environment correlates of pa_ a review of evidence about parks and recreation.pdf
Post on 03-Jun-2018
224 Views
Preview:
TRANSCRIPT
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
1/41
PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by: [B-on Consortium - 2007]On: 24 September 2008Access details: Access Details: [subscription number 778384760]Publisher RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK
Leisure SciencesPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713773100
Environmental Correlates of Physical Activity A Review of Evidence aboutParks and RecreationAndrew T. Kaczynski a; Karla A. Henderson baUniversity of Waterloo Waterloo, Ontario, Canada bNorth Carolina State University, Raleigh, North Carolina,USA
Online Publication Date: 01 July 2007
To cite this ArticleKaczynski, Andrew T. and Henderson, Karla A.(2007)'Environmental Correlates of Physical Activity: A Review ofEvidence about Parks and Recreation',Leisure Sciences,29:4,315 354
To link to this Article DOI 10.1080/01490400701394865URL http://dx.doi.org/10.1080/01490400701394865
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.
http://www.informaworld.com/smpp/title~content=t713773100http://dx.doi.org/10.1080/01490400701394865http://www.informaworld.com/terms-and-conditions-of-access.pdfhttp://www.informaworld.com/terms-and-conditions-of-access.pdfhttp://dx.doi.org/10.1080/01490400701394865http://www.informaworld.com/smpp/title~content=t713773100 -
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
2/41
Leisure Sciences, 29: 315354, 2007
Copyright C Taylor & Francis Group, LLC
ISSN: 0149-0400 print / 1521-0588 online
DOI: 10.1080/01490400701394865
Research Articles
Environmental Correlates of Physical Activity:A Review of Evidence about Parks and Recreation
ANDREW T. KACZYNSKI
University of Waterloo
Waterloo, Ontario, Canada
KARLA A. HENDERSON
North Carolina State University
Raleigh, North Carolina, USA
Research on physical activity (PA) has expanded in recent years to examine environ-mental influences that enhance or limit the opportunities people have to be active. The
purpose of this study was to review and critically examine evidence related to parks andrecreation as features of the built environment and the relationship of these settings toPA. Fifty studies were retrieved from four major databases that reported an empirical re-lationship between parks or recreation variables and PA variables. Mixed associations
with PA were observed for different types of parks or recreation settings, while proximityto parks or recreation was generally associated with increased PA. Shortcomings existin this literature and many opportunities for researching parks, recreation, and active
living are evident for the future.
Keywords active living, literature review, physical activity, social ecological models
The connection between health and physical activity (PA) is well-documented (Hardman
& Stensel, 2003; Sallis & Owen, 1999). This relationship is emphasized almost daily in
relation to issues such as childhood obesity and the physical inactivity of North American
populations. Only one-quarter of the U.S. population engages in the recommended amount
of PA, and another one-quarter are inactive (Centers for Disease Control and Prevention,2001). Estimates from the 200001 Canadian Community Health Survey indicated that
56% of Canadians were classified as insufficiently active (Canadian Fitness and Lifestyle
Research Institute, 2002). Low levels of PA have been linked to a greater prevalence of
obesity and related diseases including diabetes, cardiovascular disease, and cancer (Ball
& McCargar, 2003; Bassuk & Manson, 2005; Stein & Colditz, 2004; U.S. Department
of Health and Human Services, 1996; Westerlind, 2004). Along with poor nutrition and
smoking, lack of exercise is among the top three modifiable risk factors for chronic disease
Received 24 March 2006; accepted 12 September 2006.
We would like to thank three anonymous reviewers for their insightful comments and Deb Bialeschki for
overseeing the editorial process on this manuscript. This paper was presented at the 2006 Cooper Institute on
Parks, Recreation, and Public Health: Collaborative Frameworks for Promoting Physical Activity.
Address correspondence to Andrew T. Kaczynski, Department of Recreation and Leisure Studies, University
of Waterloo, Waterloo, Ontario, Canada, N2L 3G1. E-mail: atkaczyn@ahsmail.uwaterloo.ca
315
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
3/41
316 A. T. Kaczynski and K. A. Henderson
and premature death (World Health Organization, 2005). Therefore, improving PA levels
has been consistently identified as a top public health priority (U.S. Department of Health
and Human Services, 2000; The Integrated Pan-Canadian Healthy Living Strategy, 2005).
The role that leisure services and parks and recreation settings play in helping people
become physically active, and thus healthier, is beginning to be addressed. For example,
Healthy People 2010(U.S. Department of Health and Human Services, 2000) emphasized
that the design of communities and the presence of parks, trails, and other public recreational
facilities affect peoples abilities to reach the recommended 30 minutes per day of moderate-
intensity PA. Similar recommendations about access to places for PA, and specifically trails
and facilities, were made by a collaboration of government agencies and private partners in
theGuide to Community Preventive Services (Task Force on Community Preventive Ser-
vices, 2002). Yet, the examination and documentation of parks and recreation settings as en-
vironmental correlates of PA is in its infancy. Theinvolvement of leisure scientists in address-
ing many of these health related issues has been lacking (Henderson & Bialeschki, 2005).
The purpose of this study was to review reported empirical evidence about the asso-
ciation between parks and recreation settings (PRSs) as features of the built environmentand PA. Specifically, we aimed to answer the following questions: a) what types of PRSs
are most related to increased PA levels, and b) how is proximity to PRSs related to PA?
