bmj open is committed to open peer review. as part of this
TRANSCRIPT
BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available. When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to. The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript. BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com). If you have any questions on BMJ Open’s open peer review process please email
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review onlyProtocol for the Process Evaluation of a Complex
Intervention Delivered in Schools to Prevent Adolescent Depression: The Future Proofing Study
Journal: BMJ Open
Manuscript ID bmjopen-2020-042133
Article Type: Protocol
Date Submitted by the Author: 01-Jul-2020
Complete List of Authors: Beames, Joanne; University of New South Wales, Black Dog InstituteLingam, Raghu; University of New South Wales, School of Women’s and Children’s HealthBoydell, Katherine; University of New South Wales, Black Dog InstituteCalear, Alison L.; Australian National University, Centre for Mental Health ResearchTorok, Michelle; University of New South Wales, Black Dog InstituteMaston, Kate; University of New South Wales, Black Dog InstituteZbukvic, Isabel; Orygen The National Centre of Excellence in Youth Mental HealthHuckvale, Kit; University of New South Wales, Black Dog InstituteBatterham, Philip; Australian National University, Centre for Mental Health ResearchChristensen, Helen; University of New South Wales, Black Dog InstituteWerner-Seidler, Aliza ; University of New South Wales, Black Dog Institute
Keywords: Child & adolescent psychiatry < PSYCHIATRY, Depression & mood disorders < PSYCHIATRY, PREVENTIVE MEDICINE
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on N
ovember 15, 2021 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2020-042133 on 12 January 2021. D
ownloaded from
For peer review onlyI, the Submitting Author has the right to grant and does grant on behalf of all authors of the Work (as defined in the below author licence), an exclusive licence and/or a non-exclusive licence for contributions from authors who are: i) UK Crown employees; ii) where BMJ has agreed a CC-BY licence shall apply, and/or iii) in accordance with the terms applicable for US Federal Government officers or employees acting as part of their official duties; on a worldwide, perpetual, irrevocable, royalty-free basis to BMJ Publishing Group Ltd (“BMJ”) its licensees and where the relevant Journal is co-owned by BMJ to the co-owners of the Journal, to publish the Work in this journal and any other BMJ products and to exploit all rights, as set out in our licence.
The Submitting Author accepts and understands that any supply made under these terms is made by BMJ to the Submitting Author unless you are acting as an employee on behalf of your employer or a postgraduate student of an affiliated institution which is paying any applicable article publishing charge (“APC”) for Open Access articles. Where the Submitting Author wishes to make the Work available on an Open Access basis (and intends to pay the relevant APC), the terms of reuse of such Open Access shall be governed by a Creative Commons licence – details of these licences and which Creative Commons licence will apply to this Work are set out in our licence referred to above.
Other than as permitted in any relevant BMJ Author’s Self Archiving Policies, I confirm this Work has not been accepted for publication elsewhere, is not being considered for publication elsewhere and does not duplicate material already published. I confirm all authors consent to publication of this Work and authorise the granting of this licence.
Page 1 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
1 Protocol for the Process Evaluation of a Complex Intervention Delivered in Schools to
2 Prevent Adolescent Depression: The Future Proofing Study
3
4 Joanne R. Beames1 [email protected]
5 Raghu Lingam2 [email protected]
6 Katherine Boydell1 [email protected]
7 Alison L. Calear3 [email protected]
8 Michelle Torok1 [email protected]
9 Kate Maston1 [email protected]
10 Isabel Zbukvic4 [email protected]
11 Kit Huckvale1 [email protected]
12 Philip J. Batterham3 [email protected]
13 Helen Christensen1 [email protected]
14 Aliza Werner-Seidler1* [email protected]
15 Affiliations:
16 1 Black Dog Institute, University of New South Wales, Sydney, NSW, Australia.
17 2 Population Child Health Clinical Research Group, School of Women’s and Children’s
18 Health, University of New South Wales, NSW, Australia.
19 3 Centre for Mental Health Research, The Australian National University, Canberra, ACT,
20 Australia.
21 4 Orygen, Parkville, Victoria, Australia; Centre for Youth Mental Health, The University of
22 Melbourne, Parkville, Victoria, Australia
23 *Corresponding author:
24 Dr Aliza Werner-Seidler, Black Dog Institute, University of New South Wales, Sydney,
25 NSW, Australia. [email protected]; +61 2 9382 3808
Page 2 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 2
1 Abstract
2 Introduction: Process evaluations provide insight into how interventions are delivered across
3 varying contexts and why interventions work in some contexts and not in others. This
4 manuscript outlines the protocol for a process evaluation embedded in a cluster randomised
5 trial of a digital depression prevention intervention delivered to secondary school students
6 (The Future Proofing Study). The purpose is to describe the methods that will be used to
7 capture process evaluation data within this trial.
8 Methods and Analysis: A mixed methods design will be used with data collected in the
9 intervention arm of the Future Proofing Study. Data collection methods will include semi-
10 structured interviews with school staff and study facilitators, automatically collected
11 intervention usage data, and participant questionnaires (completed by school staff, school
12 counsellors, study facilitators, and students). Information will be collected about: i) how the
13 intervention was implemented in schools, including fidelity; ii) school contextual factors and
14 their association with intervention reach, uptake and acceptability; iii) how school staff, study
15 facilitators, and students and responded to delivering or completing the intervention. How
16 these factors relate to trial effectiveness outcomes will also be assessed. Overall synthesis of
17 the data will provide school cluster-level and individual-level process outcomes.
18 Ethics and Dissemination: Ethics approval obtained from the University of New South
19 Wales. Results will be submitted for publication in peer reviewed journals and discussed at
20 conferences. Our process evaluation will contextualise the trial findings with respect to how
21 the intervention may have worked in some schools but not in others. This evaluation will
22 inform the development of a model for rolling out digital interventions for the prevention of
23 mental illness in schools.
24 Trial Registration: The randomised controlled trial has been prospectively registered:
25 ACTRN12619000855123. Registered 31 May 2019,
Page 3 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 3
1 https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=377664&isReview=true
2 Keywords: Process evaluation, complex intervention, mixed methods, adolescent mental
3 health, depression, prevention, randomised controlled trial, implementation.
Page 4 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 4
1 Strengths and Limitations of this Study
2 This is a process evaluation embedded in a randomised clinical trial, which is critical
3 for understanding how interventions are delivered in real world contexts.
4 This protocol is the first to illustrate how to embed a process evaluation into a school-
5 based trial of a digital depression prevention intervention, including the use of flexible
6 and pragmatic quantitative and qualitative data collection methods.
7 School context factors will be rigorously assessed and analysed for contribution to
8 intervention reach and completion, uptake of study processes, and clinical outcomes.
9 Study outcomes will guide sustainability efforts in the delivery of digital depression
10 prevention interventions in schools.
Page 5 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 5
1 Introduction
2 Embedded in randomised controlled trials, process evaluations provide insight into
3 reasons why interventions work in some contexts and not in others. These process evaluations
4 can help to demystify the ‘black box’ of complex intervention trials by taking into account
5 contextual factors, differences in ways the intervention is delivered, and adaptations made for
6 intervention delivery into a particular system (1, 2). Contextual factors typically include
7 features of the organisation or broader environment that influence the delivery of the
8 intervention (e.g., leadership, engagement, culture, political landscape). Considering the
9 contribution of contextual factors is necessary to aid interpretation of trial outcomes,
10 maximise the knowledge gained from trials, identify optimal delivery processes across
11 different settings, and inform broader dissemination efforts. In recognition of this need for
12 process evaluations, the UK’s Medical Research Council (MRC) set out a framework that
13 emphasises the value of these evaluations in order to capture both contextual and
14 implementation factors associated with complex interventions (3).
15 In line with best-practice recommendations that study protocols which pre-specify
16 methods and approaches should be published to maintain research integrity (4), we describe a
17 protocol for a mixed methods process evaluation embedded within a cluster randomised
18 controlled trial (cRCT) known as the Future Proofing Study (FPS). The FPS is a large school-
19 based trial examining whether depression can be prevented using cognitive behaviour therapy
20 (CBT) delivered by smartphone application (5).
21 There has been an increase in the availability of digital mental health programs which
22 show promise in addressing the significant disease burden associated with depression (6).
23 Depression often first emerges during adolescence (7) and treatment alone cannot adequately
24 reduce this burden (8). Accumulating evidence indicates that early adolescence is the ideal
25 developmental window during which to intervene because it captures young people before
Page 6 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 6
1 the incidence of depression increases exponentially around the age of 16-17 years (9, 10).
2 Therefore, increased efforts are being directed toward prevention approaches that are
3 developed specifically for delivery among adolescents.
4 While there is evidence supporting the value and effectiveness of prevention
5 programs in schools (11, 12), their uptake has been significantly limited by low levels of
6 help-seeking (13) as well as practical constraints in cost and scalability. A lack of
7 understanding about how to deliver these interventions sustainably at scale, and engage those
8 who stand to benefit, have hampered the translation of prevention approaches into the
9 community.
10 Two ways to overcome the barriers of cost and scalability of implementing adolescent
11 depression interventions are:
12 Delivering universal prevention programs in schools, and
13 Using technology to deliver programs, which tends to be lower cost than traditional
14 face-to-face methods (14).
15 Working with schools to deliver universal prevention programs, to every student,
16 regardless of their risk, circumstance, or symptom profile, not only dramatically increases the
17 potential reach of interventions, it also means that these programs can eventually be
18 integrated into the curriculum and delivered to every student, making this approach a
19 sustainable implementation strategy for widescale delivery and dissemination. Additionally,
20 working with schools to deliver such programs reduces the need for young people to actively
21 seek professional help. This is critically important - despite having a significant need, fewer
22 than 80% of young people with mental illness seek help and receive the services they need
23 (15, 16). The scaffolding provided by schools can be leveraged to deliver prevention
24 programs to all students, which fits with a trend for schools to be designated the first point of
25 contact to support youth mental health problems when they first emerge (9). Process
Page 7 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 7
1 evaluations such as the one described in the current paper are relatively rare, particularly
2 those examining digital (as compared to face-to-face) interventions in school settings. In fact,
3 we were unable to find any process evaluation descriptions in the literature that evaluate
4 digital mental health programs in school settings. This process evaluation is an initial step
5 towards addressing that gap.
6 Technology offers a promising way to deliver mental health interventions to the
7 community. Digital mental health interventions offer two key advantages over face-to-face
8 approaches – first, they are cheaper to access and more cost effective (17, 18), and second,
9 they can be used to reach people across vast geographic areas. The latter is particularly
10 important given Australia’s geography, where remoteness and low population density meant
11 that people in regional and rural areas often do not have the same level of access to mental
12 health services as people in metropolitan areas. With more young people than ever using
13 smartphones, mental health intervention delivered online or through applications represents
14 an exciting avenue for reaching adolescents.
15 The Future Proofing Study (FPS)
16 The trial that is the subject of this process evaluation is the FPS. The FPS addresses
17 barriers of reach, cost, and scalability by delivering a depression prevention program via a
18 smartphone app to Year 8 secondary school students aged 12-13 years. This study is being
19 conducted with approximately 200 Australian schools (up to 10,000 participants), of which
20 half will be allocated to the intervention condition. The primary outcome is symptoms of
21 depression. Secondary outcomes include anxiety, psychological distress and insomnia, among
22 others. Symptom outcomes will be assessed at baseline, post-intervention, six months
23 (primary outcome only), and then annually for five years. The primary endpoint is 12 months
24 following baseline. More details are available from the Australian New Zealand Clinical
25 Trials Registry (ANZCTRN12619000855123) and protocol paper (Werner-Seidler et al.,
Page 8 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 8
1 2020).
2 The intervention being delivered is known as SPARX, a CBT-based program
3 incorporating gamification principles. The development and initial evaluation of SPARX has
4 been documented in detail previously (19). SPARX has been tested in an Australian sample
5 of secondary students (delivered via computer) and shown to prevent depression in the lead
6 up to final school exams (20). The gamified intervention teaches young people about the
7 relationship between thoughts, feelings, and behaviour. Skills learnt through SPARX include
8 emotion identification, emotion regulation, behavioural activation (being active), recognising
9 and challenging unhelpful thoughts, and practical problem solving. SPARX consists of seven
10 20-minute modules. The intervention is fully automated, and the therapeutic components are
11 standardised. See the trial protocol for full intervention details (Werner-Seidler et al., 2020).
12 School Engagement and Recruitment
13 The FPS requires multiple levels of approval, beginning with state department and
14 independent school body approval (NSW Department of Education; Catholic school
15 dioceses), followed by individual school engagement. An engagement strategy for this study
16 involves sending electronic communication material to all schools across New South Wales
17 and in other Australian capital cities, targeting school principals and wellbeing staff to invite
18 them to participate in this study opportunity. Schools are invited to submit expressions of
19 interest and are subsequently followed up by the research team over the phone to explain
20 more about the study. The schools for which the FPS is a good fit (as determined by the
21 school) are then signed onto the study, with support from the principal, the school counsellor,
22 and at least one other staff member (typically a teacher). In line with best practice in
23 implementation science (21), this group of 2-3 staff members (typically not including the
24 principal) will form the school-specific ‘study implementation team’. After signing on to the
25 study, several webinars are scheduled throughout the lead up to the study start date, so that
Page 9 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 9
1 school staff and parents can listen to a 15-minute study overview from the trial manager and
2 have their questions answered.
3 Preparation
4 In preparation for in-class assessment sessions, study facilitators (volunteer research
5 assistants) are recruited to support the study. The purpose of these facilitators is to attend
6 schools to introduce the study to students, and ensure the technology is functioning so
7 students can download the SPARX app and complete the baseline (and post-assessment)
8 questionnaires. All facilitators go through an interview and screening process prior to
9 selection, then attend a half day face-to-face training session before supporting the study in
10 schools. This training provides an overview of the study and detailed information about their
11 roles within schools, including a step-by-step guide to running the sessions. Discussion and
12 practice are core components of the training.
13 Aims and Objectives
14 1. To evaluate the reach (including completion), uptake, and acceptability of the
15 intervention (school- and student-level).
16 2. To understand the contribution of contextual factors (e.g., characteristics of the
17 outer/inner setting, intervention, individuals) on:
18 o School-level fidelity to the implementation strategy
19 o Implementation outcomes (intervention reach, uptake, acceptability), as
20 assessed from the perspectives of school staff, teachers and students.
21 3. To examine the impact of school-level variation (in implementation fidelity and
22 outcomes) on clinical effectiveness outcomes at the school- and student-level.
23 The aim of this process evaluation is to understand how SPARX is implemented and
24 delivered in schools, and to identify systematic differences and variation in delivery between
25 clusters (individual schools). The differing ways in which schools support and deliver the
Page 10 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 10
1 intervention will inevitably impact its effectiveness, and this evaluation will assess these
2 differences (3, 22). For example, degree of support from senior school leaders who would be
3 expected to understand that participating in study activities is a priority for the school will
4 likely impact effectiveness at the cluster level. A secondary aim is to evaluate individual-
5 level factors likely to impact outcome. For instance, young people’s openness to receiving
6 mental health material via an app will likely impact intervention acceptability and
7 completion. This process evaluation has been designed to capture important information from
8 teachers, school staff and students at both the school and individual level, which may
9 ultimately impact the effectiveness of the intervention on clinical mental health outcomes.
10 Findings will provide insight into factors which support and/or hinder the
11 implementation of digital universal mental health programs in school settings. Knowledge
12 gained from this process evaluation will help to inform the development of a model and guide
13 for how to best deliver digital mental health programs to young people in schools.
