked at as the whole dataset. This process was
essentially to understand “what happened first, next and last to the participants”
(202, p.51). The categorised data were then reviewed to explore similarities and
differences as well as any over time comparisons and differences, to identify
patterns and determine specific relationships between categories. This provided an
109
overall conceptual picture of the impact of the JMOF programme within the
context of its unique setting and population.
The final stage of analysis was the identification of themes. Green et al (2007)
explains, “the generation of themes requires moving beyond a description of a
range of categories; it involves shifting to an explanation or even better an
interpretation of the issue under investigation” (208, p.349). Essentially this phase
was about recognising patterns and meanings within the data to conceptualise
findings. Comparisons were made with relevant theoretical concepts and the
literature base to determine if findings resonated with existing knowledge or made
new contributions to the evidence.
’
5.5 Ethical considerations
110
Ethics approval was received from Deakin University Human Research Ethics
Committee (HEAG-H 117_11) in October 2011. Written consent was gained from
participants prior to the interviews for both participation and digital recording of
the interview. Each participant was given an opportunity to read the Plain
Language Statement (Appendix 5) and to ask any questions. All written consent
forms were kept in secure locked storage.
A number of steps were employed to maintain participants’ privacy and
all information containing participants’ names and contact
5.6 Summary
This study has described the methods used for the qualitative study. In summary,
repeated 30-minute semi-structured interviews were used. Participants were
interviewed three times, before their programme, after completion of their
programme, and at six-month follow-up. Participants were purposively selected to
ensure that a broad range of participant characteristics were met. All data were
analysed thematically. The results of this research are presented in the next two
chapters: Chapter 6, Quantitative results and Chapter 7, Qualitative results.
111
Chapter 6. Quantitative results
6.1 Chapter overview
This chapter presents the results from the quantitative evaluation of the JMoF
programme. Chapter 4 outlined the quantitative methods, whilst this chapter
reports the findings from the statistical analysis of the JMoF questionnaire
repeated at different time points. More specifically, this chapter explores a range
of effects on participants associated with from their attendance at the programme.
112
6.2 Participants: quantitative study
6.2.1 Participant response rates
Over the course of data collection, from November 2011 to December 2013, a
total of 1,690 people registered to participate in the programme and were invited
to participate in the quantitative evaluation. Based on the study design, 1,526 were
allocated to the intervention group and 434 were allocated to the wait-list control
group. Figure 10 provides details of participant response rates. After allowing for
exclusions and dropouts, a total of 694 intervention participants completed
questionnaires at T1, 383 at T2, and 259 at T3. In the intervention group, 55.2%
of participants who completed T1 questionnaires also completed T2
questionnaires; and a total of 30.8% of all participants completed the
questionnaire three times, at baseline T1, T2 and T3. An additional 45 participants
completed T3 but did not complete T2 questionnaires; these participants were
included in the analysis given their baseline information and the employment of
multilevel mixed model statistical methods which allowed for all data to be used
in the analysis. The wait-list control group had a 62.9% response rate, with 237
participants completing the questionnaire at T1 and 149 at T2.
6.2.2 Baseline demographics
Table 7 shows the demographic characteristics of participants, in both the control
and interventions groups, at baseline. Overall, the two groups were largely
similar, however because the study was not randomised there were statistically
significant differences in terms of gender, age and employment.
113
Figure 10: Questionnaire response rates
Intervention T2 analysed (n =383)
Loss to six month follow-up, n=169
Intervention T3 analysed (n =214)
Whilst 45 participants failed to complete T2, they completed T1 and T3.
Loss to follow-up, n=311
Did not respond to invitation to participate in evaluation, n=691 Excluded n=141 Underage (n=120) Incomplete surveys (n=14) Incorrect timeframe (n=7)
Intervention Registered for the programme <10
weeks before programme commencement
(n= 1526)
Intervention T1 analysed (n = 694)
Wait-list control Registered for the programme >10
weeks before programme commencement (n=434)
Did not respond to invitation to participate in evaluation, n=154 Excluded, n=43 Underage (n=38) Incomplete surveys (n=2) Incorrect timeframe (n=3)
Loss to follow-up, n=88
Wait-list control T2 analysed (n =149)
Wait-list control T1 analysed (n = 237)
114
Participants’ age and gender
The sample was not culturally diverse. Most participants at baseline spoke
English (92.2% in the intervention group and 94.7% in the control group) and
lived in the Ipswich district (82.0% in the intervention group and 78.8% in the
control group). There were relatively few indigenous Australians, with only 1.8%
of intervention group participants and less than 1% of the control group indicating
they were of Aboriginal and/or Torres Strait Islander descent.
Most participants were in full or part-time employment; however there were a
significantly higher proportion of control participants (34.7%) in full-time
employment compared to the intervention group (26.4%) (P=0.02). In addition,
23% of intervention participants were retired compared to 21% in the control
group. There was also a larger proportion of control participants in the home
duties or carer category compared to the intervention group (18.4% and 14.4%).
In terms of education levels, nearly half of all participants indicated that high
school (year 12 or less) was their highest level of education (intervention group
47.8% and control group 45.8%). Household yearly income was proportionally
higher in the control group compared to the intervention group. The intervention
115
group had a higher proportion of participants in the lower income group with
50.3% earning AUD 50,000 or less compared to the control group (38.5%).
be slightly higher as the ‘other’ category sometimes incl
grandparent living in the household. Other examples in the ‘other’ category
Over half of all participants in both control and intervention groups, indicated that
they attended the programme with another person. Of those that shared their
programme experience, more intervention participants (24.4%) attended the
programme as a member of a community group compared to control participants
(7.3%).
For the most part, the demographic characteristics did not change dramatically
over the course of the evaluation. As previously shown, there were more
participants who completed the T1 questionnaires than T2 and T3. Over time, the
older participants (aged ≥ 50 years) were more likely to remain in the study and
complete the T2 and T3 questionnaires.
116
6.3 Jamie’s Ministry of Food sample compared to the
Ipswich population
Over 500 participants contributed to this quantitative evaluation of JMoF
Australia. Participants were mostly female, spoke English at home, lived within
the Ipswich district and were in full or part-time employment. The JMoF
population reflected the Ipswich population in terms of ages of the age of persons
attending the programme. Whilst any direct comparisons with the Ipswich
population should be interpreted with care, it would appear that the JMoF
evaluation sample had a lower population of participants from Aboriginal and/or
Torres Strait Islander backgrounds than Ipswich (3.5% in Ipswich, and 0.9% to
1.8% at baseline in the JMoF sample). Also lower was the proportion who spoke a
language other than English at home (18.8% in Ipswich compared to 5.3% to
7.8% at baseline in the JMoF sample). Household composition was markedly
different in the JMoF sample compared to Ipswich. There was a lower population
of one parent families with children living at home in the JMoF sample (7% to
8.9% at baseline compared to 19.2% in the Ipswich population). There also
appeared to be a lower proportion of participants with children living at home in
the JMoF sample. Overall, at baseline the JMoF sample comprised approximately
37.2% to 42.2% of participants in households with two parent families and
children (aged 0 to 18) living at home. State-wide Ipswich population data show
that 45.2% of the population were couple families with children living at home
(143), which is higher than the JMoF sample.
117
Table 7: Demographic characteristics of participants1 by time point
Intervention % (n) Control % (n) Demographic, time point T1
N=694 T2
N=383 T3
N=259 T1
N=237 T2
N=149
Gender2
Female 77.4 (525) 79.1 (299) 80.5 (207) 87.2 (198) 87.7(128) Male 22.6 (153) 20.9 (79) 19.5 (50) 12.8 (29) 12.3 (18) Age (years) Under 502 55.6 (375) 44.1(165) 43.5 (110) 64.3 (144) 60.3 (85) 50 and over2 44.4 (300) 55.9 (209) 56.5 (143) 35.7 (80) 39.7 (56) 18-24 7.4 (50) 2.7 (10) 3.2 (8) 5.8 (13) 4.3 (6) 25-34 17.5 (118) 14.2 (53) 14.6 (37) 22.8 (51) 20.6 (29) 35-44 23.0 (155) 19.8 (74) 19.0 (48) 26.3 (59) 24.1 (34) 45-54 16.0 (108) 15.5 (58) 15.8 (40) 16.5 (37) 18.4 (26) 55-64 15.0 (101) 18.5 (69) 18.6 (47) 12.5 (28) 13.5 (19) 65-74 17.5 (118) 24.6 (92) 24.1 (61) 13.4 (30) 15.6 (22) 75+ 3.70 (25) 4.8 (18) 4.7 (12) 2.7 (6) 3.5 (5) Mean age years (SD) 48.0(16.1) 52(15.7) 52.0(15.9) 46(15.1) 47(15.2) Aboriginal or Torres Strait Islander 1.8 (12) 1.8 (7) 2.7 (7) 0.9 (2) 1.4 (2) Speaks a language other than English at home 7.8 (53) 6.9 (26) 7.4 (19) 5.3 (12) 3.4 (5) Locality Ipswich 82.0 (555) 84.7(320) 83.3 (214) 78.8 (178) 79.4 (116) Other Queensland localities 17.7 (120) 15.3 (58) 16.3 (42) 21.2 (48) 20.5 (30) NSW 0.3 (2) 0.0 (0) 0.4 (1) 0.0 (0) 0.0 (0) Highest level of education attained
High school, year 12 or less 47.8 (321) 49.3(185) 49.4 (126) 45.8 (104) 47.3 (69) Technical and Further Education, apprenticeship, diploma or certificate 22.2 (149) 20.8 (78) 21.6 (55) 22.9 (52) 19.9 (29) Tertiary, bachelor degree or higher 28.0 (188) 28.5(107) 27.4 (70) 29.1 (66) 30.1 (44) Other 2.0 (13) 1.3 (5) 1.6 (4) 2.2 (5) 2.7 (4) Employment2 Full-time 26.4 (176) 23.6 (88) 26.2 (67) 34.7 (79) 31.3 (46) Part-time or casual 18.6 (124) 16.6 (62) 18.4 (47) 14.5 (33) 17.0 (25) Retired 23.8 (159) 31.6(118) 30.5 (78) 21.5 (49) 23.8 (35) Home duties/ carer 14.4 (96) 15.3 (57) 13.3 (34) 18.4 (42) 17.7 (26) Not working (permanently ill or unable to work, unemployed) 9.9 (66) 7.0 (26) 4.3 (11) 8.8 (20) 7.5 (11) Student (full-time and part-time) 3.1 (21) 1.9 (7) 1.9 (5) 1.3 (3) 2.0 (3) Other 3.9 (26) 4.0 (15) 5.47(14) 0.9 (2) 0.7 (1) Household yearly income
$1-$6,000 2.5 (15) 1.8 (6) 2.6 (6) 2.0 (4) 1.5 (2) $6,001-$13,000 5.7 (34) 5.7 (19) 5.3 (12) 5.0 (10) 5.3 (7) $13,001-$20,000 11.9(71) 12.9 (43) 14.1 (32) 9.5 (19) 9.9 (13) $20,001-$30,000 14.8(88) 17.7 (59) 17.2 (39) 9.5 (19) 9.9 (13) $30,001-$50,000 15.4(92) 14.7 (49) 14.5 (33) 12.5 (25) 12.2 (16) $50,001-$100,000 30.0(179) 29.7 (99) 26.9 (61) 35.5 (71) 36.6 (48) $100,001-$150,000 13.6(81) 11.1 (37) 12.3 (28) 18.5 (37) 16.8 (22) >$150,000 6.0(36) 6.3 (21) 7.0 (16) 7.5 (15) 7.6 (10)
118
1 Sample size for different variables might vary from total sample size because of missing responses and the rounding of weighted frequencies. 2Significant difference between groups (P<0.05) at baseline as tested with chi-squared analysis. 3Excludes two participants living in institutional facilities. SD = standard deviation.
Intervention % (n) Control % (n) Demographic, time point T1
N=694 T2
N=383 T3
N=259 T1
N=237 T2
N=149 Household Characteristics Couple, with young children (0-17 years old) living at home 24.7 (169) 23.2 (88) 20.0(51) 32.1 (76) 30.2 (45) Couple, with adult children (18 years and over) living at home 12.5 (86) 10.8 (41) 11.0(28) 10.1 (24) 10.7 (16) Couple, without children living at home 32.9 (226) 35.3(134) 36.0(92) 24.5 (58) 27.5 (41) One parent family with children living at home 7.0 (48) 3.7 (14) 4.3(11) 8.9 (21) 6.7 (10) Live Alone 16.0 (110) 21.6 (82) 22.7(58) 17.7 (42) 20.1(30) Other 6.9 (47) 5.5 (21) 6.2(16) 6.8 (16) 4.7 (7) Mean household size (SD)3 2.8 (1.5) 2.6 (1.3) 2.5 (1.3) 3.0 (1.6) 2.9 (1.6) Median household size
(50% percentile)3 2 2 2 3 2 Shared programme experience Attending JMoF with others or group (Yes)
58.0 (393) 56.2(212) 53.5(137) 67.0 (152) 66.7 (98)
If yes, who they attending with….
Friends 32.2 (123) 31.0 (64) 25.0 (33) 45.7 (69) 49.5 (48) Family 40.3 (154) 36.9 (76) 40.1 (53) 39.1 (59) 34.0 (33) With a carer 1.6 (6) 1.0 (2) 1.5 (2) 3.3 (5) 3.1 (3) With a community group 24.4 (93) 28.7 (59) 33.3 (44) 7.3 (11) 8.3 (8) Other 1.6 (6) 2.4 (5) 0.0 (0) 4.6 (7) 5.2 (5)
119
Table 8: Prevalence of health indicators, population 18 years and over
Health indicator Ipswich, population
prevalence1 % (95% CI)
JMoF baseline sample, population
prevalence % (95% CI)
Body Mass Index (BMI) 2 Underweight (BMI <18.5) 3.4 (1.8-6.3) 2.5 (1.4-3.5) Healthy weight (BMI 18.5 to <25) 37.4 (33.1-42.0) 28.5 (25.5-31.6) Overweight (BMI 25 to <30) 34.0 (30.1-38.2) 30.3 (27.2-33.4) Obese (BMI 25+) 25.2 (21.8-28.8) 38.7 (35.4-42.0) Overweight/obese 59.2 (54.6-63.6) 69.0 (65.9-72.1) Unhealthy weight 62.6 (58.0-66.9) 71.5 (68.4–74.5) Self-rated health (rated excellent, very good or good)
80.5 (75.4-84.7) 69.4 (67.2-71.6)
Fruit and vegetable consumption Mean daily vegetable intake (serves) 2.2 (2.1-2.3) 2.5 (2.4-2.6) Mean daily fruit intake (serves) 1.5 (1.4-1.6) 1.6 (1.6-1.7) Persons aged 18 years and older (n= 941 for Ipswich, n=931 for JMoF) 1Data take from 2011-2012 Queensland Health self-reported health status survey (212). 2BMI cut-off points sourced from World Health Organization (176).
120
6.4 Programme outcomes
6.4.1 Cooking confidence
Cooking confidence was a primary outcome of the evaluation and results have
been presented in Flego et al (2014) (of which the candidate was a major
contributor and second author) (10). Intervention participants overall showed
significant positive improvements between T1 and T2 in their confidence to cook
from basic ingredients (0.81, SD 0.05, P<0.001), follow a simple recipe (0.53, SD
0.04, P<0.001), prepare and cook new foods and recipes (0.77, SD 0.05,
P<0.001), cook something that will turn out well (0.72, SD 0.04, P<0.001) and to
taste new foods (0.54, SD 0.05, P<0.001) (Table 9). There was no significant
change between T1 and T2 in all confidence outcomes amongst control group
participants. All cooking confidence scores showed a statistically significant
group by time interaction effect (P<0.001) indicating that the intervention group
significantly increased their cooking confidence compared to the control group
(Table 9). The effect size between groups and overtime for all cooking confidence
measures ranged from medium to large (0.42-0.85).
6.4.2 Cooking and healthy eating knowledge attitudes, beliefs and skills
Intervention participants significantly improved their healthy eating skills,
knowledge, attitudes and beliefs between T1 and T2 (Table 9). There were no
significant changes reported in the same measures in the control group. While
effect size was small between groups there were statistically significant
121
differences between the intervention and control groups and over time in terms of
preparing a meal from scratch in 30 minutes (P<0.001), beliefs around the ease of
changing eating habits (P=0.02), vegetables being tasty (P=0.01) and eating
enough fruit and vegetables (P<0.001). Participant attitudes around lifestyle
preventing a healthy diet did not show a significant difference between control
and interventions groups over time (P=0.07), however the mean score was high
for both groups both before and after the programme. In the intervention group,
analysis over the three time points showed that all attitudes and knowledge around
cooking and healthy eating were significantly sustained from baseline (T1) to six
months post-programme (T3) (Table 10).
122
6.4.3 Cooking enjoyment and satisfaction
123
Figure 11: Correctly answered nutrition knowledge by group and over time
Table 9: Predicted mean scores in cooking confidence, knowledge, attitudes beliefs, skills, enjoyment and satisfaction outcomes by group (T1 and T2)¹
Intervention Control Interaction effect, group by time P-value
Outcome measure baseline (T1) mean (SE)
intervention completion (T2) mean (SE)
change from baseline(T2-T1) mean (SE) P-value
baseline (T1) mean (SE)
wait-list completion (T2) mean (SE)
change from baseline(T2-T1) mean (SE) P-value
Difference between group (effect size)7
Cooking confidence Confidence to cook from basic ingredients4 3.56 (0.04) 4.36 (0.05) 0.81 (0.05) P<0.001 3.69 (0.07) 3.72 (0.08) 0.03 (0.08) P=0.70 P<0.001 0.78 Confidence to follow a simple recipe4 4.00 (0.04) 4.53 (0.05) 0.53 (0.04) P<0.001 4.11 (0.06) 4.06 (0.07) -0.06 (0.07) P=0.40 P<0.001 0.67 Confidence in preparing and cook new foods and recipes4
3.35 (0.04) 4.13 (0.05) 0.77 (0.05) P<0.001 3.45 (0.07) 3.55 (0.08) 0.10 (0.08) P=0.22 P<0.001 0.65
Confidence that what one cooks will turn out well4
3.21 (0.04) 3.93 (0.05) 0.72 (0.04) P<0.001 3.28 (0.06) 3.35 (0.07) 0.07 (0.07) P=0.30 P<0.001 0.70
Confidence to taste new foods never eaten before4
3.47 (0.04) 4.01 (0.05) 0.54 (0.05) P<0.001 3.41 (0.07) 3.51 (0.09) 0.09 (0.08) P=0.25 P<0.001 0.42
Combined confidence score5 17.59 (0.02) 20.95 (0.2) 3.36 (0.18) P<0.001 17.94 (0.03) 18.17 (0.03) 0.23 (0.28) P=0.41 P<0.001 0.85 Cooking and healthy eating knowledge, attitudes, beliefs and skills
I can put together a healthy meal from scratch in 30 minutes6
2.85 (0.031) 3.30 (0.04) 0.45 (0.04) P<0.001 2.85 (0.05) 2.89 (0.06) 0.03 (0.06) P=0.61 P < 0.001 -0.49
I find it easy to change my eating habits6 2.52 (0.03) 2.71 (0.04) 0.19 (0.04) P<0.001 2.52 (0.05) 2.53 (0.06) 0.01 (0.06) P=0.82 P = 0.02 -0.18 Vegetables can be tasty foods6 3.54 (0.02) 3.69 (0.03) 0.15 (0.03) P<0.001 3.53 (0.04) 3.51 (0.05) -0.02 (0.05) P=0.74 P = 0.01 -0.26 I eat enough fruit and vegetables6 2.66 (0.03) 3.00 (0.04) 0.34 (0.04) P<0.001 2.66 (0.06) 2.68 (0.07) 0.02 (0.06) P=0.71 P < 0.001 -0.37 My lifestyle does not prevent me eating a healthy diet6a 3.11 (0.03) 3.33 (0.04) 0.22 (0.04) P<0.001 3.04 (0.05) 3.12 (0.06) 0.08 (0.06) P=0.17 P = 0.07 -0.19 Cooking enjoyment and satisfaction I enjoy cooking6 3.05 (0.03) 3.33 (0.04) 0.28 (0.03) P<0.001 3.12 (0.05) 3.17 (0.06) 0.06 (0.05) P=0.28 P = 0.001 -0.31 I get a lot of satisfaction from cooking my meals6 2.96 (0.03) 3.31 (0.04) 0.35 (0.03) P<0.001 3.02 (0.05) 3.05 (0.06) 0.03 (0.05) P=0.60 P < 0.001 -0.44 I enjoy cooking for others6 3.01 (0.03) 3.27 (0.04) 0.26 (0.03) P<0.001 3.09 (0.06) 3.16 (0.07) 0.07 (0.06) P=0.22 P = 0.004 -0.27 I enjoy eating a meal with others6 3.51 (0.02) 3.60 (0.03) 0.09 (0.03) P = 0.01 3.47 (0.39) 3.55 (0.05) 0.07 (0.05) P=0.16 P = 0.81 -0.03 Outcomes within each group and over time were determined by a mixed linear model for repeated measures using all available data at each time point. All means and Standard Errors (SE) have been rounded to 2 decimal points. Baseline values were not significantly different between groups (independent t tests P<0.05). A significant group by time interaction effect denotes that the response over time differed between groups. 4Scale values are 1-5 (where 1= not at all confident and 5 = extremely confident). 5 The combined confidence score is equal to the sum total of all other confidence scores (scores ≥20 = confident). 6Mean predicted score indicating level of agreement with statements from a Likert Scale (1=strongly disagree, 2=somewhat disagree, 3=some agree, 4=strongly agree), a Score assignment was reversed. 7Cohen's d effect size, general guideline: small (0.2), medium (0.5) and large (0.8).
125
Table 10: Predicted mean scores in cooking confidence, knowledge, attitudes beliefs, skills, enjoyment and satisfaction outcomes for the intervention group (T1, T2, T3)¹
Outcome measure Intervention group Interaction effect over time2 P-value
Baseline
(T1) Mean (SE)
Intervention completion (T2) mean SE)
6 months follow-up (T3) mean (SE)
Change from T2-T1 mean (SE) P-value
Change from T3-T1 mean (SE) P-value
Change from T3-T2 mean (SE) P-value
Effect size (T2-T1)6
Effect size (T3-T1) 6
Cooking confidence Confidence to cook from basic ingredients3 3.56 (0.04) 4.37 (0.05) 4.43 (0.06) 0.81 (0.05) P<0.001 0.87(0.06) P<0.001 0.07(0.06) P=0.28 P<0.001 0.83 1.15 Confidence to follow a simple recipe3 4.00 (0.04) 4.53 (0.04) 4.61 (0.05) 0.53 (0.04) P<0.001 0.61(0.05) P<0.001 0.08(0.05) P=0.13 P<0.001 0.83 0.95 Confidence in preparing and cook new foods and recipes3
3.35 (0.04) 4.13 (0.05) 4.17 (0.06) 0.78 (0.05) P<0.001 0.82(0.06) P<0.001 0.05(0.06) P=0.44 P<0.001 1.09 1.16
Confidence that what one cooks will turn out well3 3.21 (0.04) 3.93 (0.05) 3.94 (0.05) 0.72 (0.04) P<0.001 0.73(0.05) P<0.001 0.01(0.06) P=0.80 P<0.001 1.07 1.12 Confidence to taste new foods never eaten before3 3.47 (0.04) 4.01 (0.05) 3.99 (0.06) 0.53 (0.05) P<0.001 0.52(0.06) P<0.001 -0.02(0.06) P=0.75 P<0.001 0.74 0.74 Combined confidence score4 17.59 (0.16) 20.95 (0.19) 21.15 (0.22) 3.36 (0.18) P<0.001 3.56(0.21) P<0.001 0.20(0.22) P=0.36 P<0.001 1.24 1.32 Cooking and healthy eating knowledge, attitudes, beliefs and skills
I can put together a healthy meal from scratch in 30 minutes5
2.85 (0.03) 3.29 (0.04) 3.31 (0.05) 0.44 (0.04) P<0.001 0.46 (0.05) P<0.001 0.02 (0.06) P=0.67 P < 0.001 0.75 0.72
I find it easy to change my eating habits5 2.52 (0.03) 2.71 (0.04) 2.70 (0.04) 0.17 (0.04) P<0.001 0.18 (0.05) P<0.001 0.00 (0.05) P=0.94 P < 0.001 0.29 0.37 Vegetables can be tasty foods5 3.54 (0.02) 3.69 (0.03) 3.69 (0.04) 0.15 (0.03) P=0.001 0.15 (0.04) P<0.001 0.00 (0.04) P=0.97 P < 0.001 0.28 0.33 I eat enough fruit and vegetables5 2.66 (0.03) 3.00 (0.04) 3.05 (0.05) 0.34 (0.04) P<0.001 0.39 (0.05) P=0.001 0.06 (0.05) P=0.26 P < 0.001 0.55 0.61 My lifestyle does not Prevent me eating a healthy diet5a
3.11 (0.03) 3.32 (0.04) 3.29 (0.05) 0.21 (0.04) P<0.001 0.18 (0.05) P<0.001 -0.03 (0.05) P=0.55 P < 0.001 0.38 0.35
Cooking enjoyment and satisfaction I enjoy cooking5 3.05 (0.03) 3.32 (0.04) 3.28 (0.04) 0.27 (0.03) P<0.001 0.23 (0.04) P<0.001 -0.04 (0.04) P=0.31 P < 0.001 0.58 0.46 I get a lot of satisfaction from cooking my meals5 2.96 (0.03) 3.31 (0.04) 3.29 (0.04) 0.35 (0.04) P<0.001 0.33 (0.04) P<0.001 -0.02 (0.04) P=0.72 P < 0.001 0.71 0.58 I enjoy cooking for others5 3.01 (0.03) 3.26 (0.04) 3.18 (0.05) 0.25 (0.04) P<0.001 0.18 (0.04) P<0.001 -0.08 (0.05) P=0.11 P < 0.001 0.53 0.29 I enjoy eating a meal with others5 3.51 (0.02) 3.60 (0.03) 3.61 (0.03) 0.09 (0.03) P = 0.01 0.10 (0.04) P = 0.01 0.01 (0.04) P=0.77 P = 0.003 0.18 0.15 Outcomes at each time point were determined by a mixed linear model for repeated measures using all available data at each time point. All means and Standard Errors (SE) have been rounded to 2 decimal points. 2A significant over time interaction effect denotes that the response over time differed between groups. 3Scale values are 1-5 (where 1= not at all confident and 5 = extremely confident).4 The combined confidence score is equal to the sum total of all other confidence scores (scores ≥ 20 = confident).5 Mean predicted score indicating level of agreement with statement from a Likert Scale (1=strongly disagree, 2=somewhat disagree, 3=some agree, 4=strongly agree), a Score assignment was reversed. 6 Cohen's d effect size, general guideline: small (0.2), medium (0.5) and large (0.8).
126
6.4.4 Healthy diets
127
Figure 12: Mean daily fruit and vegetable consumption - comparison of JMoF intervention participants and Queensland (QLD) state-wide 2012 data*
serves of fruit per day compared to the JMoF population’s mean intake at T1 of
128
Figure 13: Take-away food consumption comparison between intervention participants and Queensland*
129
Table 11 indicates that cooking behaviours positively increased after programme
attendance. Intervention participants significantly increased their frequency of
cooking the main meal from basic ingredients, from 4.05 times per week at T1 to
4.66 at T2 (P<0.001), whereas the control group did not significantly change over
time (P=0.95). The overall group by time interaction effect for cooking the main
meal from basic ingredients was statistically significant (P<0.001). The
consumption of ready-made meals (such as frozen dinners or pre-prepared pizzas)
did not show a significant difference between groups and over time (P=0.06). The
frequency of consuming a salad or vegetables with the main meal significantly
increased in the intervention group over time, but not in the control group. The
intervention group consumed an average of 0.39 more salads or vegetables per
week as part of their main meal, with total weekly serves increasing from 4.67 at
T1 to 5.06 at T2. The overall group by time interaction effect of consuming a
salad or vegetables with the main meal was significant (P=0.01).
6.4.5 Affordability of a healthy diet
Three food purchasing behaviours were measured (table 11). Total food and drink
expenditure did not significantly differ between groups or over time. On the other
hand, total fruit and vegetable expenditure did significantly increase in the
intervention group by a mean AUD 2.50 per week between T1 and T2 (P<0.001),
but not in the control group. The overall group by time interaction effect was not
130
significantly different for fruit and vegetable expenditure (P=0.10). There was a
statistically significant decrease in total weekly take-away food spending in the
intervention group (P<0.001), but not in the control group, between T1 and T2
(P=0.07). Weekly take-away food expenditure showed a significant difference
between intervention and control groups and over time (P=0.004) (Table 11).
6.4.6 Social connectedness
131
6.4.7 Health and emotional well-being
The last component of the programme logic model measured quantitatively was
self-esteem. In addition to this, perceived general health and BMI were analysed.
The JMoF programme did not impact on participants' BMI; there were no overall
significant interactions between groups and over time (Table 11). There was a
statistically significant increase in global self-esteem in the intervention group
(P<0.001) but not in the control group. Overall, there was however a significant
group by time interaction for general health (P<0.001) and self-esteem (P=0.002).
‘ ’ ‘ ’
6.5 Adjusted analyses of outcomes
Each outcome analysed was adjusted to account for any differences between the
intervention and control groups at baseline due to the non-randomised nature of
132
the evaluation design. Each variable was adjusted for age, gender and
employment, with each covariant analysed separately and then all together. These
results are presented in Appendix 9. There were small differences in the predicted
group by time mean values; however, pairwise comparisons remained similar to
the unadjusted analysis. Appendix 9 shows similar patterns of change over time in
the pairwise comparisons. Similar patterns mean that inferences between the
adjusted and unadjusted result models remain the same, therefore the results have
not been presented in this chapter.
Table 11: Predicted mean scores in behaviours and health and well-being by group (T1 and T2)¹
Intervention Control Interaction effect, group by time P-value
Difference between group (effect size)10
Outcome measure Baseline (T1) mean (SE)
Intervention completion (T2) mean (SE)
Change from T2-T1 mean (S.E) P-value
Baseline (T1) mean (SE)
Wait-list completion (T2) mean (SE)
Change from T2-T1 mean (SE) P-value
Consumption behaviours Daily vegetable consumption (serves per day) 2.46 (0.51) 2.97 (0.06) 0.52 (0.06) P<0.001 2.49 (0.09) 2.59 (0.10) 0.10 (0.10) P=0.30 P<0.001 0.29 Daily fruit consumption (serves per day) 1.65 (0.04) 1.93 (0.05) 0.28 (0.05) P<0.001 1.61 (0.07) 1.71 (0.08) 0.10 (0.08) P=0.20 P=0.06 0.16 Take-away consumption4 0.98 (0.04) 0.77 (0.04) -0.21(0.04) P<0.001 0.94 (0.06) 0.96 (0.07) 0.03 (0.06) P=0.62 P=0.001 -0.26 Cooking behaviours Cooking the main meal from basic ingredients4 4.05 (0.08) 4.66 (0.09) 0.61 (0.09) P<0.001 4.16 (0.14) 4.17 (0.16) 0.01 (0.15) P= 0.95 P<0.001 0.29 Consumption of ready- made meals at home4 1.06 (0.05) 0.95 (0.06) -0.11 (0.06) P=0.06 1.11 (0.08) 1.21 (0.10) 0.10 (0.10) P=0.30 P=0.06 -0.09 Consumption of vegetables with the main meal4 4.67(0.07) 5.06 (0.09) 0.39 (0.08) P<0.001 4.76 (0.12) 4.75 (0.14) 0.01 (0.14) P=0.94 P=0.01 0.25 Food purchasing behaviours and attitudes Total weekly food and drink expenditure (AUD)5 137.16 (2.72) 135.60 (3.15) -1.56 (2.46) P = 0.53 147.34 (4.68) 151.68 (5.20) 4.33 (3.96) P=0.27 P = 0.21 -0.10 Total weekly fruit and veg expenditure (AUD)5 20.77 (0.61) 23.28 (0.73) 2.50 (0.63) P<0.001 21.70 (1.06) 22.24 (1.20) 0.53 (1.01) P=0.60 P = 0.10 0.18 Total weekly take away expenditure (AUD)5 13.17 (0.59) 9.86 (0.69) -3.31 (0.55) P<0.001 12.395 (1.01) 12.05 (1.13) -0.34 (0.87) P=0.70 P = 0.004 -0.24 I can prepare a meal from basics that is low in Price6 2.99 (0.03) 3.41 (0.04) 0.41 (0.04) P<0.001 3.00 (0.05) 2.97 (0.06) -0.02 (0.06) P=0.71 P < 0.001 -0.55 Buying more fruit/vegetables would not be difficult on my budget6a
2.85 (0.03) 2.93 (0.04) 0.08 (0.04) P = 0.06 2.85 (0.06) 2.89 (0.07) 0.04 (0.07) P=0.59 P = 0.60 -0.03
Fruit and vegetables are cheaper when they are in season6
3.42 (0.02) 3.62 (0.03) 0.21 (0.03) P<0.001 3.43 (0.04) 3.50 (0.05) 0.07 (0.06) P=0.21 P = 0.04 -0.20
Social eating Frequency of eating together at home with others7 3.94 (0.07) 4.20 (0.08) 0.24 (0.07) P<0.001 3.97 (0.11) 4.02 (0.13) 0.06 (0.11) P=0.61 P = 0.13 0.15 Frequency of eating dinner in front of the television7 2.69 (0.08) 2.50 (0.09) -0.19 (0.07) P = 0.01 2.51 (0.14) 2.52 (0.15) 0.00 (0.11) P=0.99 P = 0.17 -0.12 Frequency of eating dinner at a dinner table7 3.12 (0.08) 3.40 (0.09) 0.29 (0.06) P<0.001 3.11 (0.13) 3.09 (0.14) -0.02(0.10) P=0.86 P = 0.01 0.23 Health and emotional well-being Global self-esteem score8 20.88 (0.22) 22.60 (0.25) 1.73 (0.20) P <0.001 20.46 (0.37) 21.02 (0.42) 0.56 (0.32) P=0.09 P = 0.002 0.26 General health9 2.77 (0.04) 3.11 (0.04) 0.34 (0.04) P <0.001 2.80 (0.06) 2.86 (0.07) 0.06 (0.06) P=0.34 P < 0.001 0.37 Body Mass Index (BMI) 28.86 (0.27) 28.78 (0.28) -0.09 (0.13) P = 0.49 29.71 (0.46) 29.70 (0.47) -0.02 (0.20) P=0.94 P = 0.76 -0.03 ¹Outcomes at each time point were determined by a mixed linear model for repeated measures using all available data at each time point. All means and Standard Errors (SE) have been rounded to 2 decimal points. ²Baseline values were not significantly different between groups (independent t tests P<0.05). ³A significant group x time interaction effect denotes that the response over time differed between groups 4times per week 5Expenditure data was collected in Australian dollars (AUD) on a 7-Point scale which was analysed by its midpoints. 6Mean predicted score indicating level of agreement with statement from a Likert Scale (1=strongly disagree, 2=somewhat disagree, 3=some agree, 4=strongly agree), aScore assignment was reversed. 7Mean frequency for a typical week was collected on a 6 or 7-Point scale which was analysed by its midpoint, with the maximum category being five or more times Per week. 8Rosenberg's global self-esteem score (Low self-esteem= 0-14, Normal self-esteem=15-25 and High self-esteem =16-30). 9Perceived general health (Poor=1, fair=2, good=3, very good=4, excellent=5). 10 Cohen's d effect size, general guideline: small (0.2), medium (0.5) and large (0.8)..
134
Table 12: Predicted mean scores in behaviours and health and well-being for the intervention group (T1, T2, T3)¹
Intervention group Interaction effect over time2 P value
Difference over time T2-T1 (effect size)9
Difference over time T3-T1 (effect size)9
Outcome measure
Baseline (T1) mean (SE)
Intervention completion (T2) mean SE)
6 months follow-up (T3) mean (SE)
Change from T2-T1 mean (S.E) P-value
Change from T3-T1 mean (S.E) P-value
Change from T3-T2 mean (S.E) P-value
Consumption behaviours Daily vegetable consumption (serves per day) 2.46 (0.05) 2.97 (0.06) 3.05 (0.07) 0.51(0.06) P<0.001 0.60(0.07) P<0.001 0.08(0.08) P=0.27 P<0.001 0.54 0.53 Daily fruit consumption (serves per day) 1.65 (0.04) 1.93 (0.05) 2.05 (0.06) 0.27(0.05) P<0.001 0.40(0.06) P<0.001 0.12(0.06) P=0.06 P<0.001 0.38 0.50 Take-away consumption3 0.98 (0.04) 0.76 (0.04) 0.73 (0.05) -0.23(0.04) P<0.001 -0.25(0.04) P<0.001 -0.02(0.05) P=0.61 P<0.001 -0.30 -0.35 Cooking behaviours Cooking the main meal from basic ingredients3 4.05 (0.08) 4.65 (0.10) 4.88 (0.11) 0.60(0.09) P<0.001 0.84 (0.10) P<0.001 0.24(0.11) P=0.03 P<0.001 0.39 0.64 Consumption of ready- made meals at home3 1.06 (0.05) 0.93 (0.06) 0.80 (0.07) -0.13(0.06) P=0.04 -0.26(0.07) P<0.001 -0.13(0.08) P=0.09 P<0.001 -0.02 -0.30 Consumption of vegetables with the main meal3 4.67 (0.07) 5.05 (0.09) 5.31 (0.10) 0.38(0.09) P<0.0011 0.64(0.09) P<0.001 0.25(0.10) P=0.02 P<0.001 0.29 0.51 Food purchasing behaviours and attitudes Total weekly food and drink expenditure (AUD)4 137.13 (2.65) 135.21 (3.08) 137.28 (3.42) -1.93(2.48) P = 0.44 0.15 (2.90) P = 0.96 2.08 (3.08) P=0.50 P = 0.70 -0.01 0.02 Total weekly fruit and veg expenditure (AUD)4 20.77 (0.62) 23.25 (0.74) 23.64 (0.83) 2.48 (0.65) P<0.001 2.86 (0.76) P<0.001 0.39 (0.81) P=0.63 P < 0.001 0.31 0.37 Total weekly take away expenditure (AUD)4 13.19 (0.59) 9.85 (0.68) 9.14 (0.76) -3.34 (0.54) P<0.001 -4.05 (0.63) P<0.001 -0.71 (0.68) P=0.29 P < 0.001 -0.35 -0.44 I can prepare a meal from basics that is low in Price5
2.99 (0.03) 3.41 (0.04) 3.42 (0.04) 0.42 (0.04) P<0.001 0.43 (0.05) P<0.001 0.01 (0.05) P=0.79 P < 0.001 0.68 0.69
Buying more fruit/vegetables would not be difficult on my budget5a
2.85 (0.03) 2.93 (0.04) 2.97 (0.05) 0.08 (0.05) P = 0.09 0.11 (0.05) P = 0.03 0.04 (0.06) P=0.52 P = 0.06 0.11 0.18
Fruit and vegetables are cheaper when they are in season5
3.42 (0.02) 3.62 (0.03) 3.66 (0.04) 0.21 (0.04) P<0.001 0.24 (0.04) P<0.001 0.04 (0.04) P=0.41 P < 0.001 0.35 0.42
Social eating Frequency of eating together at home with others6
3.92 (0.07) 4.17 (0.08) 4.20 (0.09) 0.25 (0.07) P<0.001 0.28 (0.09) P<0.001 0.04 (0.09) P=0.69 P < 0.001 0.26 0.38
Frequency of eating dinner in front of the television6
2.69 (0.08) 2.50 (0.09) 2.46 (0.10) -0.19 (0.07) P = 0.01 -0.23 (0.08) P = 0.01 -0.04 (0.09) P=0.66 P = 0.01 -0.19 -0.18
Frequency of eating dinner at a dinner table6 3.12 (0.08) 3.40 (0.09) 3.37 (0.10) 0.28 (0.65) P<0.001 0.25 (0.08) P=0.001 -0.02 (0.08) P=0.76 P < 0.001 0.33 0.25 Health and emotional well-being Global self-esteem score7 20.88 (0.22) 22.61 (0.25) 22.92 (0.28) 1.73 (0.21) P<0.001 2.04 (0.25) P<0.001 0.31 (0.26) P=0.24 P < 0.001 0.59 0.62 General Health8 2.77 (0.04) 3.11 (0.05) 3.24 (0.05) 0.34 (0.04) P<0.001 0.47 (0.05) P<0.001 0.13 (0.05) P=0.01 P < 0.001 0.59 0.75 Body Mass Index (BMI) 28.86 (0.27) 28.79 (0.28) 28.94 (0.29) -0.07 (0.14) P = 0.61 0.08 (0.16) P = 0.65 0.15 (0.17) P=0.39 P = 0.68 -0.05 0.00 ¹Outcomes at each time point were determined by a mixed linear model for repeated measures using all available data at each time point. All means and Standard Errors (SE) have been rounded to 2 decimal points. 2A significant over time interaction effect denotes that the response over time differed between groups, 3times per week. 4Expenditure data was collected in Australian dollars (AUD) on a 7-Point scale which was analyse by its midpoints. 5Mean Predicted score indicating level of agreement with statement from a Likert Scale (1=strongly disagree, 2=somewhat disagree, 3=some agree, 4=strongly agree), aScore assignment was reversed. 6Mean frequency for a typical week was collected on a 6 or 7-Point scale which was analyse by its midpoint, with the maximum category being five or more times per week. 7Rosenberg's global self-esteem score (Low self-esteem= 0-14, Normal self-esteem=15-25 and High self-esteem =16-30). 8Perceived general health (Poor=1, fair=2, good=3, very good=4, excellent=5) .9Cohen's d effect size, general guideline: small (0.2), medium (0.5) and large (0.8).