We defined PRSs as indoor and outdoor spaces and facilities in the physical and built envi-
ronments designed for leisure and recreation activities. We anticipate that this review will
provide a foundation and catalyst for new research by researchers with expertise in parks
and recreation through transdisciplinary studies with other professionals.
Review of Literature
Sallis, Linton, and Kraft (2005) described how research on PA and health has entered a fourth
major era. The first era (prior to 1970) dealt with physiological studies that examined theimpact of patterns of PA on fitness. The second era (1970s1990s) included epidemiological
studies leading to PA being viewed as a major health priority. The third era that occurred in
a similar time period focused on appropriate interventions for promoting PA. Most of this
research primarily addressed psychosocial factors such as self-efficacy, social support, and
stages of change models (Dishman, 1994; King et al., 2002; Sallis, Kraft, & Linton, 2002).
The fourth era, which started early in this century, focused on a broader range of policy and
environmental factors that promote health such as urban planning, transportation, housing,
and parks and recreation.
At the same time, parks, recreation, and leisure research has gone through research
stages that addressed inputs, outputs, benefits, and explanations and meanings of recreationand leisure. Leisure researchers have moved beyond descriptions of inputs, outputs, and
benefits to analyze how behavior occurs as a result of social and environmental factors (El-
lis, 1993). The activities leisure researchers analyze often are broader than PA. However, to
address health issues, these researchers are joining with other scholars (i.e., primarily in pub-
lic health but also exercise scientists and urban planners) to examine more comprehensively
the built environment associated with environmental correlates and ecological models.
Ecological Models
Ecology generally refers to the interrelations between organisms and their environments
(Hawley, 1950). Social ecological concepts refer to peoples transactions with their physicaland sociocultural environments (Henderson et al., 2001; Sallis & Owen, 2002; Stokols,
1992). Social ecology is derived from systems theory with people-environment transactions
characterized by cultures of mutual influence (Bronfenbrenner, 1979; Green, Richard, &
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
4/41
Parks, Recreation, and Physical Activity 317
Potvin, 1996). The general thesis of ecological models of behavior is that environments
restrict the range of behaviors by promoting and sometimes demanding certain actions and
by discouraging or prohibiting other behaviors. When environmental constraints operate
in significant ways, they emerge as the overriding determinants of behavior (Bandura,
1986). ODonnell (2005) argued, An abundance of opportunity may cancel the need for
education and motivation. An absence of opportunity will probably prevent even the most
knowledgeable and motivated person from practicing a healthy lifestyle (p. iv).
The overemphasis on psychosocial and educational approaches to PA promotion that
dominated the literature until recently is subject to increasing criticism. Marcus and Forsyth
(1999) divided PA promotion efforts into downstream (e.g., programs to increase exercise
self-efficacy), midstream (e.g., mass media campaigns), and upstream (e.g., altering build-
ing codes) interventions. Although they acknowledged at the time that evidence of the effi-
cacy of upstream interventions was limited, they concluded that downstream interventions
produced only 1025% increases in PA and improvements were short-lived. In contrast,
upstream interventions such as adding sidewalks or bike paths are more permanent strate-
gies and can affect greater numbers of people than just those individuals who are the targetsof downstream or midstream interventions such as mailings or physician counseling. Re-
searchers also have shown that psychological and social factors explain less variance in
moderate-intensity PA than vigorous activity (Sallis & Owen, 1999). This finding is signif-
icant because recent PA recommendations focus on promoting moderate-intensity PA (e.g.,
gardening, walking) that is more appealing and practical for a majority of the population
while still providing significant health benefits (Pate et al., 1995). Thus, ecological efforts
add explanatory value beyond the intrapersonal factors that influence peoples involvement
and participation in physically active leisure.
Stokols (1992) suggested that the core assumption of social ecology related to health
promotion is that the healthfulness of the environment and the well-being of people are
influenced by multiple facets of both the physical environment and the social environment.
According to McLeroy et al. (1988), five classes of factors affect how or why a person might
participate or fail to participate in a healthy behavior such as PA: intrapersonal, interpersonal,
institutional, community, and public policy. Intrapersonal is the only factor focused on the
individual and has limited value compared to the other four that acknowledge the context of
peoples lives relative to supportive others as well as enabling and constraining opportunities
and healthy environments. Richard et al. (1996) and Stokols, Allen and Bellingham (1996)
suggested that health problems result from social structure and conditions. Consequently,
constraints and motivations for PA can best be understood by examining the environments
in which people live.
The Built Environment and Physical Activity
The specific relationships between the natural and built environment in communities and
PA have not been prevalent in the leisure and recreation research literature (Henderson &
Bialeschki, 2005). A social psychological approach (Mannell & Kleiber, 1997) has domi-
nated in leisure research with the primary focus on individual behavior within that persons
environment. The leisure field has been concerned more often with the psychological pro-
cesses of activity involvement (e.g., enjoyment, perceived freedom, or social interaction)
than with the physiological product (e.g., lowered blood pressure). From a broader social
and political context the value of parks and recreation has also been related to other goals
such as environmental preservation as well as economic and community development, ratherthan focused on public health.