14
15 Methods and Analysis
16 Design
17 This study uses a hybrid type 1 approach (23), with a focus on implementation
18 process factors and outcomes in the context of an effectiveness trial. The evaluation is guided
19 by the CFIR and RE-AIM framework (24, 25). In keeping with Nilsen’s categorisation of
20 implementation theories, models, and frameworks, these frameworks help us to understand
21 different parts of the implementation process (26). The CFIR will be used to identify
22 barriers and facilitators to intervention implementation and effectiveness (26). The CFIR is
23 a widely used deterministic theoretical framework which has been applied in multiple settings
24 (27), including schools (e.g., 28). The CFIR identifies five major domains, including (i) the
25 outer setting, which includes the social, political and economic context that the organisation
Page 11 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 11
1 in which implementation is occurring exists; (ii) the inner setting, which includes features and
2 characteristics of the organisation such as leadership and relative priority; (iii) the
3 characteristics of individuals, which includes organisational staff knowledge and attitudes
4 about the intervention and their role and identification within the wider organisation; (iv) the
5 characteristics of the intervention itself and; (v) implementation processes, which includes the
6 ways that the intervention will be delivered in a given context. Normalisation Process Theory
7 will be used to provide additional insights into implementation processes (NPT; 29). NPT
8 aims to identify and explain the implementation of new interventions and how they become
9 integrated into routine care.
10 The RE-AIM framework will be used to evaluate the implementation outcomes,
11 including fidelity to the implementation strategy, intervention reach, uptake, and
12 acceptability/appropriateness (24, 30). These outcomes also map onto the framework of
13 implementation outcomes proposed by Proctor and colleagues (31, 32). Fidelity refers to the
14 extent to which the intervention was delivered as intended. Reach refers to the proportion of
15 eligible participants who opened, used, and completed the intervention, as well as the
16 proportion of students from the entire cohort of eligible students in intervention schools who
17 consented to participate. Uptake refers to the proportion of schools that were onboarded to
18 the study (intervention and control arms), and of school staff who were willing to support the
19 delivery of the intervention (intervention arm only). Reach and uptake also incorporate the
20 representativeness of the sample (school-level and individual-level). Acceptability and
21 appropriateness refer to the perceived agreeableness or fit of the intervention. This evaluation
22 will incorporate how the barriers and facilitators identified through the CFIR impacted the
23 implementation outcomes, and, in turn, how these implementation outcomes impacted
24 effectiveness outcomes.
25
Page 12 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 12
1 Logic Model
2 Following MRC guidance (3), the research team developed a logic model for the FPS
3 process evaluation in a series of participatory workshops. The logic model was prospectively
4 informed by key CFIR constructs identified in previous literature as being important in
5 school-based studies, but also feasible and appropriate to measure within the school context
6 in the FPS. The key constructs included the outer setting, inner setting, individual
7 characteristics, and intervention characteristics. The logic model (Figure 1) was developed to
8 consider the key factors (mapped to the CFIR) that would potentially impact implementation
9 of the intervention (mapped to RE-AIM), as well as its clinical effectiveness. The process
10 evaluation method, including the selection of dependent variables and design of surveys and
11 semi-structured interview guides, were derived from this logic model. See Table 1 for
12 CFIR/RE-AIM domains, key research questions, process data, and data that will be collected
13 within each domain.
14 A pilot study of 8 schools conducted in 2019 was used to assess the suitability of the
15 planned implementation strategy and process evaluation methods. We retrospectively applied
16 our logic model to collected data to evaluate whether our methodology captured relevant
17 constructs, and sufficient variation in these constructs. We integrated learnings from this pilot
18 study using a dynamic feedback process to strengthen our methodology for the full-scale FPS
19 phase. Some minor changes were made following this pilot which relate to the CFIR,
20 including a greater emphasis on sub-constructs such as school climate (inner setting).
21
22 <INSERT FIGURE 1 HERE>
23
24 Figure 1. Logic model. The model shows that CFIR constructs including school context
25 characteristics, school organisational characteristics, and individual characteristics, will
Page 13 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 13
1 influence how staff engage with the implementation strategy. The intervention itself, which
2 includes the core cognitive-behavioural therapeutic components, is conceptualised as
3 standardised across individuals because it is delivered digitally, follows a fixed schedule and
4 does not incorporate tailored content. The yellow input factors are expected to vary across
5 schools and individuals, thus influencing engagement and flexibility of the implementation
6 strategy and in turn, implementation outcomes and student-level outcomes. The logic model
7 and implementation plan were externally peer reviewed by an experienced and internationally
8 recognised implementation scientist outside the team within an implementation workshop.
9 For details on assessment of these factors, see Table 1.
Page 14 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
1 Table 1 – Process evaluation details including process data, outcome data, data type and source
CFIR and RE-AIM Constructs Research Aim Process or Outcome Data Data Type and Source
Outer SettingSchool contextual characteristics
What was the broad context of the schools in which the SPARX intervention was delivered? (Aim 2)
School socio-economic indexSchool location (metropolitan/regional)
Publicly available information (ICSEA, GPS)
Inner SettingSchool organisational characteristics
What were the characteristics of the delivery environments (schools)? (Aim 2)
What were the barriers and facilitators that affected buy-in, delivery, and student uptake? (Aim 2)
School size, type, composition, fundingSchool culture
Implementation climate, relative priority, competing demands, leadership, school counsellor availability and level of support, networks and communication, school culture and climate, school readiness for implementation
Publicly available information (school size, type, funding)SSPESH (assesses school culture) Interviews with school staff
Implementation Climate measures (staff and students)Relative Priority measure Competing Demands measureInterviews with school staffChecklists completed by trial managerOther administrative data including number of staff allocated to assist with delivery and consent process, level of communication with the research team
School leadership How supportive of delivering SPARX were school principals, deputy principals and executives? (Aim 2)
Level of support and buy in from school leaders
Interviews with school staff
Individual CharacteristicsSchool staff What were the characteristics (including attitudes,
beliefs, traits) of school staff supporting the delivery of the intervention? (Aim 2)
How supportive of delivery were school staff who were involved on-the-ground? (Aim 2)
Age, gender, current employment, role etc
Leadership, skills, motivations, expectations, self-efficacy, expectations, time available, knowledge and beliefs about the intervention, study buy in
Demographics questionnaire
Implementation Leadership ScaleInterviews with school staffQualitative feedback based on informal observation
Study facilitators How well did study facilitators attending schools support the delivery of the intervention? (Aim 2)
Age, employment Skills, self-efficacy or confidence, motivations and expectations
Demographics questionnaireSelf-Confidence measure Interviews with study facilitators
Page 15 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 15
Students What were the characteristics of young people that affected intervention uptake and effectiveness? (Aim 2)
History of mental illness Reported by Year 8 students as part of the online FPS survey
Intervention CharacteristicsSPARX Were there any barriers to intervention use? (Aim 2
& 3)
What do staff think about the efficacy and advantage of using the intervention? (Aim 2)
Technical issues
Evidence strength and quality, relative advantage
Logs of technical issues sent through schools, parents and participantsIT data pertaining to technical problemsInformal feedback provided by schools and research staff attending schools
Relative Advantages measureAnticipated Benefits measureInterviews with school staff
Implementation Processes
Normalisation and integration How did school staff perceive the implementation processes? (Aim 2)
Coherence, cognitive participation, and collective action
NoMAD
Implementation Outcomes Fidelity to the implementation strategy
To what extent was the intervention implemented as planned? (Aim 2)
School delivery of the FP program, including changes to the plan
Completed implementation checklists, emails and feedback formsInterviews with school staff
Reach What was the extent to which those who were eligible to receive SPARX used it? (Aim 1)
Proportion of eligible participants who consented to participate; proportion who opened, used and completed the SPARX interventionRepresentativeness of the student sample
Administrative data about consent. Digital analytic data including usage (app downloads, installs, opens), completion rate (modules completed) and time spent using SPARXReported by Year 8 students as part of the online FPS survey
Uptake How many eligible schools participated in the study? Within those schools, how many staff supported the delivery of SPARX? (Aim 1)
Proportion of eligible schools that were onboarded to the study; proportion of school staff (in intervention schools) who supported SPARX Representativeness of the sample
Administrative data
Publicly available information about schools and self-report demographic data from school staff
Acceptability/appropriateness How satisfied were participants with the intervention? (Aim 1)
Acceptability/appropriateness of the intervention, expectations
Acceptability/appropriateness of the Future Proofing program (including SPARX)
Reported by Year 8 students as part of the online FPS surveyInformal conversations and feedback provided by Year 8 studentsImplementation Appropriateness measureInterviews with school study staff
Page 16 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 16
How satisfied were school staff with supporting the intervention the FP program (including SPARX)? (Aim 1)
Informal conversations and feedback provided by school staff
Across DomainsHow might the relationship between the intervention, the staff supporting the program, and context of each school shape variation in outcomes (implementation strength metric)? (Aim 2 and 3)What key lessons emerge from this study that can be generalised to the implementation of digital mental health programs in schools more broadly?
1 Note. The process data and outcomes are mapped onto Figure 2. CFIR = Consolidated Framework for Intervention Research; RE-AIM = Reach, Effectiveness, Adoption, Implementation, 2 Maintenance; Aim = Aims outlined in the Aims and Objectives section; SSPESH = Survey of School Promotion of Emotional and Social Health; NoMAD = Normalisation Measure 3 Development questionnaire.
Page 17 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
1 School Implementation Strategy
2 The school implementation strategy was developed by the study authors. The authors
3 drew on their experience in school-based intervention delivery and integrated feedback from
4 teachers and school staff from several completed school-based trials that delivered digital
5 interventions to school students (20, 33). Stakeholder consultation specifically for this study
6 involved discussions with the Department of Education, consultation with several school
7 parent committees and consultation with both youth and parent Lived Experience Advisory
8 Panels. This strategy was also refined following the first pilot wave involving eight
9 intervention schools.
10 The school ‘study implementation teams’ are principally responsible for the
11 implementation of the intervention and liaising with the research team. As described earlier,
12 these teams typically incorporate at least one classroom teacher and one school counsellor to
13 assist with the intervention delivery. Study facilitators also support the delivery of the
14 SPARX intervention by attending schools for the first school session.
15 Implementation Strategy. During the active intervention phase, schools allocate a
16 minimum of 4 x 20-minute school class sessions during which students complete the SPARX
17 intervention. The additional three sessions may be completed either in class if permitted by
18 the schools, or in the students’ own time. The implementation strategy developed by the
19 project team and stakeholders to support the completion of the SPARX intervention
20 comprises:
21 Standardised facilitator training delivered face-to-face over a half-day
22 Study facilitators are present at schools to support students in downloading the
23 SPARX app and completing the baseline assessments
24 School implementation team provided with a SPARX user-guide and information
25 booklet they can refer to during the sessions
Page 18 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 18
1 Schools provide students with weekly verbal reminders in home classrooms to use the
2 app regularly
3 Schools publish brief information about the study and mental health tips in the school
4 weekly newsletter
5 Schools liaise with research team weekly to troubleshoot problems
6 This is the strategy being outlined to schools by the research team and adherence to this
7 strategy will be assessed. Whether or not schools schedule more than the four mandated in-
8 class sessions for intervention completion is flexible and can be adapted to suit the
9 preferences of the school. Student participants receive a $20AUD voucher after the
10 intervention period to cover any phone or data related costs incurred. See Figure 2 for details
11 of training and delivery structure.
12
13 <INSERT FIGURE 2 HERE>
14
15 Data Collection Methods and Participant Groups for Process Evaluation
16 There are three participant groups taking part in the process evaluation: Year 8
17 students, school staff members (e.g., teaching and counselling staff), and study facilitators
18 (see Table 2). School staff members will be members who are responsible for leading the
19 delivery of the study in their school or will be teaching staff who have a supporting role (e.g.,
20 home room teachers who provide reminders to students to complete the intervention).
21 Four types of data will be collected and triangulated: self-report questionnaire data,
22 digital analytic data, administrative data, and qualitative interview data. Self-report
23 questionnaire data specific to the process evaluation will be collected from staff and study
24 facilitators using online survey software (Qualtrics) programmed by one of the authors (JB).
25 These questionnaires assess demographic information, school organisational characteristics,
Page 19 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 19
1 individual characteristics, intervention characteristics, implementation processes and
2 implementation outcomes. Where no available published questionnaires were identified as
3 being suitable for this process evaluation, we adapted existing standardised measures or
4 developed our own items (details below). Self-report questionnaire data from Year 8 students
5 about intervention use and feedback will be collected from an online survey (details
6 described in 5) at the post-assessment time point (after the six-week intervention stage).
7 Digital analytic data about intervention use will be captured by the purpose-built Black Dog
8 Institute research platform, which is being used for the broader FPS. Administrative data will
9 be collected through a range of sources, including staff observations, and communications
10 with the research team. Qualitative data will be collected using semi-structure interviews. All
11 data collected will be from intervention schools only as no intervention is implemented in
12 control schools. Student and school staff data will be collected immediately after the
13 intervention period has been completed; facilitator data will be collected both before and after
14 the intervention period.
15
16 Table 2. Summary of data forms provided by each of the participant groups
Questionnaire Individual interview
Digital analytics
Year 8 students ✓ ✓
School staff ✓ ✓
Facilitators ✓ ✓
17
18 Implementation Predictors
19 School Organisational Characteristics (Inner Setting).
20 Publicly Available Information. Publicly available information will be collected
21 about school contextual characteristics, including socio-economic level, size, location, type,
Page 20 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 20
1 and funding.
2 General School Culture. School staff will complete the Survey of School Promotion
3 of Emotional and Social Health (SSPESH; 34) and a measure of organisational culture (35).
4 The SSPESH assesses a school’s capacity to promote social and emotional wellbeing, and
5 contains four subscales: Positive School Community, Student Social and Emotional Learning,
6 Engaging Families, and Supporting Students Experiencing Mental Health Difficulties. Items
7 are rated on a 4-point Likert scale from 0 (not yet in place) to 3 (completely in place) and
8 preliminary investigations support the scale structure and criterion-related validity (34).
9 We adapted a 9-item questionnaire about general culture within a health care setting
10 to use within the school setting (35). Items are rated on a 5-point Likert scale from 1 (strongly
11 agree) to 5 (strongly disagree). Lower scores indicate a more positive working culture, which
12 includes transparency, productive working relationships, and receptivity to feedback.
13 Previous investigation has shown that this measure has good internal consistency, although
14 overlaps somewhat with other sub-constructs within the inner setting (e.g., learning climate;
15 35).
16 Implementation Climate. School staff will complete a measure of implementation
17 climate that we adapted for use in the school context (35). Items are rated on a 5-point Likert
18 scale from 1 (strongly agree) to 5 (strongly disagree). Lower scores indicate increased staff
19 receptivity to the program and support within the school, including rewards and recognition.
20 This measure has acceptable internal consistency and good discriminant validity (35).
21 Relative Priority. School staff will complete one adapted item from the School
22 Contextual Barriers subscale of the Perceived Attributes of the Healthy Schools Approach
23 Scale (36). The item is rated on a 5-point scale Likert scale from 1 (strongly agree) to 5
24 (strongly disagree). Higher scores indicate that other activities did not interfere with
25 implementation of the FP program.
Page 21 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 21
1 Competing Demands. Staff will complete 3-items, developed specifically for this
2 study, which assess how much time they allocated, and desired to allocate, to the FP program
3 relative to other competing workload demands on visual analogue scales (anchor by 0 = no
4 work time, 100% = all of my work time). Higher scores indicate increased allocation of time
5 and prioritisation of the FP program.