6.6 Supplementary analysis
In addition to the primary analysis, a supplementary analysis was run for individual
variables as a precautionary check to see if differences in the statistical approaches
impacted on results. Ordinal variables measured on a 4 or 5-point Likert scale were
analysed using ordinal logistic regressions to produce an odds ratio. The ordinal
logistic regression was compared to the predicted mean results from the primary
multilevel mixed model analysis (Table 13). Comparison of P-values for each
analysis showed that there were no differences in the overall group by time
interaction effect between the different analysis techniques used.
analysed as a ‘confident’ or ‘not confident’
136
Table 13: Supplementary analysis: ordinal logistic regression and multilevel mixed model analysis
Ordinal variables Ordinal logistic regression1
Multilevel mixed model analysis3
OLR odds ratio (SE)
Confidence intervals
Z score P-value Interaction effect group by time P-value2
Cooking and healthy eating knowledge, attitudes, beliefs and skills
I can put together a healthy meal from scratch in 30 minutes4
2.68 (0.49) 1.87 - 3.82 5.41 P< 0.001 P< 0.001
I find it easy to change my eating habits4
1.62 (0.31) 1.12 - 2.37 2.53 P= 0.010 P= 0.020
Vegetables can be tasty foods4 1.93 (0.38) 1.31 - 2.85 3.34 P= 0.001 P= 0.010 I eat enough fruit and vegetables4 2.06 (0.34) 1.48 - 2.86 4.31 P< 0.001 P< 0.001 My lifestyle does not prevent me eating a healthy diet4a
1.33 (0.24) 0.93 - 1.89 1.57 P= 0.120 P= 0.07
Cooking enjoyment and satisfaction I enjoy cooking4 1.75 (0.26) 1.31 - 2.34 3.80 P< 0.001 P= 0.001 I get a lot of satisfaction from cooking my meals4
2.29 (0.35) 1.69 - 3.10 5.36 P< 0.001 P< 0.001
I enjoy cooking for others4 1.56 (0.23) 1.16 - 2.01 2.98 P= 0.003 P= 0.004 I enjoy eating a meal with others4 1.05 (0.21) 0.70 - 1.56 0.22 P= 0.830 P= 0.810 Food purchasing behaviours and attitudes
I can prepare a meal from basics that is low in price4
3.01 (0.53) 2.12 - 4.26 6.20 P< 0.001 P<0.001
Buying more fruit/vegetables would not be difficult on my budget4a
1.13 (0.21) 0.79 - 1.62 0.66 P= 0.512 P=0.600
Fruit and vegetables are cheaper when they are in season4
1.54 (0.30) 1.04 - 2.26 2.18 P= 0.030 P=0.040
Cooking confidence Confidence to cook from basic ingredients5
3.73 (0.59) 2.74 - 5.08 8.37 P< 0.001 P<0.001
Confidence to follow a simple recipe5
3.21(0.55) 2.29 - 4.51 6.76 P< 0.001 P<0.001
Confidence in preparing and cooking new foods and recipes5
3.61(0.6) 2.60 - 5.01 7.67 P< 0.001 P<0.001
Confidence that what one cooks will turn out well5
3.9(0.65) 2.82 - 5.41 8.19 P< 0.001 P<0.001
Confidence to taste new foods never eaten before5
2.16(0.35) 1.57 - 2.95 4.79 P< 0.001 P<0.001
OLR = Ordinal Logistic Regression. SE = Standard error. 1 Ordinal logistic regression is the supplementary analysis. Odds of a positive change occurring between T1 and T2 in the intervention group and control group for all ordinal variables (control =0 and intervention = 1). 2A significant over time interaction effect denotes that the response over time differed between groups. 3 Multilevel mixed model analysis was the original analysis technique used. P-values are presented to compare. 4Four-point Likert scale used to measure level of agreement with statement (1=strongly disagree, 2=somewhat disagree, 3=some agree, 4=strongly agree). a Score assignment was reversed.5Five point Likert scale used to measure confidence level (1= not at all confident through to 5 = extremely confident. ≥4 = confident).
137
Table 14: Supplementary analysis: secondary logistic model (generalised estimated equation) for cooking confidence1
Ordinal variables Generalised estimated equation Multilevel mixed model analysis
OLR odds ratio (SE)
Confidence intervals
Z score P-value1 Interaction effect group by time P-value1
Confidence variables Confidence to cook from basic ingredients 3.75 (0.84) 2.41 – 5.81 5.90 P<0.001 P<0.001
Confidence to follow a simple recipe 3.66 (0.92) 2.24 – 5.98 5.18 P<0.001 P<0.001
Confidence in preparing and cooking new foods and recipes 3.56 (0.74) 2.37 – 5.36 6.09 P<0.001 P<0.001
Confidence that what one cooks will turn out well 4.31 (0.89) 2.87 – 6.46 7.06 P<0.001 P<0.001
Confidence to taste new foods never eaten before 2.21 (0.44) 1.50 – 3.27 3.99 P<0.001 P<0.001
Combined confidence score 4.71 (1.23) 2.82 - 7.876 5.92 P<0.001 P<0.001
138
6.7 Summary
In the quantitative study, the targeted sample size was achieved with over 500
participants contributing to the study. There were a number of positive statistically
significant results indicating that a diversity of benefits accrued to the intervention
group. Both primary outcomes, cooking confidence and vegetable intake, were
significantly increased and the benefits were sustained after the programme.
Intervention participants self-reported vegetable intake increased by more than 0.5
serves per day after completing the programme. The magnitude of change for self-
reported mean vegetable consumption, while significant, was smaller compared to
the medium-large effect differences found in participants’ cooking confidence. A
number of secondary outcomes also significantly improved and were maintained
over time. The programme positively influenced participants’ cooking knowledge,
attitudes and beliefs around cooking easy and quick healthy meals prepared from
“scratch”. Participants were increasingly preparing more meals at home, and
consuming and incorporating more vegetables into their meals. Participants were
also consuming significantly less take-away food, which corresponded to a reduction
in their take-away food spending.
139
Chapter 7. Qualitative results
7.1 Chapter overview
This chapter discusses the qualitative results of the evaluation. The presentation of
the qualitative results differs to that of the quantitative results because qualitative
research stems from a different paradigm of inquiry (161). The key themes emerging
from the thematic data analysis are outlined in this chapter. The motivations for
attending the programme and influences on participants’ intentions to change are
investigated. Ideas around cooking and healthy eating, social standards, identity and
individual responsibility are explored, specifically looking at the role of social norms
and personal responsibility in influencing food and cooking choices. Further themes
to emerge from the data included the celebrity influence of Jamie Oliver on
participants’ food choice, the social benefits stemming from the programme, food
affordability and food purchasing changes as a consequence of the programme; and
barriers and facilitators to cooking.
Table 15: Summary of interview themes and domains
Participants’ motivations and intentions
The celebrity factor, Jamie Oliver’s influence Social benefits of the programme Socialisation and social support in class The value of eating with others and social interactions in the home
“ ” Feelings of accomplishment and achievement
140
and to illustrate participants’ experience
the reader, utterances such as ‘umm’, ‘yeah’ and ‘you know’
the use of “…” indicate
7.2 Participants: qualitative study
Fifteen participants participated in the qualitative study. All completed round one
and two interviews, whilst 13 completed a third interview (with two participants lost
to follow-up). Figure 14 shows this in more detail. After initial reflection and
analysis of the first round of interviews, it was decided that all participants who
participated in the first interview would be invited to participate in repeated
interviews. All participants provided rich and unique insights or differing
perspectives, hence the decision to invite them back for subsequent interviews.
Another participant, who had an intellectual disability, was interviewed with a
carer present. There were no instances of the carer interrupting or preventing a
141
response during the interview, therefore it was decided that the attendance of another
person did not appear to influence the responses provided by the participant.
However, it is possible that subtle influences arising from their relationship could
have influenced the response of the other person present, this was perceived to be the
case by the researcher.
142
Figure 14: Jamie’s Ministry of Food evaluation - qualitative participation, August 2012 to July 2013*
* This diagram is published in Herbert et al (2014) (9)
Six-month T3 Interview, n=13 Face-to-face (n=8) or phone interview (n=5) Interview with another person present (n=2)
Loss to six-month follow-up, (n=2) Reasons unknown (n=1) Could not be contacted (n=1)
Before programme, Interview one, n=15 Face-to-face interview, n=15. Interview with another person present (n=3)
After programme, Interview two, n=15 Face-to-face (n=11) or phone interview (n=4) Interview with another person present (n=3)
Recruitment via telephone number provided, n=12
A total of 199 participants agreed to be contacted for an interview on their completed baseline quantitative questionnaire and provided a contact telephone number. Based on data collection dates and the participant’s programme commencement date only 24/199 participants were eligible for an interview.
12 were scheduled for an interview. 12 were unavailable for an interview due to time commitments and interview scheduling constraints.
Recruitment via class observation (n=3) 4 participants were approached only 3 agreed to an interview.
Qualitative participants
143
attempted with the individual’s carer to understand the participant’s
144
Table 16: Qualitative participant profile at time of first interview
n Total participants interviewed 15 Gender
Female 13 Male 3 Age range in years
18-29 2 30-39 2 40-49 2 50-59 5 60 + 4 Household characteristics
Couple, with young children (0-17 years old) living at home 4 Couple, with adult children (18 years and over) living at home 3 Couple, without children living at home 6 Live Alone 1 Semi-independent living in care facility 1 Main cook in the home
No 4 Yes 9 Yes, shared responsibility 2
7.3 Participants’ motivations and intentions
7.3.1 Programme participants’ motivations for programme attendance
The decision to enrol in the JMoF programme occurred at a time of significant
change in life circumstances for many participants. These included health issues,
family changes or financial difficulties. Many participants were seeking an
opportunity for either self-improvement or respite from daily realities by enrolling in
the programme. Some were looking for a positive change in their life and saw the
programme as a first step in this process. Others saw it as an avenue for help and
assistance, something they were “needing” or an “outlet… to not think about what
was going on at home”. For some participants, life was particularly tough at the time
of the initial interview and they saw the programme as an opportunity to do
something positive and as an outlet from their hardship. Participant circumstances
145
evolved and changed during the evaluation period, which often led to changes in
their cooking roles. For example, retirement, moving house, being made redundant
or working different or longer hours changed people’s schedules and routines, which
in turn impacted on their domestic cooking practices.
7.3.2 Influences on intention to change
The recruitment and sampling process previously outlined in Chapter 5 aimed to
capture critical factors that may impact on the participants’ home cooking, their
willingness to learn and ability to make changes. Some factors appeared to influence
participants’ intention to change cooking and food practices more than others.
Participant life stage impacted on domestic food practices and willingness to change.
There were differences between participants living in a household with children at
home (or adult children) compared to those without children at home or those living
alone. As participants aged they appeared to cook less, which is reflective of their
life stage. Most participants whose children had grown up and left home talked
about what they “used” to do in terms of family cooking. Some participants
considered that making a big meal from scratch, which entailed lots of ingredients,
would be a waste of time and effort unless they had a big family to cook for. For
those participants the effort involved in preparing a big meal was saved for when
their children or grandchildren visited. Cooking and meal sharing were recognised as
having a social element which influenced food choices and practices. For participants
who lived alone, cooking behaviours differed between when cooking for others and
cooking for themselves.
146
7.4 Cooking social norms and cooking identity
This section discusses the ideas around social norms, responsibility and identity
towards cooking and healthy eating. Firstly, normative themes around identity and
the responsibility of being a good mother or wife are presented; this includes, for
example, feeling the need to be responsible for food provisioning and knowledge
sharing. In addition, the influence of age and life stage is explored as well as the role
that health and well-being plays in individual food choices.
7.4.1 Provisioning and being a good mother and wife
There are complex meanings placed on decision-making when it comes to food
choices and cooking practices (44). Social influences and pressures play an
influential role in decision-making. There is a normative moral notion of the identity
of being a good wife and mother and what this means in terms of food provisioning;
it is not only about providing a “proper” meal but also about the transference of
knowledge (38, 214). For JMoF participants, the concept of being a good wife was
about providing meals that their partner will enjoy and eat. The notion of
motherhood appears to be even a stronger influence on cooking changes in the JMoF
sample than being a good wife. Therefore, this concept of motherhood will be further
explored.
“ ”
147
“I would like to be able to [show my daughter], because she says ‘can I cook tea’
and I go ‘well I don’t even know what is planned, let alone what to have…”
~ Leah, mother in her 40s.
“I’m thinking it is important to be able to move forward with yourself [referring to
her children] and later on you are going to want to cook. If nobody has taught you
by the time you move out of home you get a little bit behind. Even if in the end she
148
[daughter] conquers three or four things that she is a whip at [its] still good…”
~ Leah, mother in her 40s.
“It has actually been quite good, my girls have been a bit more encouraged I
suppose. They [say], ‘mum can we help cook dinner tonight?’ And so they are
learning the knife skills and stuff like that”
~ Wanda, mother in her 40s.
7.4.2 Influence of age and life stage
Age and life stage appeared to affect cooking responsibilities and motivations to
cook. The motivational role of having children living at home was discussed
previously in Section 7.3.2. Retired or semi-retired participants, without children or
whose children had grown up and moved away, indicated different cooking
responsibilities. In these instances the responsibility to make changes rested on the
individual. Without a motivational prompt or some subjective norm to change, there
was often no adequate trigger to bring about change.
149
7.4.3 Individual responsibility towards personal health and well-being
Health is often a motivating factor that influences life changes and transitions. In the
previous section, the social pressures associated with the preparation of a proper
meal, with being a good mother, and the influences of life stage were discussed. An
additional motivation appeared to be personal responsibility for one’s own health.
For a few participants their health and well-being were motivating factors towards
changing their cooking practices.
“ ”
“I cook most days….it’s not a burden it’s pleasant.”
~ Dolly, semi-retired grandmother in her 50s.
“I have noticed that my skin feels healthier… I seem to have more energy and I
thought maybe that is because we have cut out all the extra crap or all the added
preservatives and stuff like [that] since now we are going healthy, like the healthy
option with the fresh food… Since we have cut out all the bottle jars and stuff we
150
have noticed, both of us seem to have more energy and even the kids have more
energy…”
~ Kaye, mother in her 30s.
messages within class had a role in influencing participants’
“I think it is a little bit combined with Jamie’s health approach [which] is put
through how you cook. You go away going I have cooked with so much bad stuff for
your body and for yourself and you watch his shows, and well I have only seen a
couple of episodes of the 15 Minutes Meals, you just see his exuberance and how he
explains healthy food doesn’t have to taste like crap… You can do all these things
and it was just a by-product of that. I started to think well he says so much about
obesity and all that but if we keep eating this way, we are just going to get worse and
worse and have more health issues and things like that… Do I want to be eating
these nasty foods forever and it just kind of sparked something in my head. I started
thinking about being healthier and getting active and ways to do that [is] eating
healthier. But I would never have thought about that before”
~ Katie, student in her 20s.
The programme had an influence on participants’ attitudes towards their health
ys they could improve it. These factors resonate with Jamie Oliver’s
151
7.5 The celebrity factor, Jamie Oliver’s influence
Powerful people such as Jamie Oliver, a successful celebrity chef seen by many as an
authority on cooking as well as an obesity prevention advocate, can influence how
knowledge is viewed. It was not unusual that the topic of Jamie Oliver came up
during the interviews given that his name is a key factor that sets this programme
apart from other non-celebrity promoted cooking programmes. Jamie Oliver is well
loved within Australian popular culture, as demonstrated by his outstanding book
sales in Australia (215) and the popularity of his television shows. Within the
Australian policy context, he has the backing of state governments who support and
fund the JMoF programme. In addition to his celebrity status being a political draw
card, it also helped with programme recruitment.
Jamie’s
healthy as distinct from the “ ”
s. Jamie’s philosophy of a healthy meal is
Jamie’s
“Now that I have seen him I’m a bit impressed. Not to be horrible but I hadn’t seen
much of him and thought he was a bit of a British ponce. But now I kind of [think] he
is an honest guy, he just wants to cook good stuff...”
~ Katie, student in her 20s.
and Jamie Oliver’s cooking style and
Jamie says “
”
152
to be healthy “ ”
. One person’s perception o
“You [know] how Jamie Oliver’s [cooking] is oil, all oil... I thought it’s meant to be
healthy? Is it [that] olive oil is ok or what? I’m not sure?”
~ Sally, carer in her 50s.
“ ” “ ” “ ”
enthusiasm. Perhaps Jamie’s level
’s content
increase people’s level of food interest
“ ”
153
“My wish list is to have some of Jamie’s beautiful knives and his pots and pans.
Going home to use my daggy knives that was a little bit hard...”
~ Diana, living alone in her 50s.
“I actually saw him on TV the other night for the first time ever and he was cooking
the 15 minute meals and I went out and bought the book. So he has got me
motivated…”
~ Barbara, semi-retired in her 50s.
There was discussion about the programme in the context of the Ipswich community.
The programme has been “embraced” by the community members, which was
recognised as largely due to the efforts of the Ipswich City Council and the current
mayor of Ipswich. Participants felt special that Ipswich had been chosen for the site
and appreciated using new and lovely cooking utensils and equipment and good-
quality cuts of meat.
“Ipswich is a smaller country town and it’s nice to have some of these things
available to people outside the capital city areas…”
~ Jenny, retired in her 60s.
Participants had an understanding of where other JMoF sites were, both in Australia
and internationally. Some participants still felt a connection to the JMoF staff and
programme six months after completing the programme. Some participants felt
connected to and a part of something bigger than their programme in Ipswich, by the
realisation that they were connected to others sites in Australia and around the world.
The qualitative findings included examples of Jamie’s influence on programme
154
authority that is Jamie Oliver. Oliver’s celebrity effect was viewed as providing
e participants’
7.6 Social benefits of the programme
7.6.1 Socialisation and social support in class
The social benefits of the programme were explored. There were obvious and
expected social links within the class setting. At programme onset, strong motivating
factors to attend the programme included a desire to make new social contacts. In
class there were opportunities to interact and learn from others, through class
discussion, in an encouraging and positive way. Participants were able to share, learn
and discuss food and their cooking knowledge with other class members as well as
with the JMoF volunteers and food trainers.
participants’ experiences and feelings of inclusion. Bonds were built with others that
influenced class enjoyment. As each course progressed, participants’ interactions and
155
“You expected to go in, cook, bring home a meal, whoop-di-do, but you came out,
[and] you were friends with these people for an hour and a half every week, you got
to know your leader and learn different skills and things that you can kind of take
home. So it’s more than just a meal in a box…”
~ Katie, student in her 20s.
“
take a picture and put it on…”
7.6.2 The value of eating with others and social interactions in the home
The JMoF programme encouraged commensality and valued the social significance
of eating with others. The significance of commensality was also felt within the
broader environment as these messages are engrained in many facets of society. The
JMoF programme emphasised these societal messages with embedded context
around the value of eating with others; for example, the meal cooked within class
could be taken home and was enough to feed four. Participants’ beliefs and attitudes
about the importance of eating meals together with other people, and particularly
around the value of eating as a family, were expressed by many. Before commencing
156
the programme, most participants had similar perceptions on eating socially, with
opinions and actions dependent on their life stage and everyday scheduling of
activities. Participants with children living at home noted the importance of eating
dinner together at the table, and this only changed for special occasions or when a
family member was not home. There was a consistent belief amongst participants
about the importance of eating meals together with other people.
“The kids get to tell us what they are up [to] and we don’t always get that together
time, because I’m working, so eating at the dinner table is really good because we sit
down and we go “what did you do today?'…”
~ Wanda, mother in her 40s.
For a few older participants, it was their children’s visits to the family home
circumstances commented, “ the kids are at home…”
“ ” “ ”
in front of the television was reported by a number of respondents as being “ ”
157
participants acted in ways to prepare “ ” meals with others, whether this was
home, many describing “ ”
“We've made a bit more space and… we sort of moved some things around, so now
when I'm cooking, the kids can sit up on the kitchen bench and we can still do
reading or some of the homework so there's still that interaction, whereas before we
didn't have it set up like that… I think it's changed the dynamics, which we wouldn't
have bothered to change at all if we hadn't come along [to the programme]… It was
like, well if the kids needed help I had to go and help them out. I couldn't be
cooking. The risotto needs to be stirred continuously. You can't do that if every
minute "can you help me with…" but now it's like "yeah, I'm doing this". Yeah but
putting those changes into place…”
~ Leah, mother in her 40s.
158
7.7 Food purchasing behaviours
A number of the results in this section have been reported by the candidate in a peer-
reviewed journal article, Herbert et al (2014) (9). After completing the programme,
there was an attitude shift amongst some participants who began to recognise the
importance of preparing meals from fresh ingredients compared to using prepared
foods. As a result of this attitude change, some participants indicated changes in their
food purchasing practices. Participants reported purchasing a wider variety of fresh
foods, such as fruit and vegetables, and less ‘packet’ and processed or prepared
foods; many viewed this as a direct consequence of preparing more meals from
scratch. Six months after the programme, there were examples of participants
shopping smarter, “buying more to our list” and growing vegetables and herbs.
“I was stuffing the fatty things [in the trolley], and I wouldn’t change and try new
stuff, which was costing me more money and now I’m trying all these new things, I
might spend a bit more on fresh fruit and vegetables than what I used to but…. It’s a
good thing [it] means we are not buying crap…”
~ Kaye, mother in her 30s.
home were more likely to invest energy in providing a “ ” meal made from
159
7.8 Barriers and facilitators to cooking
Cooking and eating attitudes, beliefs, knowledge and behaviours were explored to
understand the factors that hindered or helped participants’ food experiences at
home.
7.8.1 Cooking enjoyment
Enjoyment or lack of enjoyment played a significant role in participants’ attitudes
around cooking. For many, cooking was an enjoyable process which allowed
opportunities to be creative by "creating something from nothing". However, this
creative process was only enjoyable under the right circumstances. Some people did
not enjoy cooking and yet it was described as an essential task, “a necessary evil”.
Lack of cooking enjoyment stemmed from feelings of failure and was linked heavily
to other barriers such as lack of skills, confidence and time. Whilst both male and
female participants indicated a lack of enjoyment from cooking, there were gender
differences. The men commented they did not have the skills or confidence to do it
and therefore did not like it or felt they were bad at it. When women discussed their
dislike of cooking, it often related to a lack of time, feeling unappreciated, and that it
was a chore. In contrast, there were some participants that expressed a genuine love
of and satisfaction with cooking. However, these comments were often followed by
examples of barriers around cooking. In the beginning, the programme was seen as
an opportunity to convert cooking into a positive experience through increased skills
and efficiency and a broadening of their meals repertoire. Post-programme, these
expectations were met for many.
160
7.8.2 Time to cook, the “battle between this and that”
Lack of time was considered a barrier to cooking because limited cooking skills and
knowledge prevented quick meal preparation. Between work, family and social
commitments, time was typically seen as a precious commodity not to be wasted.
Many participants begrudged time spent cooking as it took them away from other
activities they preferred to do. In the past, to save time, some participants prepared
fewer meals from scratch. One participant described how in the past cooking was a
battle and she had to choose between cooking and helping with children’s homework
“which was why some of the quick foods or the instants or takeaway were easier”.
For this participant, as well as for others, the programme offered some time-saving
tips such as planning ahead and being prepared in advance. After the programme
busy schedules were still a common problem and participants’ daily commitments
did not change. Family, work and social lives continued to be demanding. The
programme did however offer solutions and ways to save time, so that cooking was
not as burdensome.
“Some nights we still don't but a lot of the times we're making extra now to freeze
into little bits which we weren't doing before. All of a sudden we're going "hey,
there's these quick meals."
~ Leah, mother in her 40s.
and Jamie Oliver’s television shows and
7.8.3 Cooking confidence, skills and knowledge
Cooking confidence is an important aspect of the cooking process, affecting skills,
meal planning and food attitudes (61). Participants described a lack of trust in their
161
ability to cook, read recipes, prepare new foods, prepare food for others and prepare
a healthy meal. Whilst participants had the ability to prepare a simple meal from
scratch at home, they were limited by their lack of confidence to go beyond that
basic level. Thus this restricted their ability to prepare certain types of foods, extend
their cooking repertoire and their willingness to cook for others beyond their
immediate family.
“I’m confident in the meals that I do on a regular basis, if it’s something new I take
hours…. ”
~ Kaye, mother in her 30s.
“ ”
participants’ domestic
“I am going to make some stock on the weekend and make the pizza sauce, so it’s
there in the freezer ready to go…”
~ Rochelle, mother with no children living at home in 50s.
162
In contrast, one retired male participant felt that learning new knowledge at his age
and in his situation was “probably a little bit too late".
"I would say if you were a health fanatic you wouldn't find it healthy enough, but if
you eat in moderation, yeah [it was] good"
~ Jackie, retired in her 60s.
"It is there in practice but there wasn't a discussion about it, you know ‘hey you are
better off cooking homemade spaghetti than buying something else’. There weren't
nutrition lessons I guess but maybe that is enough, maybe that is enough to change
people's minds..."
~ Dolly, semi-retired grandmother in her 50s.
".... I thought going into the whole idea of trying to make people healthier would
have been useful but not "I will teach you how to cook". That is lovely but tell me
how much to cook and why I should be eating it and what are the benefits. Maybe
163
look at the courses more from a weekly menu plan. Over the ten weeks look at it like
a weekly menu plan if you cook these dishes throughout the week you are getting this
amount of serves and it’s this many calories..."
~ Danni, mother in her 30s.
Some participants thought Jamie Oliver’s recipes that use
“[I] used to throw so much away that you didn’t think you could use. Just different
aspect of how to put it together and what is in certain things that maybe you
shouldn’t eat - like don’t choose this type of meat, it’s very fatty, choose this type of
beef or steak. I think what they said was ‘get the best quality for what you can
afford’… If you can get the good quality steak or rump steak get that one, if you can
afford it, and a bit more different fruit and veggies add them into your diet you know
for all the good things that we need, like the fibres and everything I think that was
very much a kind of a point over all classes”
~ Katie, student in her 20s.
7.8.4 Cooking and food attitudes
As previously mentioned, cooking confidence played a crucial role in participants’
cooking attitudes and food choices. The opportunity to prepare a successful meal in
class and take it home built participants confidence. The process of cooking in class
allowed participants the chance to experiment with personal and family food
preferences. The programme provided an opportunity for participants to try and
prepare new foods in a safe environment. There were changes in some participants’
preparedness to try and have a go at making something new without fear of failure.
164
“[In class] everything you cook turned out. It was never a failure for anybody. And
there were some people who had skills and some people who didn't have skills and
they all seemed to manage a good outcome”
~ Dolly, semi-retired grandmother in her 50s.
“He doesn’t cook every day and I don’t use the computer everyday so I can’t use a
computer.”
~ Sally, carer in her 50s.
“It’s just a big of a waste of time. If I was by myself I would eat sandwiches all day
long…”
~ Ron, retired in his 50s.
165
7.8.5 Feelings of accomplishment and achievement
Another facilitator to cooking, gained after programme attendance by many
participants, was feelings of accomplishment and achievement. Firstly, participants
experienced a sense of accomplishment within class after completing the programme
and having tangible evidence each week of what they had prepared. Long after
completing the programme, participants reported continued feelings of happiness and
increased levels of support from others in their home. Feelings of resentment towards
the task of cooking had subsided; for many, the cooking and eating process became
more inclusive of other family members, and therefore became more enjoyable.
“You find it's a little lonely task and everybody else is watching the news or doing
something… and I'm stuck in the kitchen. How did that happen? Whereas now it's
more interactive and people are sort of appreciating it and wanting to be part of it”
~ Leah, mother in her 40s.
7.9 Summary
In summary, qualitative findings revealed many of the positive experiences
stemming from programme attendance. The course facilitated ways to make cooking
easier and simpler, thereby reducing the time taken to prepare meals and improving
participants’ self-efficacy around cooking. Cooking no longer felt like a chore and it
was considered a more enjoyable process and was inclusive of other family
members, which in turn further enhanced their enjoyment. There were social benefits
from the programme, both within the class and at home, through changed
environments and attitudes to cooking. Jamie Oliver was exemplified as an expert
role model empowering participants to cook from scratch. These results are further
discussed in the next chapter, Chapter 8, where these qualitative findings are
compared and contrasted to the findings of the quantitative study presented in
Chapter 6.
166
Chapter 8. Discussion of findings
8.1 Chapter overview
8.2 Research questions and programme logic model
The overarching research question for this research was to determine the impact of
JMoF Ipswich Australia as a community-based cooking skills intervention using a
mixed methods approach.
rticipants’ skills, knowledge, attitudes,
interviewed participants’ expectations were met, and a number of motivators for
167
Figure 15: Improved programme logic model
The next section discusses the major findings in line with the major domains in the
programme logic model, namely personal development, social connectedness,
healthy eating and affordability of a healthy diet.
168
8.3 Major findings
The evaluation results showed multiple significant, positive and sustained outcomes
of the JMoF programme in terms of the personal development of participants, such
as improvements in cooking skills, cooking knowledge, cooking confidence, cooking
enjoyment and attitudes, and perceived health. There were positive social benefits
gained as a result of attending the programme, both in terms of family home life and
commensality (the social significance placed on eating with others). Positive dietary
behaviours changed in terms of improved cooking behaviours, increased
consumption of vegetables and decreased take-away food consumption. As well,
changes in food spending patterns reflected the changes in participants' diets. The
qualitative findings accorded closely with the quantitative findings and served to
strengthen the findings by providing an understanding as to why the identified
changes may have occurred, a summary of quantitative and qualitative findings are
shown in Table 17 to identify complementary findings. Findings consistently
expressed a story of improved cooking skills, confidence and behaviours amongst
participants after attending the JMoF programme were sustained six months post
programme.
169
Table 17: Summary of complementary quantitative and qualitative findings
Improved and sustained cooking confidence, belief they find it easy to change eating habits, nutrition food knowledge are salt and sugar
Improved and sustained belief and attitude that vegetables can be tasty foods
Ability to put together a healthy meal from scratch in 30 minutes Increased and sustained belief in consuming enough fruit and vegetables
Small reduction in total weekly take away expenditure
8.3.1 Increase personal capacity for successful behaviour change
The JMoF programme placed great emphasis on building participants’ confidence,
skills and knowledge to cook simple food in a supportive and non-judgemental fun
environment. For example the programme teaches participants how to cut an onion
in a unique way, how to read recipes and encourages the tasting of new foods.
During this process, participants gained the experience of producing successful
healthy and cheap meals. Results indicated that participants experienced many
changes described in Social Cognitive Theory as essential to behavioural change,
including positive reinforcement, increased capacity, and positive meal expectations
and expectancies, given the emphasis placed on cooking meals from scratch and food
appreciation (150, 151, 196).
170
relating to one’s belief in
change in participants’ cooking confidence as indicated
indicated improvements in participants’ confidence to cook and
one’s
The confidence questions in the JMoF questionnaire related to participants’
improve participants’ confidence to follow a recipe. The JMoF intervention provides
171
improvements in the JMoF participants’ cooking skills, attitudes
The quantitative questionnaire included three questions on nutrition knowledge.
Before responses to these questions are discussed, it is important to note that the
JMoF programme did not directly teach nutrition education; this was not considered
172
a primary objective of the programme. However, whilst health and nutrition were not
key components of the JMoF curriculum, the notions of good nutrition and a healthy
diet are strongly embedded in all facets of the programme.
173
Like nutrition outcomes, health outcomes were not a stated principal objective of the
JMoF programme. Nevertheless, whilst they are not specifically articulated, the word
“health” features strongly in the rhetoric around and marketing of the programme.
Furthermore, potential improvements in population health and well-being are the key
factors underpinning investment in the programme by state health departments.
participants’
participants’
JMoF participants’ i
edium improvements in participants’ general
174
Another study with a control group reported on children’s BMI after
human body’s response to weight loss is slow
175
8.3.2 Positive behaviour changes after the JMoF programme
Social connectedness was measured in the quantitative study via a series of questions
relating to where and with whom people ate their evening meals. Results showed that
the JMoF programme over time, led to small significant increases in eating at the
table (9). On the other hand, programme attendance did not appear to significantly
impact on the frequency of eating dinner together with other people or eating dinner
in front of the television. In Australia, in 2008, a national study (n=1000 parents)
found that 77% of families ate together at mealtimes five or more times per week
(36). JMoF showed that six months after the intervention, approximately 72% of
those participants who had children living at home reported eating together five
times per week; this had increased from 66% at baseline. The JMoF population is
representative of only one socially disadvantaged area in Australia. Therefore, direct
comparison with a national sample must be treated with caution. The fact remains
that there is opportunity for improvements for the JMoF population to reach this
national comparison.
176
The JMoF qualitative evaluation showed that positive social interactions in the home
environment increased through both domestic cooking and social eating. These
benefits were particularly evident in mothers with children living at home, who
reported changing their food and cooking behaviours for the benefit of their
household (Chapter 7, section 7.4.1). These changes also influenced their level of
cooking enjoyment, which in turn contributed to a further reduction in cooking
barriers (Chapter 7, section 7.8.5). The social effects in the home environment
following a cooking skills programme do not appear to be well established in the
literature.
participant who attended the JMoF programme but also on their family’s experience
d meals and less on family members’ attitudes and cooking
177
shows that increased family meal frequency and children’s involvement in cooking
s’
There were small sustained improvements in the frequency of cooking a main meal
from basic ingredients, consuming more vegetables and incorporating more
vegetables into the main meal, as well as reducing take-away food purchasing and
consumption (Chapter 6, sections 6.4.4 and 6.4.5). Most of the outcomes that
improved significantly (compared to controls and over time) remained relatively
stable and were sustained between baseline and the six-month follow-up. The
preparation of a meal from basic ingredients and the consumption of more vegetables
with the meal were the two outcomes that continued to show increases after
programme completion, with significant improvements between post-intervention
and six-month follow-up. This is also highlighted by a small to medium increase in
effect size between T1 and T3. These elements are not surprising as a key message in
the programme is to about cooking from “scratch” and teaches ways of
incorporating vegetables within the main meal, for example with the addition of a
salad.
178
and eating. Results also showed that increases in participants’ cooking skills,
in participants’ cooking, consumption and shopping behaviours, whilst the
one’s family can be a motivating factor
The act of providing a “ ” meal made from scratch and containing vegetables
179
part of the programme’s
effect differences found in participants’ cooking confidence.
180
in JMoF participants’ take
equates to less than the price of a McDonalds ‘ ’
181
’
–
8.3.3 Additional emergent findings
In addition to the alignment between the results and the domains set down in the
original programme logic model, the evaluation gave rise to several key findings that
fell outside the originally identified domains. Two emergent findings in Chapter 7
were cooking role modelling as an element of parental provisioning (section 7.4.1)
and the celebrity impact of Jamie Oliver (section 7.5).
The notion of being a “good mother” by making meals from scratch and preparing
healthy meals was strong amongst participants attending the JMoF programme.
These ideas link to the provisioning of healthy foods and the normative view that, as
a mother, one should provide children with healthy nutritious meals (214). For some
JMoF participants, a mother’s role was to foster healthy behaviours in their children.
182
After attending the programme, participants gained new knowledge and cooking
skills, which enabled them to fulfil their perceived responsibility to provide healthy
meals to their families. The programme was instrumental in shifting the attitudes of
some participants and redirecting them away from the frequent use of convenience
foods.
“ ”
compromise on their perceptions of being a “ ” and negotiate between
183
This is consistent with a shift in the population’s attit
shifted the norms and attitudes around what is satisfactory to serve to one’s family;
provide for their children’s future knowledge base. This expectation can be
“ ” their children’s
184
The qualitative evaluation findings provided examples of Jamie’s influence on
programme participants. Jamie Oliver’s track record shows that he has created his
own discourse on cooking skills and what “proper cooking” involves in order to help
people achieve quick and delicious meal outcomes (231, 232). Participants sought to
acquire cooking knowledge and skills from Jamie Oliver by their attendance.
However, those who were not attuned to his discourse before starting the programme
were more aware after the programme of his “non-fancy” cooking and basic
messages. These messages encouraged an attitudinal shift in many participants
towards cooking enjoyment. Participant Katie, for example, did not think much of
Jamie before completing the programme; however, after the programme her attitude
had changed and she actively sought healthy behaviours around cooking and food.
Jamie’s “expert” advice created individual self-surveillance and self-governance
(76), and this was evident in the sustained behaviours of JMoF participants six
months after the programme in terms of both the quantitative and qualitative results.
There was an interplay occurring between Jamie’s role as an expert and his simple
approach to cooking which was accessible to participants.
Cooking skills and what is considered “ ”, or what constitutes
“ ”,
. Jamie Oliver is one such “ ” . Jamie Oliver’s altruism has been
a presence in participants’ lives after
185
’
programme were reinforced in the participants’ broader environment. This factor
me the face of one of Australia’s
branded itself the “ ”, there was controversy; the chain used
186
in 2012, Jamie Oliver participated in the Victorian government’s launch of
’s
mething “ ”
187
8.4 Evaluation strengths and limitations
The use of a mixed methods evaluation is a major strength of this PhD because the
results not only measure changes in outcomes to demonstrate programme
effectiveness but also provide insights into participant experiences in order to
understand why changes occurred. Findings showed clear and consistent results from
both methods, painting a picture of what was going on in terms of participants'
attitudes, knowledge, cooking skills, confidence and behaviours. Furthermore, the
breadth of the outcome variables measured is a key strength, which facilitated the
building of a more comprehensive picture of impacts of the programme. Two
previous interventions have used a mixed methods design with a control group and a
follow-up period (77, 94); however, this JMoF evaluation extends their findings by
the provision of a larger sample size and the broader breadth of the outcome
variables measured. The findings relate only to the 35-45% of enrolled programme
participants, as these are the participants who participated in the evaluation. Whilst
inferences and generalisations can be made, it must be acknowledged that more than
half of the programme participants were not represented in the evaluation
188
population may be “ ”. On the other hand, the use of a
inferences about the programme’s effectiveness.
189
measure such as dietary patterns, nutrition knowledge, food security and participants’
Questions testing participants’ nutrition knowledge were considered for the
available on participants’ ability to pay for food or to take time out to attend cooking
190
to obtain true accounts of participants’ weight, height and waist, however they were
researcher’s inexperience was accounted for by training conducted with her
critical and useful feedback on the researcher’s style and techniques. This
A major strength of the qualitative study methods was the effort taken to
ensure methodological rigour, reliability and validity of the data collection and data
analysis processes used. The detailed description of the data collection methods and
systematic analysis process, presented in Chapter 5 (Qualitative methods), is a point
of rigour in its own right (248) and provides credibility to the researcher (161). The
191
level of detail specified in Chapter 5 means that others could easily replicate the
methods employed.
A number of strategies are discussed in the academic literature to help establish
reliability and validity in the qualitative research process (161, 207, 249). Morse et al
(2002) recommend the building of five verification strategies into the methods (249).
The first verification approach, methodological coherence, refers to the research
questions and the methods. As the process of data collection and analysis proceeds,
the research question and methods may need to be changed (249). In this study, this
occurred with the development of the interview structure and was built into the
research objectives. The study aimed to explain any unanticipated programme
outcomes, allowing the methods and questions to be open to exploration of any new,
unexpected outcomes. The second verification strategy was the use of a sample that
was appropriate to answer the research question (249). This was achieved by using
purposive sampling. The third recommendation was that data collection and analysis
occurred concurrently (249). As previously described in Chapter 5, data analysis
began at the very first instance of data collection, with data immersion and initial
coding beginning directly after each interview via transcription memos compiled by
the researcher. This allowed opportunities to clarify any uncertainties and follow up
prevalent issues and ideas with each participant in their follow-up interviews or over
the phone. A key strength in using a longitudinal design was that participants’
journeys throughout their programme were recorded; the researcher established a
relationship with participants and built rapport over time. The fourth strategy was to
take social theory into consideration, as outlined in the data analysis section of
Chapter 5. The use of social theory allowed opportunities to verify and compare new
ideas to the initial theoretical concepts as well as to search for other concepts that
may explain results (161, 207). The fifth aspect was to contribute to theory (249) by
results extending and enhancing the knowledge base around a cooking skills
theoretical framework.
presented as being representative of the participants’ perspectives and not limited to
the researcher’s perspective
192
In addition, during the researcher’s visits to Ipswich,
help differentiate between descriptions of what occurred in class and participants’
8.5 Summary
This chapter discusses both the quantitative and qualitative findings. This study is
one of the first evaluations of any cooking programme to include a longitudinal
mixed methods evaluation with a control group and six-month follow-up. Findings
provide strong comprehensive evidence to confidently strengthen the programme
logic pathway outlined prior to the evaluation. The use of a mixed methods approach
further strengthens this programme logic model as results align with each other to
confirm the logic pathway to behaviour change after attending the JMoF programme.
Results are grounded in the literature and social theory so that findings may be
translated and applied elsewhere. Findings also accord with the literature. Findings
offer further insights into the evidence around the effectiveness of cooking skills
intervention; few other studies have contained the range of variables that the JMoF
evaluation has. Major findings highlight that JMoF has had an influence on
participants’ capabilities for behaviour change, such as cooking confidence, skills,
attitudes and enjoyment. These capabilities have led to small behavioural changes
that were sustained six months after the programme was completed. Behaviours
included changes towards positive social connectedness, improved social
environments and positive changes in eating and food purchasing for a healthy diet,
with participants prioritising and valuing health and cooking after attending the
programme. In addition, new ideas to emerge from the qualitative data related to
participant motivations for change. These were discussed, namely cooking role
modelling as an element of parental provisioning and the celebrity impact of Jamie
Oliver. The celebrity factor may have contributed to the dose of the programme and
added to the sustained effect of the programme outcomes.
193
194
Chapter 9. Implications and conclusions
The research that is presented in this PhD represents a robust study and is the first
evaluation of JMoF in Australia. It is also the first evaluation of a cooking skills
intervention using a longitudinal mixed methods design encompassing both a control
group and a six-month follow-up period with a large sample size. Given that the
JMoF programme has been in operation since 2008 in the UK, findings will impact
on implementation of the programme both in Australia and internationally. Using
mixed methods, the change over time was measured through quantitative methods
and qualitative methods that explored participants’ experiences over the course of
their programme journey. This research adds to the limited evidence available
internationally on the effectiveness of cooking skills programmes, and offers new
insights into the factors underpinning behaviour change.
9.1 Research implications
9.1.1 Considerations for policy
To date, governments have implemented JMoF without available evidence of the
programme’s effectiveness or cost-effectiveness. As a result of this research, the
programme’s effectiveness has now been addressed but not the programme’s cost-
effectiveness. Results demonstrate the effectiveness and impacts of JMoF Australia.
Any other Australian states or territories considering the JMoF programme
intervention can reasonably assume there will be multiple benefits to arise from the
implementation of the programme. The JMoF programme has been shown to
improve vegetable consumption within the community. This is highly important as
only 8.2% of the Australian population currently meet the daily recommended five to
seven serves of vegetables per day (250).
195
–
196
ry mode and the programme’s
sufficient indicators of the programme’s effectiveness. The sustainability of the 10
the Heart Foundation’s Heart moves
197
and its ‘value money’ as a public
9.1.2 Programme considerations
Practitioners implementing JMoF can confidently claim effectiveness of the
programme via the evaluation presented. Findings are supported by transparent and
robust methods. However, there are a few standout areas where practitioners may
consider an adjustment of the programme for improvement.
responsibility within this “ ” space if its auspices wish to accept funding from
198
9.1.3 Considerations for future research
The JMoF programme has been shown to result in significant and sustained gains in
cooking skills, knowledge, confidence, attitudes and behaviours, yet questions
remain for which further research is needed.
199
analyses. Supplementary sampling of participants’ families may provide data to
investigate whether there were any ripple effects on participants’ family members
wer has been discussed and it is hypothesised that Jamie’s
200
the costs and benefits of Jamie’s
health benefits associated with Jamie Oliver’s connection to the programme include:
participant recruitment to the JMOF programme,
political engagement and funding,
public awareness and engagement,
advertising and media for health promotion messages,
policy and programme enhancement.
’
9.1.4 Implications for the theoretical framework
Finally, this research has implications for theory. The findings from this mixed
method study show a breadth of outcomes which extend our knowledge around
201
cooking skills interventions. Results paint a broader picture measuring and
conveying important findings on a number of integrated outcomes that help to
present a logical pathway to behaviour change. Both the quantitative and qualitative
longitudinal results reinforce each other, and this increases confidence that the
programme logic model framework, shown in Chapter 8, Figure 14, is a valuable
representation of what is likely to occur following programme completion. At the
beginning of this study, there were a number of unknown elements inferred from the
limited and emerging data available. Findings have shown that there is a clear
pathway for confidence and skills acquisition via experiential learning and improved
self-efficacy, which translate into a number of positive cooking-related outcomes and
behaviours. The findings demonstrate that the links between the elements have been
established in a clear pathway towards behaviour change. Results have demonstrated
that the JMoF programme logic model framework is a sound process and it may be
useful for others in understanding cooking skills intervention pathways.
202
9.2 Significance of this PhD
This PhD makes an important contribution to new understandings and extends the
evidence base in the area of cooking classes as a public health nutrition intervention.