Public health researchers, however, are examining the significant role that the built
environment plays in fostering PA. Reviews similar to this study examined a wide array of
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
5/41
318 A. T. Kaczynski and K. A. Henderson
environmental variables and found substantial support for the ecological hypothesis. Early
reviews by Sallis, Bauman, and Pratt (1998) and King et al. (1995) lamented a lack of
empirical studies and conceptual models and were the impetus for research that examined
environmental and policy interventions. Since then, researchers have developed increas-
ingly sophisticated systems for classifying PA in community settings (Ainsworth et al.,
2000; Bassett et al., 2000; McKenzie, Cohen, & Sehgal, 2005; McKenzie et al., 2000) as
well as tools for measuring environmental correlates of PA (Brownson et al., 2004; Pikora
et al., 2002; Pikora et al., 2003; Saelens, Sallis, Black, & Chen, 2003). More recent reviews
of the literature (Humpel, Owen, & Leslie, 2002; McCormack et al., 2004; Owen et al.,
2004; Saelens, Sallis, & Frank, 2003) found fairly consistent positive associations between
PA and factors in the built environment such as access to facilities, safety, and aesthet-
ics. These reviews also reported that transportation and zoning variables (e.g., mixed land
use, population density, connectivity of streets, and presence of sidewalks) exhibited strong
relationships with residents PA levels.
The role that PRSs may play in fostering increased PA has received growing attention.
For example, Sallis et al.s (1998) review concluded that children were more active outdoorsand that being outdoors was the most powerful correlate of PA. Corti, Donovan, and Holman
(1997) indicated that parks were more likely to stimulate activity if they were aesthetically
pleasing with tree-lined paths rather than empty open space. Troped et al. (2001) determined
that decreased distance between a persons home and a trail was associated with greater trail
use. Arguments have also been forwarded recently in theAmerican Journal of Preventive
Medicine that PRSs can make significant contributions to facilitating PA (Bedimo-Rung,
Mowen, & Cohen, 2005; Godbey et al., 2005). Additional recognition of the growing role
that parks and recreation has to play in addressing health and PA was also found in a recent
special issue ofLeisure Sciences(Henderson & Bialeschki, 2005), the advocacy promoted
by Payne (2002) and her colleagues (Payne et al., 2005), in National Recreation and Park
Associations (2005) Step up to Health program, and through the focus that the Robert
Wood Johnson Foundations Active Living Research Program has placed on research about
parks and recreation (Sallis & Linton, 2005).
Although research has been undertaken addressing the built environment and PA, no
analysis has systematically considered the role of PRSs. Several previous literature reviews
have included a small number of variables related to parks and recreation (Humpel et al.,
2002; Owen et al., 2004), but little detail is available about the complex associations between
these settings and community-based PA. Therefore, the purpose of this article is to review
and critically examine evidence related to PRSs as features of the built environment and the
relationship they have to PA.
Methods
To address our purpose, we identified peer-reviewed journal articles in which an association
between PRSs and PA was reported as part of the investigation. Although most articles did
not provide an explicit definition of PA, PA is commonly defined as any bodily movement
produced by skeletal muscles that results in energy expenditure (Caspersen, Powell, &
Christenson, 1985, p. 126). In December 2005, searches were conducted within four ma-
jor databasesPsycInfo, PubMed, LeisureTourism Abstracts, and Web of Scienceusing
search terms tailored to each database.1
1PsycInfo: AB = (physical activity OR exercise OR inactivity OR walking) AND AB =(environment OR neighborhood OR urban design OR park OR trail OR greenway). PubMed:Search (Motor Activity[MeSH] OR Exercise[MeSH]) AND Environment Design[MeSH]
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
6/41
Parks, Recreation, and Physical Activity 319
Only articles printed in English were requested, and the date range for articles was
delimited to the period 19982005. Although one review of environmental correlates pre-
dated our selected timeframe (King et al., 1995), as did particular studies mentioned in Sallis
et al.s (1998) review, the year 1998 was considered a reasonable starting point for identi-
fying research related to the built environment and PA for several reasons. For example, as
noted earlier, Sallis et al. (2005) described the fourth phase of PA research that occurred
since the early 2000s as being primarily concerned with understanding and altering policy
and environmental factors. The Centers for Disease Control and Prevention convened a mul-
tidisciplinary conference of health professionals in 1997 and that meeting was the dawn
of what we are now calling the active living movement (Killingsworth, Earp, & Moore,
2003, p. 1). In addition, the 1997 Cooper Institutes annual conference focused on PA in-
terventions including some environmental studies, and papers presented at that conference
were published in a special issue of the American Journal of Preventive Medicineearly the
following year (Blair & Morrow, 1998). Therefore, 1998 was an appropriate starting point
for our analysis.
The searches of the four databases returned a total of 1120 distinct articles after mergingthe results and removing all duplicate records. The primary author then scanned article
abstracts to determine each papers relevance to our current studys purpose. Articles were
initially excluded from further analysis if they failed to meet any of several criteria. First,
we were only interested in studies that included PA as a dependent variable and not research
that measured relationships between the built environment and other health measures (e.g.,
body mass index, mental health, cardiovascular disease). Second, articles that examined
other psychological or interpersonal correlates of PA (e.g., self-efficacy, social support)
without including environmental variables were omitted. Third, articles that analyzed the
built environment or PA concurrently, but only as these two behaviors related to a third
variable or condition (e.g., maximal oxygen uptake) were excluded. Fourth, studies that
simply controlled for environmental influences and/or PA while examining the relationship
between two other variables were also excluded.