6 Individual Characteristics.
7 Leadership. School staff will complete the Implementation Leadership Scale (ILS;
8 37). The ILS is a 12-item scale that assesses leadership behaviours that support
9 implementation of evidence-based practices. The ILS contains four subscales, and two will be
10 included in the current study (Knowledge Leadership and Supportive Leadership). These
11 scales assess the degree to which the staff member was knowledgeable about and offered
12 support to the program. Items are rated on a 5-point Likert scale from 0 (very much so) to 4
13 (not at all) and will be re-coded such that higher scores indicate more effective
14 implementation leadership behaviours. This measure has excellent internal validity,
15 convergent validity, and discriminant validity (37).
16 Self-Efficacy and Confidence. Study facilitators will rate their level of confidence in
17 performing 23 different tasks during school-visits on visual analogue scales, which are based
18 on the training program they completed (anchor by 0 = not at all confident, 100 = very
19 confident). Higher scores indicate higher levels of confidence. A similar questionnaire will be
20 repeated following their school-visits, along with four short answer questions that assess
21 experiences with supporting staff and students during school-visits.
22 Intervention Characteristics.
23 Relative Advantage and Evidence Strength and Quality. School staff will complete
24 the 2-item Relative Advantages subscale and one item from the Anticipated Benefits subscale
25 of the Perceived Attributes of the Healthy Schools Approach Scale (36). Adapted for the
Page 22 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 22
1 current study, items are rated on a 5-point scale Likert scale from 1 (strongly agree) to 5
2 (strongly disagree). Scores will be re-coded such that higher scores indicate greater perceived
3 advantage of the FP program compared to others and greater perceived impacts on mental
4 health, respectively.
5 Implementation Processes.
6 Normalisation and Integration into Routine Practice. School staff will complete
7 NPT’s accompanying tool, the Normalisation MeAsure Development questionnaire
8 (NoMAD; 38). The NoMAD is a 23-item measure that assesses how professionals involved
9 in the implementation of a complex intervention perceive implementation processes. It is a
10 flexible measure that can be altered to more accurately describe the adoption of new
11 interventions at the provider level. Ten items of the NoMAD will be used in this study to
12 assess intervention buy-in from school staff, and how staff members incorporated the
13 initiative into their standard work responsibilities. These items are grouped into three
14 categories: coherence (i.e., making sense of an intervention), cognitive participation (i.e.,
15 working with others to support an intervention), and collective action (i.e., the type of work
16 that people do to support an intervention). Items are rated on a 5-point Likert scale from 1
17 (strongly agree) to 5 (strongly disagree). Initial validation demonstrated that the NoMAD has
18 good face validity, construct validity, and internal consistency (39).
19 Implementation Outcomes
20 Fidelity to the Implementation Strategy. Fidelity will be captured through
21 implementation checklists, informal observations provided by staff attending schools, and
22 communications between schools and research staff.
23 Reach. SPARX app usage data from students will allow for the assessment of app use
24 (downloads, installs, and opens), completion (modules completed), and time spent using
25 SPARX. Administrative data about the number of students in intervention schools with
Page 23 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 23
1 consent to participate will also be used as an indicator of reach. Self-report data about
2 individual characteristics (e.g., gender, mental health history) will be used to gauge the
3 representativeness of the students in the intervention schools.
4 Uptake. Administrative data about the proportion of schools that were onboarded to
5 the study and the proportion of teachers who supported the intervention will be collected to
6 provide an index of uptake by schools. Representativeness of the sample will be informed by
7 publicly available information about school characteristics (e.g., socio-economic status,
8 location) and self-report data about school staff characteristics (e.g., role, gender).
9 Appropriateness. School staff will complete the Intervention Appropriateness
10 Measure (IAM; 40). The IAM is a pragmatic 4-item measure of the perceived fit, relevance
11 or compatibility of an evidence-based practice for a context, person, or problem (41). Items
12 have been adapted for this study and are rated on a 5-point Likert scale from 1 (strongly
13 agree) to 5 (strongly disagree). Scores will be re-coded such that higher scores indicate higher
14 levels of perceived appropriateness. The IAM demonstrated strong psychometric properties
15 in previous research (40). School staff will also complete one adapted item from the Agency
16 Leadership Support subscale of the Barriers and Facilitators to Implementing Survey (42).
17 The item is rated on a 5-point Likert scale from 1 (strongly agree) to 5 (strongly disagree) and
18 will be re-coded such that higher scores indicate greater compatibility of the FP program
19 within a particular school.
20 Acceptability. Year 8 students will complete an 11-item feedback questionnaire
21 about SPARX. The questionnaire assesses three domains, including: (i) reasons for non-
22 adherence; (ii) intervention acceptability; and (iii) skills learnt from the intervention. Items
23 will be quantified individually. This questionnaire has previously been used to assess the
24 acceptability of the SPARX program in the school context (43).
25 Individual Interviews
Page 24 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 24
1 Interview guides for school staff and study facilitators were derived from the logic
2 model, CFIR online resources (e.g., https://cfirguide.org/evaluation-design/qualitative-data/),
3 and the broader literature investigating the delivery of interventions in school settings (44, 45;
4 see Additional File 1 for interview guides). Interviews will provide information about both
5 implementation predictors and outcomes. For school staff, questions focus on: motivations
6 and expectations about the intervention and study; knowledge and beliefs about the
7 intervention; relative advantages of the intervention; self-efficacy; barriers and facilitators
8 affecting the delivery of the intervention, including fidelity; appropriateness and acceptability
9 of the intervention; and recommendations for future implementation. School counsellors will
10 also be asked questions about their experiences of managing high risk participants within the
11 study and compatibility with existing workload. For study facilitators, questions focus on:
12 motivations and expectations about their role; confidence and competence in supporting the
13 delivery of SPARX in schools; the quality of their training; and perceptions about their ability
14 to support the study as required.
15 The interview guides allow for flexibility in questioning and diversion in responses.
16 Questions will primarily be open-ended, with specific prompts and follow-up questions being
17 used as necessary to encourage respondents to elaborate on their ideas and provide examples.
18 Patient and Public Involvement
19 The Future Proofing Study was developed with key stakeholders including school
20 personnel, school counsellors, parents, adolescents, and individuals with a lived experience of
21 mental illness. All aspects of the study design, appropriateness of outcomes measures and
22 consent procedures were developed in consultation with these stakeholder groups. The
23 current process evaluation involves consultation with school staff members to understand
24 their experience. The SPARX intervention itself was codesigned with young people. All
25 study results will be shared directly with participants and their schools through lay summaries
Page 25 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 25
1 and infographics.
2 Procedure
3 Ethical approval was provided by the University of New South Wales Human
4 Research Ethics Committee (HC180836) and NSW Government State Education Research
5 Applications Process Approval (SERAP 2019201).
6 The FPS has two waves of delivery (April-June 2020; July-Sept 2020). Process
7 evaluation data will be collected at each wave, immediately following the intervention period.
8 Purposive sampling will be used to capture a range of diverse school settings.
9 School students will complete process measures at the same time as completing
10 primary measures for the FPS. Study facilitators will provide informed consent following the
11 compulsory face-to-face training session they attend with the research team at the Black Dog
12 Institute. Questionnaires will be completed online.
13 School staff directly involved in the study will be invited to complete the
14 questionnaire by the trial manager, following the six-week SPARX intervention period. Once
15 school staff members open the letter of invitation, they will be able to consent to a 60-minute
16 individual interview (completed either in person or by phone). The semi-structured interviews
17 will be conducted with research staff who have had no previous contact with the schools
18 during the trial, thereby minimising the risk of bias and demand effects. Face-to-face
19 interviews will be held in a quiet room on school grounds. Phone and face-to-face interviews
20 will be recorded and transcribed verbatim.
21
22 Data analysis
23 Quantitative
24 Questionnaire data and open-ended questions from surveys will be exported into data
25 analytic software for analysis. Descriptive statistics will be calculated for all participant
Page 26 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 26
1 groups and will provide information about intervention use and its acceptability
2 (questionnaire data from Year 8 participants), differential implementation, fidelity to the
3 implementation strategy, school context factors within each school (questionnaire including
4 several short answer questions from school staff), and competence and experience in schools
5 from those facilitating the study (facilitator questionnaires). Differences between school
6 clusters will be assessed using Analysis of Variance methods (ANOVA). Regression models
7 will assess the on the effects of contextual factors on implementation outcomes (e.g., reach,
8 uptake, acceptability).
9 Qualitative
10 Interviews will be digitally audio-recorded and transcribed verbatim. The transcripts
11 will be checked for accuracy against the sound files as per best practice in transcription (46,
12 47). Qualitative data will then be imported into NVivo to aid in data management and
13 analysis. Thematic analysis will be undertaken to identify, interpret and report on the
14 repeated patterns of meaning within the data, drawing from Braun and Clark’s classic six
15 phase model (48, 49).
16 Initially, one coder will code a subset of the transcripts using an inductive approach,
17 incorporating themes from the data together with topics covered in the interview guide. This
18 subset will then be double coded by an independent coder who will develop their own coding
19 scheme. Codes will be compared and discussed by the two coders, with refinement occurring
20 via an interactive process until a final coding scheme is determined. A senior qualitative
21 analysist will review the coding framework and scheme before all transcripts are coded by the
22 first coder. Research rigor will be enhanced by a team approach to analysis, reflexive field
23 notes, and prolonged engagement with the subject matter (50).
24 Triangulation of Qualitative and Quantitative Data
25 Triangulation involves the use of multiple approaches to address a research question.
Page 27 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 27
1 The combination of several approaches increases confidence in the findings and provides a
2 more comprehensive account of the results than individual approaches would do alone (51).
3 In this study, reliability, validity and confidence will be maximised through cross verification
4 and exploration of differences between the outcomes of the various methods. This takes place
5 in several ways:
6 Maximising validity in analysis of qualitative data within the research team using
7 techniques such as discussing coding, constant comparison, accounting for deviant
8 cases, and systematic coding;
9 Triangulation of school staff and research assistant interviews with results from the
10 questionnaires, exploring and accounting for differences;
11 Triangulation of self-report and interview data with publicly available information
12 relating to school contextual characteristics (e.g., school socio-economic level and
13 size) and school delivery of the program, including deviations to the implementation
14 strategy.
15 Mapping the perspectives of different stakeholders across the study (school staff,
16 study facilitators).
17 Additional Analyses
18 We will generate an ‘implementation strength’ metric for each school and its
19 relationship to primary and secondary trial outcomes. This is an emerging evaluation
20 approach, used mainly in low and middle-income countries. The approach aims to understand
21 the degree of implementation effort needed during intervention delivery to achieve desired
22 benefits (52, 53).
23 The implementation strength metric will provide funders and policy makers with an
24 objective measure to monitor effectiveness of implementation if it goes beyond this trial and
25 becomes a sustainable approach. The metric can be used to assess whether the approach to
Page 28 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 28
1 implementation meets the minimum level required to prevent the onset of mental health
2 problems in adolescents. The metric will be based on implementation inputs and contextual
3 factors informed by the intervention logic model and process evaluation frameworks (CFIR
4 and RE-AIM) (e.g., adoption from teachers and other staff in the school and fidelity to the
5 intervention strategy), and will be developed with the Future Proofing research team using
6 principal component analysis.
7
8 Discussion
9 This paper describes the design of a mixed methods process evaluation of a cluster
10 randomised controlled trial, the FPS. The FPS investigates the impact of a digital cognitive-
11 behavioural therapy intervention when delivered at scale in school settings. Digital mental
12 health programs have tremendous potential to prevent up to 22% of depression cases and,
13 when delivered at scale, could have population-level impacts (54). However, these programs
14 have not been translated into practice and policy because optimal ways to scale and deliver
15 these interventions are not yet well understood.
16 As an initial step to address this issue, the current process evaluation will attend to
17 contextual and implementation factors that vary across schools and provide a lens through
18 which to interpret trial efficacy outcomes. We expect that results will provide a richly
19 detailed and nuanced understanding of the key factors involved in the effective delivery of
20 digital mental health programs across different schools. We expect that results will not only
21 contextualise our trial findings but will also be used as a model to guide the delivery of
22 school-based interventions that focus on preventing mental illness more broadly. Findings
23 from this process evaluation will indicate whether the approach used in the FPS trial is likely
24 to be sustainable in the school environment going forward and, if so, the threshold level of
25 support required in order to prevent depression and benefit student mental health.
Page 29 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 29
1 The prospective publication of this protocol outlines our planned methodological
2 approach. It also serves as a road map for other researchers on a practical way of carrying out
3 process evaluations of complex interventions in the school setting. As is the case with the
4 delivery of interventions across different contexts, we acknowledge that our approach has
5 inbuilt flexibility to explore the data and make provisions for unexpected implementation
6 factors that arise.
7 Limitations and Strengths
8 There are several limitations to our process evaluation that warrant mention. First, we
9 are not including direct observation of teachers in their role supporting the delivery of
10 SPARX. While this would provide objective fidelity data, it requires resources beyond the
11 scope of this project and is not representative of how the program will be sustained following
12 the conclusion of the trial. Second, the qualitative interviewer is a member of the research
13 team (but not the evaluation team). Care will be taken to ensure that this staff member has no
14 contact with schools prior to the interview visit to minimise bias. However, there remains a
15 risk that demand effects may impact the information that is shared. Third, the process
16 evaluation process (questionnaires and interview) will undoubtedly add to the burden placed
17 on school staff. Given that the FPS is already placing a significant burden on the time of busy
18 school staff, this additional component might contribute to low levels of participation.
19 To the point of burden on schools, one of the strengths of the design is that we have
20 undertaken a pilot phase involving eight intervention schools and have been able to refine our
21 processes (e.g., introduce an incentive) for school staff members to participate. This process
22 evaluation also involves the combination of qualitative and quantitative methods which will
23 be triangulated to provide a coherent and comprehensive picture of the data. The use of the
24 ‘implementation strength’ metric represents a novel approach in this field, borrowed from the
25 low-middle income country implementation science sector. The inclusion of this approach
Page 30 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 30
1 will provide important information to funders and policy makers following on from this trial,
2 indicating the level of implementation support required to prevent mental illness and improve
3 wellbeing of adolescent school students.
4 Notwithstanding the limitations raised above, this process evaluation will contribute
5 to the broader knowledge base and indicate how best to deliver digital mental health
6 prevention programs in school settings.
7
8 Ethics and Dissemination
9 As noted, all procedures have received ethical approval from relevant committees. Results
10 from this study will be published in an open access peer-reviewed journal.
11
12 Additional Files
13 Additional File 1 (Semi-Structured Interview Guides).docx
14
15 List of Abbreviations
16 FPS = Future Proofing Study
17 cRCT = Cluster Randomised Controlled Trial
18 CFIR = Consolidated Framework for Implementation Research
19 RE-AIM = Reach, Effectiveness, Adoption, Implementation, Maintenance framework
20
21
22
23
24
25
Page 31 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 31
1 Declarations
2 Trial status
3 Recruitment for the trial is underway. Data collection commences in September 2020
4 (delayed from April 2020 due to COVID-19).
5
6 Ethical approval and consent to participate
7 This study has ethical approval from the University of New South Wales Human Research
8 Ethics Committee (HC180836; 21st January 2019) and NSW Government State Education
9 Research Applications Process Approval (SERAP 2019201; 19th August 2019). All schools,
10 staff, parents and participants must provide active consent to participate.