The research presented is robust and its findings both build on the existing evidence
and make new contributions to the field. It is the first evaluation of JMoF and is the
first longitudinal mixed method evaluation of any cooking skills intervention in the
literature encompassing a control group, a six-month follow-up period and a large
sample size. The comprehensiveness and concordance of the findings from both the
quantitative and qualitative studies (and their alignment with the literature) are
meaningful and offer a strong platform for robust conclusions on the impacts of
cooking skills interventions. They expand the theoretical concepts around cooking
skills programs and the logical pathway to behaviour change. The programme logic
pathway that was developed offers a new framework that is useful for understanding
cooking skills interventions. This research has discovered new findings that offer
insights into motivations for attending JMoF, and potentially other cooking skills
interventions, which may be applied for future recruitment and targeting of
programme participants. Firstly, the motivator of being a good mother may offer an
avenue of engagement of potential food providers and provide solutions for
knowledge and skills translation in the home for future generations. Secondly, the
celebrity factor may have empowered participants and contributed to sustained
effects on behaviours. These findings contribute to the understanding of why people
engage in programmes such as JMoF, and offer a valuable direction for future
research. Lastly, findings have uncovered new questions for future research, and
directions and considerations for policy makers and programme implementers going
forward.
203
9.3 Conclusions
This study has investigated the impacts of the 10-week cooking skills intervention,
JMoF Australia, in Ipswich, Queensland. It demonstrated the programme’s
effectiveness to provide sustained positive impacts on participants’ cooking and
food-related behaviours and attitudes, as well as enhanced understanding as to the
reasons for changes. Specifically, programme effectiveness was proven given that
the primary outcomes, vegetable intake and cooking confidence, significantly
improved and were sustained after the programme. There were also gains in personal
development around cooking self-efficacy, cooking skills, knowledge, attitudes and
enjoyment, as well as potential improvements in health outcomes. All results are
based on self-reported outcomes; it is important to state that in addition to the use of
more objective measures, there are other areas which should be explored such as the
programme‘s cost-effectiveness, and the effects of the celebrity factor on
participants.
impact on the population’s diet.
204
References
1. World Health Organization. Diet, nutrition and the prevention of chronic diseases: report of a joint WHO/FAO expert consultation. WHO technical report series. Geneva: World Health Organization, 2003 January Report No.: 916.
2. World Health Organization. Globalization, diets and noncommunicable diseases. Geneva: World Health Organisation 2002 Available from: http://whqlibdoc.who.int/publications/9241590416.pdf.
3. Michaud P, Condrasky M, Griffin SF. Review and application of current literature related to culinary programs for nutrition educators. Topics in Clinical Nutrition. 2007;22(4):336-348.
4. Soliah LL, Walter JM, Jones SA. Benefits and barriers to healthful eating what are the consequences of decreased food preparation Ability? American Journal of Lifestyle Medicine. 2012;6(2):152-158.
5. Larson NI, Perry CL, Story M, Neumark-Sztainer D. Food preparation by young adults is associated with better diet quality. Journal of the American Dietetic Association. 2006;106(12):2001-2007.
6. Rees R, Hinds K, Dickson K, O’Mara-Eves A, Thomas J. Communities that cook: a systematic review of the effectiveness and appropriateness of interventions to introduce adults to home cooking. London: EPPI-Centre, Social Science, Research Unit, Institute of Education, University of London, 2012.
7. Reicks M, Trofholz AC, Stang JS, Laska MN. Impact of cooking and home food preparation interventions among adults: outcomes and implications for future programs. Journal of Nutrition Education and Behavior. 2014;46(4):259–276.
8. Flego A, Herbert J, Gibbs L, Swinburn B, Keating C, Waters E, Moodie M. Methods for the evaluation of the Jamie Oliver Ministry of Food program, Australia. BMC Public Health. 2013;13(1):411.
9. Herbert J, Flego A, Gibbs L, Waters E, Swinburn B, Reynolds J, Moodie M. Wider impacts of a 10-week community cooking skills program - Jamie’s Ministry of Food, Australia. BMC Public Health. 2014;14 (1):1161.
10. Flego A, Herbert J, Waters E, Gibbs L, Swinburn B, Reynolds J, Moodie M. Jamie's Ministry of Food: quasi-experimental evaluation of immediate and sustained impacts of a cooking skills program in Australia. PloS One. 2014;9(12).
11. Sassi F, Devaux M, Cecchini M, Rusticelli E. The obesity epidemic: analysis of past and projected future trends in selected OECD countries. Paris: OECD, 2009 Report No.: OECD Health Working Paper No. 45.
205
12. Dauchet L, Amouyel P, Hercberg S, Dallongeville J. Fruit and vegetable consumption and risk of coronary heart disease: a meta-analysis of cohort studies. The Journal of Nutrition. 2006;136(10):2588-2593.
13. Hariri M, Darvishi L, Maghsoudi Z, Khorvash F, Aghaei M, Iraj B, Ghiasvand R, Askari G. Intakes of vegetables and fruits are negatively correlated with risk of stroke in Iran. International Journal of Preventive Medicine. 2013;4(Suppl 2):S300.
14. Arem H, Reedy J, Sampson J, Jiao L, Hollenbeck AR, Risch H, Mayne ST, Stolzenberg-Solomon RZ. The healthy eating index 2005 and risk of pancreatic cancer in the NIH–AARP study. Journal of the National Cancer Institute. 2013;djt185.
15. Thomson CA, McCullough ML, Wertheim BC, Chlebowski RT, Martinez ME, Stefanick ML, Rohan TE, Manson JE, Tindle HA, Ockene J. Nutrition and physical activity cancer prevention guidelines, cancer risk, and mortality in the Women's Health Initiative. Cancer Prevention Research. 2014;7(1):42-53.
16. Kelly T, Yang W, Chen CS, Reynolds K, He J. Global burden of obesity in 2005 and projections to 2030. International Journal of Obesity. 2008;32(9):1431-1437.
17. Swinburn BA, Sacks G, Hall KD, McPherson K, Finegood DT, Moodie ML, Gortmaker SL. The global obesity pandemic: shaped by global drivers and local environments. The Lancet. 2011;378(9793):804-814.
18. Australian Bureau of Statistics. 4364.0 - National Health Survey: summary of results, 2007-2008 (reissue) [Internet]. Canberra Australian Bureau of Statistics. 2009 [cited 2012 March 19 ]. Available from: http://www.abs.gov.au/AUSSTATS/[email protected]/Latestproducts/4364.0Main%20Features22007-2008%20(Reissue)?opendocument&tabname=Summary&prodno=4364.0&issue=2007-2008%20(Reissue)&num=&view=.
19. Australian Institute of Health and Welfare. Australia’s health 2010. Canberra: Australian Institute of Health and Welfare, 2010 Report No.: AUS 122.
20. Colagiuri S, Lee CM, Colagiuri R, Magliano D, Shaw JE, Zimmet PZ, Caterson ID. The cost of overweight and obesity in Australia. Medical Journal of Australia. 2010;192(5):260-4.
21. Caraher M, Lang T. Can't cook, won't cook: A review of cooking skills and their relevance to health promotion. International Journal of Health Promotion and Education. 1999;37(3):89-100.
22. Short F. Domestic cooking practices and cooking skills: findings from an English study. Food Service Technology. 2003;3(3-4):177-185.
23. Stitt S. An international perspective on food and cooking skills in education. British Food Journal. 1996;98(10):27-34.
206
24. Howard S, Adams J, White M. Nutritional content of supermarket ready meals and recipes by television chefs in the United Kingdom: cross sectional study. BMJ. 2012;345.
25. Jabs J, Devine CM. Time scarcity and food choices: An overview. Appetite. 2006;47(2):196-204.
26. Larson NI, Nelson MC, Neumark-Sztainer D, Story M, Hannan PJ. Making time for meals: meal structure and associations with dietary intake in young adults. Journal of the American Dietetic Association. 2009;109(1):72-79.
27. van der Horst K, Brunner TA, Siegrist M. Fast food and take-away food consumption are associated with different lifestyle characteristics. Journal of Human Nutrition and Dietetics. 2011;24(6):596-602.
28. Caraher M, Baker H, Burns M. Children's views of cooking and food preparation British Food Journal. 2004;106(4):255- 273.
29. Caraher M, Dixon P, Lang T, Carr-Hill R. The state of cooking in England: the relationship of cooking skills to food choice. British Food Journal. 1999;101(8):590-609.
30. Jabs J, Devine CM, Bisogni CA, Farrell TJ, Jastran M, Wethington E. Trying to find the quickest way: employed mothers’ constructions of time for food. Journal of Nutrition Education and Behavior. 2007;39(1):18-25.
31. Engler-Stringer R. Food, cooking skills, and health: a literature review. Canadian Journal of Dietetic Practice and Research. 2010;71(3):141-145.
32. Neumark-Sztainer D, Larson NI, Fulkerson JA, Eisenberg ME, Story M. Family meals and adolescents: what have we learned from Project EAT (Eating Among Teens)? Public Health Nutrition. 2010;13(07):1113-1121.
33. Eisenberg ME, Olson RE, Neumark-Sztainer D, Story M, Bearinger LH. Correlations between family meals and psychosocial well-being among adolescents. Archives of Pediatrics and Adolescent Medicine. 2004;158(8):792-796.
34. Christian MS, Evans CEL, Hancock N, Nykjaer C, Cade JE. Family meals can help children reach their 5 A Day: a cross-sectional survey of children's dietary intake from London primary schools. Journal of Epidemiology and Community Health. 2013;67(4):332-338.
35. MacFarlane A, Cleland V, Crawford D, Campbell K, Timperio A. Longitudinal examination of the family food environment and weight status among children. International Journal of Pediatric Obesity. 2009;4(4):343-352.
36. Huntley R. White Paper ‘Because family mealtimes matter'. Australia: Ipsos Australia, 2008 Available from: blogs.smh.com.au/mashup/white-paper.pdf.
37. Mennell S, Murcott A, van Otterloo AH. The sociology of food: eating, diet and culture. London: Sage Publications; 1992.
207
38. Charles N, Kerr M. Women, food and families. Manchester Manchester University Press; 1988.
39. van der Horst K, Brunner TA, Siegrist M. Ready-meal consumption: associations with weight status and cooking skills. Public Health Nutrition. 2011;14(02):239-245.
40. Nelson SA, Corbin MA, Nickols-Richardson SM. A call for culinary skills education in childhood obesity-prevention interventions: current status and peer influences. Journal of the Academy of Nutrition and Dietetics. 2013;113(8):1031-1036.
41. Short F. Domestic cooking skills- what are they? Journal of the Home Economics Institute of Australia. 2003;10(3):13-22.
42. Bava CM, Jaeger SR, Park J. Constraints upon food provisioning practices in ‘busy’ women's lives: Trade-offs which demand convenience. Appetite. 2008;50:486-498.
43. Begley A, Gallegos D. Should cooking be a dietetic competency? Nutrition and Dietetics. 2010;67(1):41-46.
44. Bisogni CA, Jastran M, Seligson M, Thompson A. How people interpret healthy eating: contributions of qualitative research. Journal of Nutrition Education and Behavior. 2012;44(4):282-301.
45. Glanz K, Bishop DB. The role of behavioral science theory in development and implementation of public health interventions. Annual Review of Public Health. 2010;31(1):399-418.
46. McLaren L, Hawe P. Ecological perspectives in health research. Journal of Epidemiology and Community Health. 2005;59(1):6-14.
47. Stokols D. Translating social ecological theory into guidelines for community health promotion. American Journal of Health Promotion. 1996;10(4):282-298.
48. Hayden J. Introduction to health behavior theory. Sudbury: Jones and Bartlett Publishers; 2009
49. McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on health promotion programs. Health Education and Behavior. 1988;15(4):351-377.
50. Sallis JF, Owen N, Fisher EB. Ecological models of health behavior. In: Glanz K, Rimer BK, Viswanath K, editors. Health behavior and health education: Theory, research, and practice. 4th ed. Hoboken: Jossy-Bass. 2008. p. 465-486.
51. Fitzgerald N, Spaccarotella K. Barriers to a healthy lifestyle: From individuals to public policy—an ecological perspective. Journal of Extension. 2009;47(1):1-8.
208
52. Robinson T. Applying the socio-ecological model to improving fruit and vegetable intake among low-income African Americans. Journal of Community Health. 2008;33(6):395-406.
53. Pettinger C, Holdsworth M, Gerber M. Meal patterns and cooking practices in Southern France and Central England. Public Health Nutrition. 2006;9(8):1020-1026.
54. Thornton LE, Bentley RJ, Kavanagh AM. Individual and area-level socioeconomic associations with fast food purchasing. Journal of Epidemiology and Community Health. 2010.
55. Winkler E, Turrell G. Confidence to cook vegetables and the buying habits of Australian households. Journal of the American Dietetic Association. 2010;110(5, Supplement 1):S52-S61.
56. Abbott PA, Davison JE, Moore LF, Rubinstein R. Effective nutrition education for Aboriginal Australians: lessons from a diabetes cooking course. Journal of Nutrition Education and Behavior. 2012;44(1):55-59.
57. Foley RM, Pollard CM. Food cent $—implementing and evaluating a nutrition education project focusing on value for money. Australian and New Zealand Journal of Public Health. 1998;22(4):494-501.
58. Turrell G, Hewitt B, Patterson C, Oldenburg B, Gould T. Socioeconomic differences in food purchasing behaviour and suggested implications for diet-related health promotion. Journal of Human Nutrition and Dietetics. 2002;15(5):355-364.
59. AbuSabha R, Achterberg C. Review of self-efficacy and locus of control for nutrition - and health-related behavior. Journal of the American Dietetic Association. 1997;97(10):1122-1132.
60. Bandura A. Health promotion by social cognitive means. Health Education and Behavior. 2004;31(2):143.
61. Stead M, Caraher M, Wrieden W, Longbottom P, Valentine K, Anderson A. Confident, fearful and hopeless cooks: Findings from the development of a food-skills initiative. British Food Journal. 2004;106(4):274-287.
62. Woodruff SJ, Kirby AR. The associations among family meal frequency, food preparation frequency, self-efficacy for cooking, and food preparation techniques in children and adolescents. Journal of Nutrition Education and Behavior. 2013;45(4):296-303.
63. Dibsdall L, Lambert N, Bobbin R, Frewer L. Low-income consumers' attitudes and behaviour towards access, availability and motivation to eat fruit and vegetables. Public Health Nutrition. 2003;6(02):159-168.
64. Sobal J, Sobal C, Bisogni. Constructing Food Choice Decisions. Annals of Behavioral Medicine. 2009;38(1):37-46.
209
65. James D. Factors influencing food choices, dietary intake, and nutrition-related attitudes among African Americans: application of a culturally sensitive model. Ethnicity and Health. 2004;9(4):349-367.
66. Spence F, van Teijlingen ER. A qualitative evaluation of community-based cooking classes in Northeast Scotland. International Journal of Health Promotion and Education. 2005;43(2):59-63.
67. Neysmith SM, Reitsma-Street M. “Provisioning”: Conceptualizing the work of women for 21st century social policy. Women's Studies International Forum. 2005;28(5):381-391.
68. DeVault ML. Feeding the family: The social organization of caring as gendered work: University of Chicago Press; 1994.
69. Slater J, Sevenhuysen G, Edginton B, O'neil J. ‘Trying to make it all come together’: structuration and employed mothers' experience of family food provisioning in Canada. Health Promotion International. 2011;27(3):405-415.
70. Caraher M, Lange T, Dixon P. The influence of TV and celebrity chefs on public attitudes and behavior among the English public. Journal for the Study of Food and Society. 2000;4(1):27-46.
71. Block K, Johnson B. Evaluation of the Stephanie Alexander Kitchen Garden Program, Final Report to: The Stephanie Alexander Kitchen Garden Foundation. Melbourne: University of Melbourne and Deakin University, 2009.
72. Gibbs L, Staiger PK, Johnson B, Block K, Macfarlane S, Gold L, Kulas J, Townsend M, Long C, Ukoumunne O. Expanding children’s experience of food: the impact of a school-based kitchen garden program. Journal of Nutrition Education and Behavior. 2013;45(2):137-146.
73. Eckermann S, Dawber J, Yeatman H, Quinsey K, Morris D. Evaluating return on investment in a school based health promotion and prevention program: the investment multiplier for the Stephanie Alexander Kitchen Garden National Program. Social Science and Medicine. 2014.
74. Rubin J, Rye D, Rabinovich L. Appetite for change: School meals policy in the limelight 2005. Dunfermline: Carnegie UK Trust, 2008.
75. Warin M. Foucault's progeny: Jamie Oliver and the art of governing obesity. Social Theory & Health. 2011;9(1):24-40.
76. Coveney J, Begley A, Gallegos D. " Savoir Fare": are cooking skills a new morality? Australian Journal of Adult Learning. 2012;52(3):617-642.
77. Wrieden W, Anderson A, Longbottom P, Valentine K, Stead M, Caraher M, Lang T, Gray B, Dowler E. The impact of a community-based food skills intervention on cooking confidence, food preparation methods and dietary choices ? an exploratory trial. Public Health Nutrition. 2007;10(2):203-211.
210
78. Hersch D. The impact of cooking classes on food-related preferences, attitudes, and behaviors of school-aged children: a systematic review of the evidence, 2003–2014. Preventing Chronic Disease. 2014;11.
79. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of Internal Medicine. 2009;151(4):264-269.
80. Armstrong R, Waters E, Jackson N, Oliver S, Popay J, Shepherd J, Petticrew M, Anderson L, Bailie R, Brunton G, Hawe P, Kristjansson E, Naccarella L, Norris S, Pienaar E, Roberts H, Rogers W, Sowden A, Thomas H. Guidelines for Systematic reviews of health promotion and public health interventions Melbourne: Melbourne University 2007.
81. Centre for Reviews and Dissemination. Systematic reviews: CRD's guidance for undertaking reviews in health care. York: University of York, 2009 Available from: http://www.york.ac.uk/inst/crd/index_guidance.htm
82. Jackson N, Waters E, Promotion ftGfSRiH, Taskforce PH. Criteria for the systematic review of health promotion and public health interventions. Health Promotion International. 2005;20(4):367-374.
83. Rychetnik L, Frommer M, Hawe P, Shiell A. Criteria for evaluating evidence on public health interventions. Journal of Epidemiology and Community Health. 2002;56(2):119-127.
84. The Effective Public Health Practice Project (EPHPP). Quality assessment tool for quantitative studies 2009 [cited 2011 November 3]. Available from: http://www.ephpp.ca/tools.html.
85. Spencer L, Ritchie J, Lewis J, Dillon L. Quality in qualitative evaluation: a framework for assessing research evidence. London: Government Chief Social Researcher’s Office, Cabinet Office., 2003.
86. Greenhalgh T, Taylor R. How to read a paper: papers that go beyond numbers (qualitative research). BMJ. 1997;315(7110):740.
87. Public Health Resource Unit. Critical Appraisal Skills Program (CASP) making sense of evidence 10 questions to help you make sense of qualitative research England: Public Health Resource Unit; 2006 [cited 2012 April 2]. Available from: http://www.sph.nhs.uk/sph-files/casp-appraisal-tools/Qualitative%20Appraisal%20Tool.pdf.
88. Curtis PJ, Adamson AJ, Mathers JC. Effects on nutrient intake of a family-based intervention to promote increased consumption of low-fat starchy foods through education, cooking skills and personalised goal setting: the Family Food and Health Project. British Journal of Nutrition. 2012;107(12):1833-1844.
89. Fulkerson JA, Rydell S, Kubik MY, Lytle L, Boutelle K, Story M, Neumark-Sztainer D, Dudovitz B, Garwick A. Healthy home offerings via the mealtime environment (HOME): feasibility, acceptability, and outcomes of a pilot study. Obesity. 2010;18(S1):S69-S74.
211
90. Gold A, Yu N, Buro B, Garden-Robinson J. Discussion map and cooking classes: testing the effectiveness of teaching food safety to immigrants and refugees. Journal of Nutrition Education and Behavior. 2014.
91. Levy J, Auld G. Cooking classes outperform cooking demonstrations for college sophomores. Journal of Nutrition Education and Behavior. 2004;36(4):197-203.
92. Wenrich TR, Brown JL, Wilson RT, Lengerich EJ. Impact of a community-based intervention on serving and intake of vegetables among low-income, rural appalachian families. Journal of Nutrition Education and Behavior. 2012;44(1):36-45.
93. Birkhead AG, Foote S, Monlezun DJ, Loyd J, Joo E, Leong B, Sarris L, Harlan TS, Monlezun DJ. Medical student–led community cooking classes: A novel preventive medicine model that’s easy to swallow. American Journal of Preventive Medicine. 2014;46(3):e41-e42.
94. Kennedy LA, Hunt C, Hodgson P. Nutrition education program based on EFNEP for low-income women in the United Kingdom:“Friends with Food”. Journal of Nutrition Education. 1998;30(2):89-99.
95. McKellar G, Morrison E, McEntegart A, Hampson R, Tierney A, Mackle G, Scoular J, Scott J, Capell H. A pilot study of a Mediterranean type diet intervention in female patients with rheumatoid arthritis living in areas of social deprivation in Glasgow. Annals of the Rheumatic Diseases. 2007;66(9):1239 1243.
96. Wrieden W, Anderson A, Longbottom P, Valentine K, Stead M, Caraher M, Lang T, Dowler E. Assisting dietary change in low income communities: assessing the impact of community-based practical food skills intervention (Cook Well Intervention). Dundee: Centre for Public Health Nutrition Research, University of Dundee, 2002 Report No.: N0911 Available from: http://foodbase.org.uk/results.php?f_category_id=&f_report_id=83.
97. Bielamowicz MK, Pope P, Rice CA. Sustaining a creative community-based diabetes education program: motivating Texans with type 2 diabetes to do well with diabetes control. The Diabetes Educator. 2013;39(1):119-127.
98. Brown BJ, Hermann JR. Cooking classes increase fruit and vegetable intake and food safety behaviors in youth and adults. Journal of Nutrition Education and Behavior. 2005;37(2):104-105.
99. Caraher M, Lloyd S. Findings from the Can Cook! (CiC) Family' Programme! London: City University London, 2013 Available from: http://www.cancook.co.uk/community-schools-cookery/cookery-for-communities/findings-from-the-can-cook-cic-family-programme-a-report-from-city-university-london/family-evaluation-report-final-2013/.
212
100. Chapman-Novakofski K, Karduck J. Improvement in knowledge, social cognitive theory variables, and movement through stages of change after a community-based diabetes education program. Journal of the American Dietetic Association. 2005;105(10):1613-1616.
101. Chavent G, Hoover-Litty H, Liu M, Berman C, Young T. Community nutrition course gains service learning designation by offering cooking matters for child care professionals curriculum. Journal of the Academy of Nutrition and Dietetics. 2012;112:A51-A51.
102. Children's Food Trust. A recipe for Healthier Communities. Sheffield: Children's Food Trust, 2011 Available from: http://www.letsgetcooking.org.uk/big-lottery-programme/impact/how-we-measure-the-impact-of-lets-get-cooking.
103. Condrasky M. Cooking with a Chef. Journal of Extension. 2006;44(4).
104. Condrasky M, Graham K, Kamp J. Cooking with a Chef: An innovative program to improve mealtime practices and eating behaviors of caregivers of preschool children. Journal of Nutrition Education and Behavior. 2006;38(5):324-325.
105. Condrasky M, Hegler M, Sharp JL. Application of cooking with a chef for school food service professionals. Topics in Clinical Nutrition. 2011;26(2):171-178.
106. Condrasky MD, Baruth M, Wilcox S, Carter C, Jordan JF. Cooks training for Faith, Activity, and Nutrition project with AME churches in SC. Evaluation and Program Planning. 2013;37:43-49.
107. Davies JA, Damani P, Margetts BM. Intervening to change the diets of low-income women. Proceedings of the Nutrition Society. 2009;68(2):210-215.
108. Dewolfe JA, Greaves G. The Basic Shelf Experience: A comprehensive evaluation. Canadian Journal of Dietetic Practice and Research. 2003;64(2):51-57.
109. Francis SL. Heart disease nutrition education program increases familiarity with heart-healthy lifestyle recommendations. Journal of Nutrition Education and Behavior. 2012;44(6):658-660.
110. Garcia AL, Vargas E, Lam PS, Smith F, Parrett A. Evaluation of a cooking skills programme in parents of young children – a longitudinal study. Public Health Nutrition. 2013;17(5):1013 - 1021.
111. Keller HH, Gibbs A, Wong S, Vanderkooy P, Hedley M. Men can cook! Development, implementation, and evaluation of a senior men's cooking group. Journal of Nutrition for the Elderly. 2004;24(1):71-87.
112. Kokkonen-May J, Brady A, Van Offelen S, Johnson B. Simply good cooking: online curriculum for the interactive SNAP-Ed classroom. Journal of Nutrition Education & Behavior. 2014;46(1):85-87.
213
113. Lawe B. Teaching university students to cook, to improve their diet: A pilot study at Nottingham Trent University. International Journal of Health Promotion and Education. 2013;51(3):161-168.
114. Lee JH, McCartan J, Palermo C, Bryce A. Process evaluation of Community Kitchens: Results from two Victorian local government areas. Health Promotion Journal of Australia. 2010;21(3):183-188.
115. Marquis S, Thomson C, Murray A. Assisting people with a low income to start and maintain their own community kitchens. Canadian Journal of Dietetic Practice and Research. 2001;62(3):130-132.
116. Newman VA, Thomson CA, Rock CL, Flatt SW, Kealey S, Bardwell WA, Caan BJ, Pierce JP. Achieving substantial changes in eating behavior among women previously treated for breast cancer—an overview of the intervention. Journal of the American Dietetic Association. 2005;105(3):382-391.
117. Porter J, Capra S, Watson G. An individualized food-skills programme: development, implementation and evaluation. Australian Occupational Therapy Journal. 2000;47(2):51-61.
118. Rustad C, Smith C. Nutrition knowledge and associated behavior changes in a holistic, short-term nutrition education intervention with low-income women. Journal of Nutrition Education and Behavior. 2013;45(6):490-498.
119. Shankar S, Klassen AC, Garrett-Mayer E, Houts PS, Wang T, McCarthy M, Cain R, Zhang L. Evaluation of a nutrition education intervention for women residents of Washington, DC, public housing communities. Health Education Research. 2007;22(3):425-437.
120. Swindle S, Baker SS, Auld GW. Operation Frontline: Assessment of longer-term curriculum effectiveness, evaluation strategies, and follow-up methods. Journal of Nutrition Education and Behavior. 2007;39(4):205-213.
121. Woodson JM, Braxton-Calhoun M, Benedict J. Food for health and soul: a curriculum designed to facilitate healthful recipe modifications to family favorites. Journal of Nutrition Education and Behavior. 2005;37(6):323-324.
122. Abbott P, Davison J, Moore L, Rubinstein R. Barriers and enhancers to dietary behaviour change for Aboriginal people attending a diabetes cooking course. Health Promotion Journal of Australia. 2010;21(1):33-38.
123. Engler-Stringer R, Berenbaum S. Food and nutrition-related learning in collective kitchens in three Canadian cities. Canadian Journal of Dietetic Practice and Research. 2006;67(4):178-183.
124. Engler-Stringer R, Berenbaum S. Exploring social support through collective kitchen participation in three Canadian cities. Canadian Journal of Community Mental Health. 2007;26(2):91-105.
214
125. Engler-Stringer R, Berenbaum S. Exploring food security with collective kitchens participants in three Canadian cities. Qualitative Health Research. 2007;17(1):75-84.
126. Foley W, Spurr S, Lenoy L, De Jong M, Fichera R. Cooking skills are important competencies for promoting healthy eating in an urban Indigenous health service. Nutrition and Dietetics. 2011;68(4):291-296.
127. Moldofsky Z. Meals made easy: A group program at a food bank. Social Work with Groups: A Journal of Community and Clinical Practice. 2000;23(1):83-96.
128. Snowdon WD. Asian cookery clubs: a community health promotion intervention. International Journal of Health Promotion and Education. 1999;37(4):135-136.
129. Tarasuk V, Reynolds R. A qualitative study of community kitchens as a response to income-related food insecurity. Canadian Journal of Dietetic Practice and Research. 1999;60(1):11-16.
130. Archuleta M, VanLeeuwen D, Halderson K, Wells L, Bock MA. Diabetes cooking schools improve knowledge and skills in making healthful food choices. Journal of Extension. 2012;50(2):2FEA6-2FEA6.
131. Barton KL, Wrieden WL, Anderson AS. Validity and reliability of a short questionnaire for assessing impact of cooking skills interventions. Journal of Human Nutrition and Dietetics. 2011;24(6):588–595.
132. Greene JC. Mixed methods in social inquiry: John Wiley & Sons; 2007.
133. The Good Foundation. Jamie's Ministry of Food Australia Melbourne The Good Foundation 2013 [cited 2013 October 28 ]. Available from: http://www.thegoodfoundation.com.au/ministry-of-food/.
134. Dobson S. Ministry of Food interim evaluation: December 2009. 2009. Available from: http://moderngov.rotherham.gov.uk/mgConvert2PDF.aspx?ID=35464.
135. Huntingtons-Kitchen. Huntington's Kitchen 2014 [cited 2014 8 August]. Available from: http://www.huntingtons-kitchen.org/.
136. The Jamie Oliver Food Foundation USA. Huntington's Kitchen n.d. [cited 2015 March 2015]. Available from: http://www.jamieoliverfoodfoundation.org/usa/huntington.
137. Prochaska JO, Velicer WF. The Transtheoretical Model of Health Behavior change. American Journal of Health Promotion. 1997;12(1):38-48.
138. Institute of Health & Society. Project:The YouCook study. Phase 2 research to support the evaluation and implementation of adult cooking skills interventions in the UK: pilot RCT with process and economic evaluation components.: Newcastle University 2013. Available from: http://www.ncl.ac.uk/ihs/research/project/4867.
215
139. Jamie's Ministry of Food. The Good Foundation 2012 [cited 2015 August 22]. Available from: http://www.jamieoliver.com/jamies-ministry-of-food-australia/good-foundation.php.
140. The Good Foundation. About us Melbourne The Good Foundation 2010 [cited 2012 February 2]. Available from: http://www.thegoodfoundation.com.au/about-us/.
141. Office of Economic and Statistical Research. Queensland Regional Profiles Ipswich City Based on local government area (2010). Queensland: Queensland Government, The State of Queensland (Queensland Treasury). 2011 May 27. Available from: http://statistics.oesr.qld.gov.au/qld-regional-profiles?region-type=LGA_10®ion-ids=6604.
142. Queensland Health. Self-reported Health Status 2009-2010: Local government area summary report. Brisbane Queensland Health, 2011 Available from: http://www.health.qld.gov.au
143. Queensland Government Statistician’s Office. Queensland regional profiles; resident profiles - people who live in the region. Ipswich City Local Government Area (LGA) compared with Queensland. Queensland Treasury and Trade, 2014 Available from: http://statistics.oesr.qld.gov.au/profiles/qrp/resident/pdf/T30SFTYN3J5L4G41CP7HIJ5B9DDUCRGUDUUXUKCAYDVZKEZWTO15F853HENRKEHHE0IZ0GCA22G3Y5731FZSZMXV8HPFYTSCEEYLVB6WQI8EDC6JUMKK0FCNI8EZA4NH/qld-regional-profiles-resident#view=fit&pagemode=bookmarks.
144. Australian Bureau of Statistics. 2011 Census QuickStats Ipswich: Australian Bureau of Statistics.,; 2013 [cited 2014 9 August]. Available from: http://www.censusdata.abs.gov.au/census_services/getproduct/census/2011/quickstat/310?opendocument&navpos=220.
145. National Institutes of Health. Theory at a glance: a guide for health promotion practice U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, 1995.
146. Glanz K, Rimer B, K., Frances ML. Health Behavior and Health Education. 3rd ed. San Francisco: Jossey-Bass; 2002.
147. Hagger MS. Theoretical integration in health psychology: Unifying ideas and complementary explanations. British Journal of Health Psychology. 2009;14(2):189-194.
148. Kolb DA. Experiential learning: experience as the source of learning and development. Englewood cliffs: Prentice Hall; 1984.
149. Bandura A. Health promotion from the perspective of social cognitive theory. Psychology & Health. 1998;13(4):623-649.
150. Baranowski T, Perry CL, Parcel GS. How individuals, environments and health behaviors interact. In: Glanz K, Rimer B, K., Frances ML, editors. Health Behavior and Health Education. 3rd ed. San Francisco: Jossey-Bass. 2002.
216
151. Bandura A. Organizational application of Social Cognitive Theory. Australian Journal of Management. 1988;13(2):275-302.
152. Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychological Review. 1977;84(2):191-215.
153. Crosby RA, Salazar LF, DiClemente RJ. Social Cognitive Theroy applied to health behavior. In: DiClemente RJ, Salazar LF, Crosby RA, editors. Health Behavior Theory for Public Health: Principles, Foundations, and Applications. Burlington: Jones & Bartlett Publishers. 2013.
154. Johnson RB, Onwuegbuzie AJ, Turner LA. Toward a definition of mixed methods research. Journal of mixed methods research. 2007;1(2):112-133.
155. Tashakkori A, Teddlie C. Sage handbook of mixed methods in social & behavioral research / edited by 2nd ed. Abbas Tashakkori, Teddlie C, editors. Los Angeles: SAGE Publications; 2010.
156. Crotty M. The foundations of social research: Meaning and perspective in the research process. St Leonards, Australia: Allen & Unwon 1998.
157. Swift JA, Tischler V. Qualitative research in nutrition and dietetics: getting started. Journal of Human Nutrition & Dietetics. 2010;23(6):559-566.
158. Greene JC, Hall JN. Dialectics and Pragmatism. In: Abbas Tashakkori, Teddlie C, editors. Sage Handbook of Mixed Methods in Social and Behavioral Research 2nd ed. Los Angeles: SAGE Publications. 2010.
159. Harris LM. Asking quantitative research questions. In: Minichiello V, Sullivan G, Greenwood K, Axford R, editors. Handbook for research methods in health sciences. Sydney: Pearson Education Australia. 1999.
160. Gerber R. The role of theory in social and health research. In: Minichiello V, Sullivan G, Greenwood K, Axford R, editors. Handbook for Research Methods in Health Sciences. Sydney: Pearson Education Australia. 1999.
161. Patton M. Qualitative research and evaluation methods 3rd ed. California: Sage; 2002.
162. Blaikie N. Designing social research : the logic of anticipation Malden: Polity Press; 2000
163. Thomas JR, Nelson JK, Silverman SJ. Research methods in physical activity: Human Kinetics; 2011.
164. Heard R, Harris LM. Experimental, quasi-experimental and correlational quanitative research designs. In: Minichiello V, Sullivan G, Greenwood K, Axford R, editors. Handbook for research methods in health sciences. Sydney: Pearson Education Australia. 1999.
217
165. Elliott SA, Brown JSL. What are we doing to waiting list controls? Behaviour Research and Therapy. 2002;40(9):1047-1052.
166. Cunningham J, Kypri K, McCambridge J. Exploratory randomized controlled trial evaluating the impact of a waiting list control design. BMC Medical Research Methodology. 2013;13(1):150.
167. Black N. Why we need observational studies to evaluate the effectiveness of health care. BMJ. 1996;312:1215-1218.
168. Pomerleau J, Lock, K, Knai, C, McKee, M. Interventions designed to increase adult fruit and vegetable intake can be effective: a systematic review of the literature. The Journal of Nutrition. 2005;135:2486-2495.
169. Queensland Health. Self-Reported Adult Health Status: Queensland 2009 survey report. Brisbane: Queensland Health 2009.
170. Salesforce. Salesforce. Sydney SFDC Australia Pty Ltd; 2014.
171. Parmenter K, Wardle J. Development of a general nutrition knowledge questionnaire for adults. European Journal of Clinical Nutrition. 1999;53(4):298-308.
172. Pettinger C, Holdsworth, M, Gerber, M. Meal patterns and cooking practices in Southern France. Public Health Nutrition. 2006;9(8):1020-1026.
173. University of Melbourne. The Household Income and Labour Dynamics in Australia (HILDA) Survey Melbourne 2011 [cited 2011 24 August]. Available from: http://melbourneinstitute.com/hilda/questionnaires/default.html.
174. Pettinger C, Holdsworth M, Gerber M. Psycho-social influences on food choice in Southern France and Central England. Appetite. 2004;42(3):307-316.
175. Rosenberg M, Schooler C, Schoenbach C. Self-Esteem and adolescent problems: modeling reciprocal effects. American Sociological Review. 1989;54(6):1004-1018.
176. World Health Organization. Obesity: preventing and managing the global epidemic: World Health Organization; 2000.
177. Australian Bureau of Statistics. 4363.0.55.002 National Health Survey: data reference package 2007-08 [Internet]. Australian Bureau of Statistics. 2010 [cited 2011 July 14]. Available from: http://www.abs.gov.au/ausstats/[email protected]/detailspage/4363.0.55.0022007-08.
178. van Gelder MMHJ, Bretveld RW, Roeleveld N. Web-based questionnaires: the future in epidemiology? American Journal of Epidemiology. 2010;172(11):1292-1298.
179. Qualtrics. Qualtics n.d. [cited 2012 14 May]. Available from: http://www.qualtrics.com/.
218
180. Fleming CM, Bowden M. Web-based surveys as an alternative to traditional mail methods. Journal of Environmental Management. 2009;90(1):284-292.
181. StataCorp. Stata: Release 12. Texas: StataCorp LP.; 2011.
182. Peacock J, Peacock P. Oxford handbook of medical statistics. Oxford Oxford University Press; 2010.
183. Kirkwood B, Sterne J. Essential medical statistics. 2nd ed. Oxford: Blackweel Science; 2005.
184. Pevalin D, Robson K. The Stata survival manual. UK: McGraw-Hill International; 2009.
185. Stata Press. Stata longitudinal/panel-data reference manual. 10 ed. Texas: Stata Press; 2007.
186. Lumley T, Diehr P, Emerson S, Chen L. The importance of the normality assumption in large public health data sets. Annual Review of Public Health. 2002;23(1):151-169.
187. Rosenberg M. Society and the adolescent self-image. edition R, editor. Middletown, CT: Wesleyan University Press.; 1989.
188. Queensland Health. Self Reported Health Status 2012 - HealthiIndicators: chronic disease and behavioural risk factors Queensland. Brisbane: Queensland Health, 2012 Available from: http://www.health.qld.gov.au/epidemiology/default.asp.
189. Magnusson K. Interpreting Cohen's d effect size an interactive visualization 2014 [cited 2015 August 21]. Available from: http://rpsychologist.com/d3/cohend/.
190. Hansen EC. Successful qualitative health research: a practical introduction. Crows Nest, NSW: Allen & Unwin; 2006.
191. Holland J, Thomson R, Henderson S. Qualitative longitudinal research: a discussion paper. London: London South Bank University, 2006.
192. Rice PL, Ezzy D. Qualitative research methods: a health focus. South Melbourne: Oxford Universtiy Press; 1999.
193. Burr V. Social constructionism. London: Routledge; 2003.
194. O'Shaughnessy M, Dtadler J. Media and society an introdution. 3rd edition ed. Melbourne: Oxford University Press; 2005.
195. Morse JM, Mitcham C. Exploring Qualitatively-derived Concepts: Inductive—Deductive Pitfalls. International Journal of Qualitative Methods. 2008;1(4):28-35.
196. Bandura A. Social foundations of thought and action: Englewood Cliffs; 1986.
219
197. Farrall S. What is qualitative longitudinal research? London School of Economics and Political Science Methodology Institute, 2006 Available from: http://www2.lse.ac.uk/methodology/aboutUs/QualitativePapers.aspx.
198. Neale B. Qualitative longitudinal reserach: an introduction to the timescape methods guides series: Economic and Social Reserach Council 2012 [cited 2012 22 October]. Available from: http://www.timescapes.leeds.ac.uk/assets/files/methods-guides/timescapes-methods-guides-introduction.pdf.
199. Shirani F, Henwood K. Continuity and change in a qualitative longitudinal study of fatherhood: relevance without responsibility. International Journal of Social Research Methodology. 2010;14(1):17-29.
200. Neale B, Flowerdew J. Time, texture and childhood: The contours of longitudinal qualitative research. International Journal of Social Research Methodology. 2003;6(3):189-199.
201. Corden A, Millar J. Time and Change: A Review of the Qualitative Longitudinal Research Literature for Social Policy. Social Policy and Society. 2007;6(04):583-592.
202. Saldana J. Longitudinal qualitative reserach. Walnut Creek: AltaMira Press 2003.
203. Gibbs L, Kealy M, Willis K, Green J, Welch N, Daly J. What have sampling and data collection got to do with good qualitative research? Australian and New Zealand Journal of Public Health. 2007;31(6):540-544.
204. Liamputtong P. Qualitative data analysis: conceptual and practical considerations. Health Promotion Journal of Australia. 2009;20(2):133-139.
205. Liamputtong P. Qualitative research methods. Forth ed. Melbourne: Oxford University Press; 2013.
206. Mason M. Sample size and saturation in PhD studies using qualitative interviews. Forum: Qualitative Social Research. 2010;11(3):1-19.
207. Harris JE, Gleason PM, Sheean PM, Boushey C, Beto JA, Bruemmer B. An introduction to qualitative research for food and nutrition professionals. Journal of the American Dietetic Association. 2009;109(1):80-90.
208. Green J, Willis K, Hughes E, Small R, Welch N, Gibbs L, Daly J. Generating best evidence from qualitative research: the role of data analysis. Australian and New Zealand Journal of Public Health. 2007;31(6):545-550.
209. Daly J, Willis K, Small R, Green J, Welch N, Kealy M, Hughes E. A hierarchy of evidence for assessing qualitative health research. Journal of Clinical Epidemiology. 2007;60(1):43-49.
210. Willis K, Daly J, Kealy M, Small R, Koutroulis G, Green J, Gibbs L, Thomas S. The essential role of social theory in qualitative public health research. Australian and New Zealand Journal of Public Health. 2007;31(5):438-443.
220
211. Flego A, Herbert J, Gibbs L, Waters E, Keating C, Swinburn B, Moodie M. The evaluation of Jamie’s Ministry of Food, Ipswich: final evaluation report. Melbourne Deakin University, 2014 Available from: http://www.jamieoliver.com/jamies-ministry-of-food-australia/.
212. Queensland Health. Self reported health status 2011–12 - Health indicators: chronic disease and behavioural risk factors, local government areas. Brisbane Queensland Health, 2013 Available from: http://www.health.qld.gov.au
213. Blignault I, Ritchie J. Revealing the wood and the trees: reporting qualitative research. Health Promotion Journal of Australia. 2009;20(2):140.
214. Ristovski-Slijepcevic S, Chapman GE, Beagan BL. Being a ‘good mother’: Dietary governmentality in the family food practices of three ethnocultural groups in Canada. Health. 2010;14(5):467-483.
215. Steger J. Who Thrived in 2012 in the Face of Fifty Shades, Hunger Games, Wimpy Kid and Jack of Diamonds. The Age. 2013 January 12.
216. Jamie's Ministry of Food. Jamie's Ministry of Food Australia 2012 [cited 2014 July 7]. Available from: http://www.jamieoliver.com/jamies-ministry-of-food-australia/.
217. Shepherd R, Raats MM. Attitudes and beliefs in food habits. In: Meiselman HL, MacFie HJ, editors. Food choice, acceptance and consumption. 1st ed. London: Blackie Academix & Professional. 1996.
218. Hartmann C, Dohle S, Siegrist M. Importance of cooking skills for balanced food choices. Appetite. 2013;65(0):125-131.
219. Spence F, Teijlingen ERv. A qualitative evaluation of community-based cooking classes in Northeast Scotland. International Journal of Health Promotion and Education. 2005;43(2):59-63.
220. Hall KD, Sacks G, Chandramohan D, Chow CC, Wang YC, Gortmaker SL, Swinburn BA. Quantification of the effect of energy imbalance on bodyweight. The Lancet. 2011;378(9793):826-837.
221. DiClemente RJ, Crosby RA. Health behavior theory for public health: principles, foundations, and applications. Burlington: Jones & Bartlett Publishers; 2013.