Some studies existed that examined worksite and school environments in relation to
PA. The former were excluded because they appeared minimally related, if at all, to PRSs.
Schools, on the other hand, often constitute an important source of indoor and outdoor
recreational opportunities for communities. However, of the few articles uncovered in our
searches that examined school environments related to PA, none were deemed relevant to our
studys purpose. For example, some addressed only physical education class participation as
opposed to the broader concept of PA. In other cases, factors not part of the built environment
were the main focus of the study (e.g., length of breaks during the day, available sports
equipment). However, many if not most studies of the built environment measured thelocation of PA resources (e.g., a playground) in a manner that was so general that school-
based PRSs, especially accessible outdoor ones, would likely be included in the tally of
available resources.
Finally, only original empirical studies were examined. Conceptual papers, review
articles, and studies that were purely methodological in purpose (e.g., validating self-report
measures of the built environment or PA) were not considered. In summary, similar to
Humpel et al.s (2002) work, Only those studies that measured environmental variables
that could be related individually and directly to measured physical activity variables were
Field: MeSH Terms.LeisureTourism Abstracts: ( (environment) in ABSTRACT OR (neighborhood)in ABSTRACT OR (park) in ABSTRACT OR (trail) in ABSTRACT)) AND ((physical activity) inABSTRACT OR (exercise) in ABSTRACT OR (walking) in ABSTRACT)). Web of Science: TS =(physical activity OR exercise OR walking) AND TS = (environment OR neighborhood OR urbandesign OR park OR trail OR greenway); Database = SSCI.
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
7/41
320 A. T. Kaczynski and K. A. Henderson
retained (p. 189). Application of these criteria temporarily reduced the original set of
articles to a list of 105 studies that described empirical relationships between some aspect
of the built environment and PA levels.
Within these 105 articles, four journals were represented by more than five papers:
American Journal of Health Promotion, American Journal of Preventive Medicine,
American Journal of Public Health, and Preventive Medicine. Subsequently, a total of
263 issues for these four journals from 2000 to 2005 were hand-searched to identify any
relevant articles that were missed by the previously described database searches. This pro-
cess revealed only two additional studies that satisfied the aforementioned criteria. One
of the 107 articles identified could not be acquired despite extensive search efforts. Four of
the remaining 106 articles were qualitative studies, and within three of these, the utility of
the contextual findings to our studys purpose was difficult to summarize. Consequently,
all qualitative studies were removed, and we analyzed only quantitative articles. However,
some of this qualitative literature is described in the discussion section of the paper.
Within the remaining set of 102 articles, we sought to identify only those studies that
reported an association between PA and some aspect of parks or recreation settings (e.g.,presence of, distance to) as features of the built environment. Consequently, although rare,
studies with parks and recreation programming (e.g., skills training) as the focus were
excluded from our review. Similarly, associations between PA levels and questionnaire
items that related solely to the presence of recreational equipment (e.g., treadmills) in
respondents homes also were not included in our study.
Some studies that examined environmental influences on PA frequently inquired about
generalized variables such as access to facilities or places to exercise. Such broadly defined
measures were not considered to be clearly related to parks and recreation because they
could easily be interpreted as referring to other PA amenities (e.g., streets). Composite
measures (e.g., aggregations of individual access ratings for multiple types of facilities or
a single summary score covering all aspects of the built environment) were often described
but only included when a large majority of the items in the measure were related to PRSs. In
the end, a total of 50 articles were reviewed that reported empirical quantitative relationships
between PRSs as features of the built environment and PA levels of the study participants.
The following section describes the nature of these relationships.
Results
Table 1 provides brief summaries of the empirical associations that were reported in the 50
primary articles in which a relationship between parks or recreation and PA was directly
analyzed. The first three columns of the table describe the age, location, and size of thestudy sample and whether it was representative of the larger population. Brief descriptions
of the parks or recreation and PA variables are provided along with the associations among
them reported by the original authors.
In 20 of the 50 studies (40%), all or most of the associations examined between parks or
recreation and PA variables were positive (Ball et al., 2001; Bauman et al., 1999; Blanchard
et al., 2005; Booth et al., 2000; Brownson et al., 2001; Chad et al., 2005; Deshpande et al.,
2005; Fisher et al., 2004; Giles-Corti et al., 2005; Giles-Corti & Donovan, 2003, 2002a;
Gordon-Larsen, McMurray, & Popkin, 2000; Humpel et al., 2004; Humpel et al., 2004;
Li, Fisher, & Brownson, 2005; Mota et al., 2005; Reed et al., 2004; Troped et al., 2001;
van Lenthe, Brug, & Mackenbach, 2005; Vernez-Moudon et al., 2005). Nine of the articles
(18%) reported that the associations examined were not significant (Atkinson et al., 2005;
Carver et al., 2005; Duncan & Mummery, 2005; Giles-Corti & Donovan, 2002b; Lund,
2003; Norman et al., 2005; Plaut, 2005; Romero, 2005; Wilcox et al., 2000), while one
study reported a negative relationship (Duncan et al., 2004).