11
12 Consent for publication
13 Not applicable.
14
15 Availability of data and materials
16 Not applicable.
17
18 Competing interests
19 The authors declare that they have no competing interests.
20
21 Funding
22 Funding for this project came from an NSW Health Early-Mid Career Fellowship awarded to
23 Aliza Werner-Seidler, and a Black Dog Institute Post-Doctoral Fellowship awarded to Joanne
24 Beames, secured by Helen Christensen. ALC is supported by NHMRC fellowships 1122544
25 and 1173146. PJB is supported by NHMRC Fellowship 1158707. Funding for the
Page 32 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 32
1 randomised controlled trial within which this process evaluation is embedded came from an
2 NHMRC Project Grant Awarded to Helen Christensen GNT1120646. The funding bodies had
3 no role in any aspect of the study design or this manuscript.
4
5 Author contributions
6 HC and AWS conceived of the study and secured the funding. AWS and JB led the design of
7 the process evaluation, with input from all authors, and expert guidance from RL and KB.
8 AWS drafted the manuscript, with assistance from JB. ALC, MT, KM, RL, KL and HC have
9 a continuing role in monitoring the conduct and outcomes of the process evaluation. All
10 named authors contributed substantially to the approved final manuscript.
11
12 Acknowledgments
13 We would like to thank Professor Melanie Barwick for her advice on this manuscript,
14 together with all the facilitators from the Training Institute for Dissemination and
15 Implementation Research in Health (TIDIRH; Australia 2020) for their comments and
16 feedback on this project.
17
18 Author information
19 Not applicable.
Page 33 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 33
1 References
2 1. Oakley A, Strange V, Bonell C, Allen E, Stephenson J, Team RS. Process evaluation
3 in randomised controlled trials of complex interventions. BMJ. 2006;332:413.
4 2. Greenhalgh T, Papoutsi C. Studying complexity in health services research:
5 Desperately seeking an overdue paradigm shift. BMC Med. 2018;16:95.
6 3. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process
7 evaluation of complex interventions: Medical research council guidance. BMJ.
8 2015;350:h1258.
9 4. Grant A, Treweek S, Dreischulte T, Foy R, Guthrie B. Process evaluations for cluster-
10 randomised trials of complex interventions: A proposed framework for design and reporting.
11 Trials. 2013;14:15.
12 5. Werner-Seidler A, Huckvale K, Larsen ME, Calear AL, Maston K, Johnston L, et al.
13 A trial protocol for the effectiveness of digital interventions for preventing depression in
14 adolescents: The future proofing study. Trials. 2020;21:2.
15 6. Tal A, Torous J. The digital mental health revolution: Opportunities and risks.
16 Psychiatric Rehabilitation Journal. 2017;40:263.
17 7. Sawyer SM, Azzopardi PS, Wickremarathne D, Patton GC. The age of adolescence.
18 Lancet Child Adolesc Health. 2018;2:223.
19 8. Andrews G, Issakidis C, Sanderson K, Corry J, Lapsley H. Utilising survey data to
20 inform public policy: Comparison of the cost-effectiveness of treatment of ten mental
21 disorders. British Journal of Psychiatry. 2004;184:526.
22 9. Lawrence D, Johnson S, Hafekost J, de Hann KB, Sawyer M, Ainley J, et al. The
23 mental health of children and adolescents: Report on the second australian child and
24 adolescent survey of mental health and wellbeing. 2015.
Page 34 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 34
1 10. Lee FS, Heimer H, Giedd JN, Lein ES, Šestan N, Weinberger DR, et al. Adolescent
2 mental health—opportunity and obligation. Science. 2014;346:547.
3 11. Werner-Seidler A, Perry Y, Calear AL, Newby JM, Christensen H. School-based
4 depression and anxiety prevention programs for young people: A systematic review and
5 meta-analysis. Clin Psychol Rev. 2017;51:30.
6 12. van Zoonen K, Buntrock C, Ebert DD, Smit F, Reynolds CF, Beekman AT, et al.
7 Preventing the onset of major depressive disorder: A meta-analytic review of psychological
8 interventions. Int J Epidemiol. 2014;43:318.
9 13. Gulliver A, Griffiths KM, Christensen H. Perceived barriers and facilitators to mental
10 health help-seeking in young people: A systematic review. BMC Psychiatry. 2010;10:113.
11 14. Donker T, Blankers M, Hedman E, Ljotsson B, Petrie K, Christensen H. Economic
12 evaluations of internet interventions for mental health: A systematic review. Psychological
13 Medicine. 2015;45:3357.
14 15. Essau CA. Frequency and patterns of mental health services utilization among
15 adolescents with anxiety and depressive disorders. Depress Anxiety. 2005;22:130.
16 16. Rickwood DJ, Deane FP, Wilson CJ. When and how do young people seek
17 professional help for mental health problems. Med J Aust. 2007;187:S35.
18 17. Donker T, Blankers M, Hedman E, Ljótsson B, Petrie K, Christensen H. Economic
19 evaluations of internet interventions for mental health: A systematic review. Psychological
20 Medicine. 2015;45:3357.
21 18. Lee YY, Barendregt JJ, Stockings EA, Ferrari AJ, Whiteford HA, Patton GA, et al.
22 The population cost-effectiveness of delivering universal and indicated school-based
23 interventions to prevent the onset of major depression among youth in australia.
24 Epidemiology and Psychiatric Sciences. 2017;26:545.
Page 35 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 35
1 19. Merry SN, Stasiak K, Shepherd M, Frampton C, Fleming T, Lucassen MF. The
2 effectiveness of sparx, a computerised self help intervention for adolescents seeking help for
3 depression: Randomised controlled non-inferiority trial. BMJ. 2012;344:e2598.
4 20. Perry Y, Werner-Seidler A, Calear A, Mackinnon A, King C, Scott J, et al. Preventing
5 depression in final year secondary students: School-based randomized controlled trial.
6 Journal of Medical Internet Research. 2017;19:e369.
7 21. Higgins MC, Weiner J, Young L. Implementation teams: A new lever for
8 organizational change. Journal of Organizational Behavior. 2012;33:366.
9 22. Clarke DJ, Godfrey M, Hawkins R, Sadler E, Harding G, Forster A, et al.
10 Implementing a training intervention to support caregivers after stroke: A process evaluation
11 examining the initiation and embedding of programme change. Implement Sci. 2013;8:96.
12 23. Curran GM, Bauer M, Mittman B, Pyne JM, Stetler C. Effectiveness-implementation
13 hybrid designs: Combining elements of clinical effectiveness and implementation research to
14 enhance public health impact. Med Care. 2012;50:217.
15 24. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health
16 promotion interventions: The re-aim framework. American Journal of Public Health.
17 1999;89:1322.
18 25. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering
19 implementation of health services research findings into practice: A consolidated framework
20 for advancing implementation science. Implementation Science. 2009;4:50.
21 26. Nilsen P. Making sense of implementation theories, models and frameworks.
22 Implementation Science. 2015;10:53.
23 27. Kirk MA, Kelley C, Yankey N, Birken SA, Abadie B, Damschroder L. A systematic
24 review of the use of the consolidated framework for implementation research. Implement Sci.
25 2016;11:72.
Page 36 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 36
1 28. Leeman J, Wiecha JL, Vu M, Blitstein JL, Allgood S, Lee S, et al. School health
2 implementation tools: A mixed methods evaluation of factors influencing their use.
3 Implement Sci. 2018;13:48.
4 29. Murray E, Treweek S, Pope C, MacFarlane A, Ballini L, Dowrick C, et al.
5 Normalisation process theory: A framework for developing, evaluating and implementing
6 complex interventions. BMC Medicine. 2010;8:63.
7 30. Glasgow RE, Harden SM, Gaglio B, Rabin B, Smith ML, Porter GC, et al. Re-aim
8 planning and evaluation framework: Adapting to new science and practice with a 20-year
9 review. Frontiers in Public Health. 2019;7.
10 31. Proctor EK, Landsverk J, Aarons G, Chambers D, Glisson C, Mittman B.
11 Implementation research in mental health services: An emerging science with conceptual,
12 methodological, and training challenges. Administration and Policy in Mental Health.
13 2009;36:24.
14 32. Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, et al. Outcomes
15 for implementation research: Conceptual distinctions, measurement challenges, and research
16 agenda. Administration and Policy in Mental Health and Mental Health Services Research.
17 2011;38:65.
18 33. Calear AL, Christensen H, Mackinnon A, Griffiths KM, O'Kearney R. The
19 youthmood project: A cluster randomized controlled trial of an online cognitive behavioral
20 program with adolescents. Journal of Consulting and Clinical Psychology. 2009;77:1021.
21 34. Dix KL, Green MJ, Tzoumakis S, Dean K, Harris F, Carr VJ, et al. The survey of
22 school promotion of emotional and social health (sspesh): A brief measure of the
23 implementation of whole-school mental health promotion. School Mental Health.
24 2019;11:294.
Page 37 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 37
1 35. Fernandez ME, Walker TJ, Weiner BJ, Calo WA, Liang S, Risendal B, et al.
2 Developing measures to assess constructs from the inner setting domain of the consolidated
3 framework for implementation research. Implement Sci. 2018;13:52.
4 36. Deschesnes M, Trudeau F, Kébé M. Psychometric properties of a scale focusing on
5 perceived attributes of a health promoting school approach. Can J Public Health.
6 2009;100:389.
7 37. Aarons GA, Ehrhart MG, Farahnak LR. The implementation leadership scale (ils):
8 Development of a brief measure of unit level implementation leadership. Implement Sci.
9 2014;9:45.
10 38. Finch TL, Girling M, May CR, Mair FS, Murray E, Treweek S, et al. Nomad:
11 Implementation measure based on normalization process theory 2015 [Available from:
12 http://www.normalizationprocess.org.
13 39. Finch TL, Girling M, May CR, Mair FS, Murray E, Treweek S, et al. Improving the
14 normalization of complex interventions: Part 2 - validation of the nomad instrument for
15 assessing implementation work based on normalization process theory (npt). BMC Med Res
16 Methodol. 2018;18:135.
17 40. Weiner BJ, Lewis CC, Stanick C, Powell BJ, Dorsey CN, Clary AS, et al.
18 Psychometric assessment of three newly developed implementation outcome measures.
19 Implement Sci. 2017;12:108.
20 41. Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, et al. Outcomes
21 for implementation research: Conceptual distinctions, measurement challenges, and research
22 agenda. Adm Policy Ment Health. 2011;38:65.
23 42. Salyers MP, Rollins AL, McGuire AB, Gearhart T. Barriers and facilitators in
24 implementing illness management and recovery for consumers with severe mental illness:
25 Trainee perspectives. Adm Policy Ment Health. 2009;36:102.
Page 38 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 38
1 43. Perry Y, Calear AL, Mackinnon A, Batterham PJ, Licinio J, King C, et al. Trial for
2 the prevention of depression (tripod) in final-year secondary students: Study protocol for a
3 cluster randomised controlled trial. Trials. 2015;16:451.
4 44. Punukollu M, Burns C, Marques M. Effectiveness of a pilot school-based intervention
5 on improving scottish students’ mental health: A mixed methods evaluation. J Youth
6 Adolesc. 2019:1.
7 45. Outhwaite LA, Gulliford A, Pitchford NJ. A new methodological approach for
8 evaluating the impact of educational intervention implementation on learning outcomes. Int J
9 Res Meth Educ. 2019:1.
10 46. Poland BD. Transcription quality. Handbook of interview research: Context and
11 method. 2002:629.
12 47. Poland BD. Transcription quality as an aspect of rigor in qualitative research.
13 Qualitative inquiry. 1995;1:290.
14 48. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol.
15 2006;3:77.
16 49. Braun V, Clarke V. Reflecting on reflexive thematic analysis. Qual Res Sport Exerc
17 Health. 2019;11:589.
18 50. Noble H, Smith J. Issues of validity and reliability in qualitative research. Evid-Based
19 Nurs. 2015;18:34.
20 51. Tashakkori A, Teddlie C. Handbook of mixed methods in social & behavioral
21 research. Thousand Oaks, CA: Sage; 2003.
22 52. Hargreaves JR, Goodman C, Davey C, Willey BA, Avan BI, Schellenberg JR.
23 Measuring implementation strength: Lessons from the evaluation of public health strategies
24 in low- and middle-income settings. Health Policy Plan. 2016;31:860.
Page 39 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 39
1 53. Victora CG, Black RE, Boerma JT, Bryce J. Measuring impact in the millennium
2 development goal era and beyond: A new approach to large-scale effectiveness evaluations.
3 Lancet. 2011;377:85.
4 54. Ebert DD, Cuijpers P. It is time to invest in the prevention of depression. JAMA Netw
5 Open. 2018;1:e180335.
Page 40 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
School Contextual CharacteristicsSociodemographic factors (school size,
socio economic status, area)
Intervention Characteristics (SPARX)
Standardised: Delivered by app over 6 weeks
Technical issues, evidence strength and quality, relative advantage
Student-Level Outcomes
Primary outcome: Lower levels of depression
Secondary outcomes: Lower levels of anxiety, distress,
insomnia
Additional outcomes:Improved quality of life, wellbeing
Lower levels of suicidality, self-harm
Improved academic performance on standardised tests
Lower levels of health service use
Implementation Outcomes
School-level fidelity to the implementation strategy
Reach (e.g., intervention received by students, including number of modules completed and time spent using SPARX)
School-level uptake of the FPS
Staff and student acceptability/appropriateness
Implementation Strategy
App download in school time, with support from
research staff
Intervention completion, with support from school
staff
Reminders and support provided by school staff to
complete intervention
Therapeutic Components
PsychoeducationCognitive restructuringBehavioural activationRelaxation strategies
Practical problem solvingEmotion regulation and management
Individual CharacteristicsSchool staff (e.g., skills, motivation, self-
efficacy, buy in, knowledge about the intervention, leadership)
Study facilitators (e.g., skills, confidence)
School Organisational CharacteristicsSchool culture, buy in and support for the
intervention and study, support from leadership, ability of school to support delivery (including
technological infrastructure, counsellor availability)
Individual CharacteristicsStudents (e.g., mental health
history, intervention expectations)
Page 41 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Research Team
Facilitators
Teachers Counsellors
Year 8 Students
Training
Other staff
Delivery Support
Direct trainingSupport to access interventionDelivery support
School Implementation Team
Page 42 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 1
Additional File 1
Semi-Structured Interview Guides
School Staff INTRODUCTION: Thank you for agreeing to take part in this interview. I appreciate your time. I am interested in hearing your thoughts and experiences of the Future Proofing program in your school. For this interview, the FP program includes the intervention app (SPARX) and everything else that was done to support the program in your school. I hope to use this information to work out what works in different schools and how to best provide support. I will ask you a few questions about your experiences. This is chance for you to reflect and offer feedback on how we could improve. The interview will last up to one hour. It will be recorded, and your responses will be transcribed to ensure we accurately reflect your words. All personal information will be removed so that your answers are non-identifiable and confidential. Nothing you personally say will be shared with anyone on the FP team that you have been liaising with or with other staff at your school. You can choose to not answer any questions. Do you have any questions before we get started? General Opening Questions:
1. How did you first get involved in teaching/counselling/psychology?
2. Could you tell me about the core values and ethos of your school?
3. To the best of your knowledge, what mental health programs or initiatives has your
school implemented to care for students besides the FP program?
Specific Questions:
4. Overall, how would you describe your experience with the FP program in your school?
5. What was your role in delivering or supporting the FP program in your school?
a. Leader/driver, decision-maker, consent/recruitment, peripheral support
CFIR DOMAIN: INDIVIDUAL CHARACTERISTICS Personal attributes (e.g., motivations, expectations, capacity, values)
6. Why were you motivated to take on this role?
a. How did you come on board with the program?
7. Why do you think your school decided to support this program?
a. To what extent do you think there was a need for a mental health program like FP in
you school (Tension for change)?
8. What were your expectations about the FP program prior to coming on board?
a. How were your expectations met (or not met)? What went differently to what
you expected? What surprised you?
b. What were your expectations about the apps?