222. Kornides ML, Nansel TR, Quick V, Haynie DL, Lipsky LM, Laffel LMB, Mehta SN. Associations of family meal frequency with family meal habits and meal preparation characteristics among families of youth with type 1 diabetes. Child: Care, Health and Development. 2014;40(3):405-411.
223. Ministry of Social Development. The Social Report 2010 Wellington: Ministry of Social Development, 2010 Available from: http://socialreport.msd.govt.nz/tools/downloads.html.
221
224. Montano D, Kasprzrk D. The theory of reseasoned action and the theory of planned behaviour. In: Glanz K, Rimer B, K., Frances ML, editors. Health Behavior and Health Education. 3rd ed. San Francisco: Jossey-Bass. 2002.
225. Meah A, Watson M. Saints and Slackers: challenging discourses about the decline of domestic cooking. Sociological Research Online. 2011;16(2).
226. Boeing H, Bechthold A, Bub A, Ellinger S, Haller D, Kroke A, Leschik-Bonnet E, Müller MJ, Oberritter H, Schulze M. Critical review: vegetables and fruit in the prevention of chronic diseases. European Journal of Nutrition. 2012;51(6):637-663.
227. D.H., R.L.W. The Big Mac index: The Economist Newspaper Limited 2014 [cited 2014 October 29]. Available from: http://www.economist.com/content/big-mac-index.
228. Devine CM, Jastran M, Jabs J, Wethington E, Farell TJ, Bisogni CA. “A lot of sacrifices:” Work–family spillover and the food choice coping strategies of low-wage employed parents. Social Science and Medicine. 2006;63(10):2591-2603.
229. Neysmith SM, Reitsma-Street M, Baker-Collins S, Porter E, Tam S. Provisioning responsibilities: how relationships shape the work that women do. Canadian Review of Sociology/Revue canadienne de sociologie. 2010;47(2):149-170.
230. Marquis M. Exploring convenience orientation as a food motivation for college students living in residence halls. International Journal of Consumer Studies. 2005;29(1):55-63.
231. Oliver J. Jamie's 30-minute meals. UK: Penguin; 2011.
232. Oliver J, Matthew M, Pointer N, Britain G. Jamie's 15 minute meals: Freemantle Media; 2012.
233. jamieoliver.com. Jamie wins animal welfare award 2013 [cited 2014 September 19]. Available from: http://www.jamieoliver.com/news-and-features/news/jamie-wins-animal-welfare-award/.
234. Jamie Oliver Foundation. Jamie's Ministry of Food UK 2013 [cited 2013 August 13]. Available from: http://www.jamieoliver.com/jamies-ministry-of-food/.
235. Hollows J, Jones S. ‘At least he’s doing something’: Moral entrepreneurship and individual responsibility in Jamie’s Ministry of Food. European Journal of Cultural Studies. 2010;13(3):307-322.
236. Short F. Kitchen secrets: The meaning of cooking in everyday life. London: Berg; 2006.
237. Woolworths. Woolworths is the official food partner for Jamie Oliver's Ministry of Food Australia: Woolworths; 2014. Available from: http://www.woolworthslimited.com.au/page/The_Newsroom/Latest_News/Woolworths_is_the_official_food_partner_for_Jamie_Olivers_Ministry_of_Food__Australia/.
222
238. ABC. Woolworths' Jamie Oliver Campaign Success 'A Furphy', According to AUSVEG: ABC News; 2014 [cited 2015 February 12]. Available from: http://www.abc.net.au/news/2014-07-05/woolworths-jamie-oliver-campaign-questioned-by-farmers/5570060.
239. Long W. Farmers slugged for Woolworths Jamie Oliver campaign ABC Rural; 2014 [cited 2015 February 12]. Available from: http://www.abc.net.au/news/2014-06-06/nrn-ausveg-vs-woolworths/5505808.
240. Long W. Jamie Oliver says he can't change Woolworths treatment of farmers: ABC Rural; 2014 [cited 2015 February 12]. Available from: http://www.abc.net.au/news/2014-06-16/nrn-jamie-responds/5526426.
241. Chapman S. Does celebrity involvement in public health campaigns deliver long term benefit? Yes. British Medical Journal. 2012;345.
242. Victoria Department of Health. Creating a vibrant healthy eating culture in Victoria The Victorian Healthy Eating Enterprise 2014 [cited 2014 19 September]. Available from: http://docs.health.vic.gov.au/docs/doc/Victorian-Healthy-Eating-Enterprise.
243. jamieoliver.com. Jamie Oliver and Victoria vovernment to tackle diet related disease together.: Jamie Oliver Food Foundation.,; 2012 [cited 2014 19 September]. Available from: http://www.jamieoliver.com/jamies-ministry-of-food/news.php?title=jamie-oliver-victorian-government-to-tac.
244. Jamie's Ministry of Food. Jamies Ministry of Food Australia launches 2010 [cited 2014 September 19]. Available from: http://www.jamieoliver.com/jamies-ministry-of-food-australia/news.php?title=jamie-s-ministry-of-food-australia-launc.
245. The Good Foundation. First Australian Ministry of Food centre opens in Ipswich [Media release]. The Good Foundation; 2011. Available from: http://www.thegoodfoundation.com.au/wp-content/uploads/2011/05/Jamie-Oliver-Launches-First-Ministry-of-Food-Australia_2.pdf.
246. Agudo A. Measuring Intake of Fruit and Vegetables Kobe: World Health Organization, 2004 September 1-3. Available from: http://whqlibdoc.who.int/publications/2005/9241592826_eng.pdf.
247. Roark RA, Niederhauser VP. Fruit and vegetable intake: issues with definition and measurement. Public Health Nutrition. 2013;16(01):2-7.
248. Mays N, Pope C. Qualitative research: rigour and qualitative research. BMJ. 1995;311(6997):109-112.
249. Morse JM, Barrett M, Mayan M, Olson K, Spiers J. Verification strategies for establishing reliability and validity in qualitative research. International Journal of Qualitative Methods. 2008;1(2):13-22.
223
250. Australian Bureau of Statistics. Daily intake of fruit and vegetables [Internet]. Canberra Australian Bureau of Statistics. 2013. Available from: http://www.abs.gov.au/ausstats/[email protected]/Lookup/by%20Subject/4338.0~2011-13~Main%20Features~Daily%20intake%20of%20fruit%20and%20vegetables~10009.
251. Council. IC. HAPI Ipswich 2013 [cited 2014 20 February]. Available from: http://www.ipswich.qld.gov.au/residents/hapi_ipswich/.
252. Ipswich. Co. Two year campaign to create happy healthy people in Ipswich 2011 [cited 2015 August 24]. Available from: http://www.ipswich.qld.gov.au/about_council/media/articles/2011/two-year-campaign-to-create-happy-healthy-people-in-ipswich.
253. The Good Foundation. Continued government funding for Jamie’s Ministry of Food Queensland [Media release]. Jamie’s Ministry of Food Australia; 2014. Available from: http://www.thegoodfoundation.com.au/wp-content/uploads/2010/08/Media-Release_Jamies-Ministry-of-Food-Qld-announces-continued-Government-funding_27-05-2014.pdf.
254. Drummond MF, Sculpher MJ, Torrance GW, O'Brien BJ, Stoddart GL. Methods for the economic evaluation of health care programmes. 3rd ed. Oxford Oxford University Press 2005.
255. Meltzer MI. Introduction to health economics for physicians. The Lancet. 2001;358(9286):993-998.
256. Lock K, Pomerleau J, Causer L, Altmann DR, McKee M. The global burden of disease attributable to low consumption of fruit and vegetables: implications for the global strategy on diet. Bulletin of the World Health Organization. 2005;83(2):100-108.
257. Haby MM, Vos T, Carter R, Moodie M, Markwick A, Magnus A, Tay-Teo KS, Swinburn B. A new approach to assessing the health benefit from obesity interventions in children and adolescents: the assessing cost-effectiveness in obesity project. International Journal of Obesity. 2006;30(10):1463-1475.
258. Carter R, Vos T, Moodie M, Haby M, Magnus A, Mihalopoulos C. Priority setting in health: origins, description and application of the Australian Assessing Cost-Effectiveness initiative. Expert Review of Pharmacoeconomics & Outcomes Research. 2008;8(6): 593–617.
259. Cobiac LJ, Vos T, Veerman JL. Cost-effectiveness of interventions to promote fruit and vegetable consumption. PloS One. 2010;5(11):e14148.
260. Backholer K, Mannan HR, Magliano DJ, Walls HL, Stevenson C, Beauchamp A, Shaw JE, Peeters A. Projected socioeconomic disparities in the prevalence of obesity among Australian adults. Australian and New Zealand Journal of Public Health. 2012;36(6):557-563.
224
261. Beauchamp A, Backholer K, Magliano D, Peeters A. The effect of obesity prevention interventions according to socioeconomic position: a systematic review. Obesity Reviews. 2014;15(7):541-554.
262. Cookson R, Drummond M, Weatherly H. Explicit incorporation of equity considerations into economic evaluation of public health interventions - reply to Richardson and Shiell. Health Economics, Policy and Law. 2009;4(2):261-263.
225
Appendices
Appendix 1. Quality Assessment Tool for Quantitative Studies
Appendix 1 presents the “Quality Assessment Tool for Quantitative Studies” (78) use to score each study.
226
QUALITY ASSESSMENT TOOL FOR QUANTITATIVE STUDIES
COMPONENT RATINGS
A) SELECTION BIAS
(Q1) Are the individuals selected to participate in the study likely to be representative of the target population? 1 Very likely
2 Somewhat likely
3 Not likely
4 Can’t tell
(Q2) What percentage of selected individuals agreed to participate? 1 80 - 100% agreement
2 60 – 79% agreement
3 less than 60% agreement
4 Not applicable
5 Can’t tell
RATE THIS SECTION STRONG MODERATE WEAK
See dictionary 1 2 3
B) STUDY DESIGN
Indicate the study design 1 Randomized controlled trial
2 Controlled clinical trial
3 Cohort analytic (two group pre + post)
4 Case-control
5 Cohort (one group pre + post (before and after))
6 Interrupted time series
7 Other specify ____________________________
8 Can’t tell
Was the study described as randomized? If NO, go to Component C. No Yes
If Yes, was the method of randomization described? (See dictionary) No Yes
If Yes, was the method appropriate? (See dictionary) No Yes
RATE THIS SECTION STRONG MODERATE WEAK
See dictionary 1 2 3
1
C) CONFOUNDERS
(Q1) Were there important differences between groups prior to the intervention? 1 Yes
2 No
3 Can’t tell
The following are examples of confounders: 1 Race
2 Sex
3 Marital status/family
4 Age
5 SES (income or class)
6 Education
7 Health status
8 Pre-intervention score on outcome measure
(Q2) If yes, indicate the percentage of relevant confounders that were controlled (either in the design (e.g. stratification, matching) or analysis)?
1 80 – 100% (most)
2 60 – 79% (some)
3 Less than 60% (few or none)
4 Can’t Tell
RATE THIS SECTION STRONG MODERATE WEAK
See dictionary 1 2 3
D) BLINDING
(Q1) Was (were) the outcome assessor(s) aware of the intervention or exposure status of participants? 1 Yes
2 No
3 Can’t tell
(Q2) Were the study participants aware of the research question? 1 Yes
2 No
3 Can’t tell
RATE THIS SECTION STRONG MODERATE WEAK
See dictionary 1 2 3
E) DATA COLLECTION METHODS
(Q1) Were data collection tools shown to be valid? 1 Yes
2 No
3 Can’t tell
(Q2) Were data collection tools shown to be reliable? 1 Yes
2 No
3 Can’t tell
RATE THIS SECTION STRONG MODERATE WEAK
See dictionary 1 2 3
2
F) WITHDRAWALS AND DROP-OUTS
(Q1) Were withdrawals and drop-outs reported in terms of numbers and/or reasons per group? 1 Yes
2 No
3 Can’t tell
4 Not Applicable (i.e. one time surveys or interviews)
(Q2) Indicate the percentage of participants completing the study. (If the percentage differs by groups, record the lowest).
1 80 -100%
2 60 - 79%
3 less than 60%
4 Can’t tell
5 Not Applicable (i.e. Retrospective case-control)
RATE THIS SECTION STRONG MODERATE WEAK
See dictionary 1 2 3 Not Applicable
G) INTERVENTION INTEGRITY
(Q1) What percentage of participants received the allocated intervention or exposure of interest? 1 80 -100%
2 60 - 79%
3 less than 60%
4 Can’t tell
(Q2) Was the consistency of the intervention measured? 1 Yes
2 No
3 Can’t tell
(Q3) Is it likely that subjects received an unintended intervention (contamination or co-intervention) that may influence the results?
4 Yes
5 No
6 Can’t tell
H) ANALYSES
(Q1) Indicate the unit of allocation (circle one) community organization/institution practice/office individual
(Q2) Indicate the unit of analysis (circle one) community organization/institution practice/office individual
(Q3) Are the statistical methods appropriate for the study design? 1 Yes
2 No
3 Can’t tell
(Q4) Is the analysis performed by intervention allocation status (i.e. intention to treat) rather than the actual intervention received?
1 Yes
2 No
3 Can’t tell
3
GLOBAL RATING COMPONENT RATINGS Please transcribe the information from the gray boxes on pages 1-4 onto this page. See dictionary on how to rate this section.
A SELECTION BIAS STRONG MODERATE WEAK
1 2 3
B STUDY DESIGN STRONG MODERATE WEAK
1 2 3
C CONFOUNDERS STRONG MODERATE WEAK
1 2 3
D BLINDING STRONG MODERATE WEAK
1 2 3
E DATA COLLECTION METHOD
STRONG MODERATE WEAK
1 2 3
F WITHDRAWALS AND DROPOUTS STRONG MODERATE WEAK
1 2 3 Not Applicable
GLOBAL RATING FOR THIS PAPER (circle one):
1 STRONG (no WEAK ratings)
2 MODERATE (one WEAK rating)
3 WEAK (two or more WEAK ratings)
With both reviewers discussing the ratings:
Is there a discrepancy between the two reviewers with respect to the component (A-F) ratings?
No Yes
If yes, indicate the reason for the discrepancy
1 Oversight
2 Differences in interpretation of criteria
3 Differences in interpretation of study
Final decision of both reviewers (circle one): 1 STRONG 2 MODERATE 3 WEAK
4
Appendix 2. Jamie’s Ministry of Food Australia manifesto
Appendix 2 presents Jamie’s Ministry of Food Australia manifesto sourced from: http://www.jamieoliver.com/jamies-ministry-of-food-australia/
This manifesto explains a really effective way to empower the public to make short and long-term changes. I believe it will radically improve the health prospects and social welfare of the public, regardless of age and class. When I set out to establish the Ministry of Food campaign, I wanted to dig into some issues which have been nagging at me since I filmed my School Dinners television series in the UK, the biggest ones being: why aren’t people cooking at home anymore and why don’t kids recognise everyday vegetables? I wanted to find out more about how people eat at home and why our diet is leading to a huge increase in obesity in so many countries around the world. I wanted to see if I could find a way to get people cooking again, to give them the tools to help themselves by learning some basic food skills, to feel confident in the kitchen and enjoy eating their own meals. I then wanted to take it a step further by asking them to take this knowledge and Pass It On; sharing their newfound skills with other people. Diet and health are massive social issues. In the UK, obesity already costs the National Health Service more than smoking: £4.2 billion vs. £2.7 billion per year. Over 9,000 people in Britain die prematurely each year due to health conditions caused by being overweight; cancer, heart disease, stroke and diabetes are the most common. This will be the first generation in which children are predicted to die at a younger age than their parents. In Australia the figures are equally shocking. Figures given by the National Preventative Health Taskforce show that being overweight or obese affects over 60% of Australian adults and 25% of Australian children. Nearly one in three Indigenous Australian adults is obese. The total financial cost in Australia of obesity alone, not including overweight people, was estimated at $8.3 billion in 2008. The most recent projections indicate that there will be an extra 6.7 million obese Australians by 2025. This frightening statistic clearly demonstrates an urgent need for action! Recent UK government research into obesity found that “lack of knowledge, confidence and skil ls is the main barrier which stops parents cooking from scratch”.
231
223
But, despite this, obesity strategies rarely contain any measures specifically designed to address the root problem of obesity: a complete lack of cooking knowledge. Cooking skills used to be passed down through the generations but now that chain has been broken. More women go out to work, fewer countries have compulsory cooking lessons at school, and we’ve now got about two generations of people who have not been taught the basics. In my Ministry of Food television series, and throughout the campaign, my goal was to help people tackle the obvious problems of lack of money and time. But I found I had to start by addressing the more fundamental issues like a lack of cooking ability and food knowledge. But my team and I found that, armed with a little bit of knowledge and confidence, people very quickly started to change the way they were eating. On the Ministry of Food pages on my website (www.jamieoliver.com/jamies-ministry-of-food), we have provided what I think are some of the basic recipes, videos and kitchen advice to help get people on the first rung of the ladder. One thing is for sure; if a better diet is one of the keys to sorting out our health and obesity problems, then governments worldwide urgently need to put cooking at the top of their agendas. The original Ministry of Food, set up in the UK during the Second World War, was a great model. Set up to manage wartime rationing and food shortages, it is to me, a great example of how a government can quickly and successfully educate its people in times of crisis. It bombarded the population with help and advice, ensuring that people knew how to feed themselves properly and as well as possible until things were back to normal. They even went as far as sending cooks and advisors out into the community to give practical demonstrations and advice on how best to use the available resources. Australia also had a wartime government department dedicated to food issues; the Commonwealth Rationing Commission. Limited rationing was introduced in 1942 and wartime recipes were circulated in magazines, pamphlets and on government-sponsored radio programmes alongside really down-to-earth, sensible advice on how to use the available ingredients. The Australian Government reportedly feared that rationing would result in deterioration in health on the home front but the outcome was positive. As in the UK, the period saw a dip in the numbers of people suffering from diet related problems. These days the food crisis isn’t about rationing, it’s about people not understanding food, or cooking. It’s time we got back to this idea of teaching the public by getting professional cookery teachers around the world trained up to teach their communities. They need to reach the people who don’t cook, don’t watch cookery programmes, don’t buy cookbooks, and simply don’t know about food. If they give it a try and people start to feel inspired about cooking, it could make a massive difference, and quickly. All it needs is proper funding and support from Federal, State or Local Governments, specifically the departments of health, skills or education.
232
224
So I’m asking governments across the world to help make this happen through a few simple steps: 1. SET UP A FOOD CENTRE IN EVERY TOWN: These should be friendly, local walk-in centres, kitted out with the sort of equipment you’d find in homes, offering basic cooking courses for the local community. These centres can then be places where local people can meet professional cookery teachers face-to-face, get hands on experience with food, and take a course in basic cooking. They can also provide practical hubs from which public health trainers and other community workers can offer advice on healthcare or local government services. For these “Ministry of Food Centres”, we’ve put together a set of simple set of recipes that cover the basic food skills: chopping, frying, roasting and baking. These recipes are inspiring and affordable. After a course of lessons, most people will know how to follow a recipe and be able to transfer what they’ve learned to other dishes and ingredients. And they’ll have those skills for the rest of their lives. Here in the UK, we estimate a food centre costs approximately £150,000 to set up. If a government were to set up150 across the nation, this would cost approximately £22.5 million. A small price to pay when you consider the rising cost of obesity. Running costs could come from local government and public health budgets. There should also be a commitment to support any new centre for a minimum of 3 years to allow it to “bed into” its community and for the scheme to have a sustained and sustainable long-term effect on the health and well-being of the population. 2. MAKE SURE THE CENTRES ARE STAFFED BY TRAINED FOOD TEACHERS: These individuals – who will ideally be from the local community – will be trained in my Ministry of Food cookery course. They’ll teach basic food skills like how to shop for ingredients, how to use kitchen equipment, how to make the most of local and seasonal food, and how to make good simple meals on a tight budget. 3. PUT COOKS OUT IN THE COMMUNITY: If we look back at the original Ministry of Food’s wartime model, it is clear that we’ve got to get cooks out into the community to spread the message by working with charitable bodies, community groups, local services, and businesses. Sending people out to teach in the community or to Pass It On, requires only a set of cooking equipment, some mobile gas burners, ingredients and wheels to get there. Governments should therefore encourage, and seek to fund, the training of a new wave of home economists to work within communities. Properly qualified teachers working in schools and the community will deliver rapid change. 4. SUPPORT BUSINESSES TO SPREAD THE CAMPAIGN: Our Ministry of Food Centres are often contacted by businesses who want to initiate cooking lessons for their workforce and their families. They see it as a great way to promote employee health, team building, community relations, and also as a tool to break down the barriers separating departments in big businesses. Offering tax incentives to businesses that are prepared to invest in the health of their staff through cookery would encourage many more to take part and could make a significant contribution to the long-term health of the nation’s workforce and economy.
233
225
5. INVEST IN MOBILE FOOD CENTRES FOR DEPRIVED NEIGHBOURHOODS AND REGIONS: In the UK a group called Focus on Food runs a small fleet of cooking buses. These buses visit schools and introduce cooking to kids, parents, and teachers. This is a fantastic scheme for communities in the remote parts of the country, or those in very deprived inner-city neighbourhoods with little access to facilities. The visits inspire people to keep cooking long after the bus has left and are a fantastic marketing tool for cooking. Operating flat out, a bus could run over 11,500 sessions a year. They’ve got great reach and are so popular that I strongly believe mobile cooking classrooms should be introduced in other countries. This sort of scheme would be hugely beneficial to larger countries, like Australia or the US, where people are often spread out across great distances. 6. FUND ADULT COOKERY CLASSES: Over the last few years, I have consistently called for better food education for young adults and parents because it’s this group of people who have been left behind. Sure, there are cookery courses out there, but the majority of them are for people who already know quite a bit about cooking and want to take it further. If you don’t know how to boil an egg, roast a chicken or chop vegetables, it’s pretty difficult to find a course for adults at that level. This is especially true for people on low incomes or in areas of high deprivation. Our governments need to recognise cooking as an essential life skill and a key part of adult learning, and start funding courses at Food Centres, and further education colleges. People working in the community on health and obesity need this training too. Knowing how to make food fun and cooking appealing would help them get better results. Without a doubt, investing in food skills now would save millions in healthcare down the road. 7. GET KIDS COOKING PROPERLY Government research in the UK already acknowledges that kids who get a chance to cook at school often end up changing the way their family eats at home (for the better). So it makes sense to put proper cooking lessons on the school curriculum to ensure kids are learning about food and how to cook as early as possible. Teacher training courses also urgently need to include cooking. Like a lot of young adults, young teachers are unlikely to know how to cook themselves. How can we expect them to teach cooking lessons if they don’t know anything about food themselves? Getting food education into schools does cost money, but if supermarkets were encouraged to adopt their local school they could then provide ingredients for free or, at least, at cost. Not only would they be doing something wonderful, they’d be educating potential future customers. There’s a serious lack of knowledge about food and cooking, and the result is the modern-day epidemic of obesity and bad health we are currently facing. I believe if the actions I’ve mentioned above are carried through they will represent a major step towards tackling these problems once and for all. We simply can’t wait any longer. Jamie Oliver February 2010
234
Appendix 3. JMoF protocol journal article The protocol paper for Jamie’s Ministry of Food Australian can be found at: http://www.biomedcentral.com/1471-2458/13/411
235
STUDY PROTOCOL Open Access
Methods for the evaluation of the Jamie OliverMinistry of Food program, AustraliaAnna Flego1*, Jessica Herbert1, Lisa Gibbs2, Boyd Swinburn3,4, Catherine Keating1, Elizabeth Waters2
and Marj Moodie1
Abstract
Background: Community-based programs aimed at improving cooking skills, cooking confidence and individualeating behaviours have grown in number over the past two decades. Whilst some evidence exists to support theireffectiveness, only small behavioural changes have been reported and limitations in study design may haveimpacted on results.This paper describes the first evaluation of the Jamie Oliver Ministry of Food Program (JMoF) Australia, in Ipswich,Queensland. JMoF Australia is a community-based cooking skills program open to the general public consisting of1.5 hour classes weekly over a 10 week period, based on the program of the same name originating in the UnitedKingdom.
Methods/Design: A mixed methods study design is proposed. Given the programmatic implementation of JMoF inIpswich, the quantitative study is a non-randomised, pre-post design comparing participants undergoing theprogram with a wait-list control group. There will be two primary outcome measures: (i) change in cookingconfidence (self-efficacy) and (ii) change in self-reported mean vegetable intake (serves per day). Secondaryoutcome measures will include change in individual cooking and eating behaviours and psycho-social measuressuch as social connectedness and self-esteem. Repeated measures will be collected at baseline, programcompletion (10 weeks) and 6 months follow up from program completion. A sample of 250 participants per groupwill be recruited for the evaluation to detect a mean change of 0.5 serves a day of vegetables at 80% power (0.5%significance level). Data analysis will assess the magnitude of change of these variables both within and betweengroups and use sub group analysis to explore the relationships between socio-demographic characteristics andoutcomes.The qualitative study will be a longitudinal design consisting of semi-structured interviews with approximately 10-15 participants conducted at successive time points. An inductive thematic analysis will be conducted to exploresocial, attitudinal and behavioural changes experienced by program participants.
Discussion: This evaluation will contribute to the evidence of whether cooking programs work in terms ofimproving health and wellbeing and the underlying mechanisms which may lead to positive behaviour change.
Trial registration: Australian and New Zealand Trial registration number: ACTRN12611001209987.
* Correspondence: [email protected] Health Economics, Faculty of Health, Deakin University,Melbourne, AustraliaFull list of author information is available at the end of the article
© 2013 Flego et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.
Flego et al. BMC Public Health 2013, 13:411http://www.biomedcentral.com/1471-2458/13/411
BackgroundCooking skills programs have been described as a prac-tical illustration of how to simultaneously change know-ledge, attitudes and behaviours around healthy eatingpractices [1]. Interest in cooking has been stimulated bymedia attention afforded to celebrity chefs and popularprime time television cooking programs. However, theneed to promote cooking skills to individuals has in partstemmed from a decline in the traditional pathways bywhich individuals learn to cook [2], and from the hy-pothesis that a decline in cooking skills may have con-tributed to the growth in nutrition-related disease incertain sub-sections of western populations [3]. InAustralia, Winkler investigated the relationship betweena lack of confidence to cook and the purchasing of fruitand vegetables. The author concluded that cooking skillsmay contribute to socio economic differences in dietaryintake and that promotion of such could be a usefulstrategy to improve fruit and vegetable intake [4].In the past two decades, there has been an increase in
the number of not-for-profit community-based cookingskills programs both in Australia and internationally[5-8]. Such programs have been conducted in a varietyof community and institutional settings, targeting differ-ent sub-populations and varying in purpose; however,they are predominantly aimed at increasing confidenceto cook, promoting healthy eating, addressing health in-equalities and increasing access to healthy food [9].Whilst there is emerging evidence of the effectiveness
of these adult programs in terms of increasing confi-dence to cook and creating positive dietary change, thisevidence, to date, has been based on small scale evalua-tions that are subject to methodological limitations [9].In a recent systematic review of the effectiveness of adultcommunity cooking programs conducted in the UnitedKingdom, only one evaluation was identified as suitablyrobust to provide reliable findings with respect to pro-gram effectiveness [10]. This highlights the need formore rigorous, larger scale studies to examine the rangeof impacts and outcomes of cooking skills programs andthe underlying potential mechanisms for change in indi-vidual behaviour. At the same time, study designs musttake account of the challenges associated with evaluationin community settings and be practical, feasible and sen-sitive to all stakeholders involved.
The Jamie Oliver Ministry of Food program, AustraliaThis methods paper describes the evaluation frameworkand design for the Jamie Oliver Ministry of Food program(JMoF) Australia, Ipswich site. The JMoF program wasoriginally developed by Jamie Oliver, a renowned celebritychef and food author based in the United Kingdom (UK).JMoF Australia has been specifically adapted for theAustralian setting. It is a community focused program that
teaches basic cooking skills and good nutrition to non-cooks. It consists of 10, weekly, 1.5 hour cooking skillsclasses aimed at getting people of all ages and back-grounds cooking simple, fresh, healthy food quickly andeasily [11]. Participants pay AUD10 per class and, wherethis may pose a barrier to entry, subsidies are madeavailable.JMoF was pioneered as a community-based cooking
skills program in Rotherham, UK in 2008 and since then,other centres have opened in Bradford and Leeds and amobile centre in the North West of the UK. These cen-tres were reliant on funding mostly from local councilsand to a lesser extent, charities and the private sector.The first Australian site opened in Ipswich in the stateof Queensland in April 2011 co-funded by a local phil-anthropic non-government organisation (NGO), TheGood Foundation (TGF), and the Queensland Depart-ment of Health. Ipswich was intentionally chosen givenits significant low socio-economic status population [12]and increasing levels of overweight and obesity [13].
Objectives of the JMoF program AustraliaConsultation occurred between TGF, Queensland Health(as program co-founder) and the program evaluationteam to describe program objectives of the JMoF pro-gram Australia in sufficient detail to be tested in an ap-plied evaluation. The following program objectivesresulted:
1. To provide opportunities, to people of different ageand demographic background, to experience andlearn how to cook healthy meals quickly andcheaply.
2. To increase program participants’ cooking skills,knowledge and self-efficacy.
3. To increase program participants’ enjoyment of foodand social connectedness.
Theoretical perspectivesA program logic model was developed as a frameworkto describe the potential pathways to behaviour change,and in turn to guide evaluative enquiry (Figure 1).Whilst some steps along the logic pathway weregrounded by emerging or convincing evidence, otherareas were backed by limited evidence, thereby requiringfurther hypothesis testing.Theoretical frameworks were not explicitly stated for
the JMoF Program. However, Carahar and Lang, 1999[2] have identified theoretical perspectives specific tocooking skills that are in keeping with the objectives ofthe JMoF program and its evaluation - cooking skillsempower individuals in preparation for healthy eating,encourage self-esteem and provide opportunities for leis-ure and enjoyment.
Flego et al. BMC Public Health 2013, 13:411 Page 2 of 8http://www.biomedcentral.com/1471-2458/13/411
Other theories that resonate with the program includeKolb’s concept of experiential learning [14] which identi-fies the importance of empowering participants withpractical “get your hands dirty” experience in learning tocook from scratch as a basis for skill acquisition, andBandura’s Social Cognitive Theory [15] which states thatchanges in attitudes and beliefs and the development ofself- efficacy (i.e. confidence in cooking) are central toinfluencing behaviour change. Bandura’s Social LearningTheory [16] also states that modelling is an importantcomponent of the learning process and that opportun-ities for practising learned behaviours and positivereinforcement are needed for learning to take place. Animportant element in the learning process is role modelcredibility [17].
Methods/DesignThe evaluation will be conducted over a 2.5 year periodfrom late 2011 to early 2014. The evaluation was ap-proved by the Deakin University Human Research EthicsCommittee (HEAG-H 117_11) in October 2011. Evalu-ation project governance will be provided by a referencegroup (comprising personnel from the TGF team andthe research team members) which will meet twice
yearly to oversee the project. A representative fromQueensland Health will be invited to attend ReferenceGroup meetings, when appropriate.The evaluation will use a questions-oriented approach
[18] derived primarily from the JMoF Program objec-tives. It will also incorporate additional economic ques-tions of relevance to potential government funders andprogram stakeholders. A longitudinal mixed methodsevaluation design will be employed. The quantitative andqualitative components will be conducted sequentially,with baseline quantitative data informing sampling forthe initial qualitative interviews. Each component willthen be analysed independently, with merging of dataoccurring at the interpretation stage [19].
Quantitative study
Research questionsThe quantitative component of the evaluation will an-swer the following research questions:
1. Does the JMoF program increase participants’ skills,knowledge, attitudes, enjoyment and satisfaction ofcooking and cooking confidence (self-efficacy)?
JMOF Program Australia
IncreaseCooking skills and Knowledge
Increase Cooking Self -Efficacy (confidence to cook)
Increase Enjoyment & Satisfaction of Cooking and Eating
Increase Global Self Esteem
Increase Social Connectedness around Cooking and Eating
Increase Frequency of Cooking meals
IncreaseAffordability of Meals
Improve Attitude to Cooking/ Willingness to try new food
Increase HealthyEating
Decrease risk factors for ill health
Key
Emerging/ convincing evidence
Limited evidence
Figure 1 JMOF Australia program logic model.
Flego et al. BMC Public Health 2013, 13:411 Page 3 of 8http://www.biomedcentral.com/1471-2458/13/411
2. Does the JMoF program result in broader positiveoutcomes for participants in terms of behaviourchange to a healthier diet, more affordable healthymeals, improved self-esteem and socialconnectedness?
Outcome measuresThere will be two primary outcome measures: a change incooking confidence (self-efficacy) and a change in self-reported mean vegetable intake (serves per day). Second-ary outcome measures will include change in self-reportedmeasures of: (i) mean daily fruit intake, (ii) mean weeklytakeaway/fast food intake, (iii) frequency of cooking themain meal from basic ingredients, (iv) nutrition know-ledge, (v) attitudes towards cooking, (vi) willingness to trynew foods and (vii) enjoyment and satisfaction of cooking.Change in psycho-social measures such as (viii) globalself-esteem and (ix) social connectedness in relation tocooking and eating will be measured as will (x) a changein participant’s total expenditure on food.
Study designA quasi-experimental pre-post design will consist of anintervention group of participants undergoing the JMoFprogram and a control group comprising of participantsfrom the program waitlist who are waiting for at least 10weeks until program entry. Recruitment to each group willbe based on program start date and will not be subject torandomisation. Randomisation was not possible as itwould not allow participants any choice as to when andwith whom they participated in the program – which areimportant aspects of the JMoF program design [20].Intervention participants will be surveyed at three
time-points: program commencement, program comple-tion (10 weeks) and at six months post program comple-tion. Controls drawn from the waitlist will be surveyedat two time-points: 10 weeks prior to program com-mencement and on completion of their 10 weeks on thewaitlist (which will correspond to their program entry).A time-three measurement will not be obtained fromcontrols as it was considered neither feasible nor accept-able for the waitlist controls to have to wait a further sixmonths before entering the program (equivalent to theintervention follow-up period); this potentially wouldlead to a high drop-out rate both from the evaluationand the program itself. However, for one of the primaryoutcome measures, vegetable intake, Queensland state-wide monitoring data will be used as a proxy time-threemeasure for the control group.
Survey instrumentIn collaboration with key stakeholders, a quantitativemeasurement tool was developed. The self- adminis-tered questionnaire was designed to be completed in
approximately 15 minutes. Given the lack of validatedand reliable survey tools which can accurately measurethe impacts and outcomes of cooking skills programs invarying population groups [21], a prototype question-naire was designed to address the unique objectives ofthe evaluation. Where suitable, specific questions wereincorporated that have been previously used to measurethe impact of cooking skills programs, particularly oncooking confidence and cooking behaviours such asthose used by Barton et al, 2011 [21].To measure the primary outcome of cooking confidence
as a proxy for cooking self-efficacy, questions were devel-oped addressing confidence in relation to specific cookingskills based on Short’s work [22] and Barton et al. 2011work [21]. These items are presented on a 5 point Likertconfidence scale ranging from ‘not at all confident’ to‘extremely confident’.The other primary outcome measure of change in vege-
table intake will be captured through self-report questionsof vegetable intake (serves per day) and aligned with base-line measurement questions of the same nature used byQueensland Health in its population-based self-reportedhealth surveys [23]. Survey items addressing specific sec-ondary outcome measures such as self- reported meandaily fruit intake and mean daily takeaway/fast food intakewere also aligned with corresponding questions from thesame baseline population health survey [23].Other secondary outcome items include measuring
change in the self-reported frequency of cooking themain meal from basic ingredients and the inclusion ofsalad or vegetables with the main meal. Nutrition know-ledge questions, aligned with the nutrition messages em-bedded within the program, will test knowledge aroundsalt, fat and sugar intake and have been adapted fromParmenter et al’s work [24].Participant attitudes towards cooking and eating
healthy foods, willingness to try new foods, enjoymentand satisfaction in cooking and eating healthy foods willbe tested, using Likert scale based questions. Questionsabout shared enjoyment of cooking, eating and norma-tive eating behaviours were adapted from questions fromThe Stephanie Alexander Kitchen Garden Evaluation(SAKG) [25]. The Rosenberg Self-Esteem Scale (RES), awidely validated and reliable measure of self-esteem, willalso be administered [26]Participants will be asked to report their total house-
hold food and drink (excluding alcohol) expenditure andexpenditure specifically on take-away food and fruit andvegetables. Height and weight will be self-reported to en-able the calculation of Body Mass Index (BMI).
PilotingThe questionnaire was piloted in a 3 step process to testthe design, content and potential delivery methods.
Flego et al. BMC Public Health 2013, 13:411 Page 4 of 8http://www.biomedcentral.com/1471-2458/13/411
Following comment by the reference group and aca-demic colleagues on content and layout/design, a paper-based version of the questionnaire was piloted with asample (N = 30) of the current JMoF population. Feed-back was invited through informal focus group sessions,facilitating the identification of any questions that wereambiguous, sensitive or required further development.The final stage of piloting involved testing the ques-
tionnaire in an online format whilst simultaneously test-ing the online survey distribution system and the likelyresponse rate. As the survey distribution required inte-gration between the JMoF participant database and thesurvey platform, Qualtrics™ [27], the piloting tested thatthese two components were effectively integrated andcapable of providing the necessary information needs.
RecruitmentAll participants registered on the JMoF Australia waitlistdatabase, aged 18 years or over, and who had received aconfirmed start date for the program, will be eligible toparticipate in the quantitative component of the evalu-ation. Whilst the program is open to all members of theIpswich community over the age of 12 years, the evalu-ation will target the adult population only as othercooking skills programs exist within the Greater Ipswichregion that are specifically targeting children and adoles-cents in educational settings.All participants will be required to consent prior to
participation in the evaluation. A link to the question-naire will be embedded within an email generated by theJMoF program database, inviting program participantsto participate in the evaluation questionnaire.Recruitment to both groups will be based on con-
firmed program start dates. As control participants arerequired to be on the program waitlist for 10 weeks andbecause the JMoF waitlist is sufficiently large to enablethis to occur naturally, participants that are allocated astart date longer than 10 weeks ahead, will be automatic-ally made a control, whilst persons commencing theprogram within 10 weeks will be assigned to the inter-vention group. Computer programmed “rules” in theJMoF participant database have been created to auto-mate this process.To show appreciation for participation, those interven-
tion participants who complete the third and final ques-tionnaire at six months post completion of the programwill be sent a $20 “Good Guys” store charge card re-deemable at any Good Guys store (white goods retailer)across Australia.
Data collectionData collection commenced in February 2012 followingcompletion of piloting. Whilst the challenges of deliver-ing online surveys are well documented [28], it was
decided to be the most feasible and practical firstmethod of delivery with subsequent delivery of a postalversion to non-respondents and persons who do nothave a working email address or access to a computer.
Sample sizeGiven that the specific questions developed for thisstudy to measure confidence to cook (cooking self-efficacy) have not been previously employed, a precisesample size calculation for this primary outcome couldnot be calculated given the absence of a priori baselinemeasures, measures of effect size and measures of vari-ance (standard deviation).Sample size calculations around the second primary
outcome of a change in vegetable intake between thetwo groups assumed the use of a split-plot anova and Ftest for interaction given the wait list design. The litera-ture does not provide clear guidance with regards to anexpected effect size for a program of this nature [29,30].However, as there is some evidence from Wrieden et althat an effect size of one serve per day may be too largefor a program of this nature [20], sample size calcula-tions are based on an effect size range likely to beachieved [29,30]. For an effect size of 0.5 serves a day,starting from a mean daily consumption at baseline of2.5 serves per day [13,23], 250 subjects per group will berequired at 80% power (0.05 significance). In the eventthat accrual is slower than expected, recruiting at least140 participants will give 80% power for an effect size of0.7 serves a day. There will be no analysis of the data be-fore accrual has closed.
Data analysisDemographic and baseline characteristics will besummarised for both intervention and control groupsusing standard summary statistics (mean and standarddeviation) and non-parametric statistics (medians andinter-quartile ranges) where applicable. Frequencies andpercentages will be reported for categorical variables.The magnitude of change both within and betweenintervention and control groups for T1, T2 and T3 time-points will be assessed. For continuous data, such asfruit and vegetable intake, two sample t tests will beemployed to compare means between intervention andcontrol groups at each time point and paired t tests forwithin group comparisons. A split-plot in time Anovawill be used as the basis for these t-tests. For categoricaldata, frequencies will be reported for intervention andcontrol groups and chi squared analysis will provide be-tween group comparison results. Regression analysis willbe conducted to determine the potential contribution ofspecific demographic factors on the outcome variablesof interest. All data analysis will be performed usingSTATA S.E. 12.0 statistical software.
Flego et al. BMC Public Health 2013, 13:411 Page 5 of 8http://www.biomedcentral.com/1471-2458/13/411
Qualitative study
Research questionsTo further understand how and why the JMoF programimpacts on participants, the qualitative investigation willexplore the following:
1. What are the expectations and experiences ofparticipants?
2. What are the moderators, facilitators and barriers tobehaviour change?
3. Were there any unanticipated outcomes?
Study designA longitudinal design will be employed for the qualita-tive study to follow program participants over the courseof their JMoF journey. This will allow for prospective ac-counts of a participant’s experience and change overtime [31,32]. Repeated semi-structured interviews will beconducted with participants. Up to three interviews willbe conducted - prior to program commencement, onprogram completion and six months after completion.
Participant interviewsThe interview structure was generated to capture participantperspectives, to explore moderators, facilitators and barriersto behaviour change, and to capture any unanticipated out-comes from program involvement. As indicated in Table 1,the purpose of the sequential interviews is to understandparticipant expectations and experiences at different stagesof program involvement. The interviews will be unstruc-tured; however questions and prompts will be used to guidethe discussion where appropriate. Table 1 lists the general
discussion points used for each participant (interviews threeand four will be based on previous discussions).Interviews of approximately 30-40 minutes duration will
take place in a public location that provides a comfortableenvironment for both interviewer and the participant. All in-terviews will be conducted by the same researcher and par-ticipants will be required to consent to both participationand recording of the interview. To show appreciation forparticipation in an interview, participants will be thankedwith a $15 Coles supermarket gift voucher at eachinterview.
Participant samplingPurposive sampling will be employed initially to capturemaximum variation [33] in factors considered likely toinfluence expectations and experiences of the JMoF pro-gram. These factors were captured in the participantquestionnaire and included socio-economic status, age,gender, family structure, and cooking confidence level.Further sampling will be informed by themes emergingfrom the initial interviews and instructor observationsabout characteristics that seem to influence participantmotivations and experience. Interview one will subse-quently be conducted with approximately 10-15 partici-pants. Participants who provide rich data in terms ofinsights and unique perspectives, and who are willing tocontinue, will be invited to progress to interviews twoand three. In the event that there are insufficient num-bers to progress (less than 6), new participants will berecruited from new enrolments.
Participant recruitmentParticipants who have completed the baseline quantitativestudy survey and have agreed to be contacted for an inter-view, will be eligible to participate as well as all partici-pants who are within the first three weeks of commencingthe program. The baseline survey data will assist in pur-posive selection for the qualitative interviews, using thecriteria based on demographic and personal circumstancesas previously described. Opportunistic purposive samplingwill also be carried out during the researcher’s time inIpswich during class observations. Selected participantswill be contacted by phone or in person (in class context)and provided with information about the qualitativecomponent of the study and invited to participate in theinterviews. Written consent will be required prior to par-ticipation in the interview process.