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
8/41
TABLE1
ArticlesReportingEmpiricalA
ssociationsbetweenParksand
RecreationAmenitiesandPhys
icalActivity
Parksorrecr
eation
Physicalactivity
Authorsa
nddate
Population1
N2
variable(s)3
variable(s)3
Association(s)4
1.Addy
etal.
(2004)
18+
yearoldsin
southeasternU.S.
county
1194
Neighborhood(w
ithin0.5
milesor10-minutewalk
ofhome)and
community(10milesor
20-minutedriv
e)
recreationfacilities,
walking/biking
trails,
swimmingpoo
ls,parks,
playgrounds,sports
fields
Sufficientlyactive(
5+
dayswith30+minutes
ofmoderatePAo
r3+
dayswith20+minutes
ofvigorousPAin
past
week)
Insufficientlyactive
(less
PA)
Inactive(nomoderateor
vigorousPA)
Usersofneighborhoodrecreation
facilitiessignificantlymorelikely
tobesufficientlyactive(OR=
4.36)orinsufficientlyactive
(OR=
7.26)thanin
active
Usersofcommunityp
arks
significantlymorelikelytobe
sufficientlyactive(O
R=
1.96)or
insufficientlyactive
(OR=
2.20)
thaninactive
2.Atkinsonet
al.(2005)
Adultsintwo
neighborhoodsin
SanDiego,CA
102
Tallyofconvenience
(5-minutedrive,
10-minutewalk,oron
frequentlytraveled
route)for18re
creational
orexercisefacilities
(yes/noforeac
h)
Numberofself-reported
episodesinpast7
days
ofmoderate,vigo
rous
andtotalPA
Minutesofmoderate,
vigorousandtotalPA
measuredby
accelerometer
Convenientrecreationalfacilities
notsignificantlyrelatedto
moderate,vigorous,
ortotal(r=
.17)self-reportedPA
Convenientrecreationalfacilities
notsignificantlyrelatedto
moderate,vigorous,
ortotal
minutesofobjective
ly-measured
PA
3.Balletal.
(2001)
AdultsinNewSouth
Wales,Australia
3392
Conveniencesum
mary
scoreofagreem
enton
5-ptscalesthat3items
arewithinwalking
distance:shops,parkor
beach,cyclepath
Walkingforexercis
ein
past2weeks(any
vs.
none)
Respondentsreporting
low(OR=
.64)andmoderate(OR=
.84)
convenienceoffacil
ities
significantlylesslikelytowalk
forexercisethantho
sereporting
highconvenienceof
facilities.
Similarresultsfoundwhen
sampledividedinto
thoseinpoor
andgoodhealth.
(Continued
onnextpage)
321
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
9/41
TABLE1
ArticlesReportingEmpiricalA
ssociationsbetweenParksandRecreationAmenitiesandPhysicalActivity(Continued)
ParksorRecreation
PhysicalActiv
ity
Authorsa
ndDate
Population1
N2
Variable(s)3
Variable(s)3
Association(s)4
4.Bauman
etal
.
(199
9)
18+
yearoldsin
NewSouthWales,
Australia
16,178
Liveinpostalcodethat
touchescoastline
Vigorouslyactive(>1600
kcal/wk)Adequately
active(>800kcal/wk)
Sedentary(30)respondents
6.Booth
etal.
(200
0)
60+
yearoldsacross
Australia
449
Accesstolocale
xercise
hall,recreation
center,
cyclepath,golfcourse,
gym,park,swimming
pool,tennisco
urt,
bowlinggreen
(yes/no;
askedindividu
ally)
Sufficientlyactive(>800
kcalskg1energy
expenditureperw
eek)
Inactive(0.25
miles)significantlym
orelikelyto
reportanincreasein
walking
sinceusingtrail
Distancetotrailnotsignificantly
relatedtoanincrease
inwalking
sinceusingthetrail
9.Carver
eta
l.
(20
05)
1214yearoldsin
westernSydney,
Australia
347
Parentsagreementthat
ourneighborhoodhas
goodsportsfacilities(1
or2onascaleranging
from-2to2)
Adolescentsreportsof
theirfrequencyand
durationofwalkin
gand
cyclingforexercis
e,
recreation,transpo
rt,and
to/fromschool
Ofallwalkingorcyclingand
purposecombination
s,sports
facilitiesasignificantpredictorof
onlyfrequency(notduration)of
cyclingfortransport
andonlyin
boys
(Continuedonnextpage)
323
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
11/41
TABLE
1
ArticlesReportingEmpiricalAssociationsbetweenParksand
RecreationAmenitiesandPhysicalActivity(Continued)
ParksorRecreation
PhysicalActivity
AuthorsandDate
Population1
N2
Variable(s)3
Variable(s)3
Association(s)4
10.Chad
etal.
(20
05)
50+
yearoldsina
midsized
Canadiancity
764
Presenceoffacilities
withinneighbo
rhood
(within5-minu
tewalkor
drive):bikingtrails,
walking/hiking
trails,
golfcourse,pu
blicpark,
skatingrink,sw
imming
pool,tenniscourts,
dancestudio,p
ublic
recreationcent
er
Summaryscorefor
participationin12
activitiesofvarying
intensitiesmorespecific
toolderadult
populations(e.g.yard
care,volunteering
,etc.)