Knowledge and beliefs about the intervention 9. How much do you know about SPARX, the intervention? E.g., principles, rationale,
content?
Page 43 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 2
Self-efficacy 10. How confident were you in your ability to lead/support/make decisions about the FP
program? a. What personal qualities do you think helped you to do that?
CFIR DOMAIN: OUTER SETTING Relative Advantage
11. How did SPARX compare to other mental health programs that your school has supported or delivered?
CFIR DOMAIN: INNER SETTING School contextual factors – Barriers and Facilitators
12. What factors influenced your ability to deliver/support the FP program in your school? a. What else helped or hindered your ability? b. Enquire into school-specific factors and program-specific factors e.g., What
parts of the program were easy for you to do or support (e.g., the consent process, organising rooms, weekly team meetings)? OR What aspects of your school made it difficult for you to support the program?
c. What factors got in the way of students completing the mental health app/s? d. Were you (or others) able to overcome any barriers? How did you manage to do
this? School contextual factors – Leadership
13. Tell me about the involvement of your school executive in the FP program. a. What role did this person have (e.g., principal/vice principal)? Did you meet
regularly with this person or provide updates to them? 14. How do you think the school executive saw this initiative?
Readiness for implementation
15. How would you describe the readiness of your school to take on the FP program? a. What was the preparation phase like for you/other key staff?
IMPLEMENTATION METRICS Appropriateness and Acceptability
16. How does the FP program align with your school’s approach toward student mental health?
17. How suited is the FP program to address depression prevention in your school? 18. How well does an app delivering therapy work to prevent depression in your students? 19. How do you feel about using technology to deliver psychological interventions in your
school? a. What worked well about this and what did not? b. Do you think SPARX met the student’s mental health needs? In what way?
20. What were the by-products of the intervention that you didn’t expect? a. What parts of the program do/do not suit your school?
**ADDITIONAL QUESTIONS FOR SCHOOL COUNSELLORS ONLY 21. What was your experience with responding to students identified as needing follow-up
during the FP program? a. Was there anything that could have been done to make this experience better for
you and the student? 22. What do you think of the FP risk protocol?
Page 44 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 3
KEY LEARNING AND FUTURE IMPLEMENTATION 23. What parts of the FP program, if any, could be made more attractive to your school to
increase uptake or support? a. In what ways could we make these parts more attractive?
24. To what extent do you think that the FP program could be integrated into your school’s existing mental health strategy?
25. What advice would you give another school contemplating taking on the FP program? 26. What’s been your major learning or insight from your school’s involvement in the FP
program? Closing
27. Is there anything else you would like to add?
Page 45 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 4
Volunteer Research Assistants (Study Facilitators) INTRODUCTION: Thank you for offering to take part in this interview. I appreciate your time. I am interested in your thoughts about and experiences with supporting the FP program as a Research Assistant Volunteer. I hope to use this information to work out what works in different schools and how we can help volunteers to best provide support. I will ask you a few questions about your experiences. This is chance for you to reflect and offer feedback on how we could improve. The interview will last up to one hour. It will be recorded and your responses will be transcribed to ensure we accurately reflect your words. All personal information will be removed so that your answers are non-identifiable and confidential. Nothing you personally say will be shared with the volunteer team or the Black Dog Institute more generally, so it will not affect your relationship with them. You can choose to not answer any questions. Do you have any questions before we get started? General Opening Questions:
1. What schools have you visited so far? 2. Roughly how many school visits have you attended?
Specific Questions: CFIR DOMAIN: INDIVIDUAL CHARACTERISTICS Personal attributes (e.g., motivation and expectations)
3. Why were you motivated you to volunteer? 4. What were your expectations coming on board with the FP program?
a. How were your expectations met (or not met)? b. What went differently to what you expected? What surprised you?
Perceptions of Training
5. What are your overall impressions of the training you received? a. How helpful and appropriate was the training? b. What parts were the most useful? c. What was missing from the training?
CFIR DOMAIN: INNER SETTING School visits – Barriers/Facilitators
6. What was your experience with completing the school visits? 7. Did anything unexpected happen during the school visits?
a. If so, what? b. How did you manage this? c. How well do you think you managed this?
8. In general, how was the FP program received by school staff during visits? 9. To what extent was it useful having a volunteer partner/s at the visits?
a. Were there any other benefits/disadvantages? School contextual factors - Barriers/Facilitators
10. What factors influenced your ability to assist students and teachers at the school-visits? a. Were these common across all the visits or specific to a particular visit? b. What aspects of the school made it difficult to support the program?
Page 46 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 5
c. What else helped or hindered your ability? (e.g., research platform/app specific factors)
d. How did you overcome these barriers? e. Were there any noticeable differences between different schools? If so, what were
they? KEY LEARNING AND FUTURE IMPLEMENTATION
11. What advice would you give a school contemplating taking on the FP program? 12. What advice would you give to others who are thinking about volunteering? 13. What parts of the FP program could be made more attractive to volunteers to increase
uptake or support? 14. What’s been your major learning or insight from taking part in the volunteer program?
Closing
15. Is there anything else you would like to add?
Page 47 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
1
Standards for Reporting Implementation Studies: the StaRI checklist for completion
Checklist itemReported on page # Implementation Strategy
Reported on page # Intervention
“Implementation strategy” refers to how the intervention was implemented
“Intervention” refers to the healthcare or public health intervention that is being implemented.
Title and abstractTitle 1 1-2 Identification as an implementation study, and description of the methodology in the title and/or keywords
Abstract 2 2-3 Identification as an implementation study, including a description of the implementation strategy to be tested, the evidence-based intervention being implemented, and defining the key implementation and health outcomes.
IntroductionIntroduction 3 5 Description of the problem, challenge or deficiency in healthcare or public health that the intervention being implemented aims
to address.Rationale 4 9-11 The scientific background and rationale for the
implementation strategy (including any underpinning theory/framework/model, how it is expected to achieve
its effects and any pilot work).
8 The scientific background and rationale for the intervention being implemented (including evidence
about its effectiveness and how it is expected to achieve its effects).
Aims and objectives
5 9 The aims of the study, differentiating between implementation objectives and any intervention objectives.
Methods: descriptionDesign 6 10 The design and key features of the evaluation, (cross referencing to any appropriate methodology reporting standards) and any
changes to study protocol, with reasonsContext 7 11-13 The context in which the intervention was implemented. (Consider social, economic, policy, healthcare, organisational barriers
and facilitators that might influence implementation elsewhere).Targeted
‘sites’8 13 The characteristics of the targeted ‘site(s)’ (e.g
locations/personnel/resources etc.) for implementation and any eligibility criteria.
8 The population targeted by the intervention and any eligibility criteria.
Description 9 14 A description of the implementation strategy 8 A description of the intervention
Page 48 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
2
Sub-groups 10 N/A Any sub-groups recruited for additional research tasks, and/or nested studies are described
Methods: evaluationOutcomes 11 19 Defined pre-specified primary and other outcome(s) of
the implementation strategy, and how they were assessed. Document any pre-determined targets
N/A Defined pre-specified primary and other outcome(s) of the intervention (if assessed), and how they were assessed. Document any pre-determined targets
Process evaluation
12 15 Process evaluation objectives and outcomes related to the mechanism by which the strategy is expected to work
Economic evaluation
13 N/A Methods for resource use, costs, economic outcomes and analysis for the implementation strategy
N/A Methods for resource use, costs, economic outcomes and analysis for the intervention
Sample size 14 7 Rationale for sample sizes (including sample size calculations, budgetary constraints, practical considerations, data saturation, as appropriate)
Analysis 15 21-22 Methods of analysis (with reasons for that choice)
Sub-group analyses
16 23 Any a priori sub-group analyses (e.g. between different sites in a multicentre study, different clinical or demographic populations), and sub-groups recruited to specific nested research tasks
ResultsCharacteristics 17 N/A Proportion recruited and characteristics of the recipient
population for the implementation strategyN/A Proportion recruited and characteristics (if appropriate)
of the recipient population for the interventionOutcomes 18 N/A Primary and other outcome(s) of the implementation
strategyN/A Primary and other outcome(s) of the Intervention (if
assessed)Process
outcomes19 N/A Process data related to the implementation strategy mapped to the mechanism by which the strategy is expected to work
Economic evaluation
20 N/A Resource use, costs, economic outcomes and analysis for the implementation strategy
N/A Resource use, costs, economic outcomes and analysis for the intervention
Sub-group analyses
21 N/A Representativeness and outcomes of subgroups including those recruited to specific research tasks
Page 49 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
3
Fidelity/ adaptation
22 N/A Fidelity to implementation strategy as planned and adaptation to suit context and preferences
N/A Fidelity to delivering the core components of intervention (where measured)
Contextual changes
23 N/A Contextual changes (if any) which may have affected outcomes
Harms 24 N/A All important harms or unintended effects in each group
DiscussionStructured discussion
25 24-25 Summary of findings, strengths and limitations, comparisons with other studies, conclusions and implications
Implications 26 N/A Discussion of policy, practice and/or research implications of the implementation strategy (specifically
including scalability)
N/A Discussion of policy, practice and/or research implications of the intervention (specifically including
sustainability)General
Statements 27 27 Include statement(s) on regulatory approvals (including, as appropriate, ethical approval, confidential use of routine data, governance approval), trial/study registration (availability of protocol), funding and conflicts of interest
Page 50 of 49
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review onlyProtocol for the Process Evaluation of a Complex
Intervention Delivered in Schools to Prevent Adolescent Depression: The Future Proofing Study
Journal: BMJ Open
Manuscript ID bmjopen-2020-042133.R1
Article Type: Protocol
Date Submitted by the Author: 25-Oct-2020
Complete List of Authors: Beames, Joanne; University of New South Wales, Black Dog InstituteLingam, Raghu; University of New South Wales, School of Women’s and Children’s HealthBoydell, Katherine; University of New South Wales, Black Dog InstituteCalear, Alison L.; Australian National University, Centre for Mental Health ResearchTorok, Michelle; University of New South Wales, Black Dog InstituteMaston, Kate; University of New South Wales, Black Dog InstituteZbukvic, Isabel; Orygen The National Centre of Excellence in Youth Mental HealthHuckvale, Kit; University of New South Wales, Black Dog InstituteBatterham, Philip; Australian National University, Centre for Mental Health ResearchChristensen, Helen; University of New South Wales, Black Dog InstituteWerner-Seidler, Aliza ; University of New South Wales, Black Dog Institute
<b>Primary Subject Heading</b>: Public health
Secondary Subject Heading: Mental health
Keywords: Child & adolescent psychiatry < PSYCHIATRY, Depression & mood disorders < PSYCHIATRY, PREVENTIVE MEDICINE
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on N
ovember 15, 2021 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2020-042133 on 12 January 2021. D
ownloaded from
For peer review onlyI, the Submitting Author has the right to grant and does grant on behalf of all authors of the Work (as defined in the below author licence), an exclusive licence and/or a non-exclusive licence for contributions from authors who are: i) UK Crown employees; ii) where BMJ has agreed a CC-BY licence shall apply, and/or iii) in accordance with the terms applicable for US Federal Government officers or employees acting as part of their official duties; on a worldwide, perpetual, irrevocable, royalty-free basis to BMJ Publishing Group Ltd (“BMJ”) its licensees and where the relevant Journal is co-owned by BMJ to the co-owners of the Journal, to publish the Work in this journal and any other BMJ products and to exploit all rights, as set out in our licence.
The Submitting Author accepts and understands that any supply made under these terms is made by BMJ to the Submitting Author unless you are acting as an employee on behalf of your employer or a postgraduate student of an affiliated institution which is paying any applicable article publishing charge (“APC”) for Open Access articles. Where the Submitting Author wishes to make the Work available on an Open Access basis (and intends to pay the relevant APC), the terms of reuse of such Open Access shall be governed by a Creative Commons licence – details of these licences and which Creative Commons licence will apply to this Work are set out in our licence referred to above.
Other than as permitted in any relevant BMJ Author’s Self Archiving Policies, I confirm this Work has not been accepted for publication elsewhere, is not being considered for publication elsewhere and does not duplicate material already published. I confirm all authors consent to publication of this Work and authorise the granting of this licence.
Page 1 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 1
1 Protocol for the Process Evaluation of a Complex Intervention Delivered in Schools to
2 Prevent Adolescent Depression: The Future Proofing Study
3
4 Joanne R. Beames1 [email protected]
5 Raghu Lingam2 [email protected]
6 Katherine Boydell1 [email protected]
7 Alison L. Calear3 [email protected]
8 Michelle Torok1 [email protected]
9 Kate Maston1 [email protected]
10 Isabel Zbukvic4 [email protected]
11 Kit Huckvale1 [email protected]
12 Philip J. Batterham3 [email protected]
13 Helen Christensen1 [email protected]
14 Aliza Werner-Seidler1* [email protected]
15 Affiliations:
16 1 Black Dog Institute, University of New South Wales, Sydney, NSW, Australia.
17 2 Population Child Health Clinical Research Group, School of Women’s and Children’s
18 Health, University of New South Wales, NSW, Australia.
19 3 Centre for Mental Health Research, The Australian National University, Canberra, ACT,
20 Australia.
21 4 Orygen, Parkville, Victoria, Australia; Centre for Youth Mental Health, The University of
22 Melbourne, Parkville, Victoria, Australia
23 *Corresponding author:
24 Dr Aliza Werner-Seidler, Black Dog Institute, University of New South Wales, Sydney,
25 NSW, Australia. [email protected]; +61 2 9382 3808
Page 2 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 2
1 Abstract
2 Introduction: Process evaluations provide insight into how interventions are delivered across
3 varying contexts and why interventions work in some contexts and not in others. This
4 manuscript outlines the protocol for a process evaluation embedded in a cluster randomised
5 trial of a digital depression prevention intervention delivered to secondary school students
6 (The Future Proofing Study). The purpose is to describe the methods that will be used to
7 capture process evaluation data within this trial.
8 Methods and Analysis: A mixed methods design will be used with data collected in the
9 intervention arm of the Future Proofing Study. Data collection methods will include semi-
10 structured interviews with school staff and study facilitators, automatically collected
11 intervention usage data, and participant questionnaires (completed by school staff, school
12 counsellors, study facilitators, and students). Information will be collected about: i) how the
13 intervention was implemented in schools, including fidelity; ii) school contextual factors and
14 their association with intervention reach, uptake and acceptability; iii) how school staff, study
15 facilitators, and students and responded to delivering or completing the intervention. How
16 these factors relate to trial effectiveness outcomes will also be assessed. Overall synthesis of
17 the data will provide school cluster-level and individual-level process outcomes.
18 Ethics and Dissemination: Ethics approval obtained from the University of New South
19 Wales. Results will be submitted for publication in peer reviewed journals and discussed at
20 conferences. Our process evaluation will contextualise the trial findings with respect to how
21 the intervention may have worked in some schools but not in others. This evaluation will
22 inform the development of a model for rolling out digital interventions for the prevention of
23 mental illness in schools.
24 Trial Registration: The randomised controlled trial has been prospectively registered:
25 ACTRN12619000855123. Registered 31 May 2019,
Page 3 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 3
1 https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=377664&isReview=true
2 Keywords: Process evaluation, complex intervention, mixed methods, adolescent mental
3 health, depression, prevention, randomised controlled trial, implementation.
Page 4 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 4
1 Strengths and Limitations of this Study
2 This protocol is the first to illustrate how to embed a process evaluation into a school-
3 based randomised controlled trial of a digital depression prevention intervention.
4 Flexible and pragmatic quantitative and qualitative data collection methods will be
5 used to balance research rigor with feasibility within the school delivery context.