Data analysisThe interviews will be transcribed verbatim. The datawill be managed with the assistance of qualitative soft-ware package NVivo 9 (NVivo 9 [program]: QSR Inter-national Pty Ltd 2011). Concurrent data collection andanalysis will be conducted to allow for confirmation of
Table 1 Interview structure
Interview timing Discussion topics
Interview 1: Prior tocommencement of theprogram
Motivations for registering for theprogram, expectations of the program.Discussion of previous and current foodand cooking experiences.
Contact: During program Phone conversation to recruitparticipants to repeat interviews and toenquire how the course is going.
Interview 2: After programcompleted
Discussion around their programexperience and whether programexpectations were met. If participantsexperienced any changes in food andcooking behaviour and anyunanticipated changes as a result of theprogram.
Interview 3: Six months afterprogram completion
Discussion around whether any changesas a result of the program have beensustained in terms of food and cookingbehaviour. Any unanticipated changesas a result of the program. Reflection onwhat was talked about in the lastinterview.
Flego et al. BMC Public Health 2013, 13:411 Page 6 of 8http://www.biomedcentral.com/1471-2458/13/411
emerging themes and clarification of any contradictoryfindings [34]. To contribute to the analytical process, theinterviewer will record post-interview memos as reflec-tions including contextual information, non-verbal fac-tors of note, reflections on the interview process, andthoughts about emerging patterns or contradictions inthe data [33,35].The analysis of interview transcripts and interview
memos will be conducted by the interviewer using in-ductive thematic analysis. The data will first be codedand then categorised to allow themes and patterns toemerge [34,36]. A second researcher will independentlygenerate codes on a sub-sample of transcripts [37].Comparisons will then be conducted and any differenceswill be discussed to achieve consensus in the final codes.The categorised data will then be reviewed to exploresimilarities and differences, to identify patterns and todetermine whether there are specific relationships occur-ring between categories that together provide an overallconceptual picture of the impact of the JMOF programwithin the context of its unique setting and population.The resultant conceptual analysis will then be comparedboth to relevant theoretical frameworks and to the lit-erature base to determine if it resonates with existingknowledge or makes a new contribution to the evidence.
Integrated analysisIn addition to the separate analysis of the quantitative andqualitative results, integration of the respective findingswill be conducted. This will involve examining consisten-cies and inconsistencies in the findings from each method[19,38] to build a more nuanced and comprehensive un-derstanding of the JMoF program impacts and outcomes.This added depth and breadth will inform the conclusionsdrawn from the evaluation.
DiscussionThe evaluation of the JMoF program will contribute to thegrowing body of literature on the effectiveness ofcommunity-based cooking skills programs. It will employ amixed methodology to draw on the strengths of bothquantitative and qualitative study design to best captureand measure experiences, impacts and outcomes ofcooking skills programs. The methods described herein willinform the research community about program outcomesand facilitate comparisons of results with other cookingskills programs conducted in comparable populations.This study will also provide insights into practical consid-
erations required when designing program evaluations incommunity settings. These include factors such as recruit-ment of a comparison group, minimisation of participantdata collection burden, and the suitability and feasibility ofselected data collection modes, which must be consideredwithout compromising study design or program integrity.
There are both strengths and limitations to the evalu-ation design. Mixed methods studies as a paradigm canrisk compromising methodological rigour if integrationoccurs at point of data collection and/or analysis andpotentially undermines paradigm and process consider-ations [18,19]. This is not an issue in the current evalu-ation with integration only occurring in relation tosample identification and final integration of findings.Whilst it is acknowledged that the use of a non-
randomised quasi-experimental design makes the quantita-tive study vulnerable to sampling bias, practical limitationsprevented the application of a randomised design. Despitethis potential bias, the use of a waitlist control and pre andpost measures support attribution of any changes to theprogram.Potential selection bias associated with choice to par-
ticipate or not in the quantitative study may also occur.However various methods were employed in an attemptto address this issue: providing participants with mul-tiple options for survey completion, follow up of non-responders and the use of incentives.In the quantitative study, there is no direct measure of
cooking skills despite the JMoF program being a cookingskills program per se. However, there is currently no goldstandard for the measurement of cooking skills in an adultpopulation nor consensus on the definition of cookingskills or whether changes in it alone will predict the likeli-hood of changes in cooking behaviour [3]. Therefore con-fidence to cook which reflects self-efficacy, a relativelystrong predictor of behaviour change, was the chosenmeasure for the evaluation as suggested by Winkler,Wrieden and Barton et al [3,20,21]. It is noted that evenBarton et al’s confidence questions upon which some ofthe current survey questions are based, whilst consideredreliable, have yet to be formally tested in the communitysetting [21]. Another limitation of the quantitative study isthe reliance on self-reported measures. Yet lessons learntfrom previous evaluations [20,21] suggest that the use ofmore intensive methods would likely overburden partici-pants and lead to low participation rates.In summary, the use, in this evaluation, of a mixed
method, pre-post design with a waitlist control group willprovide sufficient strength of evidence to assess the im-pact of the JMoF program on participants’ attitudes andbehaviours. It will also make a contribution to the limitedevidence base about the effectiveness of community-basedcooking programs.
Competing interestsThe authors wish to declare that the evaluation has been commissioned byThe Good Foundation.
Authors’ contributionsAll authors have contributed to project design and development, whilst AFand JH are responsible for project management and data collection. AFdrafted the manuscript with JH and LG contributing to the qualitative
Flego et al. BMC Public Health 2013, 13:411 Page 7 of 8http://www.biomedcentral.com/1471-2458/13/411
sections. All listed authors reviewed the draft manuscript, then read andapproved the final manuscript.
AcknowledgmentsThe authors would like to acknowledge Alicia Peardon and all the staff ofThe Good Foundation as well as Christina Stubbs of the QueenslandDepartment of Health for their contributions in relation to evaluation designand implementation. The authors also thank Associate Professor JohnReynolds, Deakin University for his contribution to the quantitative dataanalysis design. Moodie and Swinburn are researchers within an NHMRCCentre for Research Excellence in Obesity Policy and Food Systems(APP1041020).
Author details1Deakin Health Economics, Faculty of Health, Deakin University,Melbourne, Australia. 2Jack Brockhoff Child Health and WellbeingProgram, The McCaughey Centre, Melbourne School of PopulationHealth, The University of Melbourne, Melbourne, Australia. 3WHOCollaborating Centre for Obesity Prevention, Faculty of Health, DeakinUniversity, Melbourne, Australia. 4School of Population Health, Faculty ofMedical and Health Sciences, University of Auckland, Auckland, NewZealand.
Received: 9 January 2013 Accepted: 15 April 2013Published: 30 April 2013
References1. Caraher M, Dixon P, Lang T, Carr-Hill R: The state of cooking in England:
the relationship of cooking skills to food choice. Brit Food J 1999,101(8):590–609.
2. Caraher M, Lang T: Can't cook, won't cook: A review of cooking skills andtheir relevance to health promotion. Int J Health Promot Educ 1999,37(3):89–100.
3. Winkler E: Food Accessibility affordability, cooking skills, and socioeconomicdifferences in fruit and vegetable purchasing in Brisbane, Australia. Australia:Queensland University of Technology, Institute of Health and BiomedicalInnovation School of Public Health; 2008.
4. Winkler E, Turrell G: Confidence to Cook Vegetables and the BuyingHabits of Australian Households. J Am Diet Assoc 2009, 109(10):1759–1768.
5. Block K, Johnson B: Evaluation of the Stephanie Alexander KitchenGarden Program. In Final Report to: The Stephanie Alexander Kitchen GardenFoundation. Edited by Wellbeing TMCVCftPoMHaC. Melbourne: University ofMelbourne and Deakin University; 2009.
6. Fulkerson JA, Rydell S, Kubik MY, Lytle L, Boutelle K, Story M, Neumark-Sztainer D, Dudovitz B, Garwick A: Healthy Home Offerings via theMealtime Environment (HOME): feasibility, acceptability, and outcomesof a pilot study. Obesity 2010, 18(1s):S69–S74.
7. Keller HH, Gibbs A, Wong S, Vanderkooy PD, Hedley M: Men can cook!Development, implementation, and evaluation of a senior men'scooking group. J Nutr Elderly 2004, 24(1):71–87.
8. Foley W, Spurr S, Lenoy L, De Jong M, Fichera R: Cooking skills areimportant competencies for promoting healthy eating in an urbanIndigenous health service. Nutr Diet 2011, 68(4):291–296.
9. Engler-Stringer R: Food, cooking skills, and health: a literature review. CanJ Diet Pract Res 2010, 71(3):141–145.
10. Rees R, Hinds K, Dickson K, O'Mara-Eves A, Thomas J: Communities thatcook. A systematic review of the effectiveness and appropriateness ofinterventions to introduce adults to home cooking. In EPPI-Centre report2004. London: EPPI-Centre Social Science Research Unit, Institute ofEducation, University of London; 2012.
11. Ministry of Food Mission. http://www.thegoodfoundation.com.au/about-us/.12. Office of Economic and Statistical Research: Queensland Regional Profiles
Ipswich City Based on local government area (2010). In Profile generatedon 27 May 2011; 2011.
13. Queensland Health: Self- reported Health Status 2009-2010: Local governmentArea Summary Report. Brisbane; 2011.
14. Kolb DA: Experiential learning: experience as the source of learning anddevelopment. Englewood cliffs, NJ: Prentice Hall; 1984.
15. Bandura A: Self-efficacy: Toward a unifying theory of behavioral change.Psychol Rev 1977, 84(2):191–215.
16. Bandura A: Social Learning Theory. Englewood Cliffs, NJ: Prentice-Hall; 1977.
17. Turner G, Shepherd J: A method in search of a theory: peer educationand health promotion. Health Educ Res 1999, 14(2):235–247.
18. Stufflebeam DL: Evaluation models, New Directions for Evaluation, Volume89. San Francisco, CA: Jossey-Bass; 2001.
19. Nastasi B, Jea H (Eds): An Inclusive Framework for Conceptualizing MixedMethod Design Typologies. London: Sage; 2010.
20. Wrieden WL, Anderson AS, Longbottom PJ, Valentine K, Stead M, Caraher M,Lang T, Gray B, Dowler E: The impact of a community-based food skillsintervention on cooking confidence, food preparation methods and dietarychoices - an exploratory trial. Public Health Nutr 2006, 10(2):203–211.
21. Barton KL, Anderson AS, Wrieden WL: Validity and realiability of a shortquestionnaire for assessing impact of cooking skills interventions. J HumNutr Diet 2011, 24:588–595.
22. Short F: Domestic cooking skills- what are they? J HEIA 2003, 10(3):13–22.23. Queensland Health: Self-Reported Adult Health Status: Queensland. In
2009 Survey Report. Edited by Pollard G, White D, Harper C. Brisbane:Queensland Health; 2009.
24. Parmenter K, Wardle J: Development of a general nutrition knowledgequestionnaire for adults. Eur J Clin Nutr 1999, 53(4):298–308.
25. Block K, Johnson B, Gibbs L, et al: In Final Report to: The Stephanie AlexanderKitchen Garden Foundation. Edited by Wellbeing TMCVCftPoMHaC. Australia:Melbnourne University of Melbourne and Deakin University; 2009.
26. Ang RP, Neubronner M, Oh S-A, Leong V: Dimensionality of Rosenberg'sSelf-Esteem Scale among Normal-Technical Stream Students inSingapore. Curr Psychol 2006, 25(2):121–131.
27. Qualtrics online survey software. http://www.qualtrics.com.28. Lefever S, Dal M, Matthíasdóttir A: Online data collection in academic
research: advantages and limitations. Brit J Educ Technol 2007,38(4):574–582.
29. Brunner EJ, Rees K, Ward K, Burke M, Thorogood M: Dietary advice forreducing cardiovascular risk. Cochrane Database Syst Rev 2007(4):CD002128.
30. Pomerleau J, Lock K, Knai C, McKee M: Interventions Designed to IncreaseAdult Fruit and Vegetable Intake Can Be Effective: a Systematic Reviewof the Literature. J Nutr 2005, 135:2486–2495.
31. Shirani F, Henwood K: Continuity and change in a qualitative longitudinalstudy of fatherhood: relevance without responsibility. Int J Soc ResMethodol 2010, 14(1):17–29.
32. Farrall S: What is qualitative longitudinal research? In Papers in SocialResearch Methods Qualitative Series No 11. London: London School ofEconomics and Political Science Methodology Institute; 2006.
33. Patton M (Ed): Qualitative Research and Evaluation Methods. 3rd edition.California: Sage; 2002.
34. Green J, Willis K, Hughes E, Small R, Welch N, Gibbs L, Daly J: Generatingbest evidence from qualitative research: the role of data analysis. Aust NZ J Public Health 2007, 31(6):545–550.
35. Gibbs L, Kealy M, Willis K, Green J, Welch N, Daly J: What have samplingand data collection got to do with good qualitative research? Aust N Z JPublic Health 2007, 31(6):540–544.
36. Liamputtong P: Qualitative data analysis: conceptual and practicalconsiderations. Health Promot J Austr 2009, 20(2):133–139.
37. Liamputtong P: Qualitative research methods. 3rd edition. Melbourne: OxfordUniversity Press; 2009.
38. Tashakkori A, Teddlie C: Sage handbook of mixed methods in social &behavioral research, Volume 2. Los Angeles: SAGE Publications; 2010.
doi:10.1186/1471-2458-13-411Cite this article as: Flego et al.: Methods for the evaluation of the JamieOliver Ministry of Food program, Australia. BMC Public Health 201313:411.
Flego et al. BMC Public Health 2013, 13:411 Page 8 of 8http://www.biomedcentral.com/1471-2458/13/411
Appendix 4. Jamie’s Ministry of Food Australia questionnaire
IDT1 _________________________
We appreciate and thank you for your time in completing this survey. All the information you provide us will be completely confidential. Only the survey team will have access to this form. Your full name and address will never be linked with any of the information you provide. When completed, please seal in the envelope provided and post to Deakin University. Today’s Date (Day /Month /Year) / / Q1. Are you 18 years old or over? 1 Yes 2 No, if no you do not need to complete this survey. Q2. In a typical week….? How often do you prepare and cook a main meal from basic ingredients? For example starting with meat and vegetables (Please select one answer) 1 Never 2 Less than once 3 Once 4 2-3 times 5 4-6 times 6 Daily
Q3. In a typical week….? How often do you eat ready-made meals at home? For example a frozen roast dinner, lasagne or pizza that was purchased (Please select one answer) 1 Never 2 Less than once 3 Once 4 2-3 times 5 4-6 times 6 Daily
Q4. In a typical week….? How often do you include salad or vegetables with the main meal? For example, chicken with vegetables or pasta with a salad. (Please select one answer) 1 Never 2 Less than once 3 Once 4 2-3 times 5 4-6 times 6 Daily
244
Q5. How many serves of vegetables including fresh, dried, frozen and tinned vegetables do you usually eat each day, where a serve is half a cup of cooked vegetables or 1 cup of salad? (Please select one answer) 1 None 2 Less than 1 serve a day 3 1 serve
4 2 serves 5 3 serves 6 4 serves
7 5 serves 8 6 or more serves 9 Don’t know
Q6. How many serves of fruit including fresh, dried, frozen and tinned fruit do you usually eat each day, where a serve is a medium piece or two small pieces of fruit, or a cup of diced pieces? (Please select one answer) 1 None 2 Less than 1 serve a day 3 1 serve
4 2 serves 5 3 serves 6 4 serves
7 5 serves 8 6 or more serves 9 Don’t know
Q7. How many times do you eat take-away or ‘fast foods’ from places such as McDonalds, Red-rooster, fish and chips shop, Asian take-away or local take-away pizza places? (Please select one answer) 1 Never 2 Less than once a month 3 Once a week
4 2 times a week 5 3- 4 times a week 6 5 or more times a week
7 Don’t know
Q8. Which of the following choices do you think is not high in salt and could be used to enhance the flavours in food? (Please select one answer) 1 Bacon 2 Tomato Sauce
3 Herbs and Spices 4 Cheese
5 Not Sure
Q9. Out of the items listed below, which foods do you think health experts recommend eating the least? (Please select one answer) 1 Olive Oil 2 Nuts & Seeds
3 Avocados 4 Butter
5 Vegetable Oil 6 Not sure
Q10. If a person felt like something sweet, but was trying to cut down on sugar, which would be the best choice? (Please select one answer) 1 Glass of fruit juice 2 A cereal snack bar
3 A plain biscuit 4 A piece of fruit
5 Not sure
245
Q11. How confident do you feel ….? (Please circle one response for each question)
The next section asks about your household Q12. What is your current household characteristic? (Please select one answer) 1 Couple, with young children (0-17 years old) living at home 2 Couple, with adult children (18 years and over) living at home 3 Couple, without children living at home 4 One parent family with children living at home 5 Live Alone 6 Other, please specify ………………………………………… Q13. How many persons (including yourself) live in your household?
Q14. How many adults (18 years and over) do you usually prepare food for on a daily basis?
Q15. How many children (0-17 years old), living at home, do you usually prepare food for on a daily basis?
Q16. Do you have access to a kitchen to prepare your own meals? 1 Yes 2 No Q17. Do you do the food shopping in your household? 1 Yes 2 No Q18. In a typical week how much does your household spend in total on food and drink? Including take away food, eating out, food and drinks for cooking and eating at home (excluding alcohol). (Please select one answer) 1 $0 2 $1 - $29 3 $30 - $59
4 $60 - $99 5 $100 - $149 6 $150 - $199
7 $200 or more
Not at all Confident Extremely Confident
a) About being able to cook from basic ingredients? 1 2 3 4 5
b) About following a simple recipe? 1 2 3 4 5 c) About preparing and cooking new foods
and recipes? 1 2 3 4 5
d) That what you cook will ‘turn out’ well? 1 2 3 4 5 e) About tasting foods that you have not
eaten before? 1 2 3 4 5
246
Q19. In a typical week, how much does your household spend on fruit and vegetables? (Please select one answer) 1 $0 2 $1- $9 3 $10 - $19
4 $20 - $39 5 $40 - $49 6 $50 - $99
7 $100 or more
Q20. In a typical week, how much does your household spend on take away/fast foods? (Please select one answer) 1 $0 2 $1- $9 3 $10 - $19
4 $20 - $39 5 $40 - $49 6 $50 - $99
7 $100 or more
The next section asks about preparing and eating food Q21. In a typical week, how often do you eat together at home, with the other people in your household in the evenings? (Please select one answer) 1 Never 2 Less than once a month 3 Once a week
4 2 times a week 5 3- 4 times a week 6 5 or more times a week
7 Don’t know 8 I live alone
Q22. In a typical week, how often do you eat your dinner in front of the TV? (Please select one answer) 1 Never 2 Less than once a month 3 Once a week
4 2 times a week 5 3- 4 times a week 6 5 or more times a week
7 Don’t know
Q23. In a typical week, how often do you eat your dinner sitting at a dinner table, or equivalent? (Please select one answer) 1 Never 2 Less than once a month 3 Once a week
4 2 times a week 5 3- 4 times a week 6 5 or more times a week
7 Don’t know
247
Q24. Please choose the answer which best indicates how much you agree or disagree with each of the following statements. (Please select one answer for each question) Strongly
disagree Somewhat disagree
Somewhat agree
Strongly agree
a) I find it easy to change my eating habits 1 2 3 4
b) My lifestyle prevents me eating a healthy diet 1 2 3 4
c) Vegetables can be tasty foods 1 2 3 4 d) I eat enough fruit and vegetables 1 2 3 4
e) I can put together a healthy meal from scratch in 30 minutes 1 2 3 4
f) I enjoy cooking 1 2 3 4 g) I enjoy cooking for others 1 2 3 4
h) I get a lot of satisfaction from cooking my meals 1 2 3 4
i) I enjoy eating a meal with others 1 2 3 4
j) I can prepare a meal from basic ingredients that is low in price 1 2 3 4
k) Fruit and vegetables are cheaper when they are in season 1 2 3 4
l) Buying more fruit/vegetables would be difficult on my budget 1 2 3 4
The next section ask about how you feel about your health Q25. In general how do you feel about your health? (Please select one answer) 1 Excellent 2 Very good 3 Good 4 Fair 5 Poor
Q26. What is your current weight? Please answer in either kilogrammes (kg) or stone and pounds.
kg OR
stone and pounds (Note there are 14 pounds in a stone)
Q27. How tall are you (without shoes)? Please answer in either centimetres or feet and inches
cm OR
feet and inches (note there are 12 inches in a foot)
248
Q28. Below is a list of statements dealing with your general feelings about yourself. Please choose the answer which best indicates how much you agree or disagree with each of the following statements. (Please select one answer for each question)
Strongly disagree
Somewhat disagree
Somewhat agree
Strongly agree
a) I feel that I am a person of worth, at least on an equal plane with others
1 2 3 4
b) I feel that I have a number of good qualities 1 2 3 4
c) All in all, I am inclined to feel that I am a failure 1 2 3 4
d) I am able to do things as well as most other people 1 2 3 4
e) I feel I do not have much to be proud of 1 2 3 4
f) I take a positive attitude toward myself 1 2 3 4
g) On the whole, I am satisfied with myself 1 2 3 4
h) I wish I could have more respect for myself 1 2 3 4
i) I certainly feel useless at times 1 2 3 4
j) At times I think I am no good at all 1 2 3 4
The next section ask for some details about you Q29. What is your postcode? Q30. What is your date of birth? (Day /Month /Year) / / Q31. What is your sex? Male1 Female 2 Q32. Are you of Aboriginal or Torres Strait Islander origin? (Please select one answer) 1 No 2 Yes, Aboriginal 3 Yes, Torres Strait Islander 4 Yes, both Aboriginal and Torres Strait Islander
249
Q33. Do you speak a language other than English at home? (Please select one answer) 1 No 2 Yes, Italian
3 Yes, Greek 4 Yes, Cantonese
5 Yes, Arabic 6 Yes, Vietnamese 7 Yes, Mandarin 8 Yes, Other please specify …………………………
Q34. Please select which one of the following best describes your current employment? (Please select one answer) 1 Employed full-time 2 Employed part-time / casual 3 Home duties / carer 4 Unemployed 5 Full-time student
6 Part-time student 7 Retired 8 Permanently ill / unable to work 9 Other
Q35. What is the highest level of education you have completed? (Please select one answer) 1 Year 10 or less 2 Year 11 3 Year 12 4 Technical apprenticeship
5 Technical diploma 6 Tertiary degree 7 Post graduate degree 8 Other (please specify) ………………
Q36. Before tax is taken out, which of the following ranges best describes your household income, from all sources over the last 12 months? This includes wages/salaries, government benefits, pensions, allowances and other income usually received. (Please select one answer) 0 Nil Income 1 $1 - $6,000 per year 2 $6,001 - $13,000 per year 3 $13,001 - $20,000 per year 4 $20,001 - $30,000 per year
5 $30,001 - $50,000 per year 6 $50,001 - $100,000 per year 7 $100,001 - $150,000 per year 8 More than $150,000 per year 9 Don’t Know
Q37. Are you attending Jamie's Ministry of Food programme with someone else or as part of a group or organisation? 1 No, please continue to question 39 2 Yes, please answer question 38 Q38. Who are you attending the programme with? 1 Friend 2 Family 3 With a Carer 4 As part of a community group, please specify …………………………………………. 5 Other, please specify ………………………………………….
250
Q39. Are you happy to be contacted in the future for an interview about your experiences with the Ministry of Food Programme? 1 No 2 Yes, if yes please provide a contact number Thank you for your time When completed, please seal this survey in the envelope provided and post to Deakin University
251
TO: Programme Participants [questionnaire] Plain Language Statement Date: 25/11/2011 Full Project Title: The Jamie Oliver Ministry of Food (JMoF) Programme Evaluation Principal Researcher: Associate Professor Marj Moodie, Catherine Keating, Professor Boyd Swinburn, Professor Elizabeth Waters, Dr. Lisa Gibbs Researchers: Anna Flego, Jessica Herbert This document includes Plain Language Statement and consent forms. Please make sure you have all the pages. You are invited to take part in the evaluation of Jamie’s Ministry of Food programme in Ipswich, Queensland. Programme participants under the age of 18 do not need to complete this questionnaire. Deakin University with the University of Melbourne, in Melbourne Victoria, have been funded by The Good Foundation to conduct the programme evaluation. Some aspects of this study are being completed as part of the requirements for a Doctor of Philosophy by Jessica Herbert. This evaluation will help us to understand if the programme makes a difference for people who participate. What we learn from this evaluation will help to support the extension of Jamie’s Ministry of Food Programme and guide any future improvements in Australia and internationally. This Plain Language Statement contains detailed information about the evaluation. Its purpose is to explain to you as openly and clearly as possible all the procedures involved in this evaluation so that you can decide whether you are going to participate. Please read it carefully. Feel free to ask questions about any information in the document. You do not have to participate if you don’t want to. You can end your involvement in the research at any time without explanation. If you decide not to be part of the evaluation it will not affect your involvement in Jamie’s Ministry of Food Programme in any way and any information you have previously provided will not be used. Once you understand what the evaluation is about and if you agree to take part in it, you will be asked to sign and return the Consent Form provided on page 3. By signing the Consent Form, you indicate that you understand the information and that you give your consent to participate in the research project.
253
If you agree to participate in the evaluation, you will be asked to complete a short questionnaire on two to three occasions about your cooking and eating behaviours. It should take less than 15 minutes to complete. All of the information you provide will be grouped and reported together with other responses. It will not be possible to identify anyone in any of the research reports or publications. All data will be kept secure at Deakin University and only the research team will be able to access it. When the evaluation is finished, the data will be kept on a computer CD and stored for six years. After that it will be destroyed. We will send you a link to an online summary of the research findings when available if you are interested. The research results will also be reported through conference presentations and academic publications. If you have any complaints about any aspect of the project, the way it is being conducted or any questions about your rights as a research participant, then you may contact The Manager, Office of Research Integrity, Deakin University, 221 Burwood Highway, Burwood Victoria 3125, Telephone: 9251 7129, Facsimile: 9244 6581; [email protected]. Please quote project number [HEG-H117_2011]. If you require further information, wish to withdraw your participation or if you have any problems concerning this evaluation please feel free to contact a member of the two evaluation team members listed below: Associate Professor Marj Moodie Deakin University Phone: (03) 9251 7367 Email: [email protected]
Anna Flego, Project Coordinator Deakin University Phone: (03) 9251 7138, Email: [email protected]
254
TO: Programme Participants [questionnaire] Consent Form Date: 25/11/2011 Full Project Title: The Evaluation of Jamie Oliver’s Ministry of Food programme Reference Number: HEAG-H117_2011 I have read, and I understand the attached Plain Language Statement. I freely agree to participate in this evaluation according to the conditions in the Plain Language Statement, and understand I may be contacted more than once to participate in this evaluation. I have been given a copy of the Plain Language Statement and Consent Form to keep. The researcher has agreed not to reveal my identity and personal details, including where information about this project is published, or presented in any public form. Participant’s Name (printed) …………………………………………………………………… Signature ……………………………………………………… Date ………………………… Associate Professor Marj Moodie Deakin University, 221 Burwood Road, Burwood, VIC 3121, Australia Tel: 613 9251 7367, email: [email protected]
255
TO: Participants [questionnaire]
Revocation of Consent Form [questionnaire] (To be used for participants who wish to withdraw from the project) Date: 25/11/2011 Full Project Title: The Ministry of Food Evaluation Reference Number: HEAG-H117_2011 I hereby wish to WITHDRAW my consent to participate in the above evaluation and understand that such withdrawal WILL NOT jeopardise my relationship with Deakin University or Jamie’s Ministry of Food. Participant’s Name (printed) ……………………………………………………. Signature …………………………………………………. Date …………………… Please mail, fax or scan and email a signed copy of this form to: Jessica Herbert Population Health Strategic Research Centre Deakin University 221 Burwood Highway Burwood, VIC, 3125 Phone: (03) 9244 6258 Fax: (03) 9244 6624
256
Qualitative Study
TO: Programme Participants [interview] Plain Language Statement Date: 30/08/2011 Full Project Title: The Jamie Oliver Ministry of Food (JMoF) Programme Evaluation Principal Researcher: Associate Professor Marj Moodie, Catherine Keating, Professor Boyd Swinburn, Professor Elizabeth Waters, Dr. Lisa Gibbs Researchers: Anna Flego, Jessica Herbert You are invited to take part in the evaluation of Jamie’s Ministry of Food programme in Ipswich, Queensland. Deakin University with the University of Melbourne, in Melbourne Victoria, have been funded by The Good Foundation to conduct the programme evaluation. Some aspects of this study are being completed as part of the requirements for a Doctor of Philosophy by Jessica Herbert. This evaluation will help us to understand if the programme makes a difference for people who participate. Please read this Plain Language Statement carefully as it contains detailed information about this stage of the evaluation. Feel free to ask questions about any information in the document. With your consent, you will be interviewed for approximately 30-45 minutes. You can stop the interview at any point and any information you have previously provided will not be used. If you agree to participate you will be asked to sign the consent form provided on page 2. By signing the Consent Form, you indicate that you understand the information and that you give your consent to participate in the evaluation. The interviewer will ask you about your expectation and experiences with JMoF programme. You may be contacted again to clarify information or to participate in a repeat interview (after your programme and six months after your programme). They will also ask to voice record the interview. If you do not wish this to occur we will take handwritten notes of the interview. All of the information you provide will be kept secure at Deakin University and only the research team will be able to access it. When the evaluation is finished, the data will be securely stored for six years. After that it will be destroyed. We will send you a link to an online summary of the research findings when available if you are interested. The research results will also be reported through conference presentations and academic publications. If you have any complaints about any aspect of the project, the way it is being conducted or any questions about your rights as a research participant, then you may
257
contact The Manager, Office of Research Integrity, Deakin University, 221 Burwood Highway, Burwood Victoria 3125, Telephone: 9251 7129, Facsimile: 9244 6581; [email protected]. Please quote project number [HEAG-H117_2011]. If you require further information, wish to withdraw your participation or if you have any problems concerning this evaluation please feel free to contact a member of the two evaluation team members listed below: Associate Professor Marj Moodie Deakin University Phone: (03) 9251 7367 Email: [email protected]
Anna Flego, Project Coordinator Deakin University Phone: (03) 9251 7138, Email: [email protected]
258
TO: Programme Participants [interviews] Consent Form Date: 30/08/2011 Full Project Title: The Evaluation of Jamie Oliver’s Ministry of Food programme Reference Number: HEAG-H117_2011 I have read, and I understand the attached Plain Language Statement. I freely agree to participate in this interview according to the conditions in the Plain Language Statement and understand I may be contacted more than once to participate in this evaluation I have been given a copy of the Plain Language Statement and Consent Form to keep. The researcher has agreed not to reveal my identity and personal details, including where information about this project is published, or presented in any public form. Participant’s Name (printed) …………………………………………………………………… Signature ……………………………………………………… Date ………………………… Associate Professor Marj Moodie Deakin University, 221 Burwood Road, Burwood, VIC 3121, Australia Tel: 613 9251 7367, email: [email protected]
259
TO: Participants [interviews] Revocation of Consent Form (To be used for participants who wish to withdraw from the project) Date: 30/08/2011 Full Project Title: The Ministry of Food Evaluation Reference Number: HEAG-H117_2011 I hereby wish to WITHDRAW my consent to participate in the interview for the above evaluation and understand that such withdrawal WILL NOT jeopardize my relationship with Deakin University or Jamie’s Ministry of Food. Participant’s Name (printed) ……………………………………………………. Signature ………………………………………………………………. Date …………………… Please mail, fax or scan and email a signed copy of this form to: Jessica Herbert Population Health Strategic Research Centre Deakin University 221 Burwood Highway Burwood, VIC, 3125 Phone: (03) 9244 6258 Fax: (03) 9244 6624
260
Appendix 6. Interview guide Interview 1 Why they registered for the programme? Main motivations. How they found out about the programme? Jamie Oliver’s influence? ~ What is it about him? ~ What does he represent to them? Role at home? In terms of food, cooking, shopping and growing food. How they feel about cooking? ~ Prompts: confidence, trying new foods. How they eat their meal? ~ Who they eat with? ~ Do they ever eat together? How they feel about mealtime? Who they prepare food for? Socially? Do they have people over for a meal? Interview 2 Programme experience How was their experience of the programme? – encourage detail. What was their experience like in class? How did they feel when they were there? Did feelings change over the weeks? What was good/what was difficult? What kept them going and wanting to finish or why didn’t they complete (if applicable)? Programme Learning’s (if any). Did they learn anything or get anything from the programme? Was this what was talked about in the previous interview (reflection of programme expectations). Expectations. Were expectations met? Are there any other unexpected outcomes? Behaviours (cooking and food). What is happening at home now? Has anything changed since the last interview? ~ Last time we talked about your cooking preferences and mealtimes. Description of their food and cooking experience. Who they cook for, thoughts on healthy eating? (Has this changed as a result of the programme?) Reflection from last interview. Has anything changed as a result of the programme? (Reflection on key themes discussed in the previous interview). Changes: when and how changes occurred (if any). Were there any other changes not specific to food and cooking?
261
Interview 3 Food and cooking discussion. What is happening at home now? Description of their food and cooking experience. ~ Prompts: what they cook, who they cook for, healthy eating, shopping, cost of food? Did they feel there has been changed since we last talked? How? Why? How do they feel about food and cooking? Has food spending’s changed at all? Tell me about that? Confidence. What makes them feel confident as a cook? ~If they don’t, what would it take to make you feel confident as a cook? Social connections. Have they made any connections with others from your class? Have they joined any other programmes? Has the programme changed the way they socialise with family and friends? Willingness to pay for the course. What would they pay for the course and how much? Barriers/enablers to cooking. Explore what influences their “readiness to cook” – i.e. daily barriers and enablers. What have they indicted make it easy/hard to cook at home… Changes from the programme. Did they see any changes in themselves? When and how changes happened (if any). Other changes not specific to food and cooking. Has anything changed or is different as a result of the programme? (Reflection on key themes discussed in the previous interview). Unanticipated outcomes. Were there any unanticipated outcomes? * Talk about the last six months – reflect on what was said in the last interview with each participant
262
Appendix 7. JMoF Australia primary outcomes journal article
The primary outcomes paper for Jamie’s Ministry of Food Australia evaluation can be found at: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0114673
263
RESEARCH ARTICLE
Jamie’s Ministry of Food:Quasi-Experimental Evaluation ofImmediate and Sustained Impacts of aCooking Skills Program in AustraliaAnna Flego1*, Jessica Herbert1, Elizabeth Waters2, Lisa Gibbs2,Boyd Swinburn3,4, John Reynolds5, Marj Moodie1
1. Deakin Health Economics, Faculty of Health, Deakin University, Melbourne, Australia, 2. Jack BrockhoffChild Health and Wellbeing Program, Melbourne School of Population and Global Health, The University ofMelbourne, Melbourne, Australia, 3. World Health Organisation Collaborating Centre for Obesity Prevention,Faculty of Health, Deakin University, Melbourne, Australia, 4. School of Population Health, Faculty of Medicaland Health Sciences, University of Auckland, Auckland, New Zealand, 5. Biostatistics Unit, Faculty of Health,Deakin University, Melbourne, Australia
Abstract
Objective: To evaluate the immediate and sustained effectiveness of the first
Jamie’s Ministry of Food Program in Australia on individuals’ cooking confidence
and positive cooking/eating behaviours.
Methods: A quasi- experimental repeated measures design was used
incorporating a wait-list control group. A questionnaire was developed and
administered at baseline (T1), immediately post program (T2) and 6 months post
completion (T3) for participants allocated to the intervention group, while wait -list
controls completed it 10 weeks prior to program commencement (T1) and just
before program commencement (T2). The questionnaire measured: participants’
confidence to cook, the frequency of cooking from basic ingredients, and
consumption of vegetables, vegetables with the main meal, fruit, ready-made meals
and takeaway. Analysis used a linear mixed model approach for repeated
measures using all available data to determine mean differences within and
between groups over time.
Subjects: All adult participants ($18 years) who registered and subsequently
participated in the program in Ipswich, Queensland, between late November 2011-
December 2013, were invited to participate.
Results: In the intervention group: 694 completed T1, 383 completed T1 and T2
and 214 completed T1, T2 and T3 assessments. In the wait-list group: 237
completed T1 and 149 completed T1 and T2 assessments. Statistically significant
increases within the intervention group (P,0.001) and significant group*time
OPEN ACCESS
Citation: Flego A, Herbert J, Waters E, Gibbs L,Swinburn B, et al. (2014) Jamie’s Ministry of Food:Quasi-Experimental Evaluation of Immediate andSustained Impacts of a Cooking Skills Program inAustralia. PLoS ONE 9(12): e114673. doi:10.1371/journal.pone.0114673
Editor: Martyn Kirk, The Australian NationalUniversity, Australia
Received: June 26, 2014
Accepted: November 12, 2014
Published: December 16, 2014
Copyright: � 2014 Flego et al. This is an open-access article distributed under the terms of theCreative Commons Attribution License, whichpermits unrestricted use, distribution, and repro-duction in any medium, provided the original authorand source are credited.
Data Availability: The authors confirm that all dataunderlying the findings are fully available withoutrestriction. All relevant data are within the paperand its Supporting Information files.
Funding: This evaluation was jointly commis-sioned by the Queensland Government and TheGood Foundation, co-funders of the JMOFprogram, Ipswich. Whilst the project funders wereinvolved in initial discussions around study design,they had no role in data collection and analysis,decision to publish or preparation of the manu-script.
Competing Interests: The authors have declaredthat no competing interests exist.
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 1 / 18
interaction effects (P,0.001) were found in all cooking confidence measures
between T1 and T2 as well as cooking from basic ingredients, frequency of eating
vegetables with the main meal and daily vegetable intake (0.52 serves/day
increase). Statistically significant increases at T2 were sustained at 6 months post
program in the intervention group.
Conclusions: Jamie’s Ministry of Food Program, Australia improved individuals’
cooking confidence and cooking/eating behaviours contributing to a healthier diet
and is a promising community-based strategy to influence diet quality.
Introduction
Healthy food choices and eating patterns are essential for promoting good health
and well-being and preventing a wide range of chronic diseases [1] such as type
two diabetes [2], cardiovascular disease(CVD) [3], stroke [4]some cancers [5, 6]
and obesity [7, 8]. Yet it is increasingly more difficult for westernized societies to
achieve population adherence to recommended nutrition guidelines for healthy
eating [9]. As such, nutrition-related chronic disease and obesity is on the rise
[10], and, in addition to placing considerable burden on healthcare budgets, is
also having far wider impacts within societies [11, 12]. Poor diet has been
identified as a leading contributing risk factor for global disease burden, with
obesity, in part a product of unhealthy eating, being the leading risk factor in
Australasia [13].
One concerning trend is the increasing consumption of energy dense foods of
lower nutritional value prepared outside of the home [14] rather than the
preparation and consumption of home cooked meals that are associated with
higher vegetable intake [15] and overall higher diet quality [16]. This shift towards
outsourcing of meal preparation [14] is likely to be the product of a multitude of
varying, competing and inter-related factors [17, 18] including social, economic,
cultural, environmental and technological influences, that serve to potentially
constrain home cooking practices [14]. One hypothesis is that a decline in
cooking skills in adults is contributing to the problem [19]. Caraher and Lang
[19], in their review of the state of cooking skills in England, identified many
possible reasons for declining cooking skills including a reduction in
opportunities to learn to cook. In Australia, cooking is not always taught in
schools and fundamental changes in the ways in which individuals and families
function, on a day-to-day basis, in a modern society have reduced the traditional
opportunities for learning to cook from family members. Furthermore there is
growing evidence of a relationship between a lack of cooking skills, low cooking
confidence and poor food choices [20], including the likelihood of higher ready
prepared meal consumption [21]; inversely, higher levels of cooking skills and
confidence are associated with higher vegetable purchasing [22] and healthier
eating [15].
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 2 / 18
This evidence underpins the proliferation of community-based cooking skills
education programs which are being used as a strategy, both by government and
non -government organisations, to promote cooking skills and cooking
confidence as a vehicle to healthy eating. For example, the recent World Cancer
Research Fund’s (WCRF) ‘‘NOURISHING’’ food policy framework specifically
identifies cooking skills education within its nutrition education policy area,
which is one of the ten key policy areas of action to foster healthy eating and
prevent obesity [23]. The Cooking Matters program in the United States (US) and
the Jamie Ministry of Food programs in the United Kingdom (UK) and now
Australia are recognisable community-based cooking skills programs. The latter,
in particular, has attracted attention, in part due to their development by Jamie
Oliver, a UK based celebrity chef. However, the evidence to support the efficacy of
these and other community-based cooking programs, in adult populations, is still
emerging [24, 25]; the Jamie Ministry of Food program itself is yet to be formally
evaluated. Individual programs have reported positive findings in relation to
improved cooking confidence [26, 27], healthy cooking and food consumption
patterns [27, 28, 29] with some sustained effects up to one year post intervention
[27]. A recent systematic review concluded that interventions involving home
food preparation and/or cooking may result in improved food choices, dietary
behaviours and other health related outcomes [25]. Still the authors were cautious
in their interpretation of the results given the dearth of suitably robust evaluations
from which to make definitive statements about program effectiveness [25]. Given
growing pressure for evidence-based public health investment [30], evaluations of
program effectiveness should be both robust and practicable.
This study aims to evaluate the immediate impacts and longer term outcomes
of Jamie’s Ministry of Food program, in Ipswich, Australia using a quasi-
experimental, repeated measures design. This paper reports the impacts of the
program in terms of cooking confidence and cooking and eating behaviours. This
quantitative component is embedded within a larger mixed methods evaluation
[31]; other secondary outcomes (including a range of attitudinal, knowledge,
affordability metrics and broader psychosocial impacts) and qualitative study
findings are reported elsewhere [32].
Methods
The intervention
Jamie’s Ministry of Food programs have been running since 2008 in the UK
funded by a mixture of private and public sources. In 2010, Jamie’s Ministry of
Food was brought to Australia by The Good Foundation (TGF), a not-for-profit,
health promotion organisation funded by The Good Guys, a major Australian
electrical goods retailer, in partnership with Jamie Oliver. In April 2011, the first
Australian Jamie’s Ministry of Food Centre opened in Ipswich, Queensland
(QLD). The Ipswich Centre is predominantly funded by philanthropist Mr
Andrew Muir (owner of The Good Guys) and QLD State Government, as well as
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 3 / 18
local partners. Whilst the program is all inclusive and not targeted towards a
specific demographic group, Ipswich was deliberately chosen as the location due
to its population’s lower socioeconomic status [33] and higher prevalence of
overweight and obesity than the QLD state average [34]. Given the socioeconomic
gradient apparent in obesity prevalence within Australia [35], wherein health
inequalities related to obesity and its risk factors such as unhealthy eating [36] are
more evident in lower income groups, Ipswich was considered a suitable setting
for the program.
Jamie Oliver’s manifesto is to inspire individuals to cook simple basic meals
from scratch both for themselves and their families [37].The 10 week program,
consisting of weekly 1.5 hour classes, runs from a fixed site located in the main
Ipswich shopping precinct. Participants are taught recipes from 50 ‘‘Jamie Oliver’’
recipes which have been adapted to the Australian context from the UK program
so that over the 10 week course they learn to prepare and cook a variety of dishes
while learning specific cooking techniques such as chopping, frying, roasting and
baking. Messages about good nutrition, meal planning and budgeting are
embedded in the program and are discussed in an informal manner during the
skills sessions. There is an emphasis on cooking from scratch using fresh
ingredients such as fish, meat and seasonal vegetables and fruit. Further details of
the program, its specific objectives, and the full evaluation protocol are reported
elsewhere [31].