SignificantlyhigherPA
scoresfor
respondentsreportingthe
presenceofbikingtrails,
walking/hikingtrails,golfcourse,
publicpark,skatingrink,
swimmingpool,and
tennis
courts.Someminordifferencesin
significanceoffacilitieswhen
samplesplitinto5064,6579,
and80+
agegroups.
11.De Bourdeau
dhuijetal.
(20
03)
1865yearoldsin
Ghent,Belgium
521
Tallyofconvenience
(5-minutedrivefrom
workorhomeoron
frequentlytraveled
route)for18re
creational
orexercisefacilities
(yes/noforeac
h)
Minutesofsitting,
walking,
moderate-intensity,and
vigorous-intensity
activitiesduringp
ast
week(measured
separately)
Forbothmalesandfem
ales,
convenienceoffacilitiesscore
significantlyrelatedtoamountof
vigorousactivityonly
12.Desh
pandeet
al.(2005)
20+
yearoldsin
Missouri,
Tennesseeand
Arkansas
278
Useofcommunity
facilitiesinpast30days
(used/didnotu
se):park,
recreationcent
er,
biking/walking
trail,
publicswimmingpool,
healthclub
Numberofminu
testo
walkfromhom
etoeach
ofabovefacilities
EngageinregularPA
(30+minutesatleast5
daysperweek)
Significantlyincreased
oddsof
engaginginregularP
Afor
respondentswhohad
usedapark
(OR=
4.21),recreationcenter
(OR=
12.20),trail(OR=
3.81),
orhealthclub(OR=
7.48).Odds
increaseddramaticallywithuse
of3+facilities.
Shorterwalkingtimes
topark,
recreationcenter,trail,andhealth
clubassociatedwithmoreregular
PA
324
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
12/41
13.Dunc
anand
Mu
mm-
ery(20
05)
18+
yearoldsin
Rockhampton,
Australia
1281
Streetnetworkdis
tanceto
nearestparkland
Achievesrecommen
ded
PAlevel(150minutesin
pastweek)
Anyrecreationalwalking
inpastweek
Participantswithparklandbeyond
600msignificantlym
orelikelyto
achieverecommende
dPAlevel
(OR=
1.41)thantho
selessthan
600mfromparkland
Parklandproximitynotrelatedto
recreationalwalking
14.Dunc
an
eta
l.
(20
04)
1014yearold
siblingsin58
neighborhoodsin
PacificNorthwest
U.S.
930
Thereareplaygro
unds,
parks,orgymscloseto
myhomeortha
tIcan
gettoeasily(1
5,
disagree-agree)
Numberofparks
and
exerciseand
recreationalfac
ilitiesin
neighborhood
Numberofdaysinp
ast
weekthateachsib
ling
tookpartin:vigorous
exercisefor20+
minutes;stretching
exercises;strength
ening
exercises
Numberofdaysof
vigorousPAinatypical
weekforeachsibl
ing
Perceptionsofneighbo
rhood
recreationalfacilities
andcountof
numberofneighborh
oodPA
facilitieswerebothn
egatively
andsignificantlyrela
tedtofamily
levelsofPA
15.Eyleretal.
(20
03)
18+
yearoldsacross
U.S.
1818
Nowalking/joggingtrails
Regularwalker(5x/week
for30min)
Occasionalwalker(walk
10+
minatleasto
nce
duringpastweek)
Neverwalker(didn
ot
walk10+
minatleast
onceinpastweek)
Neverwalkerssignificantlymore
likelytoreportalack
of
walking/joggingtrailsthan
regularwalkers(OR
=
1.59)
Occasionalwalkersno
t
significantlymorelik
elytoreport
alackofwalking/jog
gingtrails
thanregularwalkers(OR=
1.18)
(Continuedonnextpage)
325
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
13/41
TABLE1
ArticlesReportingEmpiricalA
ssociationsbetweenParksandRecreationAmenitiesandPhysicalActivity(Continued)
ParksorRecreation
PhysicalActiv
ity
Authorsa
ndDate
Population1
N2
Variable(s)3
Variable(s)3
Association(s)4
16.Fisher
etal.
(200
4)
6494yearolds
from56
neighborhoodsin
Portland
582
Totalparks,path
s,trails
perneighborhoodacre
Neighborhoodwalking
activity(scorede
rived
fromindividuals
responsesto3be
havior
questionsratedon5-pt
scale)
Walkingfacilitiesper
neighborhood
acresignificantlyrelatedto
neighborhoodwalkingactivity
17.Foster
etal.
(200
4)
1674yearolds
acrossEngland
4157
Apark/openspaceis
withinwalking
distance
(agree/disagree)
Aleisurecenter
iswithin
walkingdistan
cefrom
myhome(yes/no)
Walking>150min
utesper
weekinpastfour
weeks
Walkingatleast15
minutesperweek
inpast
fourweeks
Inbivariateanalyses,neitherthe
parknorleisurecentervariable
weresignificantlyrelatedtoeither
walkingmeasurein
eithermenor
women
Inmultivariateanalys
es,formen,
havingaparkwithinwalking
distancewasonlyenvironmental
variableassociatedwithhigher
oddsofwalking>1
50minutes
perweek(OR=2.2
2)
18.Giles-Corti
etal
.