6 Process data from a range of key stakeholders will be collected, including school
7 staff, study facilitators, and students.
8 To minimise burden on schools, fidelity data from teachers and in-depth qualitative
9 data from students will not be collected.
10 The methodology and study processes were pilot tested to ensure appropriateness
11 within the school context.
Page 5 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 5
1 Introduction
2 Embedded in randomised controlled trials, process evaluations provide insight into
3 reasons why interventions work in some contexts and not in others. These process evaluations
4 can help to demystify the ‘black box’ of complex intervention trials by taking into account
5 contextual factors, differences in ways the intervention is delivered, and adaptations made for
6 intervention delivery into a particular system (1, 2). Contextual factors typically include
7 features of the organisation or broader environment that influence the delivery of the
8 intervention (e.g., leadership, engagement, culture, political landscape). Considering the
9 contribution of contextual factors is necessary to aid interpretation of trial outcomes,
10 maximise the knowledge gained from trials, identify optimal delivery processes across
11 different settings, and inform broader dissemination efforts. In recognition of this need for
12 process evaluations, the UK’s Medical Research Council (MRC) set out a framework that
13 emphasises the value of these evaluations in order to capture both contextual and
14 implementation factors associated with complex interventions (3).
15 In line with best-practice recommendations that study protocols that pre-specify
16 methods and approaches should be published to maintain research integrity (4), we describe a
17 protocol for a mixed methods process evaluation embedded within a cluster randomised
18 controlled trial (cRCT) known as the Future Proofing Study (FPS). The FPS is a large school-
19 based trial examining whether depression can be prevented using cognitive behaviour therapy
20 (CBT) delivered by smartphone application (5).
21 There has been an increase in the availability of digital mental health programs which
22 show promise in addressing the significant disease burden associated with depression (6).
23 Depression often first emerges during adolescence (7) and treatment alone cannot adequately
24 reduce this burden (8). Accumulating evidence indicates that early adolescence is the ideal
25 developmental window during which to intervene because it captures young people before
Page 6 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 6
1 the incidence of depression increases exponentially around the age of 16-17 years (9, 10).
2 Therefore, increased efforts are being directed toward prevention approaches that are
3 developed specifically for delivery among adolescents.
4 While there is evidence supporting the value and effectiveness of prevention
5 programs in schools (11, 12), their uptake has been significantly limited by low levels of
6 help-seeking (13) as well as practical constraints in cost and scalability. A lack of
7 understanding about how to deliver these interventions sustainably at scale, and engage those
8 who stand to benefit, have hampered the translation of prevention approaches into the
9 community.
10 Two ways to overcome the barriers of cost and scalability of implementing adolescent
11 depression interventions are:
12 Delivering universal prevention programs in schools, and
13 Using technology to deliver programs, which tends to be lower cost than traditional
14 face-to-face methods (14).
15 Working with schools to deliver universal prevention programs, to every student,
16 regardless of their risk, circumstance, or symptom profile, not only dramatically increases the
17 potential reach of interventions, it also means that these programs can eventually be
18 integrated into the curriculum and delivered to every student, making this approach a
19 sustainable implementation strategy for widescale delivery and dissemination. Additionally,
20 working with schools to deliver such programs reduces the need for young people to actively
21 seek professional help. This is critically important - despite having a significant need, fewer
22 than 80% of young people with mental illness seek help and receive the services they need
23 (15, 16). The scaffolding provided by schools can be leveraged to deliver prevention
24 programs to all students, which fits with a trend for schools to be designated the first point of
25 contact to support youth mental health problems when they first emerge (9). Process
Page 7 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 7
1 evaluations of face-to-face mental health programs have been documented, some of which
2 have been delivered in school settings (e.g., 17). However, we were unable to find any
3 process evaluation descriptions in the literature that evaluate digital mental health programs
4 in school settings. Conducting process evaluations specific to digital methods of delivery are
5 important because the contextual barriers and facilitators, as well as concerns around fidelity
6 of the intervention, are likely to have unique characteristics. For example, facilitator or
7 training in intervention delivery will be different in supporting the use of an automated
8 program relative to a face-to-face program. This process evaluation is an initial step towards
9 addressing that gap.
10 Technology offers a promising way to deliver mental health interventions to the
11 community. Digital mental health interventions offer two key advantages over face-to-face
12 approaches – first, they are cheaper to access and more cost effective (18, 19), and second,
13 they can be used to reach people across vast geographic areas. The latter is particularly
14 important given Australia’s geography, where remoteness and low population density meant
15 that people in regional and rural areas often do not have the same level of access to mental
16 health services as people in metropolitan areas. With more young people than ever using
17 smartphones, mental health intervention delivered online or through applications represents
18 an exciting avenue for reaching adolescents.
19 The Future Proofing Study (FPS)
20 The trial that is the subject of this process evaluation is the FPS. The FPS addresses
21 barriers of reach, cost, and scalability by delivering a depression prevention program via a
22 smartphone app to Year 8 secondary school students aged 12-13 years. This study is being
23 conducted with approximately 200 Australian schools (up to 10,000 participants), of which
24 half will be allocated to the intervention condition. The primary outcome is symptoms of
25 depression. Secondary outcomes include anxiety, psychological distress and insomnia, among
Page 8 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 8
1 others. Symptom outcomes will be assessed at baseline, post-intervention, six months
2 (primary outcome only), and then annually for five years. The primary endpoint is 12 months
3 following baseline. More details are available from the Australian New Zealand Clinical
4 Trials Registry (ANZCTRN12619000855123) and protocol paper (5).
5 The intervention being delivered is known as SPARX, a CBT-based program
6 incorporating gamification principles. The development and initial evaluation of SPARX has
7 been documented in detail previously (20). SPARX has been tested in an Australian sample
8 of secondary students (delivered via computer) and shown to prevent depression in the lead
9 up to final school exams (21). The gamified intervention teaches young people about the
10 relationship between thoughts, feelings, and behaviour. Skills learnt through SPARX include
11 emotion identification, emotion regulation, behavioural activation (being active), recognising
12 and challenging unhelpful thoughts, and practical problem solving. SPARX consists of seven
13 20-minute modules. The intervention is fully automated, and the therapeutic components are
14 standardised. See the trial protocol for full intervention details (5).
15 School Engagement and Recruitment
16 The FPS requires multiple levels of approval, beginning with state department and
17 independent school body approval (NSW Department of Education; Catholic school
18 dioceses), followed by individual school engagement. An engagement strategy for this study
19 involves sending electronic communication material to all schools across New South Wales
20 and in other Australian capital cities, targeting school principals and wellbeing staff to invite
21 them to participate in this study opportunity. Schools are invited to submit expressions of
22 interest and are subsequently followed up by the research team over the phone to explain
23 more about the study. The schools for which the FPS is a good fit (as determined by the
24 school) are then signed onto the study, with support from the principal, the school counsellor,
25 and at least one other staff member (typically a teacher). In line with best practice in
Page 9 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 9
1 implementation science (22), this group of 2-3 staff members (typically not including the
2 principal) will form the school-specific ‘study implementation team’. After signing on to the
3 study, several webinars are scheduled throughout the lead up to the study start date, so that
4 school staff and parents can listen to a 15-minute study overview from the trial manager and
5 have their questions answered.
6 Preparation
7 In preparation for in-class assessment sessions, study facilitators (volunteer research
8 assistants) are recruited to support the study. The purpose of these facilitators is to attend
9 schools to introduce the study to students, and ensure the technology is functioning so
10 students can download the SPARX app and complete the baseline (and post-assessment)
11 questionnaires. All facilitators go through an interview and screening process prior to
12 selection, then attend a half day face-to-face training session before supporting the study in
13 schools. This training provides an overview of the study and detailed information about their
14 roles within schools, including a step-by-step guide to running the sessions. Discussion and
15 practice are core components of the training.
16 Aims and Objectives
17 We adapted the MRC framework for complex interventions to focus on effectiveness of
18 an evidence-based intervention within a specific context of delivery (i.e., schools). Overall,
19 the objective of this process evaluation is to understand how SPARX is implemented and
20 delivered in schools, and to identify systematic differences and variation in delivery.
21 Specifically, the aims are:
22 1. To evaluate the reach (including completion), uptake and acceptability of the
23 intervention (school- and student-level).
24 2. To understand the contribution of contextual factors (e.g., characteristics of the
25 outer/inner setting, intervention, individuals) on:
Page 10 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 10
1 o School-level fidelity to the implementation strategy. For example, different
2 schools will likely provide different levels of study support based on available
3 resourcing.
4 o Implementation outcomes (intervention reach, uptake, acceptability), as
5 assessed from the perspectives of school staff, teachers, and students. For
6 example, young people’s openness to receiving mental health material via an
7 app will likely impact intervention acceptability and completion.
8 3. To examine the impact of school-level variation (in implementation fidelity and
9 outcomes) on clinical effectiveness outcomes at the school- and student-level. School-
10 level clinical effectiveness is defined as changes in clinical outcomes (e.g., self-
11 reported depression) for different schools. The differing ways in which schools
12 support and deliver the intervention will inevitably impact its effectiveness, and this
13 evaluation will assess these differences (3, 23). For example, degree of support from
14 senior school leaders who would be expected to understand that participating in study
15 activities is a priority for the school will likely impact effectiveness at the cluster
16 level.
17 This process evaluation has been designed to capture important information from
18 teachers, school staff and students at both the school and individual level, which may
19 ultimately impact the effectiveness of the intervention on clinical mental health outcomes for
20 students. Findings will provide insight into factors which support and/or hinder the
21 implementation of digital universal mental health programs in school settings. Knowledge
22 gained from this process evaluation will help to inform the development of a model and guide
23 for how to best deliver digital mental health programs to young people in schools.
24
25 Methods and Analysis
Page 11 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 11
1 Design
2 This study uses a hybrid type 1 approach (24), with a focus on implementation
3 process factors and outcomes in the context of an effectiveness trial. The evaluation is guided
4 by the CFIR and RE-AIM framework (25, 26). In keeping with Nilsen’s categorisation of
5 implementation theories, models, and frameworks, these frameworks help us to understand
6 different parts of the implementation process (27). The CFIR will be used to identify
7 barriers and facilitators to intervention implementation and effectiveness (27). The CFIR is
8 a widely used deterministic theoretical framework which has been applied in multiple settings
9 (28), including schools (e.g., 29). The CFIR identifies five major domains, including (i) the
10 outer setting, which includes the social, political and economic context that the organisation
11 in which implementation is occurring exists; (ii) the inner setting, which includes features and
12 characteristics of the organisation such as leadership and relative priority; (iii) the
13 characteristics of individuals, which includes organisational staff knowledge and attitudes
14 about the intervention and their role and identification within the wider organisation; (iv) the
15 characteristics of the intervention itself and; (v) implementation processes, which includes the
16 ways that the intervention will be delivered in a given context (including fidelity to the
17 implementation strategy). Normalisation Process Theory will be used to provide additional
18 insights into implementation processes (NPT; 30). NPT aims to identify and explain the
19 implementation of new interventions and how they become integrated into routine care.
20 The RE-AIM framework will be used to evaluate the implementation outcomes,
21 including intervention reach, uptake, and acceptability/appropriateness (25, 31). These
22 outcomes also map onto the framework of implementation outcomes proposed by Proctor and
23 colleagues (32, 33). Reach refers to the proportion of eligible participants who opened, used,
24 and completed the intervention, as well as the proportion of students from the entire cohort of
25 eligible students in intervention schools who consented to participate. Uptake refers to the
Page 12 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 12
1 proportion of schools that were onboarded to the study (intervention and control arms), and
2 of school staff who were willing to support the delivery of the intervention (intervention arm
3 only). Reach and uptake also incorporate the representativeness of the sample (school-level
4 and individual-level). Acceptability and appropriateness refer to the perceived agreeableness
5 or fit of the intervention. This evaluation will incorporate how the barriers and facilitators
6 identified through the CFIR impacted the implementation outcomes, and, in turn, how these
7 implementation outcomes impacted effectiveness outcomes.
8
9 Logic Model
10 Following MRC guidance (3), the research team developed a logic model for the FPS
11 process evaluation in a series of participatory workshops. The logic model was prospectively
12 informed by key CFIR constructs identified in previous literature as being important in
13 school-based studies, but also feasible and appropriate to measure within the school context
14 in the FPS. The key constructs included the outer setting, inner setting, individual
15 characteristics, and intervention characteristics. The logic model (Figure 1) was developed to
16 consider the key factors (mapped to the CFIR) that would potentially impact implementation
17 of the intervention (mapped to RE-AIM), as well as its clinical effectiveness. The process
18 evaluation method, including the selection of dependent variables and design of surveys and
19 semi-structured interview guides, were derived from this logic model. See Table 1 for
20 CFIR/RE-AIM domains, key research questions, process data, and data that will be collected
21 within each domain.
22 A pilot study of 8 schools conducted in 2019 was used to assess the suitability of the
23 planned implementation strategy and process evaluation methods. We retrospectively applied
24 our logic model to collected data to evaluate whether our methodology captured relevant
25 constructs, and sufficient variation in these constructs. We integrated learnings from this pilot
Page 13 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 13
1 study using a dynamic feedback process to strengthen our methodology for the full-scale FPS
2 phase. Some minor changes were made following this pilot which relate to the CFIR,
3 including a greater emphasis on sub-constructs such as school climate (inner setting).
4
5 <INSERT FIGURE 1 HERE>
6
7 Figure 1. Logic model. The model shows that CFIR constructs including school context
8 characteristics, school organisational characteristics, and individual characteristics, will
9 influence how staff engage with the implementation strategy. The intervention itself, which
10 includes the core cognitive-behavioural therapeutic components, is conceptualised as
11 standardised across individuals because it is delivered digitally, follows a fixed schedule and
12 does not incorporate tailored content. The yellow input factors are expected to vary across
13 schools and individuals, thus influencing engagement and flexibility of the implementation
14 strategy and in turn, implementation outcomes and student-level outcomes. The logic model
15 and implementation plan were externally peer reviewed by an experienced and internationally
16 recognised implementation scientist outside the team within an implementation workshop.