Evaluation Objectives
To provide a basis for evaluation enquiry, a program logic model was developed
to determine the potential pathways to behaviour change [31]. Whilst some
pathways such as the effect of cooking skills programs on individuals’ cooking
confidence/self -efficacy and cooking and eating behaviours were based on an
emerging evidence base, and were directly aligned with program objectives, others
around psychosocial, social connectedness, attitudinal and other broader impacts
were viewed as more exploratory [31]. This paper draws on the quantitative
evaluation to explore the former pathways and determine the impact of Jamie’s
Ministry of Food on participants’ cooking confidence (self-efficacy), and cooking
and eating behaviours.
Evaluation design
The evaluation used a quasi-experimental design with a wait-list control group.
All program participants aged 18 years and over were invited to participate in the
evaluation. Participants who signed up to the program .10 weeks before their
course start date, due to the length of the wait-list at the time, were allocated to
the control group whilst participants who signed up to the program ,10 weeks
before their course start date, were allocated to the intervention group.
Intervention participants were surveyed at three time points (before the program
start (T1), on program completion (T2) and six months after program
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 4 / 18
completion (T3), whilst control participants were surveyed 10 weeks before
program start date (T1) and just prior to beginning their first class (T2). An
online questionnaire was used followed by a postal version sent to all non-
respondents or to anyone who did not have access to a computer or the internet.
Data collection ran from December 2011 to December 2013. Six month follow up
data were not collected from controls; instead, data from Queensland State-wide
monitoring of vegetable intake [38] collected and reported during the evaluation
data collection period was used as a proxy T3 control measurement to enable
comparison with the self- reported vegetable intake measured in the intervention
group at 6 months post intervention (T3). Further details of the methods are
provided in Flego et al 2013 [31]. The evaluation was approved by the Deakin
University Human Research Ethics Committee (HEAG-H 117_11) in October
2011 and was registered with the Australian and New Zealand Clinical Trials
Registry (registration number: ACTRN12611001209987).
Questionnaire and outcome measures
A self -administered quantitative measurement tool was developed [31]. The
primary outcome measures for the quantitative evaluation were a change in self-
reported cooking confidence and daily vegetable intake. Cooking confidence (self-
efficacy) was measured by five questions assessing confidence in a number of
generalised skills required to do basic cooking at home on a scale from 1-5
anchored from ‘‘not at all confident’’51 to ‘‘extremely confident’’55. Four of the
confidence questions were adapted from a validated cooking skills questionnaire
by Barton et al, 2011 [39] and the fifth adapted from Keller et al, 2004 [40]. In
addition to the separate confidence items, all item scores were combined to create
an overall confidence score. Self-reported daily vegetable intake was measured as
serves per day (a serve was equal to half a cup of cooked vegetables or 1 cup of
salad where cup 5250 ml) anchored from 05 no serves per day to 6 or more
serves per day 56. This question used the same wording as that used in the
Queensland Self-Reported Health Status Survey question to enable comparison
[41].
Secondary outcomes included changes in self-reported cooking and eating
measures including: (i) frequency of cooking the main meal from basic
ingredients; (ii) frequency of eating ready-made meals at home; (ii) frequency of
eating vegetables with the main meal, all measured on a scale from 05 never to
75 daily; (iv)serves of fruit per day measured as per daily vegetable intake and (v)
frequency of eating takeaway per week measured from 05 never to 555 or more
times per week.
Sample size
Sample size calculations were based on the primary outcome of a change in self-
reported vegetable intake. The study was powered to detect a mean change in self-
reported vegetable intake of 0.5 serves per day, at 80% power (0.05 significance)
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 5 / 18
starting from a baseline measure of 2.5 serves a day [41]. This assumed the use of a
nested anova or mixed model analysis with an F test for the group by time
interaction. Additional sample size calculations suggested at least 140 participants
would be required in each group to detect a mean difference of 0.5–0.7 serves per
day. For further details of sample size calculations, see Flego et al, 2013 [31].
Statistical Analysis
All continuous outcomes were analysed using linear mixed models for repeated
measures to determine mean differences between groups over time. This type of
analysis copes with unbalanced groups, missing (at random) follow up data and
enables all available data to be utilised to determine the estimated changes in
mean outcomes. Predicted means from the mixed model analyses are reported in
all tables in the results section. Each analysis was also subsequently run with
separate adjustments for gender, age (dichotomised at below 50 or 50 years and
above) and categories of employment status then with all three covariates
together, to account for the potential effect of baseline differences (assessed by
chi-squared tests between groups, refer to Table 1) on the estimated effect of the
intervention.
Ratings on the cooking confidence score from ‘‘not at all confident’’ (1) to
‘‘extremely confident’’ (5) were regarded as a continuous variable in the primary
analysis - an acceptable technique for larger data sets [42]. However,
supplementary analyses using ordinal logistic and logistic models, to account for
the ordered response categories, and the dichotomization of the categories (‘‘not
confident’’ and ‘‘confident’’), were also conducted to determine if results from the
differing statistical approaches would provide similar inferences with respect to
differences between groups in their changes over time (see S3 and S4 Tables).
To explore the sustainability of the intervention group effect over time, a
repeated measures analysis was performed using the intervention group data
collected at all three time points. For the primary outcome measure of self-
reported vegetable intake, a two-sample t-test comparison between the 6-month
post-program intervention group mean and the reported state-wide mean was
performed.
All analyses were performed using STATA software (version 12.0). Results were
deemed significant at the P,0.05 level.
Results
Participant recruitment and retention
Over the two-year data collection period, approximately 1960 participants
registered for the program and were invited to participate in the evaluation; 1526
were allocated to the intervention group and 434 to the control group. Fig. 1
provides a summary of group allocation and response numbers including
specification of exclusions and loss to follow up at each time point.
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 6 / 18
Table 1. Demographic characteristics of evaluation participants by time point a.
Group, Time-point Intervention T1 Intervention T2 Intervention T3 Control T1 Control T2
% (n) % (n)
Gender b
Female 77.4 (525) 79.1(299) 80.5(207) 87.2 (198) 87.7(128)
Male 22.6 (153) 20.9(79) 19.5(50) 12.8 (29) 12.3(18)
Age (years)
Under 50b 55.6 (375) 44.1(165) 43.5(110) 64.3 (144) 60.3(85)
50 and over b 44.4 (300) 55.9 (209) 56.5(143) 35.7 (80) 39.7(56)
18–24 7.4 (50) 2.7(10) 3.2(8) 5.8 (13) 4.3(6)
25–34 17.5 (118) 14.2(53) 14.6(37) 22.8 (51) 20.6(29)
35–44 23.0 (155) 19.8(74) 19.0(48) 26.3 (59) 24.1(34)
45–54 16.0 (108) 15.5(58) 15.8(40) 16.5 (37) 18.4(26)
55–64 15.0 (101) 18.45(69) 18.6(47) 12.5 (28) 13.5(19)
65–74 17.5 (118) 24.6(92) 24.1(61) 13.4 (30) 15.6(22)
75+ 3.70 (25) 4.8(18) 4.7(12) 2.70 (6) 3.5(5)
Mean age years (SD) 48(16.1) 52(15.7) 52(15.9) 46(15.1) 47(15.2)
Aboriginal and/or Torre Strait Islander 1.8 (12) 1.8(7) 2.7(7) 0.9 (2) 1.4(2)
Speaks a language other than English at home 7.8 (53) 6.9(26) 7.4(19) 5.3 (12) 3.4(5)
Locality
Ipswich 82.0 (555) 84.7(320) 83.3(214) 78.8 (178) 79.4(116)
Other Queensland localities 17.7 (120) 15.3 (58) 16.3(42) 21.2 (48) 20.5 (30)
NSW 0.3 (2) 0.0(0) .4(1) 0.0 (0) 0.0
Highest Level of education attained
High school, year 12 or less 47.8 (321) 49.3(185) 49.4(126) 45.8 (104) 47.3(69)
TAFE, apprenticeship, diploma or certificate 22.2 (149) 20.8(78) 21.6(55) 22.9 (52) 19.9(29)
Tertiary, bachelor degree or higher 28.0 (188) 28.5(107) 27.4(70) 29.1 (66) 30.1(44)
Other 2.0 (13) 1.3(5) 1.6(4) 2.2 (5) 2.7(4)
Employment b
Full-time 26.4 (176) 23.6(88) 26.2(67) 34.7 (79) 31.3(46)
Part-time/casual 18.6 (124) 16.6(62) 18.4(47) 14.5 (33) 17.0(25)
Retired 23.8 (159) 31.6(118) 30.5(78) 21.5 (49) 23.8(35)
Home duties/carer 14.4 (96) 15.3(57) 13.3(34) 18.4 (42) 17.7(26)
Not working (Permanently ill/unable to work, unem-ployed)
9.9 (66) 7.0(26) 4.3(11) 8.8 (20) 7.5(11)
Student (full-time and part-time) 3.1 (21) 1.9(7) 1.9(5) 1.3 (3) 2.0(3)
Other 3.9 (26) 4.0(15) 5.47(14) 0.9 (2) .70(1)
Household yearly income
1– 6,000 2.5(15) 1.8(6) 2.6(6) 2.0(4) 1.5(2)
6,001– 13,000 5.7(34) 5.7(19) 5.3(12) 5.0(10) 5.3(7)
13,001– 20,000 11.9(71) 12.9(43) 14.1(32) 9.5(19) 9.9(13)
20,001– 30,000 14.8(88) 17.7(59) 17.2(39) 9.5(19) 9.9(13)
30,001– 50,000 15.4(92) 14.7(49) 14.5(33) 12.5(25) 12.2(16)
50,001– 100,000 30.0(179) 29.7(99) 26.9(61) 35.5(71) 36.6(48)
100,001– 150,000 13.6(81) 11.1(37) 12.3(28) 18.5(37) 16.8(22)
.150,000 6.0(36) 6.3(21) 7.0(16) 7.5(15) 7.6(10)
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 7 / 18
Some 45% of intervention and 55% of control participants invited to
participate in the evaluation completed the baseline assessment (T1). Of those
participants who completed baseline, 55% of intervention and 63% of control
participants completed the T2 assessment. In the intervention group, 31% of
those participants who completed baseline, also completed both T2 and T3
assessments.
Baseline demographic profile
At baseline (Table 1), the majority of participants (,80%) were female, with a
statistically significant higher proportion of females in the control group (87.2%)
than the intervention group (77.4%) (P,0.05). Both groups had more
participants aged under 50 years than over, with the controls having significantly
more in the younger category than the intervention group (64% and 55%
respectively P,0.05). Employment status differed between groups (P,0.05) with
the intervention group less likely to be in full time employment (26%) and more
likely to be retired (24%) compared to the control group (35% and 21%
respectively). Losses to follow-up in each group over time significantly differed by
age (P,0.05) but not for gender (P.0.05) (Table 1).
Change in cooking confidence (self-efficacy)
In the intervention group but not the control group, there was a statistically
significant increase between T1 and T2 in all cooking confidence measures
(P,0.001) both individually and when combined into a cooking confidence score
(Table 2). Statistically significant group by time interactions for all confidence
measures (P,0.001) demonstrate that the change over time differed between
Table 1. Cont.
Group, Time-point Intervention T1 Intervention T2 Intervention T3 Control T1 Control T2
Household Characteristics
Couple, with young children (0–17 years old) living athome
24.7 (169) 23.2(88) 20.0(51) 32.1 (76) 30.2(45)
Couple, with adult children (18 years and over) living athome
12.5 (86) 10.8(41) 11.0(28) 10.1 (24) 10.7(16)
Couple, without children living at home 32.9 (226) 35.3(134) 36.0(92) 24.5 (58) 27.5(41)
One parent family with children living at home 7.0 (48) 3.7(14) 4.3(11) 8.9 (21) 6.7(10)
Live Alone 16.0 (110) 21.6(82) 22.7(58) 17.7 (42) 20.1(30)
Other 6.9 (47) 5.5(21) 6.2(16) 6.8 (16) 4.7(7)
Mean household size (SD) c 2.8 (1.5) 2.6(1.3) 2.5(1.3) 3.0(1.6) 2.9(1.6)
Median household size c (50%centile) 2 2 2 3 2
aSample size for different variables might vary from total sample size because of missing responses and rounding of weighted frequencies.bsignificant difference between groups (p,0.05) at baseline as tested with chi squared analysis.cexcludes 2 participants living in institutional facilities.SD 5 standard deviation.
doi:10.1371/journal.pone.0114673.t001
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 8 / 18
Fig. 1. Evaluation participation over time.
doi:10.1371/journal.pone.0114673.g001
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 9 / 18
groups. These results remained significant at the P,0.001 level for all covariate-
adjusted analyses performed for each confidence variable (see S1 Table). The
supplementary ordinal logistic and logistic analyses (S3 and S4 Tables) also
Table 2. Cooking and eating measures at baseline and immediately post program1.
Intervention group Control group
Differencebetween groupsin changes frombaseline(interactioneffect)3 P value
Outcome measure
Baselinevalue(T1)mean (S.E)2
Follow upvalue(T2)mean (S.E)
Change frombaseline(T2-T1)mean (S.E) P value
Baselinevalue(T1)mean (S.E)2
Follow upvalue(T2)mean (S.E)
Change frombaseline(T2-T1)mean (S.E) P value P value
Cooking confidence
Confidence to cookfrom basic ingredients4
3.56(0.04) 4.36(0.05) 0.81(0.05) p,0.001 3.69(0.07) 3.72(0.08) 0.03(0.08) p50.70 p,0.001
Confidence to follow asimple recipe4
4.00(0.04) 4.53(0.05) 0.53(0.04) p,0.001 4.11(0.06) 4.06(0.07) 20.06(0.07)p50.40
p,0.001
Confidence in prepar-ing and cooking newfoods and recipes4
3.35(0.04) 4.13(0.05) 0.77(0.05) p,0.001 3.45(0.07) 3.55 (0.08) 0.10(0.08) p50.22 p,0.001
Confidence that whatone cooks will turn outwell4
3.21(0.04) 3.93(0.05) 0.72(0.04) p,0.001 3.28(0.06) 3.35(0.07) 0.07(0.07) p50.30 p,0.001
Confidence to tastenew foods never eatenbefore4
3.47(0.04) 4.01(0.05) 0.54(0.05) p,0.001 3.41(0.07) 3.51(0.09) 0.09(0.08) p50.25 p,0.001
Combined confidencescore5
17.6(0.02) 21.0(0.2) 3.36(0.18) p,0.001 17.9(0.03) 18.2(0.03) 0.23(0.28) p50.41 p,0.001
Cooking and eatingmeasures
Cooking the main mealfrom basic ingredients6
4.05 (0.08) 4.66 (0.09) 0.61 (0.09) p,0.001 4.16 (0.14) 4.17(0.16) 0.01(0.15) p5 0.95 p,0.001
Consumption of ready-made meals at home6
1.06 (0.05) 0.95 (0.06) 20.11 (0.06)p50.06 1.11(0.08) 1.21(0.10) 0.10(0.10) p50.30 p50.06
Consumption of vege-tables with the mainmeal6
4.67(0.07) 5.06(0.09) 0.39 (0.08) p,0.001 4.76(0.12) 4.75(0.14) 0.01(0.14) p50.94 p50.01
Daily vegetable con-sumption (serves perday)
2.46 (0.51) 2.97(0.06) 0.52 (0.06) p,0.001 2.49(0.09) 2.59(0.10) 0.10(0.10) p50.30 p,0.001
Daily fruit consumption(serves per day)
1.65(0.04) 1.93(0.05) 0.28 (0.05) p,0.001 1.61(0.07) 1.71(0.08) 0.10(0.08) p50.20 p50.06
Take-away consump-tion6
0.98(0.04) 0.77(0.04) 20.21(0.04)p,0.001
0.94(0.06) 0.96(0.07) 0.03(0.06) p50.62 p50.001
1Outcomes within each group and over time were determined by a mixed linear model for repeated measures using all available data at each time point. Allmeans and Standard Errors (S.E) have been rounded to 2 decimal points.2Baseline values were not significantly different between groups (independent t tests P,0.05).3A significant group x time interaction effect denotes that the response over time differed between groups.4Scale values are 1–5 (where 15 not at all confident and 55 extremely confident and 4 or. 5 confident).5The combined confidence score is equal to the sum total of all other confidence scores (scores 20 or. 5 confident).6Times per week.
doi:10.1371/journal.pone.0114673.t002
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 10 / 18
resulted in similar findings regarding the differences between the groups for all
cooking confidence measures (P,0.001).
When analysis was restricted to the intervention group only to test the
sustained effect of the program (i.e. to T3), statistically significant increases in all
cooking confidence measures (P,0.001) were reported for pairwise comparisons
between T1 and T2 (P,0.001) and T1 and T3 (6 months post intervention)
(P,0.001) but not between T2 and T3 (Table 3). Statistically significant results
were also noted (P,0.001) for the overall or main effects of time.
Self -reported Vegetable intake
Self-reported daily vegetable intake increased significantly between T1 and T2 in
the intervention group by just over a half serve (0.52 serves, SD 0.06, P,0.001)
but not in the control group (0.10 serves, SD 0.1, P50.30). A statistically
significant group by time interaction (P,0.001) was found (Table 2). All adjusted
analyses found very similar results (see S2 Table).
For the intervention group only, daily vegetable consumption increased
significantly from T1 (2.46, SD 0.05) to T2 (2.97, SD 0.06, P,0.001) and T3 (3.05,
SD 0.07), although the change between T2 and T3 was not significant (P50.273).
The change between T1 and T3 was significant (0.60 serves, SD 0.07, P,0.001) as
was the overall effect of change over time (P,0.001) (Table 3).
Results of the t-test comparison between the self-reported mean vegetable
intake at T3 for the intervention group (mean53.13, SD51.39) and state-wide
monitoring data from the 2012 Queensland Self-Reported Health Status Survey
(mean52.39, SD 5 2.39) [38] showed a statistically significant mean difference of
0.74 serves (SD 0.09, P,0.001).
Changes in other cooking and eating measures
Cooking the main meal from basic ingredients increased significantly from T1 to
T2 in the intervention group (0.61, SD 0.09, P,0.001) and the overall group by
time interaction effect was statistically significant (P,0.001) (Table 2). Both the
consumption of takeaway food and of ready -made meals reduced in the
intervention group although only the reduction in takeaway consumption was
statistically significant (-0.21, SD 0.04, P,0.001) and showed a significant
interaction effect (P,0.001). Consumption of vegetables with the main meal and
daily fruit intake also increased significantly in the intervention group (P,0.001)
but daily fruit intake did not demonstrate a significant group by time interaction
effect (P50.06). No significant differences in any of the cooking and eating
measures were found in the control group between T1 and T2. All adjusted
analyses performed found very similar results (see S2 Table).
Analysis of data from the intervention group only (Table 3) showed an overall
statistically significant positive effect over time for the aforementioned cooking
and eating measures (P,0.001) and between T1 and T2 (P,0.05), however
change between T2 and T3 did not remain significant for reductions in
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 11 / 18
Table 3. Cooking and eating measures for the intervention group only at baseline, post intervention and 6 months follow up1.
intervention group
overalleffect ofchangeover timeP value
Outcomemeasure
Baselinevalue(T1)mean (S.E)
Follow upvalue(T2)mean (S.E)
6 months postinterventionfollow up (T3)mean (S.E)
Change frombaseline(T2-T1) mean(S.E) P value
Change frombaseline(T3-T1) mean(S.E) P value
Change betweenfollow up (T3-T2)mean (S.E) P value P value
Cooking confi-dence
Confidence tocook from basicingredients2
3.56(0.04) 4.37(0.05) 4.43(0.06) 0.81(0.05)p,0.001 0.87(0.06)p,0.001 0.07(0.06)p50.280 p,0.001
Confidence to fol-low a simplerecipe2
4.00(0.04) 4.53(0.04) 4.61(0.05) 0.53(0.04)p,0.001 0.61(0.05)p,0.001 0.08(0.05)p50.133 p,0.001
Confidence inpreparing andcooking new foodsand recipes2
3.35(0.04) 4.13(0.05) 4.17(0.06) 0.78(0.05)p,0.001 0.82(0.06)p,0.001 0.05(0.06)p50.439 p,0.001
Confidence thatwhat one cookswill turn out well2
3.21(0.04) 3.93(0.05) 3.94(0.05) 0.72(0.04)p,0.001 0.73(0.05)p,0.001 0.01(0.06)p50.803 p,0.001
Confidence totaste new foodsnever eatenbefore2
3.47(0.04) 4.01(0.05) 3.99(0.06) 0.53(0.05)p,0.001 0.52(0.06)p,0.001 20.02(0.06)p50.746 p,0.001
Combined confi-dence score3
17.59(0.16) 20.95(0.19) 21.15(0.22) 3.36(0.18)p,0.001 3.56(0.21)p,0.001 0.20(0.22)p50.363 p,0.001
Cooking and eat-ing measures
Cooking the mainmeal from basicingredients4
4.05(0.08) 4.65(0.10) 4.88(0.11) 0.60(0.09) p,0.001 0.84 (0.10) p,0.001 0.24(0.11) p50.03 p,0.001
Consumption ofready- mademeals at home4
1.06(0.05) 0.93(0.06) 0.80(0.07) 20.13(0.06)p50.04 20.26(0.07)p,0.001 20.13(0.08) p50.09 p,0.001
Consumption ofvegetables withthe main meal4
4.67(0.07) 5.05(0.09) 5.31(0.10) 0.38(0.09) p,0.001 0.64(0.09) p,0.001 0.25(0.10) p,0.018 p,0.001
Daily vegetableconsumption(serves per day)
2.46(0.05) 2.97(0.06) 3.05(0.07) 0.51(0.06)p,0.001 0.60(0.07)p,0.001 0.08(0.08)p50.273 p,0.001
Daily fruit con-sumption (servesper day)
1.65(0.04) 1.93(0.05) 2.05(0.06) 0.27(0.05) p,0.001 0.40(0.06) p,0.001 0.12(0.06) p50.055 p,0.001
Take-away con-sumption4
0.98(0.04) 0.76(0.04) 0.73(0.05) 20.23(0.04)p,0.001 20.25(0.04)p,0.001 20.02(0.05)p50.607 p,0.001
1Outcomes at each time point were determined by a mixed linear model for repeated measures using all available data at each time point. All means andStandard Errors (S.E) have been rounded to 2 decimal points.2Scale values are 1–5 (where 15 not at all confident and 55 extremely confident).3The combined confidence score is equal to the sum total of all other confidence scores (scores 20 or. 5 confident).4Times per week.
doi:10.1371/journal.pone.0114673.t003
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 12 / 18
consumption of ready-made meals or takeaway, or for increases in daily fruit
intake.
Discussion
This is the first published quantitative evaluation of Jamie’s Ministry of Food
Program providing evidence of the program’s impact on participants’ cooking
and eating behaviours. While evaluation recruitment was relatively modest at
baseline (,47% response rate), the retention rates were similar to other recent
studies in this area [27]. Results showed that the program significantly increased
participants’ cooking confidence in all generalised cooking skill areas tested from
baseline (T1) to program completion (T2) and the increase in the intervention
group was also sustained six months later which is indicative of an enduring
program effect. These findings resonate with those from Garcia et al, 2012 who
also found sustained positive effects of a cooking skills program on participants’
cooking confidence at one year post intervention [27]. Statistically significant
increases between T1 and T2 were also found in the weekly frequency of cooking
the main meal from basic ingredients and in the consumption of vegetables with
the main meal together with a reduction in weekly take-away consumption,
although effect sizes were modest. Whilst daily fruit consumption increased in the
intervention group between baseline and both T2 and T3, there was no significant
difference between groups at T2. Daily vegetable intake however increased
significantly by just over a half a serve per day within the intervention group from
baseline to T2, but not in the control group, and continued to increase (but not
significantly), between T2 to T3. The T3 intervention mean vegetable intake was
significantly higher, by 0.74 serves per day, than the reported Queensland State-
wide monitoring result [38].
These findings are encouraging and align with evaluation of other cooking skills
programs that have found positive results with respect to improved cooking
confidence [26, 27] and/or healthy eating results [25]. In particular, the 0.5 serve
increase in vegetable intake, which was maintained at 6 months follow up in the
intervention group, is comparable to outcomes achieved by other nutrition
education programs exclusively targeting healthy eating in community settings
[43] or low income communities [43]. In light of published relative risk
reductions for coronary heart disease of 7% associated with a one serve per day
increase in vegetable consumption [44] as well as reported risk reductions for
stroke [45] with increased vegetable consumption, an increase of 0.5 serves of
vegetables is likely to have small but positive protective benefits for individuals
and arguably be of public health benefit if achieved across whole populations. It
has been estimated that inadequate fruit and vegetable consumption alone, cost
the Australian health sector AUD 206 million in 2008 [46]; therefore small shifts
in the right direction are a constructive first step.
Additionally, the apparent agreement between results around daily vegetable
intake, consumption of vegetables with the main meal, cooking confidence and
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 13 / 18
cooking from basic ingredients, all of which reported statistically significant
increases, suggests that the program influences many aspects of cooking and
eating behaviours simultaneously and the parallel increases in these variables are
consistent with the causal logic model proposed for the program [31]. The results
of the intervention only analysis over time also implies a sustaining of program
effect 6 months post program for most variables reported. However, there was
very little change from completion (T2) to 6 months post program (T3), therefore
in agreement with Garcia et al [27], a refresher or booster class around this time
may be warranted to enhance program effect in the long run.
Furthermore, acknowledging that self-reported daily fruit intake did not result
in a statistically significant difference between groups at T2 (P50.06), a 0.40
serves per day increase was still found between T1 and T3 in the intervention
group only analysis (P,0.001). Nevertheless the statistically significant results
between groups for self-reported vegetable intake and not for self-reported fruit
intake possibly reflect the emphasis within the Jamie’s Ministry of Food program
to promote and teach inclusion of vegetables or salad with the main meal.
The results of this study offer the strongest quantitative evidence at this point in
time to specifically support the premise that Jamie’s Ministry Food program
improves cooking confidence and leads to healthier cooking and eating
behaviours. The evaluation also adds to the literature more generally around the
effectiveness of cooking skills programs per se, particularly given the relatively
large sample size, use of a control group and six month follow up measures, all
design elements which have been lacking in many other studies of community
cooking skills interventions [24]. It is acknowledged that there is a reliance on self-
reported data which is not as valid as other more objective measurement tools
used in traditional nutrition evaluation [47]. Yet the self-administered
questionnaire, with a focus on simple English, was well completed, which was an
important consideration, given the program was based in an area of social
disadvantage. The use of objective measurement tools would have been difficult in
this context because of the participant burden, as shown in previous studies [26].
Efforts were also made to increase data validity: guidance was provided to
respondents on portion size, plus a question was included on vegetable intake
within the main meal which is considered good practice [48]; also, importantly,
questions used in Government preventive health surveillance surveys were
included for comparison. As an aside, baseline mean daily vegetable intake
reported in both groups was similar (2.46, 2.49 for each group) to that reported by
the Queensland surveillance survey (2.4) [38] suggesting that our sample was a
valid representation of the population. A wait-list, non-randomised study design
was selected as the most suitable for this context and ensured participants were
able to participate with family and friends at a convenient time. This led to some
disparity between groups in terms of numbers and demographic characteristics at
baseline including gender, age and employment status. However by adjusting for
these covariates, the sensitivity of the group comparison to between- group
differences at baseline, was able to be tested and proven to have little impact on
results. The inclusion of 6 month follow up results in this study demonstrates the
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 14 / 18
sustainability of effects on intervention participants. While it would have been
ideal to extend the control comparison to this point, it was not feasible to expect
wait-list participants to wait a further 6 months before starting the program. It has
been recommended that studies into community cooking skills interventions
ideally use a community control group [24] to avoid this issue all together, but
this was not considered a feasible option for this evaluation.
Jamie’s Ministry of Food Program in Australia, has succeeded in recruiting
large numbers in the communities in which it is operating and has attracted both
private and public investment even when it precedes evidence of program
effectiveness. This may speak to the influence of Jamie Oliver himself as program
ambassador and his capacity to mobilise various players within communities.
Whether or not this celebrity endorsement augments program effectiveness and
what role such programs should play in the promotion of public health nutrition,
remain questions for future research as does determination of the efficiency of the
program measured against known alternatives through cost-effectiveness analysis.
However the results suggest that the Jamie’s Ministry of Food Program holds
promise as a community-based strategy and should certainly be considered as part
of a comprehensive approach to improving diet quality.
Conclusion
This evaluation has, for the first time, demonstrated that the Jamie’s Ministry of
Food Program has positive personal, dietary and likely health impacts for
participants through improvements in cooking confidence, cooking and eating
behaviours towards a healthier diet. These benefits were sustained at 6 months
post program suggesting the program does have an enduring effect on program
participants and should be considered as a component in any suite of
interventions targeting healthy eating.
Supporting Information
S1 Table. Cooking confidence measures at baseline and follow up adjusted by age,
gender and levels of employment independently and all together.
doi:10.1371/journal.pone.0114673.s001 (DOCX)
S2 Table. Cooking and eating measures at baseline and follow up adjusted by age
gender and levels of employment independently and all together.
doi:10.1371/journal.pone.0114673.s002 (DOCX)
S3 Table. Results of ordinal logistic regression for confidence to cook questions.
doi:10.1371/journal.pone.0114673.s003 (DOCX)
S4 Table. Results of the logistic models for dichotomised confidence questions
(‘‘not confident’’ and ‘‘confident’’).
doi:10.1371/journal.pone.0114673.s004 (DOCX)
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 15 / 18
Acknowledgments
The authors acknowledge Alicia Peardon and staff of The Good Foundation
(including staff at the Ipswich site) for facilitating implementation of the
evaluation, and all participants who generously offered their time to complete the
evaluation surveys. Moodie and Swinburn are researchers within an NHMRC
Centre for Research Excellence in Obesity Policy and Food Systems
(APP1041020).
Author ContributionsConceived and designed the experiments: AF JH EW LG BS MM JR. Performed
the experiments: JH AF. Analyzed the data: AF JH JR. Wrote the paper: AF JH EW
LG BS JR MM.
References
1. von Ruesten A, Feller S, Bergmann MM, Boeing H (2013) Diet and risk of chronic diseases: resultsfrom the first 8 years of follow-up in the EPIC-Potsdam study. Eur J Clin Nutr 67: 412–419.
2. Asif M (2014) The prevention and control the type-2 diabetes by changing lifestyle and dietary pattern.J Educ Health Promot 3: 1.
3. Woodside JV, Young IS, McKinley MC (2013) Fruit and vegetable intake and risk of cardiovasculardisease. Proc Nutr Soc 72: 399–406.
4. Larsson SC, Akesson A, Wolk A (2014) Overall diet quality and risk of stroke: a prospective cohortstudy in women. Atherosclerosis 233: 27–29.
5. Arem H, Reedy J, Sampson J, Jiao L, Hollenbeck AR, et al. (2013) The Healthy Eating Index 2005and risk for pancreatic cancer in the NIH-AARP study. J Natl Cancer Inst 105: 1298–1305.
6. Thomson CA, McCullough ML, Wertheim BC, Chlebowski RT, Martinez ME, et al. (2014) Nutritionand physical activity cancer prevention guidelines, cancer risk, and mortality in the women’s healthinitiative. Cancer Prev Res (Phila) 7: 42–53.
7. Jennings A, Welch A, van Sluijs EM, Griffin SJ, Cassidy A (2011) Diet quality is independentlyassociated with weight status in children aged 9-10 years. J Nutr 141: 453–459.
8. Schulze MB, Fung TT, Manson JE, Willett WC, Hu FB (2006) Dietary patterns and changes in bodyweight in women. Obesity (Silver Spring) 14: 1444–1453.
9. Australian Institute of Health and Welfare (2012) Australia’s Health 2012. Canberra.
10. Nishida C, Uauy R, Kumanyika S, Shetty P (2004) The joint WHO/FAO expert consultation on diet,nutrition and the prevention of chronic diseases: process, product and policy implications. Public HealthNutr 7: 245–250.
11. Janssen I (2013) The public health burden of obesity in Canada. Can J Diabetes 37: 90–96.
12. Lal A, Moodie M, Ashton T, Siahpush M, Swinburn B (2012) Health care and lost productivity costs ofoverweight and obesity in New Zealand. Aust N Z J Public Health 36: 550–556.
13. Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, et al. (2012) A comparative risk assessment ofburden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380: 2224–2260.
14. Soliah LAL, Walter JM, Jones SA (2011) Benefits and Barriers to Healthful Eating: What Are theConsequences of Decreased Food Preparation Ability? Am J Lifestyle Med 6: 152–158.
15. Gillman MW, Rifas-Shiman SL, Frazier AL, Rockett HR, Camargo CA, Jr., et al. (2000) Family dinnerand diet quality among older children and adolescents. Arch Fam Med 9: 235–240.
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 16 / 18
16. Woodruff SJ, Hanning RM, McGoldrick K, Brown KS (2010) Healthy eating index-C is positivelyassociated with family dinner frequency among students in grades 6-8 from Southern Ontario, Canada.Eur J Clin Nutr 64: 454–460.
17. Gough B, Conner MT (2006) Barriers to healthy eating amongst men: a qualitative analysis. Soc SciMed 62: 387–395.
18. Chang MW, Nitzke S, Guilford E, Adair CH, Hazard DL (2008) Motivators and barriers to healthfuleating and physical activity among low-income overweight and obese mothers. J Am Diet Assoc 108:1023–1028.
19. Caraher M, Dixon P, Lang T, Carr-Hill R (1999) The state of cooking in England: the relationship ofcooking skills to food choice. Brit Food J 101: 590–609.
20. Hartmann C, Dohle S, Siegrist M (2013) Importance of cooking skills for balanced food choices.Appetite 65: 125–131.
21. van der Horst K, Brunner TA, Siegrist M (2011) Ready-meal consumption: associations with weightstatus and cooking skills. Public Health Nutr 14: 239–245.
22. Winkler E, Turrell G (2010) Confidence to Cook Vegetables and the Buying Habits of AustralianHouseholds. J Am Diet Assoc 110: S52–S61.
23. Hawkes C, Jewell J, Allen K (2013) A food policy package for healthy diets and the prevention ofobesity and diet-related non-communicable diseases: the NOURISHING framework. Obes Rev 14 Suppl2: 159–168.
24. Rees R, Hinds K, Dickson K, O’Mara-Eves A, Thomas J (2012) Communities that cook. A systematicreview of the effectiveness and appropriateness of interventions to introduce adults to home cooking.London: EPPI-Centre Social Science Research Unit, Institute of Education, University of London.
25. Reicks M, Trofholz AC, Stang JS, Laska MN (2014) Impact of Cooking and Home Food PreparationInterventions Among Adults: Outcomes and Implications for Future Programs. J Nutr Educ Behav.
26. Wrieden WL, Anderson AS, Longbottom PJ, Valentine K, Stead M, et al. (2006) The impact of acommunity-based food skills intervention on cooking confidence, food preparation methods and dietarychoices - an exploratory trial. Public Health Nutr 10: 203–211.
27. Garcia AL, Vargas E, Lam PS, Smith F, Parrett A (2013) Evaluation of a cooking skills programme inparents of young children - a longitudinal study. Public Health Nutr: 1–9.
28. Flynn MM, Reinert S, Schiff AR (2013) A Six-Week Cooking Program of Plant-Based RecipesImproves Food Security, Body Weight, and Food Purchases for Food Pantry Clients. J Hung & Env Nutr8: 73–84.
29. Condrasky M, Graham K, Kamp J (2006) Cooking with a Chef: An Innovative Program to ImproveMealtime Practices and Eating Behaviors of Caregivers of Preschool Children. J Nutr Educ Behav 38:324–325.
30. Swinburn B, Bell C, King L, Magarey A, O’Brien K, et al. (2007) Obesity prevention programs demandhigh-quality evaluations. Aust N Z J Public Health 31: 305–307.
31. Flego A, Herbert J, Gibbs L, Swinburn B, Keating C, et al. (2013) Methods for the evaluation of theJamie Oliver Ministry of Food program, Australia. BMC Public Health 13: 411.
32. Herbert J, Flego A, Gibbs L, Waters E, Swinburn B, et al. (2014) Wider impacts of a 10-weekcommunity cooking skills program - Jamie’s Ministry of Food, Australia. BMC Public Health (in press).
33. Office of Economic and Statistical Research. (2011) Queensland Regional Profiles Ipswich Citybased on local government area (2010).
34. Queensland Health (2011) Self- reported Health Status 2009–2010: Local government Area SummaryReport. Brisbane.
35. Backholer K, Mannan HR, Magliano DJ, Walls HL, Stevenson C, et al. (2012) Projectedsocioeconomic disparities in the prevalence of obesity among Australian adults. Aust N Z J PublicHealth 36: 557–563.
36. Bambra CL, Hillier FC, Moore HJ, Summerbell CD (2012) Tackling inequalities in obesity: a protocolfor a systematic review of the effectiveness of public health interventions at reducing socioeconomicinequalities in obesity amongst children. Syst Rev 1: 16.
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 17 / 18
37. Jamie Oliver Food Foundation UK (2014) Jamie’s Ministry of Food. Available: http://www.jamieoliver.com/jamies-ministry-of-food/. Date accessed 10 March 2014.
38. Queensland Health (2012) Self Reported Health Status 2012: Survey report. Brisbane.
39. Barton KL, Anderson AS, Wrieden WL (2011) Validity and realiability of a short questionnaire forassessing impact of cooking skills interventions. J Hum Nutr Diet.
40. Keller HH, Gibbs A, Wong S, Vanderkooy PD, Hedley M (2004) Men can cook! Development,implementation, and evaluation of a senior men’s cooking group. J Nutr Elder 24: 71–87.
41. Queensland Health (2009) Self-Reported Adult Health Status: Queensland. 2009 Survey Report.Brisbane: Queensland Health
42. Lumley T, Diehr P, Emerson S, Chen L (2002) The importance of the normality assumption in largepublic health data sets. Annu Rev Public Health 23: 151–169.
43. Pomerleau J, Lock K, Knai C, McKee M (2005) Interventions Designed to Increase Adult Fruit andVegetable Intake Can Be Effective: a Systematic Review of the Literature. J Nutr 135: 2486–2495.
44. Dauchet L, Amouyel P, Hercberg S, Dallongeville J (2006) Fruit and vegetable consumption and riskof coronary heart disease: a meta-analysis of cohort studies. J Nutr 136: 2588–2593.
45. Boeing H, Bechthold A, Bub A, Ellinger S, Haller D, et al. (2012) Critical review: vegetables and fruitin the prevention of chronic diseases. Eur J Nutr 51: 637–663.
46. Cadilhac D, Magnus A, Cumming T, L S, Pearce D, et al. (2009) The health and economic benefits ofreducing disease risk factors: Research Report Melbourne: Vic Health.
47. Roark RA, Niederhauser VP (2013) Fruit and vegetable intake: issues with definition andmeasurement. Public Health Nutr 16: 2–7.
48. Agudo A (2004) Measuring intake of fruit and vegetables [electronic resource] Background paper for theJoint FAO/WHO Workshop on Fruit and Vegetables for Health, 1–3 September 2004, Kobe, Japan.Geneva: World Health Organisation. Available: http://whqlibdoc.who.int/publications/2005/9241592826_eng.pdf. Date accessed 10 April 2014.
Jamie’s Ministry of Food, Australia: Quasi-Experimental Evaluation
PLOS ONE | DOI:10.1371/journal.pone.0114673 December 16, 2014 18 / 18
Appendix 8. JMoF Australia secondary
outcomes journal article The secondary outcomes paper for Jamie’s Ministry of Food Australia evaluation can be found at: http://www.biomedcentral.com/1471-2458/14/1161
282
RESEARCH ARTICLE Open Access
Wider impacts of a 10-week community cookingskills program - Jamie’s Ministry of Food, AustraliaJessica Herbert1*, Anna Flego1, Lisa Gibbs2, Elizabeth Waters2, Boyd Swinburn3,4, John Reynolds5 and Marj Moodie1
Abstract
Background: Jamie’s Ministry of Food (JMoF) Australia is a 10-week community-based cooking skills program whichis primarily aimed at increasing cooking skills and confidence and the promotion of eating a more nutritious diet.However, it is likely that the program influences many pathways to behaviour change. This paper explores whetherJMoF impacted on known precursors to healthy cooking and eating (such as attitudes, knowledge, beliefs, cookingenjoyment and satisfaction and food purchasing behaviour) and whether there are additional social and healthbenefits which arise from program participation.
Methods: A mixed method, quasi-experimental longitudinal evaluation with a wait-list control was conducted.Intervention participants were measured using repeated questionnaires at three time points; before and after theprogram and at six-month follow-up. Control participants completed the questionnaire 10 weeks before theirprogram and at program commencement. Quantitative analysis used a linear mixed model approach and generalisedlinear models for repeated measures using all available data. Qualitative methods involved 30-minute repeatedsemi-structured interviews with a purposively selected sample, analysed thematically.
Results: Statistically significant differences between groups and over time were found for a reduction of take away/fastfood weekly purchasing (P = 0.004), and increases in eating meals at the dinner table (P = 0.01), cooking satisfaction(P = 0.01), and the ability to prepare a meal in 30 minutes (P < 0.001) and from basics that was low in cost (P < 0.001).The qualitative findings supported the quantitative results. Repeat qualitative interviews with fifteen participantsindicated increased confidence and skills gained from the program to prepare meals from scratch as well as increasesin family involvement in cooking and meal times at home.
Conclusions: Jamie’s Ministry of Food, Australia resulted in improvements in participants’ food and cooking attitudesand knowledge, food purchasing behaviours and social interactions within the home environment, which weresustained six months after the program.
Trial registration: Australian and New Zealand Trial registration number: ACTRN12611001209987.
Keywords: Cooking skills, Healthy eating, Health promotion, Evaluation, Nutrition education
BackgroundThere is a common discourse about a lack of homecooking and food skills in westernized societies today.Factors contributing to this lack of cooking and declinein skills include competing time demands, busy sche-dules, daily stressors, lack of cooking knowledge andconfidence and an increased reliance on prepared food[1-3]. This problem has raised interest from both the
media and health promotion sectors, resulting in anincreased number of not-for-profit community-basedcooking skills programs, both in Australia and inter-nationally [4-9].There are many factors that influence whether indivi-
duals and families will cook and eat healthy meals. Theseinclude cooking and eating knowledge, attitudes and be-liefs, and enjoyment and satisfaction with the cookingexperience. The perceived cost of healthy food can be abarrier to a healthy diet [5,10]. Healthy food choices areinfluenced by food purchasing behaviours, particularlyaround vegetable purchasing. Results from a Brisbane
* Correspondence: [email protected] Health Economics, Faculty of Health, Deakin University, Melbourne,Victoria, AustraliaFull list of author information is available at the end of the article
© 2014 Herbert et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.