(200
5)
1859yearoldsin
Perth,Australia
1803
Threemodelsofaccessto
publicopenspace(each
dividedintoqu
artiles:
verypoor,poor,good,
verygood):
1)Distanceonlymodel
2)Distanceand
attractivenessmodel
3)Distance,
attractiveness,
andsize
model
Achievessufficient
PA
(30+minutesof
moderatePAonmost
daysofweek)
Highlevelsofwalking
(6+
walkingsess
ions
perweektotaling
180+
minutes)
Fordistance-only(OR
=
0.69)and
distanceplusattractiveness
(OR=
0.71)models,pooraccess
topublicopenspacesignificantly
decreasedoddsofachieving
sufficientPAcompa
redtothose
withverypooraccess
Fordistance,attractiv
eness,and
sizemodel,havingverygood
accesstopublicope
nspace
significantly
326
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
14/41
increasedoddsofengagingin
highlevelsofwalkingcompared
tothosewithverypo
oraccess
(OR=
1.50)
19.Giles
-Corti
andDonovan
(20
03)
1859yearoldsin
Perth,Australia
1803
Accesstoattractivepublic
openspace,river,beach,
golfcourse(dividedinto
quartiles)
Walkingatrecommended
levels(12+sessionsin
previous2weeks
totaling360minutesor
more)
Participantsintopquartileofaccess
exhibitedsignificantlyhigher
oddsofsufficientwalkingthan
thoseinbottomquartileofaccess
(OR=
1.47)
20.Giles
-Corti
andDonovan
(20
02a)
1859yearoldsin
Perth,Australia
1803
Accesstoopenspace(top
quartilevs.othe
rthree
quartilescombined)
Accesstobeach(
top
quartilevs.othe
rthree
quartilescombined)
Inpasttwoweeks:a
ny
walkingfortransp
ort;
anywalkingfor
recreation;anyvig
orous
exercise
Walkingasrecomm
ended
(6+
timesperwee
kfor
30+
minutes)
Exercisingvigorous
lyat
recommendedlevel(3+
timesperweekfor20+
minutes)
Beingintopquartileofaccessto
openspacesignifican
tlyincreased
oddsofwalkingfortransport(OR
=
1.35)andwalking
as
recommended(OR=
1.43)
Beingintopquartileofaccessto
beachsignificantlydecreased
oddsofwalkingfortransport(OR
=
0.62),butsignifica
ntly
increasedoddsofwa
lkingfor
recreation(OR=1.4
9),
exercisingvigorouslyatallin
pasttwoweeks(OR=
1.38),and
exercisingvigorouslyat
recommendedlevel(
OR=
1.58)
21.Giles
-Corti
andDonovan
(20
02b)
1859yearoldsin
Perth,Australia
1803
Accesstobuiltfacilities:
e.g.golfcourse,health
club(dividedinto
quartiles)
Accesstonatural
facilities:e.g.beach,
river(dividedin
to
quartiles)
Exercisingas
recommended(30+
minutesofmoderatePA
onmostdaysofw
eek)
Neitheraccesstobuiltfacilitiesnor
accesstonaturalfacilities
significantlyrelatedtoexercising
asrecommended
(Continuedonnextpage)
327
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
15/41
TABLE1
ArticlesReportingEmpiricalA
ssociationsbetweenParksandRecreationAmenitiesandPhysicalActivity(Continued)
ParksorRecreation
PhysicalActiv
ity
Authorsa
ndDate
Population1
N2
Variable(s)3
Variable(s)3
Association(s)4
22.Gomez
etal
.
(200
4)
Grade7studentsat5
schoolsinSan
Antonio,TX
177
Straightlinedistancefrom
participantsh
ometo
nearestopenplayarea
(playground,p
ool,
athleticfield)
Boutsperweekofoutdoor,
non-schoolrelate
dPA
(basedonrecallo
f
numberofdaysp
er
monthandnumberof
monthsperyear)
Distancetonearestop
enplayarea
inverselyandsignifi
cantlyrelated
toboutsperweekofoutdoorPA
inboys,butnoting
irlsortotal
sample
23.Gordo
n-
Lars
en
etal
.
(200
0)
Grade712students
acrossU.S.
17,766
Useofneighborhood
recreationcenter(use/do
notuse)
Numberofepisode
sof
moderatetovigorousPA
perweek(basedon
7-dayactivityrec
all
questions)
Hoursperweekof
physicalinactivity
(TV/videowatchingand
videogameplaying)
Usingrecreationcenter
significantlyincreas
edoddsof
fallingintohighest(5+
episodes/wk)PAcategory(OR=
1.75),butwasnotassociatedwith
beinginthehighest
(25+hrs/wk)
inactivitycategory(OR=
1.01)
24.Hoehn
er
etal
.