17 For details on assessment of these factors, see Table 1.
Page 14 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 14
1 Table 1 – Process evaluation details including process data, outcome data, data type and source
CFIR and RE-AIM Constructs Research Aim Process or Outcome Data Data Type and Source
Outer SettingSchool contextual characteristics
What was the broad context of the schools in which the SPARX intervention was delivered? (Aim 2)
School socio-economic indexSchool location (metropolitan/regional)
Publicly available information (ICSEA, GPS)
Inner SettingSchool organisational characteristics
What were the characteristics of the delivery environments (schools)? (Aim 2)
What were the barriers and facilitators that affected buy-in, delivery, and student uptake? (Aim 2)
School size, type, composition, fundingSchool culture
Implementation climate, relative priority, competing demands, leadership, school counsellor availability and level of support, networks and communication, school culture and climate, school readiness for implementation
Publicly available information (school size, type, funding)SSPESH (assesses school culture) Interviews with school staff
Implementation Climate measures (staff and students)Relative Priority measure Competing Demands measureInterviews with school staffChecklists completed by trial managerOther administrative data including number of staff allocated to assist with delivery and consent process, level of communication with the research team
School leadership How supportive of delivering SPARX were school principals, deputy principals and executives? (Aim 2)
Level of support and buy in from school leaders
Interviews with school staff
Individual CharacteristicsSchool staff What were the characteristics (including attitudes,
beliefs, traits) of school staff supporting the delivery of the intervention? (Aim 2)
How supportive of delivery were school staff who were involved on-the-ground? (Aim 2)
Age, gender, current employment, role etc
Leadership, skills, motivations, expectations, self-efficacy, expectations, time available, knowledge and beliefs about the intervention, study buy in
Demographics questionnaire
Implementation Leadership ScaleInterviews with school staff and study facilitators
Study facilitators How well did study facilitators attending schools support the delivery of the intervention? (Aim 2)
Age, employment Skills, self-efficacy or confidence, motivations and expectations
Demographics questionnaireSelf-Confidence measure Interviews with study facilitators
Page 15 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 15
Students What were the characteristics of young people that affected intervention uptake and effectiveness? (Aim 2)
History of mental illness Reported by Year 8 students as part of the online FPS survey
Intervention CharacteristicsSPARX Were there any barriers to intervention use? (Aim 2
& 3)
What do staff think about the efficacy and advantage of using the intervention? (Aim 2)
Technical issues
Evidence strength and quality, relative advantage
Logs of technical issues sent through schools, parents and participantsIT data pertaining to technical problemsInformal feedback provided by schools and research staff attending schools
Relative Advantages measureAnticipated Benefits measureInterviews with school staff
Implementation Processes
Normalisation and integration How did school staff perceive the implementation processes? (Aim 2)
Coherence, cognitive participation, and collective action
NoMAD
Fidelity to the implementation strategy
To what extent was the intervention implemented as planned? (Aim 2)
School delivery of the FP program, including changes to the plan
Completed implementation checklists, emails and feedback formsInterviews with school staff
Implementation Outcomes Reach What was the extent to which those who were eligible
to receive SPARX used it? (Aim 1)Proportion of eligible participants who consented to participate; proportion who opened, used and completed the SPARX interventionRepresentativeness of the student sample
Administrative data about consent. Digital analytic data including usage (app downloads, installs, opens), completion rate (number of modules completed) and time spent using SPARXReported by Year 8 students as part of the online FPS survey
Uptake How many eligible schools participated in the study? Within those schools, how many staff supported the delivery of SPARX? (Aim 1)
Proportion of eligible schools that were onboarded to the study; proportion of school staff (in intervention schools) who supported SPARX Representativeness of the sample
Administrative data
Publicly available information about schools and self-report demographic data from school staff
Acceptability/appropriateness How satisfied were participants with the intervention? (Aim 1)
Acceptability/appropriateness of the intervention, expectations
Acceptability/appropriateness of the Future Proofing program (including SPARX)
Reported by Year 8 students as part of the online FPS surveyInformal conversations and feedback provided by Year 8 studentsImplementation Appropriateness measureInterviews with school study staff
Page 16 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 16
How satisfied were school staff with supporting the intervention the FP program (including SPARX)? (Aim 1)
Informal conversations and feedback provided by school staff
Across DomainsHow might the relationship between the intervention, the staff supporting the program, and context of each school shape variation in outcomes (implementation strength metric)? (Aim 2 and 3)What key lessons emerge from this study that can be generalised to the implementation of digital mental health programs in schools more broadly?
1 Note. The process data and outcomes are mapped onto Figure 2. CFIR = Consolidated Framework for Intervention Research; RE-AIM = Reach, Effectiveness, Adoption, Implementation, 2 Maintenance; Aim = Aims outlined in the Aims and Objectives section; SSPESH = Survey of School Promotion of Emotional and Social Health; NoMAD = Normalisation Measure 3 Development questionnaire.
Page 17 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 17
1 School Implementation Strategy
2 The school implementation strategy was developed by the study authors. The authors
3 drew on their experience in school-based intervention delivery and integrated feedback from
4 teachers and school staff from several completed school-based trials that delivered digital
5 interventions to school students (21, 34). Stakeholder consultation specifically for this study
6 involved discussions with the Department of Education, consultation with several school
7 parent committees and consultation with both youth and parent Lived Experience Advisory
8 Panels. This strategy was also refined following the first pilot wave involving eight
9 intervention schools.
10 The school ‘study implementation teams’ are principally responsible for the
11 implementation of the intervention and liaising with the research team. As described earlier,
12 these teams typically incorporate at least one classroom teacher and one school counsellor to
13 assist with the intervention delivery. Study facilitators also support the delivery of the
14 SPARX intervention by attending schools for the first school session.
15 Implementation Strategy. During the active intervention phase, schools allocate a
16 minimum of 4 x 20-minute school class sessions during which students complete the SPARX
17 intervention. The additional three sessions may be completed either in class if permitted by
18 the schools, or in the students’ own time. The implementation strategy developed by the
19 project team and stakeholders to support the completion of the SPARX intervention
20 comprises:
21 Standardised facilitator training delivered face-to-face over a half-day
22 Study facilitators are present at schools to support students in downloading the
23 SPARX app and completing the baseline assessments
24 School implementation team provided with a SPARX user-guide and information
25 booklet they can refer to during the sessions
Page 18 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 18
1 Schools provide students with weekly verbal reminders in home classrooms to use the
2 app regularly
3 Schools publish brief information about the study and mental health tips in the school
4 weekly newsletter
5 Schools liaise with research team weekly to troubleshoot problems
6 This is the strategy being outlined to schools by the research team and adherence to this
7 strategy will be assessed. Whether or not schools schedule more than the four mandated in-
8 class sessions for intervention completion is flexible and can be adapted to suit the
9 preferences of the school. Student participants receive a $20AUD voucher after the
10 intervention period to cover any phone or data related costs incurred. See Figure 2 for details
11 of training and delivery structure.
12
13 <INSERT FIGURE 2 HERE>
14
15 Data Collection Methods and Participant Groups for Process Evaluation
16 There are three participant groups taking part in the process evaluation: Year 8
17 students, school staff members (e.g., teaching and counselling staff), and study facilitators
18 (see Table 2). School staff members will be members who are responsible for leading the
19 delivery of the study in their school or will be teaching staff who have a supporting role (e.g.,
20 home room teachers who provide reminders to students to complete the intervention).
21 Four types of data will be collected and triangulated: self-report questionnaire data,
22 digital analytic data, administrative data, and qualitative interview data. Self-report
23 questionnaire data specific to the process evaluation will be collected from staff and study
24 facilitators using online survey software (Qualtrics) programmed by one of the authors (JB).
25 These questionnaires assess demographic information, school organisational characteristics,
Page 19 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 19
1 individual characteristics, intervention characteristics, implementation processes and
2 implementation outcomes. Where no available published questionnaires were identified as
3 being suitable for this process evaluation, we adapted existing standardised measures or
4 developed our own items (details below). Self-report questionnaire data from Year 8 students
5 about intervention use and feedback will be collected from an online survey (details
6 described in 5). Digital analytic data about intervention use will be captured by the purpose-
7 built Black Dog Institute research platform, which is being used for the broader FPS.
8 Administrative data will be collected through a range of sources, including communications
9 with the research team. Qualitative data will be collected using semi-structure interviews. All
10 data collected will be from intervention schools only as no intervention is implemented in
11 control schools. Student and school staff data will be collected immediately after the
12 intervention period has been completed (i.e., after the six-week intervention stage); facilitator
13 data will be collected both before and after the intervention period.
14
15 Table 2. Summary of data forms (and collection point) provided by each of the participant
16 groups
Questionnaire Individual interview Digital analytics
Year 8 students ✓ (post-intervention) ✓ (ongoing)
School staff ✓ (post-intervention) ✓ (post-intervention)
Facilitators ✓ (before first school visit and after final post-intervention visit)
✓ (after final school post-intervention visit)
17
18 Implementation Predictors
19 School Organisational Characteristics (Inner Setting).
20 Publicly Available Information. Publicly available information will be collected
Page 20 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 20
1 about school contextual characteristics, including socio-economic level, size, location, type,
2 and funding.
3 General School Culture. School staff will complete the Survey of School Promotion
4 of Emotional and Social Health (SSPESH; 35) and a measure of organisational culture (36).
5 The SSPESH assesses a school’s capacity to promote social and emotional wellbeing, and
6 contains four subscales: Positive School Community, Student Social and Emotional Learning,
7 Engaging Families, and Supporting Students Experiencing Mental Health Difficulties. Items
8 are rated on a 4-point Likert scale from 0 (not yet in place) to 3 (completely in place) and
9 preliminary investigations support the scale structure and criterion-related validity (35).
10 We adapted a 9-item questionnaire about general culture within a health care setting
11 to use within the school setting (36). Items are rated on a 5-point Likert scale from 1 (strongly
12 agree) to 5 (strongly disagree). Lower scores indicate a more positive working culture, which
13 includes transparency, productive working relationships, and receptivity to feedback.
14 Previous investigation has shown that this measure has good internal consistency, although
15 overlaps somewhat with other sub-constructs within the inner setting (e.g., learning climate;
16 35).
17 Implementation Climate. School staff will complete a measure of implementation
18 climate that we adapted for use in the school context (36). Items are rated on a 5-point Likert
19 scale from 1 (strongly agree) to 5 (strongly disagree). Lower scores indicate increased staff
20 receptivity to the program and support within the school, including rewards and recognition.
21 This measure has acceptable internal consistency and good discriminant validity (36).
22 Relative Priority. School staff will complete one adapted item from the School
23 Contextual Barriers subscale of the Perceived Attributes of the Healthy Schools Approach
24 Scale (37). The item is rated on a 5-point scale Likert scale from 1 (strongly agree) to 5
25 (strongly disagree). Higher scores indicate that other activities did not interfere with
Page 21 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 21
1 implementation of the FP program.
2 Competing Demands. Staff will complete 3-items, developed specifically for this
3 study, which assess how much time they allocated, and desired to allocate, to the FP program
4 relative to other competing workload demands on visual analogue scales (anchor by 0 = no
5 work time, 100% = all of my work time). Higher scores indicate increased allocation of time
6 and prioritisation of the FP program.
7 Individual Characteristics.
8 Leadership. School staff will complete the Implementation Leadership Scale (ILS;
9 38). The ILS is a 12-item scale that assesses leadership behaviours that support
10 implementation of evidence-based practices. The ILS contains four subscales, and two will be
11 included in the current study (Knowledge Leadership and Supportive Leadership). These
12 scales assess the degree to which the staff member was knowledgeable about and offered
13 support to the program. Items are rated on a 5-point Likert scale from 0 (very much so) to 4
14 (not at all) and will be re-coded such that higher scores indicate more effective
15 implementation leadership behaviours. This measure has excellent internal validity,
16 convergent validity, and discriminant validity (38).
17 Self-Efficacy and Confidence. Study facilitators will rate their level of confidence in
18 performing 23 different tasks during school-visits on visual analogue scales, which are based
19 on the training program they completed (anchor by 0 = not at all confident, 100 = very
20 confident). Higher scores indicate higher levels of confidence. A similar questionnaire will be
21 repeated following their school-visits, along with four short answer questions that assess
22 experiences with supporting staff and students during school-visits.
23 Intervention Characteristics.
24 Relative Advantage and Evidence Strength and Quality. School staff will complete
25 the 2-item Relative Advantages subscale and one item from the Anticipated Benefits subscale
Page 22 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 22
1 of the Perceived Attributes of the Healthy Schools Approach Scale (37). Adapted for the
2 current study, items are rated on a 5-point scale Likert scale from 1 (strongly agree) to 5
3 (strongly disagree). Scores will be re-coded such that higher scores indicate greater perceived
4 advantage of the FP program compared to others and greater perceived impacts on mental
5 health, respectively.
6 Implementation Processes.
7 Normalisation and Integration into Routine Practice. School staff will complete
8 NPT’s accompanying tool, the Normalisation MeAsure Development questionnaire
9 (NoMAD; 39). The NoMAD is a 23-item measure that assesses how professionals involved
10 in the implementation of a complex intervention perceive implementation processes. It is a
11 flexible measure that can be altered to more accurately describe the adoption of new
12 interventions at the provider level. Ten items of the NoMAD will be used in this study to
13 assess intervention buy-in from school staff, and how staff members incorporated the
14 initiative into their standard work responsibilities. These items are grouped into three
15 categories: coherence (i.e., making sense of an intervention), cognitive participation (i.e.,
16 working with others to support an intervention), and collective action (i.e., the type of work
17 that people do to support an intervention). Items are rated on a 5-point Likert scale from 1
18 (strongly agree) to 5 (strongly disagree). Initial validation demonstrated that the NoMAD has
19 good face validity, construct validity, and internal consistency (40).
20 Fidelity to the Implementation Strategy. Fidelity will be captured through
21 implementation checklists, communications between schools and research staff, and
22 interviews with school staff.
23 Implementation Outcomes
24 Reach. SPARX app usage data from students will allow for the assessment of app use
25 (downloads, installs, and opens), completion (number of modules completed, out of a total of
Page 23 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 23
1 seven), and time spent using SPARX. Administrative data about the number of students in
2 intervention schools with consent to participate will also be used as an indicator of reach.
3 Self-report data about individual characteristics (e.g., gender, mental health history) will be
4 used to gauge the representativeness of the students in the intervention schools.
5 Uptake. Administrative data about the proportion of schools that were onboarded to
6 the study and the proportion of teachers who supported the intervention will be collected to
7 provide an index of uptake by schools. Representativeness of the sample will be informed by
8 publicly available information about school characteristics (e.g., socio-economic status,
9 location) and self-report data about school staff characteristics (e.g., role, gender).
10 Appropriateness. School staff will complete the Intervention Appropriateness
11 Measure (IAM; 41). The IAM is a pragmatic 4-item measure of the perceived fit, relevance,
12 or compatibility of an evidence-based practice for a context, person, or problem (42). Items
13 have been adapted for this study and are rated on a 5-point Likert scale from 1 (strongly
14 agree) to 5 (strongly disagree). Scores will be re-coded such that higher scores indicate higher
15 levels of perceived appropriateness. The IAM demonstrated strong psychometric properties
16 in previous research (41). School staff will also complete one adapted item from the Agency
17 Leadership Support subscale of the Barriers and Facilitators to Implementing Survey (43).
18 The item is rated on a 5-point Likert scale from 1 (strongly agree) to 5 (strongly disagree) and
19 will be re-coded such that higher scores indicate greater compatibility of the FP program
20 within a particular school.
21 Acceptability. Year 8 students will complete an 11-item feedback questionnaire
22 about SPARX. The questionnaire assesses three domains, including: (i) reasons for non-
23 adherence; (ii) intervention acceptability; and (iii) skills learnt from the intervention. Items
24 will be quantified individually. This questionnaire has previously been used to assess the
25 acceptability of the SPARX program in the school context (44).
Page 24 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 24
1 Individual Interviews
2 Interview guides for school staff and study facilitators were derived from the logic
3 model, CFIR online resources (e.g., https://cfirguide.org/evaluation-design/qualitative-data/),
4 and the broader literature investigating the delivery of interventions in school settings (45, 46;
5 see Additional File 1 for interview guides). Interviews will provide information about both
6 implementation predictors and outcomes. For school staff, questions focus on: motivations
7 and expectations about the intervention and study processes more broadly; knowledge and
8 beliefs about the intervention; relative advantages of the intervention; self-efficacy; barriers
9 and facilitators affecting the delivery of the intervention, including fidelity; appropriateness
10 and acceptability of the intervention; and recommendations for future implementation.
11 School counsellors will also be asked questions about their experiences of managing high risk
12 participants within the study and compatibility with existing workload. For study facilitators,
13 questions focus on: motivations and expectations about their role; confidence and
14 competence in supporting the delivery of SPARX in schools; the quality of their training; and
15 perceptions about their ability to support the study as required.
16 The interview guides allow for flexibility in questioning and diversion in responses.
17 Questions will primarily be open-ended, with specific prompts and follow-up questions being
18 used as necessary to encourage respondents to elaborate on their ideas and provide examples.