Herbert et al. BMC Public Health 2014, 14:1161http://www.biomedcentral.com/1471-2458/14/1161
food study in Australia found that low socio-economic po-sitioned (SEP) groups purchased fewer types of fruit andvegetables, fewer foods high in fibre and low in fat com-pared to high SEP groups [11]. Those with low educationand income levels were less likely to comply with dietaryguidelines in terms of fruit and vegetable consumption asreflected by their fruit and vegetable purchasing [11]. Inaddition, increased confidence to prepare vegetables hasbeen found to be related to purchasing a greater variety ofvegetables and more often [12]. One could assume that byimproving cooking confidence around fruit and vegetablepreparation this may impact on food purchasing attitudesand behaviours.Changes to traditional family meal patterns have been
reported. Busy schedules and competing priorities im-pact on the frequency of family meals [13,14]. The tra-ditional family meal around the dinner table with foodprepared from fresh ingredients or ‘from scratch’ is nolonger a cultural norm [15]. These factors are cause forconcern because the frequency of family meals has beenfound to have a protective health factor through im-proved dietary outcomes [16]. There is evidence to sug-gest that children of families who always eat dinnertogether at the table consume more fruit and vegetablescompared to children of families who never eat together[17]. Eating meals in front of the television is associatedwith negative health impacts such as higher body massindex, particularly in older children [18]. In a nationwidesurvey of 1,011 Australians in 2008, 60% of respondentsreported that the television was always or often on du-ring meal time [19].Jamie’s Ministry of Food (JMoF) was originally developed
by celebrity chef Jamie Oliver in the United Kingdom (UK)[20], and in recent years brought to Australia and adaptedto an Australian setting. The program is a community-based cooking skills program consisting of ten weekly90-minute classes, aimed at getting people of all ages andbackgrounds cooking simple, fresh, healthy food quicklyand easily [21]. The JMoF program focuses on buildingpositive attitudes and increasing knowledge, skills andself-efficacy related to healthy eating, food and cooking.It was brought to Australia by a not-for-profit, healthpromotion organisation, The Good Foundation (TGF),and The Good Guys, a major Australian electrical goodsretailer, in partnership with Jamie Oliver. Ipswich inQueensland was the first Australian centre to open andcommenced operation in April 2011. The Ipswich Centreis primarily funded by philanthropist Mr Andrew Muir(owner of The Good Guys) and Queensland State Govern-ment, as well as local partners. Ipswich was intentionallyselected as a target site given its low SEP [22] and increa-sing levels of overweight and obesity [23] within the popu-lation. This program has been evaluated both in terms ofthe primary aims of increasing cooking confidence and
vegetable consumption (reported elsewhere [24]) and itswider dietary, health and psychosocial impacts.A program logic model was developed to explore the
pathways that might impact on behaviour change interms of cooking and food behaviours [25]. It is under-stood that changes in attitude, beliefs and self-efficacyare important pre-cursors to behaviour change [26].There is evidence to suggest that improved cooking con-fidence may impact on cooking behaviours [9] as well ashealthy diets [12], however the evidence around othereffects on cooking behaviours is less strong and warrantsmore research [27]. This paper explores the 10-weekprogram’s impacts on the other influences on cookingand eating (specifically, cooking and healthy eating atti-tudes, beliefs and knowledge, food purchasing beha-viours, cooking enjoyment and satisfaction, and socialand health benefits).
MethodsThe evaluation methods are detailed in Flego et al. [25],but are briefly summarised here. A longitudinal mixedmethods design was adopted for the evaluation. Thequantitative component measured changes in participantskills, knowledge, attitudes, and food purchasing beha-viours, cooking enjoyment and satisfaction, social con-nectedness around food and health effects as a result ofthe JMoF program. The qualitative component aimed toprovide a deeper understanding of participant expe-riences of the program and to explore the barriers andfacilitators to cooking. The qualitative and quantitativestudies ran concurrently and results were analysed sepa-rately before being combined.
Quantitative studyThe quantitative component used a quasi-experimentaldesign with a wait-list control group. Intervention partici-pants were measured at program commencement (T1), atprogram completion (T2) approximately 10 weeks aftercommencement, and six months later (T3) approximatelysix months after program completion. The control group(participants who registered for the program ten weeks inadvance) were measured ten weeks prior to (T1) and atprogram commencement (T2). Controls were not mea-sured at six month follow-up (T3) because it was deemedimpractical to make participants wait six months to attendthe program. At each time point, participants completeda 15-minute self-administered questionnaire designedto elicit self-reported information around key programdomains.Recruitment to the evaluation was restricted to par-
ticipants over the age of 18 years. The study was designedto detect a change in mean daily vegetable intake (primaryoutcome) of 0.5 serves per day, from a baseline of 2.5serves per day, with 80% power using a two-sided test at
Herbert et al. BMC Public Health 2014, 14:1161 Page 2 of 14http://www.biomedcentral.com/1471-2458/14/1161
the 5% significance level – this required a minimum of140 participants in each group [25]. Recruitment to thestudy continued until the sample reached the target of 140participants in each group at T2. Statistical analyses werebased on the set of individuals who registered for theprogram, responded to an invitation to participate inthe evaluation and subsequently completed the baselinequestionnaire.Multilevel mixed linear models were used to analyse all
continuous-scale, repeated measures data [28,29]. Resultsfrom these analyses are presented in the form of predictedmeans, recovered from the fitted mixed linear model, andtheir associated standard errors (SE). Differences in thesepredicted means over time and within each participantgroup are also presented. Responses to questions on nutri-tional knowledge were dichotomised into correct or incor-rect responses. The proportions of correct responses insubgroups are presented and comparisons of subgroupswere based on generalized linear models, fitted using themethod of generalized estimating equations (GEE) whichallows for longitudinal binary data [29].Two analyses of each outcome variable were con-
ducted: (i) comparisons between groups of their changesover time from T1 to T2 (equivalent to testing for a timeby group interaction), and, (ii) comparisons of the threetime points (T1, T2 and T3) within the interventiongroup. Each model-based analysis was also re-run ad-justing for covariates (such as age, gender, and employ-ment status) when the covariates exhibited baselinedifferences (i.e. differences at T1) between the controland intervention groups. Analyses was performed usingSTATA (version 12.0) [30]. Results were considered sig-nificant if P < 0.05.
Qualitative studyA selected number of JMoF program participants wereapproached to be involved in the qualitative study. Pur-posive sampling was employed, utilising maximum vari-ation [31] to ensure a diverse group of participants interms of socio-demographic characteristics such as socio-economic status, age, gender, family structure, and coo-king confidence level.Recruitment of participants occurred via two ways
[25]. Firstly, in the quantitative questionnaire, partici-pants were asked about their willingness to be contactedfor a future interview. Secondly, purposive sampling wasundertaken by the researcher whilst observing groups intheir first week of the program.The interviews were conducted face-to-face or by phone
by a trained researcher, Figure 1. The first interview wasconducted before the program had commenced or nogreater than two weeks into the program. Repeat inter-views with the same people were conducted at programcompletion, and at six months post completion.
The three repeat semi-structured interviews were ap-proximately 30–40 minutes long and were conducted in apublic location in Ipswich. The interviews explored par-ticipant expectations and experiences of the program andreflections on the impact of the program on their attitudesand behaviours, described in more detail elsewhere [25].Participants were given a $15 supermarket gift voucher atthe end of each interview in appreciation of their time.All interviews were recorded and transcribed verbatim,
with participant consent. The data was managed with theassistance of a qualitative software package NVivo 9 (NVivo9 [program]: QSR International Pty Ltd 2011). Transcriptsand data memos were coded, then categorised to identifythemes and emerging patterns. A second researcher inde-pendently generated codes on a sub-sample of transcriptsand any differences between codes were discussed [32,33].Analysis was conducted concurrently with data collectionallowing for ongoing clarification of emerging findings [32].Further analysis of the themes was conducted in compari-son with relevant literature to determine alignment withexisting evidence [32,33].
EthicsEthics approval was received from Deakin UniversityHuman Research Ethics Committee (HEAG-H 117_11).Informed consent was received for both quantitative andqualitative components of the study [25].
ResultsQuantitative resultsParticipantsFigure 1 provides a summary of the participants involvedin both the quantitative and qualitative components of theevaluation. In the quantitative analysis, a total of 694 inter-vention participants completed T1 measurements, 383 atT2 and 259 at T3. In the wait-list control group, 237 par-ticipants completed the survey at T1and 149 at T2. Furtherdetails of participant dropout rates are presented elsewhere,as is a detailed description of the demographic profile [24].At baseline, most participants spoke English at home, werefemale, resided within the Ipswich district, and were in fullor part-time employment. There were three significant dif-ferences between the control and intervention groups - thecontrol group was younger (64.3% aged below 50 yearscompared to 55.6% in the intervention group and medianages of 48 and 46 years in the control and interventiongroups respectively); was comprised of fewer males (12.8%compared to 22.6% males in the intervention group) andwas more likely to have participants in full employment(34.7% compared to 26.4% in the intervention group).
Food purchasing behaviours and attitudesThere was a statistically significant decrease in totalweekly take away/fast food expenditure in the intervention
Herbert et al. BMC Public Health 2014, 14:1161 Page 3 of 14http://www.biomedcentral.com/1471-2458/14/1161
group (P < 0.001) but not in the control group betweenT1 and T2. This was the only food purchasing behaviourto show a statistically significant group by time interaction(P = 0.004) (Table 1). However total fruit and vegetable ex-penditure did significantly increase by a mean AUD2.50over time in intervention group between T1 and T2(P < 0.001). There was a statistically significant increase inthe numbers believing that they could prepare a mealfrom basics that was low in price between T1 and T2 inthe intervention group (P < 0.001) but not the control.There was also a small significant increase in knowledgearound cost of fruit and vegetables being cheaper when in
season in the intervention group (P < 0.001) but not in thecontrol.When the analysis was restricted to the intervention
group across the three time points (Table 2), overall ex-penditure on food and drink did not change, but therewere significant increases between T1 and T3 in fruitand vegetable expenditure, preparing low cost meal fromscratch, attitudes around buying more fruit and vegeta-bles and the knowledge that fruit and vegetables arecheaper when in season. There was also a significant de-crease in take away/fast food expenditure between T1and T3 (P < 0.001).
Recruitment via telephone number provided, n=12
•A total of 199 participants agreed to be contacted for an interview on their completed baseline quantitative questionnaire and provided a contact telephone number.
•Based on data collection dates and the participant’s program commencement date only 24/199 participants were eligible for an interview. o 12 were scheduled for an interview. o 12 were unavailable for an interview due to time
commitments and interview scheduling constraints.
Recruitment via class observation (n=3) • 4 participants were approached only 3 agreed to
Intervention, n= 1526
<10 weeks before program commencement
Wait-list control, n=434
>10 weeks before program commencement
Quantitative participants, n=1,960 Qualitative participants, n=15
Six-month T3 Interview, n=13
• Face-to-face (n=8) or phone interview (n=5) • Interview with another person present (n=2)
Intervention T3 analysed (n =214)
An additional 45 participants failed to
complete T2 but completed T1 and
T3 (n=259)
Loss to six-month follow-up, (n=2)
• Reasons unknown (n=1) • Could not be contacted (n=1)
T1 Interview, n=15
• Face-to-face interview (n=15) • Interview with another person present (n=3)
T2 Interview, n=15
• Face-to-face (n=11) or phone interview (n=4) • Interview with another person present (n=3)
Wait-list control T2 analysed (n=149)
Intervention T2 analysed
(n=383)
Wait-list control T1 analysed (n = 237)
Intervention T1 analysed
(n= 694)
Figure 1 Jamie’s ministry of food mixed methods evaluation, quantitative and qualitative participation. The quantitative componentbegan in November 2011 – December 2013, the qualitative evaluation began in August 2012 – July 2013.
Herbert et al. BMC Public Health 2014, 14:1161 Page 4 of 14http://www.biomedcentral.com/1471-2458/14/1161
Table 1 Secondary outcome measures by group at baseline and follow up1
Outcome measure Intervention Control Difference betweengroups in changes
over time (interactioneffect)3 P value
Baseline (T1)mean (S.E)2
Follow up(T2) mean (S.E)
Change frombaseline (T2-T1)
mean (S.E) P value
Baseline (T1)mean (S.E)2
Follow up (T2)mean (S.E)
Change frombaseline(T2-T1) mean
(S.E) P value
Food purchasing behaviours and attitudes
Total weekly food and drinkexpenditure (AUD)4
137.16 (2.72) 135.60 (3.15) −1.56 (2.46) P = 0.53 147.34 (4.68) 151.68 (5.20) 4.33 (3.96) P = 0.27 P = 0.21
Total weekly fruit and vegexpenditure (AUD)4
20.77 (0.61) 23.28 (0.73) 2.50 (0.63) P < 0.001 21.70 (1.06) 22.24 (1.20) 0.53 (1.01) P = 0.60 P = 0.10
Total weekly take away/fast food expenditure (AUD)4
13.17 (0.59) 9.86 (0.69) −3.31 (0.55) P < 0.001 12.395 (1.01) 12.05 (1.13) −0.34 (0.87) P = 0.70 P = 0.004
I can prepare a meal frombasics that is low in price5
2.99 (0.03) 3.41 (0.04) 0.41 (0.04) P < 0.001 3.00 (0.05) 2.97 (0.06) −0.02 (0.06) P = 0.71 P <0.001
Buying more fruit/vegetableswould not be difficult on mybudget5a
2.85 (0.03) 2.93 (0.04) 0.08 (0.04) P = 0.06 2.85 (0.06) 2.89 (0.07) 0.04 (0.07) P = 0.59 P = 0.60
Fruit and vegetables arecheaper when they are inseason5
3.42 (0.02) 3.62 (0.03) 0.21 (0.03) P < 0.001 3.43 (0.04) 3.50 (0.05) 0.07 (0.06) P = 0.21 P = 0.04
Cooking and healthy eating knowledge, attitudes beliefs and behaviours
I can put together a healthymeal from scratch in30 minutes5
2.85 (0.031) 3.30 (0.04) 0.45 (0.04) P < 0.001 2.85 (0.05) 2.89 (0.06) 0.03 (0.06) P = 0.61 P <0.001
I find it easy to change myeating habits5
2.52 (0.03) 2.71 (0.04) 0.19 (0.04) P < 0.001 2.52 (0.05) 2.53 (0.06) 0.01 (0.06) P = 0.82 P = 0.02
Vegetables can be tastyfoods5
3.54 (0.02) 3.69 (0.03) 0.15 (0.03) P < 0.001 3.53 (0.04) 3.51 (0.05) −0.02 (0.05) P = 0.74 P = 0.01
I eat enough fruit andvegetables5
2.66 (0.03) 3.00 (0.04) 0.34 (0.04) P < 0.001 2.66 (0.06) 2.68 (0.07) 0.02 (0.06) P = 0.71 P <0.001
My lifestyle does not Preventme eating a healthy diet5a
3.11 (0.03) 3.33 (0.04) 0.22 (0.04) P < 0.001 3.04 (0.05) 3.12 (0.06) 0.08 (0.06) P = 0.17 P = 0.07
Cooking enjoyment and satisfaction
I enjoy cooking5 3.05 (0.03) 3.33 (0.04) 0.28 (0.03) P < 0.001 3.12 (0.05) 3.17 (0.06) 0.06 (0.05) P = 0.28 P = 0.001
I get a lot of satisfactionfrom cooking my meals5
2.96 (0.03) 3.31 (0.04) 0.35 (0.03) P < 0.001 3.02 (0.05) 3.05 (0.06) 0.03 (0.05) P = 0.60 P <0.001
I enjoy cooking for others5 3.01 (0.03) 3.27 (0.04) 0.26 (0.03) P < 0.001 3.09 (0.06) 3.16 (0.07) 0.07 (0.06) P = 0.22 P = 0.004
I enjoy eating a meal withothers5
3.51 (0.02) 3.60 (0.03) 0.09 (0.03) P = 0.01 3.47 (0.39) 3.55 (0.05) 0.07 (0.05) P = 0.16 P = 0.81
Herbert
etal.BM
CPublic
Health
2014,14:1161Page
5of
14http://w
ww.biom
edcentral.com/1471-2458/14/1161
Table 1 Secondary outcome measures by group at baseline and follow up1 (Continued)
Social eating
Frequency of eating togetherat home with others6
3.94 (0.07) 4.20 (0.08) 0.24 (0.07) P < 0.001 3.97 (0.11) 4.02 (0.13) 0.06 (0.11) P = 0.61 P = 0.13
Frequency of eating dinner infront of the television6
2.69 (0.08) 2.50 (0.09) −0.19 (0.07) P = 0.01 2.51 (0.14) 2.52 (0.15) 0.00 (0.11) P = 0.99 P = 0.17
Frequency of eating dinnerat a dinner table6
3.12 (0.08) 3.40 (0.09) 0.29 (0.06) P < 0.001 3.11 (0.13) 3.09 (0.14) −0.02(0.10) P = 0.86 P = 0.01
Health and emotional well-being
Global self-esteem score7 20.88 (0.22) 22.60 (0.25) 1.73 (0.20) P <0.001 20.46 (0.37) 21.02 (0.42) 0.56 (0.32) P = 0.09 P = 0.002
General Health8 2.77 (0.04) 3.11 (0.04) 0.34 (0.04) P <0.001 2.80 (0.06) 2.86 (0.07) 0.06 (0.06) P = 0.34 P <0.001
Body Mass Index (BMI) 28.86 (0.27) 28.78 (0.28) −0.09 (0.13) P = 0.49 29.71 (0.46) 29.70 (0.47) −0.02 (0.20) P = 0.94 P = 0.761Outcomes within each group and over time were determined by a mixed linear model for repeated measures using all available data at each time point using STATA (version 12.0).2Baseline values were not significantly different between groups (independent t tests). 3A significant group and time interaction effect denotes that the response over time differed between groups (P = 0.05).4Expenditure data was collected in Australian dollars (AUD) on a 7-point scale which was analyse by its midpoints. 5Mean predicted score indicating level of agreement with statement from a Likert Scale (1 = stronglydisagree, 2 = somewhat disagree, 3 = somewhat agree, 4 = strongly agree), aScore assignment was reversed. 6Mean frequency for a typical week was collected on a 6 or 7-point scale which was analyse by its midpoint,with the maximum category being five or more times per week. 7Rosenberg’s global self-esteem score (Low self-esteem = 0-14, Normal self-esteem = 15-25, High self-esteem = 16-30). 8Perceived general health (poor = 1,fair = 2, good = 3, very good = 4, excellent = 5).
Herbert
etal.BM
CPublic
Health
2014,14:1161Page
6of
14http://w
ww.biom
edcentral.com/1471-2458/14/1161
Table 2 Secondary outcome measures for the intervention group only at baseline (T1), post intervention (T2) and 6 months follow up (T3)1
Outcome measure Baseline(T1) mean
(S.E)
Follow up(T2) mean
(S.E)
6 months postintervention followup (T3) mean (S.E)
Change frombaseline (T2-T1) mean
(S.E) P value
Change frombaseline (T3-T1) mean
(S.E) P value
Change frombaseline (T3-T2) mean
(S.E) P value
Overall effectof change overtime P value2
Food purchasing behaviours and attitudes
Total weekly food and drinkexpenditure (AUD)3
137.13 (2.65) 135.21 (3.08) 137.28 (3.42) −1.93(2.48) P = 0.44 0.15 (2.90) P = 0.96 2.08 (3.08) P = 0.50 P = 0.70
Total weekly fruit and vegexpenditure (AUD)3
20.77 (0.62) 23.25 (0.74) 23.64 (0.83) 2.48 (0.65) P < 0.001 2.86 (0.76) P < 0.001 0.39 (0.81) P = 0.63 P <0.001
Total weekly take away/fast foodexpenditure (AUD)3
13.19 (0.59) 9.85 (0.68) 9.14 (0.76) −3.34 (0.54) P < 0.001 −4.05 (0.63) P < 0.001 −0.71 (0.68) P = 0.29 P <0.001
I can Prepare a meal from basicsthat is low in Price4
2.99 (0.03) 3.41 (0.04) 3.42 (0.04) 0.42 (0.04) P < 0.001 0.43 (0.05) P < 0.001 0.01 (0.05) P = 0.79 P <0.001
Buying more fruit/vegetableswould not be difficult on mybudget4a
2.85 (0.03) 2.93 (0.04) 2.97 (0.05) 0.08 (0.05) P = 0.09 0.11 (0.05) P = 0.03 0.04 (0.06) P = 0.52 P = 0.06
Fruit and vegetables are cheaperwhen they are in season4
3.42 (0.02) 3.62 (0.03) 3.66 (0.04) 0.21 (0.04) P < 0.001 0.24 (0.04) P < 0.001 0.04 (0.04) P = 0.41 P <0.001
Cooking and healthy eating knowledge, attitudes beliefs and behaviours
I can Put together a healthymeal from scratch in30 minutes4
2.85 (0.03) 3.29 (0.04) 3.31 (0.05) 0.44 (0.04) P < 0.001 0.46 (0.05) P < 0.001 0.02 (0.06) P = 0.67 P <0.001
I find it easy to change myeating habits4
2.52 (0.03) 2.71 (0.04) 2.70 (0.04) 0.17 (0.04) P < 0.001 0.18 (0.05) P < 0.001 0.00 (0.05) P = 0.94 P <0.001
Vegetables can be tasty foods4 3.54 (0.02) 3.69 (0.03) 3.69 (0.04) 0.15 (0.03) P = 0.001 0.15 (0.04) P < 0.001 0.00 (0.04) P = 0.97 P <0.001
I eat enough fruit andvegetables4
2.66 (0.03) 3.00 (0.04) 3.05 (0.05) 0.34 (0.04) P < 0.001 0.39 (0.05) P = 0.001 0.06 (0.05) P = 0.26 P <0.001
My lifestyle does not Prevent meeating a healthy diet4a
3.11 (0.03) 3.32 (0.04) 3.29 (0.05) 0.21 (0.04) P < 0.001 0.18 (0.05) P < 0.001 −0.03 (0.05) P = 0.55 P <0.001
Cooking enjoyment and satisfaction
I enjoy cooking4 3.05 (0.03) 3.32 (0.04) 3.28 (0.04) 0.27 (0.03) P < 0.001 0.23 (0.04) P < 0.001 −0.04 (0.04) P = 0.31 P <0.001
I get a lot of satisfaction fromcooking my meals4
2.96 (0.03) 3.31 (0.04) 3.29 (0.04) 0.35 (0.04) P < 0.001 0.33 (0.04) P < 0.001 −0.02 (0.04) P = 0.72 P <0.001
I enjoy cooking for others4 3.01 (0.03) 3.26 (0.04) 3.18 (0.05) 0.25 (0.04) P < 0.001 0.18 (0.04) P < 0.001 −0.08 (0.05) P = 0.11 P <0.001
I enjoy eating a meal withothers4
3.51 (0.02) 3.60 (0.03) 3.61 (0.03) 0.09 (0.03) P = 0.01 0.10 (0.04) P = 0.01 0.01 (0.04) P = 0.77 P = 0.003
Herbert
etal.BM
CPublic
Health
2014,14:1161Page
7of
14http://w
ww.biom
edcentral.com/1471-2458/14/1161
Table 2 Secondary outcome measures for the intervention group only at baseline (T1), post intervention (T2) and 6 months follow up (T3)1 (Continued)
Social eating
Frequency of eating together athome with others5
3.92 (0.07) 4.17 (0.08) 4.20 (0.09) 0.25 (0.07) P < 0.001 0.28 (0.09) P < 0.001 0.04 (0.09) P = 0.69 P <0.001
Frequency of eating dinner infront of the television5
2.69 (0.08) 2.50 (0.09) 2.46 (0.10) −0.19 (0.07) P = 0.01 −0.23 (0.08) P = 0.01 −0.04 (0.09) P = 0.66 P = 0.01
Frequency of eating dinnerat a dinner table5
3.12 (0.08) 3.40 (0.09) 3.37 (0.10) 0.28 (0.65) P < 0.001 0.25 (0.08) P = 0.001 −0.02 (0.08) P = 0.76 P <0.001
Health and emotional well-being
Global self-esteem score6 20.88 (0.22) 22.61 (0.25) 22.92 (0.28) 1.73 (0.21) P < 0.001 2.04 (0.25) P < 0.001 0.31 (0.26) P = 0.24 P <0.001
General Health7 2.77 (0.04) 3.11 (0.05) 3.24 (0.05) 0.34 (0.04) P < 0.001 0.47 (0.05) P < 0.001 0.13 (0.05) P = 0.01 P <0.001
Body Mass Index (BMI) 28.86 (0.27) 28.79 (0.28) 28.94 (0.29) −0.07 (0.14) P = 0.61 0.08 (0.16) P = 0.65 0.15 (0.17) P = 0.39 P = 0.681Outcomes within each group and over time were determined by a mixed linear model for repeated measures using all available data at each time Point using STATA (version 12.0).2A significant group and time interaction effect denotes that the response over time differed between groups (P < 0.05). 3Expenditure data was collected in Australian dollars (AUD) on a 7-Point scale which wasanalyse by its midpoints. 4Mean Predicted score indicating level of agreement with statement from a Likert Scale (1 = strongly disagree, 2 = somewhat disagree, 3 = somewhat agree, 4 = strongly agree), aScore assignmentwas reversed. 5Mean frequency for a typical week was collected on a 6 or 7-Point scale which was analyse by its midpoint, with the maximum category being five or more times Per week. 6Rosenberg’s globalself-esteem score (Low self-esteem = 0–14, Normal self-esteem = 15-25 and High self-esteem = 16-30). 7Perceived general health (Poor = 1, fair = 2, good = 3, very good = 4, excellent = 5).
Herbert
etal.BM
CPublic
Health
2014,14:1161Page
8of
14http://w
ww.biom
edcentral.com/1471-2458/14/1161
Cooking and healthy eating knowledge, attitudes, beliefsThe belief that participants could prepare a healthy mealfrom scratch in 30 minutes increased significantly in theintervention group but not in the control group betweenT1 and T2. A statistically significant group by time in-teraction (P < 0.001) also indicated a difference betweengroups in their changes and over time (Table 1). Table 1also indicates statistically significant group by time inter-actions in beliefs around ease of changing eating habits(P = 0.02), vegetables being tasty (P = 0.01), and eatingenough fruit and vegetables (P < 0.001). In the inter-vention group only analyses of the three time points, allattitudes and knowledge around cooking and healthyeating were significantly sustained from baseline (T1) to6 months post program (T3) (Table 2).Participants were asked to select the healthiest option
from a range of food choices to test their knowledgearound salt, sugar and fat content (results presented intext and not shown in a table). Based on the test for agroup by time interaction in the GEE analysis, there wasa significant increase in salt knowledge in the interven-tion group with 89.2% of participants indicating the cor-rect answer at baseline and 94.75% at T2 and no changein the control group between T1 and T2 (91.45% and90.41%; P = 0.04. Sugar knowledge increased in the inter-vention group between T1 (87.1%) and T2 (92.2%), andthis was significantly different (P = 0.02) from the changeover time in the control group between T1 and T2(88.94% and 86.49% respectively). Changes in fat know-ledge over time were also significantly different betweenthe control and intervention groups (P = 0.03), with alarger increase in the control group (67.7% at T1, 79.9%at T2) compared to the intervention group (71.0% at T1,74.5% at T2). When the analysis was restricted to partici-pants in the intervention group, between T2 and T3, therewas a significant increase (P = 0.02) in salt knowledge(93.4% at T3). But there was a significant (P = 0.001) de-crease in sugar knowledge (90.3% at T3). Lastly, know-ledge around fat appeared to decrease at T3 (69.4%) butnot significantly (P = 0.42).
Cooking enjoyment and satisfactionThere were small but statistically significant differences inthe increases over time and between groups in cookingenjoyment (P = 0.001), cooking satisfaction (P < 0.001) andcooking for others (P = 0.004) (Table 1). In the inter-vention group, all improvements in the level of cookingenjoyment and satisfaction were sustained at T3 (Table 2).
Social eatingWeekly frequency of eating dinner at the dinner table in-creased significantly in the intervention group (P < 0.001)but not in the control group. The overall group by timeinteraction was statistically significant (P = 0.01) (Table 1).
The improvements in the intervention group in terms ofbehaviours around meal location and eating with otherswere statistically significant over time (Table 2). Betweenbaseline (T1) and six months after the program (T3) therewere significant increases in frequency of eating withothers (P < 0.001) and decreased eating in front of thetelevision (P = 0.01).
Health and emotional well-beingThe JMoF program did not impact on participants’ self-reported BMI between groups and over time (Table 1).There was however a significant group by time inter-action in both self-reported general health (P < 0.001)and self-esteem (P = 0.002). There was a statistically sig-nificant increase in global self-esteem in the interventiongroup (P < 0.001) but not in the control group. Therewas also a significant increase in general health in theintervention group (P < 0.001), but not in the controlgroup. The improvement in general health continued toT3 and the improvement in self-esteem was maintainedat T3 (Table 2).
Adjusted analyses of outcomesAnalyses of each outcome adjusted for age, gender andemployment separately, then all together, to account fordifferences in composition between the non-randomizedintervention and control groups, showed small diffe-rences in the predicted group by time means howeverpairwise comparisons remained similar to those in theresults of the unadjusted analyses (results not shown,see [Additional file 1]).
Qualitative resultsParticipantsFifteen participants participated in the qualitative study.All completed T1 and T2 interviews, whilst 13 competedT3 interviews (with 2 lost to follow up) (Figure 1). Theinterviewees represented a heterogeneous cross-sectionof people from various stages of life. They varied in age(from 21–69 years old), household characteristics andlevels of food preparation, responsibility and confidence,i.e. key factors which impacted on their home cookingand their willingness to learn and ability to makechanges. There were more females interviewed thanmales, which reflects overall program enrolment. Twoparticipants were interviewed together and one partici-pant was interviewed with a carer present. There wereno instances of one person interrupting, correcting orotherwise changing the response of another. However, itis possible that subtle influences arising from the rela-tionship could have influenced participants’ responseswhen another was present. The qualitative sample in-cluded a young adult living at home, both working andstay-at-home mothers, a young adult with an intellectual
Herbert et al. BMC Public Health 2014, 14:1161 Page 9 of 14http://www.biomedcentral.com/1471-2458/14/1161
disability and retired or semi-retired males and females,whose children had left home.
Qualitative findingsThe qualitative findings facilitate a deeper understandingof the quantitative results from the program participants’perspectives. Two key themes to emerge from the datawere changes in food shopping behaviours and in socialinteractions at home through domestic cookingpractices.Participants reported purchasing a wider variety of
fresh foods, such as fruit and vegetables, and less ‘packet’and processed/prepared foods; many viewed this as aconsequence of preparing more meals from scratch. Sixmonths after the program, there were examples of par-ticipants shopping smarter, “buying more to our list” andgrowing their own vegetables and herbs.
“I was stuffing the fatty things [in the trolley] and Iwouldn’t change and try new stuff, which was costingme more money and now I’m trying all these newthings, I might spend a bit more on fresh fruit andvegetables than what I used to but…. It’s a good thing[it] means we are not buying crap…”
However, for a number of participants, shopping be-haviour did not appear to change. Several retired olderparticipants claimed they were “set in their ways” andhad not made many changes to their food purchasing,nor noticed many distinct changes in their food prefer-ences. Some older participants indicated that the effortinvolved and the cost of ingredients, prevented themfrom making, for example a curry paste from scratch,when they could either go out for a meal or purchasethe prepared version. Premixed sauces, for example,were seen as “getting close to authentic” compared towhat they were like in the past. For older participants,prepared meals continued to be an easy option for easeand convenience. However this attitude did not persistamongst younger participants, particularly those withchildren living at home. The decision to make a mealfrom scratch appeared to be more influenced by whomthey were cooking for. Those with children or youngadults residing at home were more likely to invest en-ergy in providing a “proper” meal made from scratchand containing vegetables. There was limited discussionaround the consumption of takeaway food. Howeverthere was discussion around the role that time playswhen cooking at home and how increases in their cook-ing confidence and skills to prepare meals quickly fromscratch may have contributed to a decrease in take-awayexpenditure.As participants gained confidence in their cooking
abilities and found enjoyment in attending the program,
the benefits gained in the class were taken home andshared with others. Firstly, they described sharing theirprogram experience with others, through sharing theknowledge they had acquired and the food they had pre-pared with friends, family and colleagues. For many, thisbrought about feelings of accomplishment and encour-agement and interest from others that did not attend theprogram. Secondly, a small number of participants de-scribed changes after the program in their ability andconfidence to prepare a meal for others. Some partici-pants endeavoured to prepare “fun” meals with others,whether this was sharing the food they made in class, orpreparing and serving a new meal like Jamie would onshared platters, or moving away from eating in front ofthe television to eating at the table. For many, sharingtheir program experience provided a positive experiencethat added to their cooking enjoyment and satisfaction.All interview participants were unanimous about the
importance of eating meals together with other people.There were reports of more social interactions in thehome environment after attending the JMoF program,with many describing “an opportunity to have familytime cooking”. Social cooking and eating as a family re-sulted in shared food enjoyment and special memoriesfor some. There were reported changes to the family en-vironment and in family interactions. It was commonlymentioned, that working as a household team to preparemeals was still occurring six months after finishing theprogram. Some participants reported talking more aboutfood and cooking practices and meal planning than be-fore doing the program. Some reported physical changesto the home environment.
“We've made a bit more space and… we sort of movedsome things around, so now when I’m cooking, the kidscan sit up on the kitchen bench and we can still doreading or some of the homework so there’s still thatinteraction, whereas before we didn’t have it set uplike that… I think it’s changed the dynamics, which wewouldn’t have bothered to change at all if we hadn’tcome along [to the program]… It was like, well if thekids needed help I had to go and help them out. Icouldn’t be cooking. The risotto needs to be stirredcontinuously. You can’t do that if every minute “canyou help me with…” but now it’s like “yeah, I’m doingthis”. Yeah but putting those changes into place…”
DiscussionThis study which uses a longitudinal mixed method de-sign including a wait-list control group, is the first toevaluate a JMoF program. The evaluation adds to thebody of literature around cooking skills interventionswhich to date has been limited in showing effectivenessof impacts [34]. This study has demonstrated that the
Herbert et al. BMC Public Health 2014, 14:1161 Page 10 of 14http://www.biomedcentral.com/1471-2458/14/1161
10-week JMoF program has provided small but positivesustained effect on intervention participants’ attitudes,beliefs, knowledge and enjoyment around cooking andhealthy eating. The strongest changes in attitudes con-cerned being able to prepare a meal in 30 minutes andable to prepare a meal from basics that was low in cost.This was also reflected in participants’ pride in beingable to discuss and demonstrate their program expe-riences to others. There were also small improvementsin eating at the dinner table, expenditure on take-awayor fast foods, as well as self-perceived health and self-esteem. Attitudes and beliefs are understood to be goodpredictors of behaviour [35], so improved attitudes asso-ciated with participation in the program is a positiveoutcome. The confidence and skills gained by partici-pants from attending the JMoF program gave them afresh attitude to cooking, which in turn enabled them toreview and change cooking and meal practices at home.The baseline values suggest that the evaluation partici-pants started in the mid-range of cooking skills, attitudesand knowledge. So there is a possibility that the programcaptured people who may not have needed it and mayattract people with at least some cooking skills whichthey wanted to improve. While each individually mea-sured change was small, together they represent a movetowards positive behaviour change.Six months after finishing the program, JMoF partici-
pants were spending on average 4.15AUD less on takeaway/fast foods per week, which is consistent with thefinding that participants were consuming less take away/fast food [24]. To put this into context the cost of a “BigMac” in Australia according to the Big Mac index is ap-proximately AUD5.15 [36]. Whilst participants’ overallweekly food and drink expenditure did not change, theywere spending more on fruit and vegetables. This alignswith qualitative findings which indicated that there was achange in attitude around food spending with some par-ticipants favouring cooking from scratch using fresh in-gredients rather than packaged convenience foods. Thisappeared to be a direct consequence of their improvedcooking confidence and knowledge. This attitude shift wasaligned with life stage, with younger participants and thosewho had children living at home more motivated tochange. In summary, food expenditure changes appearedto be driven from a prioritisation of cooking from scratchin the act of providing a ‘proper’ meal to those living athome.Another cooking skills intervention program con-
ducted in Australia, The Food Cent$ program, aimed toimprove diets and change food purchasing behavioursthrough developing budgeting, cooking and shoppingskills and knowledge [5]. Through the provision of bud-geting skills the program led to significant dietary beha-viour changes - more vegetables consumed, decreased
consumption of confectionary items and fewer purchasesof cakes. The Can Cook Family programme in the UK[37] showed that participants increased their mean per-centage weekly spend on fruit and vegetable expenditureby 2.55% after the program, which is similar to the esti-mated increase in the percentage (2.07% = 100 × ( (23.64/137.28) – (20.77/137.13)) six months after participation inthe JMoF program. The Can Cook Family Programme alsoled to a reduction of takeaway meals purchased and indi-cated this was due to improvements in participants’ coo-king confidence [37]. These two programs (Food Cent$and Can Cook programs) however had small sample sizesand no control group. The JMoF program has a largersample size and a control group and therefore makes astronger case that cooking skills interventions have an im-pact on food purchasing behaviours and this can lead tohealthier diets.Results showed that over time the JMoF program led
to an increase in eating at the table and an associateddecrease in eating in front of the TV. These changeswere also reflected in the qualitative findings. Partici-pants, particularly those with children living at home,found more enjoyment from cooking and involvingothers in planning, cooking and sharing meals. Thereare a number of consequences arising from changes insocial and mealtime behaviours. Firstly, social connec-tedness can be fostered through positive family relation-ships [38]. Cooking and eating together at home resultsin families spending more quality time together, therebyproviding an opportunity for social support and im-proved family relationships. Increased family meal fre-quency and children’s involvement in cooking is alsolikely to improve the nutritional quality of a meal [6,17].Children’s involvement in the cooking process has beenfound to increase consumption of fruit and vegetables[17]. Participants’ greater confidence and enjoyment ofcooking translated to more positive shared experiencesof both preparing meals and eating together as a family.An Australia wide study conducted in 2008 looking at
family dinners found 77% of families ate together at meal-time five or more times per week [19]. To make a com-parison to the JMoF program proportions showed that, atT3, approximately 72% of JMoF participants who had chil-dren living at home reported eating together 5 times perweek. This had increased from approximately 66% atbaseline. Direct comparison must be treated with cautionbecause the JMOF sample represents greater social dis-advantage, and geographical differences than the Huntley(2008) sample. However, it does indicate scope for furtherimprovements.Typically, evaluations of cooking skills interventions
report on social outcomes in terms of social connectionsand support experienced through attendance of the pro-gram such as social support, friendship building and
Herbert et al. BMC Public Health 2014, 14:1161 Page 11 of 14http://www.biomedcentral.com/1471-2458/14/1161
information sharing [8,39,40]. Keller et al. reported fin-dings from a men’s cooking skills intervention in whichqualitative findings suggested the program had a positiveeffect on participants’ sense of self-worth and connectionwith others [8]. Engler-Stringer and Berenbaum [40] ex-plored social support developed through participation incollective kitchens through qualitative participant ob-servations and interviews. Findings suggested that therewere improvements in social isolation, increased socialsupport, participation and sharing resources in theircommunity as well as knowledge of where to find help[40]. There is limited evidence available on the social ef-fect beyond the program. The JMoF evaluation offersnew findings that highlight improved social interactionsthrough domestic cooking within the home.BMI did not show an overall effect either between
groups or over time and this was not unexpected giventhat the program did not focus on weight reduction dur-ing its 10-week intervention. BMI is rarely reported ormeasured in evaluations of cooking skills interventions.One study that did report on children’s BMI after a five90-minute parent/child cooking intervention addressingobesity revealed that BMI did not significantly changeafter program attendance [6]. Another intervention calledRaising the Bar on Nutrition, aimed at food pantry clients,did show a significant reduction in BMI after a six weekcooking program, however this study had low participantnumbers (n = 54) and did not have a control group [41].Hall et al. (2011) states that weight loss response is slow[42] and therefore it would not be expected in a programof this nature and duration. If the study had a T3 control,there is a possibility that results may have shown a differ-ence between control and intervention groups, howeverspeculating this is not possible within the current studydesign and beyond the data presented.While this study has strengths, it is not without its limi-
tations. Qualitative findings are based on a purposive sub-sample of JMoF participants willing to be interviewed. Itsuccessfully captured a range of perspectives but the find-ings, shaped by their personal insights and experiences,may not be translatable to all JMoF participants. Bothquantitative and qualitative findings may possess an ele-ment of social desirability bias which is often common inself-report measures and interview data [43]. On the otherhand such measures are commonly used, until alternativesare devised. Another issue is the possibility of biases re-lated to “Readiness to change” that may impact on thecomparison of wait-list control groups with interventiongroups, a recent paper [44] has attempted to investigatethis (in the context of problem drinking). If the highlyready are forced to wait they might start changing and theeffect of the intervention is underestimated. Accidentalconfounding of readiness-to-change with the controlgroup is then a problem. Randomization would reduce
confounding however randomisation within real life set-tings often poses barriers to the intervention [45]. A ran-domised control would have also been unsuitable forparticipation in the JMoF program because participantsoften attend at a convenient time with family and friends.The low literacy level of some participants may also haveimpacted on results, however literacy of participants werenot measured in this evaluation, which may be required infuture studies. While the questionnaire used clear simplelanguage and was piloted within Ipswich, there may bemisinterpretations of questions within the population.Results presented in this paper suggest there were
sustained improvements in attitudes, knowledge and pur-chasing behaviours around the consumption and prepa-ration of vegetables after the program. The program had asustained impact on participants’ cooking enjoyment andsatisfaction, which linked heavily to improved social inter-action around cooking and meal consumption within thefamily home. Many changes resulting from the programwere statistically significant but small and sustained. Theprogram implementers need to explore ways in which theparticipant benefits gained as a consequence of the pro-gram can be sustained over time.
ConclusionsThis study is the first rigorous evaluation of the JMoFprogram that incorporates a control group, a mixedmethods design, and a follow-up period. Results showedmultiple improvements in participants’ food and cookingattitudes and knowledge, food purchasing behavioursand social interactions within the home environment,which were sustained six months after the program, ad-ding to the limited evidence of the wider impacts ofcooking skills interventions.
Additional file
Additional file 1: Secondary outcome measures betweenintervention and control group at baseline and follow up adjustedfor age, gender, employment and combined. Additional analysesto show adjusted results for intervention and control at baseline andfollow up.
Competing interestsThe evaluation has been commissioned by The Good Foundation.
Authors’ contributionsJH drafted the manuscript, performed the analysis and interpretation ofresults, with AF and LG contributing to the analysis and interpretation ofquantitative and qualitative sections respectively. Authors LG, MM, EW andBS substantially contributed to conception and design of the research,critically reviewed the draft manuscript and provided intellectual content.JH and AF were responsible for project management and data collection.JR provided statistical guidance and performed secondary statistical analysisto validate all data presented and critically reviewed the draft manuscript toprovide intellectual content. All authors read and approved the finalmanuscript.
Herbert et al. BMC Public Health 2014, 14:1161 Page 12 of 14http://www.biomedcentral.com/1471-2458/14/1161
AcknowledgementsThe authors acknowledge Alicia Peardon and staff of The Good Foundation(including staff at the Ipswich site) for facilitating implementation of theevaluation, and all participants who generously offered their time tocomplete the evaluation surveys. Thanks are extended to Ipswich CityCouncil for providing interview space We also thank Catherine Keating, ChristinaStubbs for early contributions to the study design and Dr. MohammadrezaMohebbi for providing additional statistical advice. Authors Moodie andSwinburn are researchers within an NHMRC Centre for Research Excellence inObesity Policy and Food Systems (APP1041020).
Author details1Deakin Health Economics, Faculty of Health, Deakin University, Melbourne,Victoria, Australia. 2Jack Brockhoff Child Health and Wellbeing Program,Centre for Health Equity, Melbourne School of Population Health, TheUniversity of Melbourne, Melbourne, Victoria, Australia. 3WHO CollaboratingCentre for Obesity Prevention, Faculty of Health, Deakin University,Melbourne, Victoria, Australia. 4School of Population Health, Faculty ofMedical and Health Sciences, University of Auckland, Auckland, New Zealand.5Deakin Biostatistics Unit, Faculty of Health, Deakin University, Melbourne,Victoria, Australia.
Received: 1 August 2014 Accepted: 3 November 2014Published: 12 December 2014
References1. Bava CM, Jaeger SR, Park J: Constraints upon food provisioning practices
in ‘busy’ women’s lives: Trade-offs which demand convenience. Appetite2008, 50:486–498.
2. Bisogni CA, Jastran M, Seligson M, Thompson A: How people interprethealthy eating: contributions of qualitative research. J Nutr Educ Behav2012, 44(4):282–301.