(200
5)
1896yearoldsin
areasofSt.Louis,
MOand
Savannah,GA
1073
Therearemanyplacesto
beactiveinmy
community,no
t
includingstree
ts
Park,walkingtrail,private
fitnessfacility
within
5-minutewalk
ofhome
(yes/noindividually)
Numberofrecre
ation
facilitieswithin
5-minutewalk
ofhome
RecreationalPA(not
includingtranspo
rt)
duringleisure-tim
e:
MeetsPA
recommendation
(5+
dayswith30+m
inutes
ofmoderatePAo
r3+
dayswith20+m
inutes
ofvigorousPAin
past
week)
Doesnotmeet
recommendation
Respondentswhoagreedthatthere
aremanyplacestobeactive
(OR=
2.0)andthatreported23
recreationfacilities
within
5-minutewalk(OR
=
1.6)
significantlymorelikelytomeet
PArecommendation
(buttrends
notclear)
Havinganyofpark,w
alkingtrail,
orprivatefitnessfac
ilitywithin
5-minutewalknota
ssociated
withmeetingPA
328
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
16/41
(7total)Within400mof
respondentsho
me:
Countofparkswith
facilities;Anyp
ark,trail
orfitnessfacility;Count
ofrecreationalfacilities
recommendation
Noneoftheobjectivemeasuresof
parksorrecreationfa
cilities
significantlyrelatedtomeeting
PArecommendation
25.Humpel,
Ma
rshall
eta
l.
(20
04)
Facultyandstaffat
anAustralian
university
800and
512at
10-wk
followup
Summaryconvenience
scoreof(eachitemrated
110forunfavo
rableto
favorable):
Walkingdistance
to
park/beachAcc
essibility
ofpath/cycleway
Overallconvenienceof
neighborhoodfor
walking
Numberofminutesper
weekofneighborh
ood
walking
Inbothmenandwome
n,increased
perceptionsofconvenience
relatedtosignificantlyincreased
oddsofanyincrease
inwalking
(OR=
1.95and2.58
,
respectively),increas
eof30or
moreminutesofwalking(OR=
2.02and2.31,respec
tively),and
increaseof60ormoreminutesof
walking(OR=
1.98
and2.01,
respectively)
26.Humpel,
Ow
en,
Ive
rson
eta
l.
(20
04)
40+
yearoldsfrom
acoastal
Australiancity
399
Liveinpostalcod
ethat
touchescoastlin
e
Lakeorbeachwithineasy
walkingdistanc
e
Numberofminutesper
weekofneighborh
ood
walking,walkingfor
exercise,forpleasure,
andtogettoandfrom
places(separately)
Inbivariateanalyses,participants
livingincoastalpostalcode
reportedsignificantly
more
minuteswalkinginn
eighborhood
(189vs.149)andfor
exercise
(139vs.109)thanthosein
non-coastalpostalco
de
Inbivariateanalyses,p
articipants
withalakeorbeachwithin
walkingdistancereported
(Continuedonnextpage)
329
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
17/41
TABLE1
ArticlesReportingEmpiricalA
ssociationsbetweenParksandRecreationAmenitiesandPhysicalActivity(Continued)
ParksorRecreation
PhysicalActiv
ity
Authorsa
ndDate
Population1
N2
Variable(s)3
Variable(s)3
Association(s)4
significantlymorem
inutes
walkinginneighborhood(224vs.
139),forexercise(1
63vs.100),
andforpleasure(33
vs.21)
Inmultivariateanalys
es,livingin
coastalpostalcodenot
significantlyassocia
tedwith
increasedoddsofanytypeof
walkingformen,bu
tgreaterodds
ofneighborhoodwa
lkingfor
women(OR=
3.32
)
27.Hump
el,
Owen,
Leslie
etal
.
(200
4)
Facultyandstaffat
anAustralian
university
800
Summaryconvenience
scoresplitinto
low,
moderate,high
tertiles
basedon(each
item
rated110for
unfavorableto
favorable):
Walkingdistanc
eto
park/beach
Accessibilityofpath/cycle
way
Overallconvenienceof
neighborhood
for
walking
Numberofminutes
per
weekofneighborhood
walking(splitintohigh
andlowgroupsat
median)
Numberofminutesper
weekoftotalwalking
(splitatmedian)
Numberofminutesper
weekoftotalPA
(splitat
median)
Inmen,participantsinhigh
conveniencetertileexhibited
significantlyhigher
oddsof
neighborhoodwalking(OR=
2.20)andtotalPA(OR=
1.82)
thanthoseinlowco
nvenience
tertile
Inwomen,thoseinhigh(OR=
3.78)andmoderate
(OR=
3.19)
conveniencetertiles
exhibited
significantlyhigher
oddsof
neighborhoodwalkingthanthose
inlowconvenience
tertile
28.Huston
etal
.
(200
3)
18+
yearoldsin6
countiesinNorth
Carolina
1796
Trailsinneighbo
rhood
(yes/no)
Anyleisure-timePAin
pastmonth
MeetsPA
recommendation
(5+
Inbivariateanalyses,respondents
reportingpresenceoftrails
significantlymorelikely
330
-
8/12/2019 Environment correlates of PA_ a review of evidence about parks and recreation.pdf
18/41
dayswith30+
minutes
ofmoderatePAor
3+
dayswith20+minutes
ofvigorousPAinpast
week)
toengageinanyPA(77.8%vs.
70.3%)andrecommendedPA
(31.3%vs.23.8%)th
anthose
reportingnotrails
Inmultivariateanalyse
s,reported
presenceoftrailsnot
associated
withhigheroddsofa
nyPA,but
marginallyassociatedwith
recommendedPA(O
R=
1.46,p
top related