3. Soliah LL, Walter JM, Jones SA: Benefits and barriers to healthful eatingwhat are the consequences of decreased food preparation ability?Am J Lifestyle Med 2012, 6(2):152–158.
4. Abbott P, Davison J, Moore L, Rubinstein R: Barriers and enhancers todietary behaviour change for Aboriginal people attending a diabetescooking course. Health Promot J Aust 2010, 21(1):33–38.
5. Foley RM, Pollard CM: Food cent $—implementing and evaluating anutrition education project focusing on value for money. Aust N Z JPublic Health 1998, 22(4):494–501.
6. Fulkerson JA, Rydell S, Kubik MY, Lytle L, Boutelle K, Story M, Neumark-SztainerD, Dudovitz B, Garwick A: Healthy Home Offerings via the MealtimeEnvironment (HOME): feasibility, acceptability, and outcomes of a pilotstudy. Obesity 2010, 18(S1):S69–S74.
7. Garcia AL, Vargas E, Lam PS, Smith F, Parrett A: Evaluation of a cookingskills programme in parents of young children – a longitudinal study.Public Health Nutr 2013, 17(5):1013–1021.
8. Keller HH, Gibbs A, Wong S, Vanderkooy P, Hedley M: Men can cook!Development, implementation, and evaluation of a senior Men’sCooking Group. J Nutr Elder 2004, 24(1):71–87.
9. Wrieden WL, Anderson AS, Longbottom PJ, Valentine K, Stead M, Caraher M,Lang T, Gray B, Dowler E: The impact of a community-based food skillsintervention on cooking confidence, food preparation methods and dietarychoices? An exploratory trial. Public Health Nutr 2007, 10(2):203–211.
10. Abbott PA, Davison JE, Moore LF, Rubinstein R: Effective nutritioneducation for aboriginal Australians: lessons from a diabetes cookingcourse. J Nutr Educ Behav 2012, 44(1):55–59.
11. Turrell G, Hewitt B, Patterson C, Oldenburg B, Gould T: Socioeconomicdifferences in food purchasing behaviour and suggested implications fordiet-related health promotion. J Hum Nutr Diet 2002, 15(5):355–364.
12. Winkler E, Turrell G: Confidence to cook vegetables and the buying habits ofAustralian households. J Am Dietetic Assoc 2010, 110(5, Supplement 1):S52–S61.
13. Jabs J, Devine CM: Time scarcity and food choices: An overview. Appetite2006, 47(2):196–204.
14. Neumark-Sztainer D, Larson NI, Fulkerson JA, Eisenberg ME, Story M: Familymeals and adolescents: what have we learned from Project EAT (EatingAmong Teens)? Public Health Nutr 2010, 13(07):1113–1121.
15. Caraher M, Lang T: Can’t cook, won’t cook: A review of cooking skills andtheir relevance to health promotion. Int J Health Promot Educ 1999,37(3):89–100.
16. Eisenberg ME, Olson RE, Neumark-Sztainer D, Story M, Bearinger LH:Correlations between family meals and psychosocial well-being amongadolescents. Arch Pediatr Adolesc Med 2004, 158(8):792–796.
17. Christian MS, Evans CEL, Hancock N, Nykjaer C, Cade JE: Family meals canhelp children reach their 5 A Day: a cross-sectional survey of children’sdietary intake from London primary schools. J Epidemiol CommunityHealth 2013, 67(4):332–338.
18. MacFarlane A, Cleland V, Crawford D, Campbell K, Timperio A: Longitudinalexamination of the family food environment and weight status amongchildren. Int J Pediatr Obes 2009, 4(4):343–352.
19. Huntley R: White Paper: ‘Because Family Mealtimes Matter’. In Prepared forContinental by Ipsos Australia. Australia: Ipsos Australia; 2008.
20. Jamie Oliver Foundation: Jamie’s Ministry of Food UK. 2013. http://www.jamieoliver.com/jamies-ministry-of-food/.
21. Jamie Oliver Foundation: Jamie’s Ministry of Food Australia. 2013.http://www.thegoodfoundation.com.au/ministry-of-food/.
22. Office of Economic and Statistical Research: Queensland Regional ProfilesIpswich City Based on local government area (2010). In Profile generatedon 27 May 2011. 2011.
23. Queensland Health: Self- reported Health Status 2009–2010: Local GovernmentArea Summary Report. Brisbane: Queensland Health; 2011.
24. Flego A, Herbert J, Waters E, Gibbs L, Swinburn B, Reynolds J, Moodie M:Jamie’s Ministry of Food: Quasi-experimental evaluation of immediateand sustained impacts of a cooking skills program in Australia. PloS one2014. In Press.
25. Flego A, Herbert J, Gibbs L, Swinburn B, Keating C, Waters E, Moodie M:Methods for the evaluation of the Jamie Oliver Ministry of Foodprogram, Australia. BMC Public Health 2013, 13(1):411.
26. Bandura A: Health promotion from the perspective of social cognitivetheory. Psychol Health 1998, 13(4):623–649.
27. Reicks M, Trofholz AC, Stang JS, Laska MN: Impact of cooking and homefood preparation interventions among adults: outcomes andimplications for future programs. J Nutr Educ Behav 2014,46(4):259–276.
28. Pevalin D, Robson K: The Stata Survival Manual. McGraw-Hill International:UK; 2009.
29. Stata Press: Stata User’s Guide Release 12. StataCorp LP: Texas; 2011.30. StataCorp: Stata: Release 12. StataCorp LP: Texas; 2011.31. Patton M: Qualitative Research and Evaluation Methods. 3rd edition.
California: Sage; 2002.32. Green J, Willis K, Hughes E, Small R, Welch N, Gibbs L, Daly J: Generating
best evidence from qualitative research: the role of data analysis. AustN Z J Public Health 2007, 31(6):545–550.
33. Liamputtong P: Qualitative data analysis: conceptual and practicalconsiderations. Health Promot J Aust 2009, 20(2):133–139.
34. Rees R, Hinds K, Dickson K, O’Mara-Eves A, Thomas J: Communities thatcook: a systematic review of the effectiveness and appropriateness ofinterventions to introduce adults to home cooking. In EPPI-Centre report2004. London: EPPI-Centre, Social Science, Research Unit, Institute ofEducation, University of London; 2012.
35. Shepherd R, Raats MM: Attitudes and beliefs in food habits. In Food choice,Acceptance and Consumption. 1st edition. Edited by Meiselman HL, MacFieHJ. London: Blackie Academix & Professional; 1996.
36. D.H., R.L.W: The Big Mac index. [http://www.economist.com/content/big-mac-index]
37. Caraher M, Lloyd S: Findings from the Can Cook! (CiC) Family’Programme! London: City University London; 2013.
38. Ministry of Social Development: The Social Report 2010. Wellington:Ministry of Social Development; 2010 [http://socialreport.msd.govt.nz/tools/downloads.html]
39. Dewolfe JA, Greaves G: The Basic Shelf Experience: A comprehensiveevaluation. Can J Diet Pract Res 2003, 64(2):51–57.
40. Engler-Stringer R, Berenbaum S: Exploring social support throughcollective kitchen participation in three Canadian cities. Can J CommMental Health 2007, 26(2):91–105.
41. Flynn MM, Reinert S, Schiff AR: A six-week cooking program of plant-basedrecipes improves food security, body weight, and food purchases for foodpantry clients. J Hunger Environ Nutr 2013, 8(1):73–84.
42. Hall KD, Sacks G, Chandramohan D, Chow CC, Wang YC, Gortmaker SL,Swinburn BA: Quantification of the effect of energy imbalance onbodyweight. Lancet 2011, 378(9793):826–837.
Herbert et al. BMC Public Health 2014, 14:1161 Page 13 of 14http://www.biomedcentral.com/1471-2458/14/1161
43. de Vaus D: Structured questionnaires and interviews. In Handbook forResearch Methods in Health Sciences. 1st edition. Edited by Minichiello V,Sullivan G, Greenwood K, Axford R. Sydney: Addison-Wesley LongmanAustralia; 1999.
44. Cunningham J, Kypri K, McCambridge J: Exploratory randomizedcontrolled trial evaluating the impact of a waiting list control design.BMC Med Res Methodol 2013, 13(1):150.
45. Black N: Why we need observational studies to evaluate theeffectiveness of health care. BMJ 1996, 312:1215–1218.
doi:10.1186/1471-2458-14-1161Cite this article as: Herbert et al.: Wider impacts of a 10-week communitycooking skills program - Jamie’s Ministry of Food, Australia. BMC PublicHealth 2014 14:1161.
Submit your next manuscript to BioMed Centraland take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at www.biomedcentral.com/submit
Herbert et al. BMC Public Health 2014, 14:1161 Page 14 of 14http://www.biomedcentral.com/1471-2458/14/1161
Appendix 9. Adjusted results for age, gender, employment and combined
297
Adjusted results for age, gender, employment and combined
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Cooking confidence Confidence to cook from basic ingredients4 3.56 (0.04) 4.36 (0.05) 0.81 (0.05) P<0.001 3.69 (0.07) 3.72 (0.08) 0.03 (0.08) P=0.70 P<0.001 Age 3.56 (0.04) 4.37 (0.04) 0.81 (0.05) P<0.001 3.72 (0.08) 3.76 (0.07) 0.04 (0.08) P= 0.61 P<0.001 Gender 3.56 (0.04) 4.38 (0.04) 0.81 (0.05) P<0.001 3.70 (0.08) 3.74 (0.07) 0.05 (0.08) P=0.53 P<0.001 Employment 3.56 (0.04) 4.38 (0.04) 0.81 (0.05) P<0.001 3.70 (0.08) 3.74 (0.07) 0.04 (0.08) P= 0.62 P<0.001 Age, gender, employment 3.57 (0.04) 4.38 (0.04) 0.81 (0.05) P<0.001 3.70 (0.08) 3.75 (0.07) 0.05 (0.08) P= 0.54 P<0.001 Confidence to follow a simple recipe4 4.00 (0.04) 4.53 (0.05) 0.53 (0.04) P<0.001 4.11 (0.06) 4.06 (0.07) -0.06 (0.07) P=0.40 P<0.001 Age 4.00 (0.04) 4.53 (0.04) 0.54 (0.04) P<0.001 4.15 (0.07) 4.07 (0.06) -0.08 (0.07) P=0.21 P<0.001 Gender 4.00 (0.04) 4.54 (0.04) 0.54 (0.04) P<0.001 4.12 (0.07) 4.07 (0.06) -0.05 (0.07) P=0.48 P<0.001 Employment 4.00 (0.04) 4.54 (0.04) 0.53 (0.04) P<0.001 4.13 (0.07) 4.07 (0.06) -0.06 (0.07) P=0.35 P<0.001
Age, gender, employment 4.01 (0.04) 4.55 (0.04) 0.53 (0.04) P<0.001 4.12 (0.07) 4.05 (0.06) -0.07 (0.07) P=0.276 P<0.001
Confidence in preparing and cooking new foods and recipes4 3.35 (0.04) 4.13 (0.05) 0.77 (0.05) P<0.001 3.45 (0.07) 3.55 (0.08) 0.10 (0.08) P=0.22 P<0.001 Age 3.35 (0.04) 4.13 (0.04) 0.77 (0.05) P<0.001 0.07 3.55 (0.07) 0.10 (0.08) P=0.19 P<0.001 Gender 3.36 (0.04) 4.13 (0.04) 0.77 (0.05) P<0.001 3.45 (0.07) 3.54 (0.07) 0.10 (0.08) P=0.22 P<0.001 Employment 3.36 (0.04) 4.14 (0.04) 0.78 (0.05) P<0.001 3.44 (0.07) 3.54 (0.07) 0.10 (0.08) P=0.21 P<0.001 Age, gender, employment 3.37 (0.04) 4.12 (0.04) 0.77 (0.05) P<0.001 3.44 (0.08) 3.53 (0.07) 0.10 (0.08) P=0.22 P<0.001 Confidence that what one cooks will turn out well4 3.21 (0.04) 3.93 (0.05) 0.72 (0.04) P<0.001 3.28 (0.06) 3.35 (0.07) 0.07 (0.07) P=0.30 P<0.001 Age 3.21 (0.04) 3.93 (0.04) 0.72 (0.04) P<0.001 3.30 (0.07) 3.37 (0.06) 0.08 (0.07) P=0.26 P<0.001 Gender 3.21 (0.04) 3.93 (0.04) 0.72 (0.04) P<0.001 3.30 (0.07) 3.36 (0.07) 0.06 (0.07) P=0.35 P<0.001 Employment 3.22 (0.04) 3.94 (0.04) 0.72 (0.04) P<0.001 3.28 (0.07) 3.35 (0.07) 0.07 (0.07) P=0.33 P<0.001 Age, gender, employment 3.22 (0.04) 3.94 (0.04) 0.72 (0.04) P<0.001 3.30 (0.07) 3.38 (0.07) 0.08 (0.07) P=0.28 P<0.001 Confidence to taste new foods never eaten before4 3.47 (0.04) 4.01 (0.05) 0.54 (0.05) P<0.001 3.41 (0.07) 3.51 (0.09) 0.09 (0.08) P=0.25 P<0.001 Age 3.46 (0.05) 4.01 (0.05) 0.55 (0.05) P<0.001 3.43 (0.08) 3.52 (0.08) 0.09 (0.08) P=0.27 P<0.001 Gender 3.46 (0.05) 4.01 (0.05) 0.54 (0.05) P<0.001 3.46 (0.08) 3.56 (0.08) 0.10 (0.08) P=0.18 P<0.001 Employment 3.47 (0.05) 4.02 (0.05) 0.55 (0.05) P<0.001 3.44 (0.08) 3.54 (0.08) 0.10 (0.08) P=0.22 P<0.001 Age, gender, employment 3.46 (0.05) 4.02 (0.05) 0.55 (0.05) P<0.001 3.44 (0.08) 3.54 (0.08) 0.10 (0.08) P=0.22 P<0.001
298
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Combined confidence score5 17.59(0.02) 20.95 (0.2) 3.36 (0.18) P<0.001 17.94 (0.03) 18.17 (0.03) 0.23 (0.28) P=0.41 P<0.001 Age 17.59 (0.17) 20.98 (0.17) 3.39 (0.17) P<0.001 18.05 (0.30) 18.27 (0.28) 0.23 (0.28) P=0.42 P<0.001 Gender 17.60 (0.17) 20.99 (0.17) 3.39 (0.17) P<0.001 18.01 (0.30) 18.27 (0.28) 0.26 (0.28) P=0.34 P<0.001 Employment 17.63 (0.17) 21.02 (0.17) 3.39 (0.17) P<0.001 17.99 (0.30) 18.23 (0.28) 0.24 (0.27) P=0.39 P<0.001 Age, gender, employment 17.64 (0.17) 21.04 (0.17) 3.39 (0.17) P<0.001 17.99 (0.30) 18.24 (0.28) 0.24 (0.28) P=0.387 P<0.001 Cooking and healthy eating knowledge, attitudes and beliefs
I can put together a healthy meal from scratch in 30 minutes6
2.85 (0.031) 3.30 (0.04) 0.45 (0.04) P<0.001 2.85 (0.05) 2.89 (0.06) 0.03 (0.06) P=0.61 P < 0.001
Age 2.86 (0.03) 3.30 (0.03) 0.44 (0.04) P<0.001 2.87 (0.06) 2.89 (0.06) 0.02 (0.06) P=0.76 P < 0.001 Gender 2.86 (0.03) 3.30 (0.03) 0.44 (0.04) P<0.001 2.85 (0.06) 2.87 (0.06) 0.02 (0.06) P=0.75 P < 0.001 Employment 2.86 (0.03) 3.29 (0.03) 0.43 (0.04) P<0.001 2.86 (0.06) 2.88 (0.06) 0.02 (0.06) P=0.72 P < 0.001 Age, gender, employment 2.87 (0.03) 3.30 (0.03) 0.44 (0.04) P<0.001 2.86 (0.06) 2.88 (0.06) 0.01 (0.06) P=0.82 P < 0.001 I find it easy to change my eating habits6 2.52 (0.03) 2.71 (0.04) 0.19 (0.04) P<0.001 2.52 (0.05) 2.53 (0.06) 0.01 (0.06) P=0.82 P = 0.02 Age 2.52 (0.03) 2.71 (0.04) 0.19 (0.04) P<0.001 2.53 (0.05) 2.54 (0.06) 0.01 (0.07) P=0.89 P = 0.02 Gender 2.51 (0.03) 2.70 (0.04) 0.19 (0.04) P<0.001 2.53 (0.05) 2.55 (0.06) 0.02 (0.06) P=0.71 P = 0.03 Employment 2.51 (0.03) 2.70 (0.04) 0.19 (0.04) P<0.001 2.53 (0.05) 2.55 (0.06) 0.01 (0.06) P=0.82 P = 0.02 Age, gender, employment 2.51 (0.03) 2.71 (0.04) 0.20 (0.04) P<0.001 2.54 (0.05) 2.56 (0.06) 0.02 (0.07) P=0.78 P = 0.02 Vegetables can be tasty foods6 3.54 (0.02) 3.69 (0.03) 0.15 (0.03) P<0.001 3.53 (0.04) 3.51 (0.05) -0.02 (0.05) P=0.74 P = 0.01 Age 3.54 (0.02) 3.68 (0.03) 0.14 (0.03) P<0.001 3.54 (0.04) 3.51 (0.05) -0.03 (0.05) P=0.52 P = 0.003 Gender 3.54 (0.02) 3.69 (0.03) 0.15 (0.03) P<0.001 3.53 (0.04) 3.51 (0.04) -0.02 (0.05) P=0.75 P = 0.01 Employment 3.54 (0.02) 3.69 (0.03) 0.15 (0.03) P<0.001 3.53 (0.04) 3.51 (0.05) -0.01 (0.05) P=0.77 P = 0.01 Age, gender, employment 3.54 (0.02) 3.69 (0.03) 0.14 (0.03) P<0.001 3.54 (0.04) 3.51 (0.05) -0.03 (0.05) P=0.57 P = 0.003 I eat enough fruit and vegetables6 2.66 (0.03) 3.00 (0.04) 0.34 (0.04) P<0.001 2.66 (0.06) 2.68 (0.07) 0.02 (0.06) P=0.71 P < 0.001 Age 2.66 (0.03) 2.98 (0.04) 0.32 (0.04) P<0.001 2.70 (0.06) 2.70 (0.06) 0.00 (0.06) P=0.96 P < 0.001 Gender 2.65 (0.03) 2.99 (0.04) 0.34 (0.04) P<0.001 2.68 (0.06) 2.69 (0.06) 0.01 (0.06) P=0.83 P < 0.001 Employment 2.67 (0.03) 3.00 (0.04) 0.33 (0.04) P<0.001 2.68 (0.06) 2.69 (0.06) 0.01 (0.06) P=0.89 P < 0.001 Age, gender, employment 2.67 (0.03) 2.99 (0.04) 0.32 (0.04) P<0.001 2.70 (0.06) 2.70 (0.06) 0.01 (0.06) P=0.93 P < 0.001 My lifestyle does not prevent me eating a healthy diet6a 3.11 (0.03) 3.33 (0.04) 0.22 (0.04) P<0.001 3.04 (0.05) 3.12 (0.06) 0.08 (0.06) P=0.17 P = 0.07
299
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Age 3.11 (0.03) 3.32 (0.04) 0.21 (0.04) P<0.001 3.06 (0.05) 3.14 (0.06) 0.07 (0.06) P=0.17 P = 0.10 Gender 3.11 (0.03) 3.33 (0.04) 0.21 (0.04) P<0.001 3.04 (0.05) 3.12 (0.06) 0.08 (0.06) P=0.19 P = 0.07 Employment 3.12 (0.03) 3.32 (0.04) 0.20 (0.04) P<0.001 3.05 (0.05) 3.13 (0.06) 0.08 (0.06) P=0.18 P = 0.09 Age, gender, employment 3.17 (0.03) 3.32 (0.04) 0.21 (0.04) P<0.001 3.06 (0.05) 3.14 (0.06) 0.08 (0.06) P=0.21 P = 0.08 Cooking enjoyment and satisfaction I enjoy cooking6 3.05 (0.03) 3.33 (0.04) 0.28 (0.03) P<0.001 3.12 (0.05) 3.17 (0.06) 0.06 (0.05) P=0.28 P = 0.001 Age 3.05 (0.03) 3.34 (0.03) 0.29 (0.03) P<0.001 3.08 (0.06) 3.15 (0.06) 0.06 (0.05) P= 0.24 P<0.001
Gender 3.06 (0.03) 3.33 (0.03) 0 .27 (0.03) P<0.001 3.12 (0.06) 3.16 (0.06) 0.04 (0.05) P=0.41 P<0.001
Employment 3.06 (0.03) 3.34 (0.03) 0.28 (0.03) P<0.001 3.10 (0.06) 3.16 (0.06) 0.06 (0.05) P=0.27 P<0.001 Age, gender, employment 3.06 (0.03) 3.35 (0.03) 0.30 (0.03) P<0.001 3.09 (0.06) 3.14 (0.06) 0.05 (0.05) P=0.32 P<0.001 I get a lot of satisfaction from cooking my meals6 2.96 (0.03) 3.31 (0.04) 0.35 (0.03) P<0.001 3.02 (0.05) 3.05 (0.06) 0.03 (0.05) P=0.60 P < 0.001 Age 2.97 (0.03) 3.33 (0.03) 0.36 (0.03) P<0.001 3.00 (0.06) 3.02 (0.06) 0.02 (0.05) P=0.70 P<0.001 Gender 2.97 (0.03) 3.32 (0.04) 0.35 (0.03) P<0.001 3.03 (0.06) 3.03 (0.06) 0.01 (0.05) P=0.89 P<0.001 Employment 2.97 (0.03) 3.32 (0.04) 0.35 (0.03) P<0.001 3.00 (0.06) 3.03 (0.06) 0.02 (0.05) P=0.68 P<0.001 Age, gender, employment 2.97 (0.03) 3.33 (0.03) 0.36 (0.03) P<0.001 3.01 (0.06) 3.02 (0.06) 0.01 (0.05) P=0.83 P<0.001 I enjoy cooking for others6 3.01 (0.03) 3.27 (0.04) 0.26 (0.03) P<0.001 3.09 (0.06) 3.16 (0.07) 0.07 (0.06) P=0.22 P = 0.004 Age 3.01 (0.03) 3.28 (0.04) 0.27 (0.03) P<0.001 3.06 (0.06) 3.13 (0.06) 0.07 (0.06) P=0.19 P = 0.003 Gender 3.02 (0.03) 3.27 (0.04) 0.25 (0.03) P<0.001 3.09 (0.06) 3.14 (0.06) 0 .06 (0.05) P=0.31 P = 0.003 Employment 3.01 (0.03) 3.27 (0.04) 0.26 (0.03) P<0.001 3.07 (0.06) 3.14 (0.06) 0.07 (0.05) P=0.20 P = 0.004 Age, gender, employment 3.02 (0.03) 3.28 (0.04) 0.27 (0.03) P<0.001 3.06 (0.06) 3.12 (0.06) 0.06 (0.06) P=0.26 P = 0.002 I enjoy eating a meal with others6 3.51 (0.02) 3.60 (0.03) 0.09 (0.03) P = 0.01 3.47 (0.39) 3.55 (0.05) 0.07 (0.05) P=0.16 P = 0.81 Age 3.51 (0.02) 3.60 (0.03) 0.09 (0.03) P=0.004 3.47 (0.04) 3.54 (0.04) 0.07 (0.05) P=0.17 P = 0.75 Gender 3.52 (0.02) 3.60 (0.03) 0.08 (0.03) P = 0.01 3.47 (0.04) 3.54 (0.04) 0.07 (0.05) P=0.16 P = 0.83 Employment 3.51 (0.02) 3.60 (0.02) 0.09 (0.03) P=0.004 3.47 (0.04) 3.54 (0.04) 0.07 (0.05) P=0.16 P = 0.76 Age, gender, employment 3.52 (0.02) 3.61 (0.03) 0.09 (0.03) P = 0.01 3.47 (0.04) 3.53 (0.05) 0.07 (0.05) P=0.17 P = 0.75 Consumption behaviours Daily vegetable consumption (serves per day) 2.46 (0.51) 2.97 (0.06) 0.52 (0.06) P<0.001 2.49 (0.09) 2.59 (0.10) 0.10 (0.10) P=0.30 P<0.001 Age 2.46 (0.05) 2.98 (0.06) 0.52 (0.06) P<0.000 2.55 (0.09) 2.60 (0.10) 0.06 (0.10) P= 0.51 P<0.001
300
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Gender 2.47 (0.05) 2.99 (0.06) 0.52 (0.06) P< 0.000 2.48 (0.09) 2.57 (0.10) 0.09 (0.10) P=0.36 P<0.001
Employment 2.48 (0.05) 2.99 (0.06) 0.51 (0.06) P<0.001 2.50 (0.09) 2.59 (0.10) 0.08 (0.10) P=0.39 P<0.001 Age, gender, employment 2.49 (0.05) 3.00 (0.06) 0.51 (0.05) P<0.001 2.51 (0.09) 2.58 (0.10) 0.07 (0.10) P=0.48 P<0.001 Daily fruit consumption (serves per day) 1.65 (0.04) 1.93 (0.05) 0.28 (0.05) P<0.001 1.61 (0.07) 1.71 (0.08) 0.10 (0.08) P=0.20 P=0.06 Age 1.64 (0.04) 1.92 (0.0) 0.28 (0.05) p<0.001 1.65 (0.07) 1.75 (0.09) 0.10 (0.08) P=0.24 P=0.07 Gender 1.64 (0.04) 1.92 (0.05) 0.28 (0.05) p<0.001 1.61 (0.07) 1.72 (0.09) 0.10 (0.08) P=0.21 P=0.06 Employment 1.64 (0.04) 1.91 (0.05) 0.27 (0.05) p<0.001 1.63 (0.07) 1.73 (0.09) 0.10 (0.08) P=0.23 P=0.07 Age, gender, employment 1.64 (0.04) 1.91 (0.05) 0.27 (0.05) P<0.001 1.65 (0.07) 1.75) (0.09) 0.10 (0.08) P=0.25 P=0.07
Take-away consumption7 0.98 (0.04) 0.77 (0.04) -0.21 (0.04) P<0.001 0.94 (0.06) 0.96 (0.07) 0.03 (0.06) P=0.62 P=0.001
Age 0.98 (0.04) 0.74 (0.04) -0.23 (0.04) P<0.001 0.87 (0.07) 0.92 (0.06) 0.04 (0.06) P=0.48 P<0.001
Gender 0.98 (0.04) 0.73 (0.04) -0.25 (0.04) P<0.001 0.94 (0.07) 0.95 (0.06) 0.02 (0.06) P=0.80 P<0.001
Employment 0.97 (0.04) 0.73 (0.04) -0.24 (0.04) P<0.001 0.90 (0.07) 0.92 (0.06) 0.02 (0.06) P=0.72 P<0.001
Age, gender, employment 0.96 (0.04) 0.73 (0.04) -0.25 (0.04) P<0.001 0.89 (0.06) 0.93 (0.06) 0.04 (0.06) P=0.50 P<0.001
Cooking behaviours Cooking the main meal from basic ingredients7 4.05 (0.08) 4.66 (0.09) 0.61 (0.09) p<0.001 4.16 (0.14) 4.17 (0.16) 0.01 (0.15) P= 0.95 P=0.001
Age 4.05 (0.06) 4.68 (0.09) 0.63 (0.09) P<0.001 4.21 (0.15) 4.23 (0.15) 0.02 (0.15) P=0.89 P<0.001
Gender 4.08 (0.08) 4.71 (0.09) 0.63 (0.09) P<0.001 4.11 (0.14) 4.11 (0.15) 0.00 (0.15) P=1.0 P<0.001
Employment 4.07 (0.09) 4.70 (0.09) 0.63 (0.09) P<0.001 4.16 (0.14) 4.17 (0.15) 0.00 (0.15) P=0.99 P<0.001
Age, gender, employment 4.11 (0.08) 4.72 (0.09) 0.62 (0.09) P<0.001 4.13 (0.14) 4.15 (0.15) 0.02 (0.15) P=0.91 P=0.001
Consumption of ready-made meals at home7 1.06 (0.05) 0.95 (0.06) -0.11 (0.06) P=0.06 1.11 (0.08) 1.21 (0.10) 0.10 (0.10) P0.30 P=0.06
301
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Age 1.06 (0.05) 0.96 (0.06) -0.10 (0.06) P=0.10 1.05 (0.08) 1.19 (0.10) 0.14 (0.10) P=0.18 P=0.05 Gender 1.06 (0.05) 0.94 (0.06) -0.12 (0.06) P=0.05 1.09 (0.08) 1.17 (0.10) 0.08 (0.10) P=0.41 P=0.08 Employment 1.04 (0.05) 0.93 (0.06) -0.10 (0.06) P=0.09 1.09 (0.08) 1.20 (0.10) 0.11 (0.10) P=0.26 P=0.06 Age, gender, employment 1.04 (0.05) 0.94 (0.06) -0.09 (0.06) P=0.12 1.06 (0.08) 1.17 (0.10) 0.11 (0.10) P=0.28 P=0.08 Consumption of vegetables with the main meal7 4.67 (0.07) 5.06 (0.09) 0.39 (0.08) P<0.001 4.76 (0.12) 4.75 (0.14) -0.01 (0.14) P=0.94 P=0.01 Age 4.67 (0.07) 5.06 (0.08) 0.39 (0.08) P<0.001 4.87 (0.13) 4.82 (0.14) -0.04 (0.14) P=0.75 P=0.01 Gender 4.67 (0.07) 5.07 (0.08) 0.40 (0.08) P<0.001 4.83 (0.13) 4.81 (0.13) -0.02 (0.14) P=0.86 P=0.01 Employment 4.69 (0.07) 5.08 (0.08) 0.39 (0.08) P<0.001 4.84 (0.13) 4.81 (0.13) -0.03 (0.14) P=0.80 P=0.01 Age, gender, employment 4.70 (0.07) 5.08 (0.08) 0.38 (0.08) P<0.001 4.85 (0.13) 4.82 (0.14) -0.03 (0.14) P=0.86 P=0.01 Food Purchasing behaviours and attitudes Total weekly food and drink expenditure (AUD) 8 137.16
(2.72) 135.60 (3.15) -1.56 (2.46) P =
0.53 147.34 (4.68)
151.68 (5.20)
4.33 (3.96) P=0.27 P = 0.21
Age 136.98 (2.63) 136.42 (2.99)
-0.56 (2.44) P = 0.82
144.15 (4.61)
148.26 (5.01) 4.11 (4.00) P=0.30 P = 0.32
Gender 137.25 (2.75) 135.40 (3.14)
-1.85 (2.46 ) P = 0.45
149.03 (4.79)
151.97 (5.22) 2.94 (3.99) P=0.46 P = 0.31
Employment 137.61 (2.63) 136.46 (2.98)
-1.14 (2.45) P = 0.64
144.82 (4.54)
148.52 (4.89) 3.70 (3.94) P=0.35 P = 0.30
Age, gender, employment 136.84 (2.60) 136.31 (2.95)
-0.53 (2.44) P = 0.83
144.96 (4.56)
148.81 (4.95) 3.85 (4.00) P=0.34 P = 0.35
Total weekly fruit and veg expenditure (AUD) 8 20.77 (0.61) 23.28 (0.73) 2.50 (0.63) P<0.001 21.70 (1.06) 22.24 (1.20) 0.53 (1.01) P=0.60 P = 0.10
Age 20.77 (0.60) 23.53 (9.73) 2.76 (0.63) P<0.001
21.28 (1.06) 21.44 (1.21) 0.16 (1.04) P=0.88 P = 0.03
Gender 20.73 (0.61) 23.31 (0.74) 2.58 (0.63) P<0.001 21.92 (1.07) 22.20 (1.23) 0.28 (1.02) P=0.79 P = 0.06 Employment 20.75 (0.61) 23.56 (0.73) 2.81 (0.63) P<0.001 21.33 (1.05) 21.63 (1.20) 0.29 (1.01) P=0.77 P = 0.04 Age, gender, employment 20.66 (0.60) 23.53 (0.73) 2.87 (0.64) P<0.001 21.37 (1.06) 21.66 (1.22) 0.28 (1.04) P=0.79 P = 0.03 Total weekly take away expenditure (AUD) 8 13.17 (0.59) 9.86 (0.69) -3.31 (0.55)
P<0.001 12.395 (1.01)
12.05 (1.13) -0.34 (0.87) P=0.70 P = 0.004
Age 13.17 (0.58) 9.99 (0.60) -3.19 (0.54) P<0.001 11.50 (1.01) 11.26 (1.0) -0.24 (0.88) P=0.78 P = 0.005
302
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Gender 13.28 (0.62) 9.65 (0.62) -3.63 (0.53) P<0.001 12.47 (1.07) 11.73 ( 1.03) -0.73 (0.86) P=0.40 P = 0.004
Employment 13.22 (0.58) 9.79 (0.59) -3.43 (0.54)P<0.001 11.78 ( 1.01) 11.33 (0.97) -0.45 (0.87) P=0.60 P = 0.004
Age, gender, employment 13.15 (0.57) 9.93 (0.59) -3.22 (0.54) P<0.001 11.56 ( 1.01) 11.07 (0.98) -0.48 (0.88) P=0.57 P = 0.01
I can Prepare a meal from basics that is low in price6 2.99 (0.03) 3.41 (0.04) 0.41 (0.04) P<0.001 3.00 (0.05) 2.97 (0.06) -0.02 (0.06) P=0.71 P < 0.001 Age 3.00 (0.03) 3.40 (0.03) 0.40 (0.04) P<0.001 3.01 (0.05) 2.98 (0.05) -0.02 (0.06) P=0.70 P < 0.001 Gender 3.00 (0.03) 3.41 (0.03) 0.41 (0.04) P<0.001 2.99 (0.05) 2.97 (0.05) -0.03 (0.06) P=0.67 P < 0.001 Employment 3.00 (0.03) 3.40 (0.03) 0.40 (0.04) P<0.001 3.01 (0.05) 2.98 (0.05) -0.03 (0.06) P=0.63 P < 0.001 Age, gender, employment 3.00 (0.03) 3.40 (0.03) 0.40 (0.04) P<0.001 3.00 (0.05) 2.97 (0.05) -0.03 (0.06) P=0.68 P < 0.001 Buying more fruit/vegetables would not be difficult on my budget6a
2.85 (0.03) 2.93 (0.04) 0.08 (0.04) P = 0.06 2.85 (0.06) 2.89 (0.07) 0.04 (0.07) P=0.59 P = 0.60
Age 2.86 (0.03) 2.93 (0.04) 0.07 (0.04) P = 0.10 2.86 (0.06) 2.87 (0.07) 0.01 (0.07) P=0.94 P = 0.43 Gender 2.86 (0.03) 2.93 (0.04) 0.06 (0.04) P = 0.15 2.87 (0.06) 2.90 (0.07) 0.04 (0.07) P=0.60 P = 0.75 Employment 2.87 (0.03) 2.93 (0.04) 0.06 (0.04) P = 0.17 2.85 (0.06) 2.89 (0.07) 0.03 (0.07) P=0.65 P = 0.74
Age, gender, employment 2.87 (0.03) 2.93 (0.04) 0 .06 (0.05) P = 0.16 2.85 (0.06) 2.86 (0.07) 0.01 (0.07) P=0.84 P = 0.58
Fruit and vegetables are cheaper when they are in season6
3.42 (0.02) 3.62 (0.03) 0.21 (0.03) P<0.001 3.43 (0.04) 3.50 (0.05) 0.07 (0.06) P=0.21 P = 0.04
Age 3.41 (0.03) 3.62 (0.03) 0.21 (0.03) P<0.001 3.44 (0.05) 3.51 (0.05) 0.07 (0.06) P=0.23 P = 0.03 Gender 3.41 (0.03) 3.62 (0.03) 0.21 (0.03) P<0.001 3.43 (0.05) 3.50 (0.05) 0.07 (0.06) P=0.19 P = 0.03 Employment 3.41 (0.03) 3.62 (0.03) 0.21 (0.03) P<0.001 3.44 (0.05) 3.50 (0.05) 0.07 (0.06) P=0.24 P = 0.03
Age, gender, employment 3.41 (0.03) 3.62 (0.03) 0.21 (0.03) P<0.001 3.42 (0.05) 3.50 (0.05) 0.07 (0.06) P=0.19 P = 0.05
Social eating Frequency of eating together at home with others9 3.94 (0.07) 4.20 (0.08) 0.24 (0.07) P<0.001 3.97 (0.11) 4.02 (0.13) 0.06 (0.11) P=0.61 P = 0.13 Age 3.93 (0.07) 4.20 (0.08) 0.27 (0.07) P<0.001 3.97 (0.12) 4.08 (0.12) 0.11 (0.11) P=0.31 P = 0.23 Gender 3.93 (0.07) 4.19 (0.08) 0.27 (0.07) P<0.001 4.00 (0.12) 4.07 (0.12) 0.07 (0.12) P=0.51 P = 0.13 Employment 3.95 (0.07) 4.21 (0.08) 0.27 (0.07) P<0.001 3.98 (0.12) 4.08 (0.12) 0.11 (0.11) P=0.33 P = 0.20 Age, gender, employment 3.95 (0.07) 4.21 (0.08) 0.27 (0.07) P<0.001 3.98 (0.12) 4.06 (0.12) 0.08 (0.11) P=0.45 P = 0.15
303
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Frequency of eating dinner in front of the television9 2.69 (0.08) 2.50 (0.09) -0.19 (0.07) P = 0.01
2.51 (0.14) 2.52 (0.15) 0.00 (0.11) P=0.99 P = 0.17
Age 2.69 (0.08) 2.49 (0.09) -0.19 (0.07) P = 0.01 2.49 (0.14) 2.45 (0.15) 0.01 (0.12) P=0.92 P = 0.14
Gender 2.69 (0.08) 2.52 (0.09) -0.18 (0.07) P = 0.01 2.48 (0.14) 2.48 (0.15) 0.00 (0.12) P=0.98 P = 0.18
Employment 2.69 (0.08) 2.52 (0.09) -0.17 (0.07) P = 0.02 2.46 (0.14) 2.48 (0.15) 0.01 (0.12) P=0.90 P = 0.17
Age, gender, employment 2.70 (0.08) 2.51 (0.09) -0.19 (0.07) P = 0.01 2.44 (0.14) 2.44 (0.15) 0.00 (0.12) P=0.97 P = 0.16
Frequency of eating dinner at a dinner table9 3.12 (0.08) 3.40 (0.09) 0.29 (0.06) P<0.001 3.11 (0.13) 3.09 (0.14) -0.02 (0.10) P=0.86 P = 0.01 Age 3.12 (0.08) 3.40 (0.08) 0.28 (0.06) P<0.001 3.14 (0.14) 3.17 (0.14) 0.03 (0.10) P=0.76 P = 0.03 Gender 3.12 (0.08) 3.40 (0.08) 0.28 (0.06) P<0.001 3.13 (0.14) 3.12 (0.14) -0.01 (0.10) P=0.89 P = 0.01 Employment 3.13 (0.08) 3.41 (0.08) 0.28 (0.06) P<0.001 3.15 (0.13) 3.14 (0.14) -0.01 (0.10) P=0.89 P = 0.01 Age, gender, employment 3.13 (0.08) 3.41 (0.08) 0.28 (0.06) P<0.001 3.17 (0.14) 3.20 (0.14) 0.02 (0.10) P=0.82 P = 0.03 Health and emotional well-being Global self-esteem score10 20.88 (0.22) 22.60 (0.25) 1.73 (0.20) P<0.001 20.46 (0.37) 21.02 (0.42) 0.56 (0.32) P=0.09 P = 0.002 Age 20.90 (0.22) 22.58 (0.25) 1.68 (0.20) P<0.001 20.53 (0.38) 21.07 (0.42) 0.54 (0.33) P=0.10 P = 0.003 Gender 20.85 (0.22) 22.58 (0.25) 1.73 (0.20) P<0.001 20.50 (0.38) 21.12 (0.42) 0.62 (0.32) P=0.06 P = 0.003 Employment 20.94 (0.22) 22.65 (0.25) 1.71 (0.20) P<0.001 20.47 (0.37) 21.05 (0.41) 0.58 (0.32) P=0.07 P = 0.003 Age, gender, employment 20.91 (0.21) 22.62 (0.25) 1.71 (0.20) P<0.001 20.60 (0.37) 21.17 (0.42) 0.58 (0.33) P=0.08 P = 0.004 General health11 2.77 (0.04) 3.11 (0.04) 0.34 (0.04) P<0.001 2.80 (0.06) 2.86 (0.07) 0.06 (0.06) P=0.34 P < 0.001 Age 2.77 (0.04) 3.11 (0.04) 0.34 (0.04) P<0.001 2.83 (0.06) 2.90 (0.07) 0.07 (0.06) P=0.26 P < 0.001 Gender 2.76 (0.04) 3.10 (0.04) 0.34 (0.04) P<0.001 2.82 (0.07) 2.89 (0.07) 0.07 (0.06) P=0.24 P < 0.001 Employment 2.77 (0.04) 3.12 (0.04) 0.34 (0.04) P<0.001 2.83 (0.06) 2.89 (0.07) 0.06 (0.06) P=0.33 P < 0.001 Age, gender, employment 2.77 (0.04) 3.11 (0.04) 0.34 (0.04) P<0.001 2.84 (0.07) 2.92 (0.07) 0.07 (0.06) P=0.22 P < 0.001 Body Mass Index (BMI) 28.86 (0.27) 28.78 (0.28) -0.09 (0.13) P =
0.49 29.71 (0.46) 29.70 (0.47) -0.02 (0.20) P=0.94 P = 0.76
Age 28.86 (0.27) 28.77 (0.29) -0.09 (0.13) P = 0.49 29.50 (0.47) 29.47 (0.49) -0.04 (0.21) P=0.86 P = 0.83
304
Intervention Control Interaction effect, group by time P value Outcome measure
Baseline (T1) mean (S.E)
Intervention completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Baseline (T1) mean (S.E)
Wait-list completion (T2) mean (S.E)
Change from T2-T1 mean (S.E) P value
Gender 28.83 (0.27) 28.74 (0.28) -0.09 (0.13) P = 0.49 29.66 (0.46) 29.63 (0.48) -0.03 (0.20) P=0.90 P = 0.80
Employment 28.90 (0.27) 28.80 (0.29) -0.10 (0.13) P = 0.42 29.56 (0.46) 29.53 (0.48) -0.03 (0.20) P=0.90 P = 0.75
Age, gender, employment 28.85 (0.28) 28.75 (0.29) -0.10 (0.13) P = 0.43 29.54 (0.47) 29.50 (0.49) -0.03 (0.21) P=0.87 P = 0.79
Outcomes within each group and over time were determined by a mixed linear model for repeated measures using all available data at each time point. All means and Standard Errors (S.E) have been rounded to 2 decimal points. Baseline values were not significantly different between groups (independent t tests P<0.05). A significant group x time interaction effect denotes that the response over time differed between groups. 4Scale values are 1-5 (where 1= not at all confident and 5 = extremely confident) 5 6Mean Predicted score indicating level of agreement with statement from a Likert Scale (1=strongly disagree, 2=somewhat disagree, 3=some agree, 4=strongly agree), a Score assignment was reversed. 7times per week. 8Expenditure data was collected in Australian dollars (AUD) on a 7-point scale which was analysed by its midpoints. 9Mean frequency for a typical week was collected on a 6 or 7-point scale which was analysed by its midpoint, with the maximum category being five or more times per week. 10Rosenberg's global self-esteem score (Low self-esteem= 0-14, Normal self-esteem=15-25 and High self-esteem =16-30). 11Perceived general health (Poor=1, fair=2, good=3, very good=4, excellent=5).
305