19 Patient and Public Involvement
20 The Future Proofing Study was developed with key stakeholders including school
21 personnel, school counsellors, parents, adolescents, and individuals with a lived experience of
22 mental illness. All aspects of the study design, appropriateness of outcomes measures and
23 consent procedures were developed in consultation with these stakeholder groups. The
24 current process evaluation involves consultation with school staff members to understand
25 their experience. The SPARX intervention itself was codesigned with young people. All
Page 25 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 25
1 study results will be shared directly with participants and their schools through lay summaries
2 and infographics.
3 Procedure
4 Ethical approval was provided by the University of New South Wales Human
5 Research Ethics Committee (HC180836) and NSW Government State Education Research
6 Applications Process Approval (SERAP 2019201).
7 The FPS has two waves of delivery (October-December 2020; April-July 2021).
8 Process evaluation data will be collected at each wave. All school students will complete
9 relevant survey questions at the same time as completing primary measures for the FPS. All
10 study facilitators will be asked to provide informed consent for their data to be used for
11 research purposes following the compulsory face-to-face training session they attend with the
12 research team at the Black Dog Institute. Data provided by facilitators will be from two
13 surveys, one completed immediately following the training and another completed after their
14 final school visit. All school staff from intervention schools (on average, three from each
15 school, estimated number of intervention schools = 100) who were directly involved in the
16 study will be invited to complete one survey following the six-week SPARX intervention
17 period. All surveys will be completed online.
18 After completing their respective online questionnaires, all school staff and study
19 facilitators will be given the option to participate in a 60-minute individual interview
20 (completed either in person or remotely). Purposive sampling will be used to capture a range
21 of diverse school settings and experiences. Face-to-face interviews will be held in a quiet
22 room on school grounds or at the Black Dog Institute. Virtual interviews, which may be
23 required due to COVID-19 restrictions, will also be conducted. All interviews will be audio-
24 recorded and transcribed verbatim. All contact with study facilitators and school staff,
25 including the semi-structured interviews, will be made by research staff who have had no
Page 26 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 26
1 previous contact with them during the trial. This independence minimises the risk of bias and
2 demand effects.
3
4 Data analysis
5 Quantitative
6 Questionnaire data and open-ended questions from surveys will be exported into data
7 analytic software for analysis. Descriptive statistics will be calculated for all participant
8 groups and will provide information about intervention use and its acceptability
9 (questionnaire data from Year 8 participants), differential implementation, fidelity to the
10 implementation strategy, school context factors within each school (questionnaire including
11 several short answer questions from school staff), and competence and experience in schools
12 from those facilitating the study (facilitator questionnaires). Differences between school
13 clusters will be assessed using Analysis of Variance methods (ANOVA). Regression models
14 will assess the on the effects of contextual factors on implementation outcomes (e.g., reach,
15 uptake, acceptability).
16 Qualitative
17 Interviews will be digitally audio-recorded and transcribed verbatim. The transcripts
18 will be checked for accuracy against the sound files as per best practice in transcription (47,
19 48). Qualitative data will then be imported into NVivo to aid in data management and
20 analysis. Thematic analysis will be undertaken to identify, interpret and report on the
21 repeated patterns of meaning within the data, drawing from Braun and Clark’s classic six
22 phase model (49, 50). An iterative and reflexive approach will be used to analyse the data,
23 incorporating themes from the data together with topics covered in the interview guide. Two
24 coders will independently engage in a familiarisation phase before generating codes and
25 initial themes for a subset of the data. These codes and themes will be reviewed and
Page 27 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 27
1 discussed by the two coders, with refinement occurring via an interactive process. A senior
2 qualitative analysist will also review the first-stage coding framework and scheme before all
3 transcripts are coded by the first coder. Refinement will continue to occur via an interactive
4 process until final codes and themes are realised and defined across the whole data set.
5 Research rigor will be enhanced by a team approach to analysis, reflexive field notes, and
6 prolonged engagement with the subject matter (51).
7 Triangulation of Qualitative and Quantitative Data
8 Triangulation involves the use of multiple approaches to address a research question.
9 The combination of several approaches increases confidence in the findings and provides a
10 more comprehensive account of the results than individual approaches would do alone (52).
11 In this study, reliability, validity and confidence will be maximised through cross verification
12 and exploration of differences between the outcomes of the various methods. This takes place
13 in several ways:
14 Maximising validity in analysis of qualitative data within the research team using
15 techniques such as discussing coding, constant comparison, accounting for deviant
16 cases, and systematic coding;
17 Triangulation of school staff and research assistant interviews with results from the
18 questionnaires, exploring and accounting for differences;
19 Triangulation of self-report and interview data with publicly available information
20 relating to school contextual characteristics (e.g., school socio-economic level and
21 size) and school delivery of the program, including deviations to the implementation
22 strategy.
23 Mapping the perspectives of different stakeholders across the study (school staff,
24 study facilitators).
25 Additional Analyses
Page 28 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 28
1 We will generate an ‘implementation strength’ metric for each school and its
2 relationship to primary and secondary trial outcomes. This is an emerging evaluation
3 approach, used mainly in low and middle-income countries. The approach aims to understand
4 the degree of implementation effort needed during intervention delivery to achieve desired
5 benefits (53, 54).
6 The implementation strength metric will provide funders and policy makers with an
7 objective measure to monitor effectiveness of implementation if it goes beyond this trial and
8 becomes a sustainable approach. The metric can be used to assess whether the approach to
9 implementation meets the minimum level required to prevent the onset of mental health
10 problems in adolescents. The metric will be based on implementation inputs and contextual
11 factors informed by the intervention logic model and process evaluation frameworks (CFIR
12 and RE-AIM) (e.g., adoption from teachers and other staff in the school and fidelity to the
13 intervention strategy), and will be developed with the Future Proofing research team using
14 principal component analysis.
15
16 Discussion
17 This paper describes the design of a mixed methods process evaluation of a cluster
18 randomised controlled trial, the FPS. The FPS investigates the impact of a digital cognitive-
19 behavioural therapy intervention when delivered at scale in school settings. Digital mental
20 health programs have tremendous potential to prevent up to 22% of depression cases and,
21 when delivered at scale, could have population-level impacts (55). However, these programs
22 have not been translated into practice and policy because optimal ways to scale and deliver
23 these interventions are not yet well understood.
24 As an initial step to address this issue, the current process evaluation will attend to
25 contextual and implementation factors that vary across schools and provide a lens through
Page 29 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 29
1 which to interpret trial efficacy outcomes. We expect that results will provide a richly
2 detailed and nuanced understanding of the key factors involved in the effective delivery of
3 digital mental health programs across different schools. We expect that results will not only
4 contextualise our trial findings but will also be used as a model to guide the delivery of
5 school-based interventions that focus on preventing mental illness more broadly. Findings
6 from this process evaluation will indicate whether the approach used in the FPS trial is likely
7 to be sustainable in the school environment going forward and, if so, the threshold level of
8 support required in order to prevent depression and benefit student mental health.
9 The prospective publication of this protocol outlines our planned methodological
10 approach. It also serves as a road map for other researchers on a practical way of carrying out
11 process evaluations of complex interventions in the school setting. As is the case with the
12 delivery of interventions across different contexts, we acknowledge that our approach has
13 inbuilt flexibility to explore the data and make provisions for unexpected implementation
14 factors that arise.
15 Limitations and Strengths
16 There are several limitations to our process evaluation that warrant mention. First, we
17 are not including direct observation of teachers in their role supporting the delivery of
18 SPARX. While this would provide objective fidelity data, it requires resources beyond the
19 scope of this project and is not representative of how the program will be sustained following
20 the conclusion of the trial. Second, given the complexity of the study and high demand placed
21 on students (e.g., engaging with the study apps and completing the online surveys at multiple
22 time points over five years), we are not collecting in-depth qualitative data by way of
23 interviews. Instead, we collect information about students’ perceptions of the intervention
24 (e.g., acceptability) through short self-report questions in the online survey. Second, the
25 qualitative interviewer is a member of the research team (but not the evaluation team). Care
Page 30 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 30
1 will be taken to ensure that this staff member has no contact with schools prior to the
2 interview visit to minimise bias. However, there remains a risk that demand effects may
3 impact the information that is shared. Third, the process evaluation process (questionnaires
4 and interview) will undoubtedly add to the burden placed on school staff. Given that the FPS
5 is already placing a significant burden on the time of busy school staff, this additional
6 component might contribute to low levels of participation.
7 To the point of burden on schools, one of the strengths of the design is that we have
8 undertaken a pilot phase involving eight intervention schools and have been able to refine our
9 processes (e.g., introduce an incentive) for school staff members to participate. This process
10 evaluation also involves the combination of qualitative and quantitative methods which will
11 be triangulated to provide a coherent and comprehensive picture of the data. The use of the
12 ‘implementation strength’ metric represents a novel approach in this field, borrowed from the
13 low-middle income country implementation science sector. The inclusion of this approach
14 will provide important information to funders and policy makers following on from this trial,
15 indicating the level of implementation support required to prevent mental illness and improve
16 wellbeing of adolescent school students.
17 Notwithstanding the limitations raised above, this process evaluation will contribute
18 to the broader knowledge base and indicate how best to deliver digital mental health
19 prevention programs in school settings.
20
21
22
23
24
25
Page 31 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 31
1 Ethics and Dissemination
2 As noted, all procedures have received ethical approval from relevant committees. Results
3 from this study will be published in an open access peer-reviewed journal.
4
5 Additional Files
6 Additional File 1 (Semi-Structured Interview Guides).docx
7 List of Abbreviations
8 FPS = Future Proofing Study
9 cRCT = Cluster Randomised Controlled Trial
10 CFIR = Consolidated Framework for Implementation Research
11 RE-AIM = Reach, Effectiveness, Adoption, Implementation, Maintenance framework
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Page 32 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 32
1 Declarations
2 Trial status
3 Recruitment for the trial is underway. Data collection commences in October 2020 (delayed
4 from April 2020 due to COVID-19).
5
6 Ethical approval and consent to participate
7 This study has ethical approval from the University of New South Wales Human Research
8 Ethics Committee (HC180836; 21st January 2019) and NSW Government State Education
9 Research Applications Process Approval (SERAP 2019201; 19th August 2019). All schools,
10 staff, parents and participants must provide active consent to participate.
11
12 Consent for publication
13 Not applicable.
14
15 Availability of data and materials
16 Not applicable.
17
18 Competing interests
19 The authors declare that they have no competing interests.
20
21 Funding
22 Funding for this project came from an NSW Health Early-Mid Career Fellowship awarded to
23 Aliza Werner-Seidler, and a Black Dog Institute Post-Doctoral Fellowship awarded to Joanne
24 Beames, secured by Helen Christensen. ALC is supported by NHMRC fellowships 1122544
25 and 1173146. PJB is supported by NHMRC Fellowship 1158707. Funding for the
Page 33 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 33
1 randomised controlled trial within which this process evaluation is embedded came from an
2 NHMRC Project Grant Awarded to Helen Christensen GNT1120646. The funding bodies had
3 no role in any aspect of the study design or this manuscript.
4
5 Author contributions
6 HC and AWS conceived of the study and secured the funding. AWS and JRB led the design
7 of the process evaluation, with input from all authors (including KH, IZ, PJB), and expert
8 guidance from RL and KB. AWS drafted the manuscript, with assistance from JRB. ALC,
9 MT, KM, RL, KB and HC have a continuing role in monitoring the conduct and outcomes of
10 the process evaluation. All named authors contributed substantially to the approved final
11 manuscript.
12
13 Acknowledgments
14 We would like to thank Professor Melanie Barwick for her advice on this manuscript,
15 together with all the facilitators from the Training Institute for Dissemination and
16 Implementation Research in Health (TIDIRH; Australia 2020) for their comments and
17 feedback on this project.
18
19 Author information
20 Not applicable.
Page 34 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 34
1 References
2 1. Oakley A, Strange V, Bonell C, Allen E, Stephenson J, Team RS. Process evaluation
3 in randomised controlled trials of complex interventions. BMJ. 2006;332:413.
4 2. Greenhalgh T, Papoutsi C. Studying complexity in health services research:
5 Desperately seeking an overdue paradigm shift. BMC Med. 2018;16:95.
6 3. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process
7 evaluation of complex interventions: Medical research council guidance. BMJ.
8 2015;350:h1258.
9 4. Grant A, Treweek S, Dreischulte T, Foy R, Guthrie B. Process evaluations for cluster-
10 randomised trials of complex interventions: A proposed framework for design and reporting.
11 Trials. 2013;14:15.
12 5. Werner-Seidler A, Huckvale K, Larsen ME, Calear AL, Maston K, Johnston L, et al.
13 A trial protocol for the effectiveness of digital interventions for preventing depression in
14 adolescents: The future proofing study. Trials. 2020;21:2.
15 6. Tal A, Torous J. The digital mental health revolution: Opportunities and risks.
16 Psychiatric Rehabilitation Journal. 2017;40:263.
17 7. Sawyer SM, Azzopardi PS, Wickremarathne D, Patton GC. The age of adolescence.
18 Lancet Child Adolesc Health. 2018;2:223.
19 8. Andrews G, Issakidis C, Sanderson K, Corry J, Lapsley H. Utilising survey data to
20 inform public policy: Comparison of the cost-effectiveness of treatment of ten mental
21 disorders. British Journal of Psychiatry. 2004;184:526.
22 9. Lawrence D, Johnson S, Hafekost J, de Hann KB, Sawyer M, Ainley J, et al. The
23 mental health of children and adolescents: Report on the second australian child and
24 adolescent survey of mental health and wellbeing. 2015.
Page 35 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 35
1 10. Lee FS, Heimer H, Giedd JN, Lein ES, Šestan N, Weinberger DR, et al. Adolescent
2 mental health—opportunity and obligation. Science. 2014;346:547.
3 11. Werner-Seidler A, Perry Y, Calear AL, Newby JM, Christensen H. School-based
4 depression and anxiety prevention programs for young people: A systematic review and
5 meta-analysis. Clin Psychol Rev. 2017;51:30.
6 12. van Zoonen K, Buntrock C, Ebert DD, Smit F, Reynolds CF, Beekman AT, et al.
7 Preventing the onset of major depressive disorder: A meta-analytic review of psychological
8 interventions. Int J Epidemiol. 2014;43:318.
9 13. Gulliver A, Griffiths KM, Christensen H. Perceived barriers and facilitators to mental
10 health help-seeking in young people: A systematic review. BMC Psychiatry. 2010;10:113.
11 14. Donker T, Blankers M, Hedman E, Ljotsson B, Petrie K, Christensen H. Economic
12 evaluations of internet interventions for mental health: A systematic review. Psychological
13 Medicine. 2015;45:3357.
14 15. Essau CA. Frequency and patterns of mental health services utilization among
15 adolescents with anxiety and depressive disorders. Depress Anxiety. 2005;22:130.
16 16. Rickwood DJ, Deane FP, Wilson CJ. When and how do young people seek
17 professional help for mental health problems. Med J Aust. 2007;187:S35.
18 17. Kösters MP, Chinapaw MJM, Zwaanswijk M, van der Wal MF, Utens EMWJ,
19 Koot HM. FRIENDS for Life: Implementation of an indicated prevention program
20 targeting childhood anxiety and depression in a naturalistic setting. Mental Health &
21 Prevention. 2017:6,44.
22 18. Donker T, Blankers M, Hedman E, Ljótsson B, Petrie K, Christensen H. Economic
23 evaluations of internet interventions for mental health: A systematic review. Psychological
24 Medicine. 2015;45:3357.
Page 36 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 36
1 19. Lee YY, Barendregt JJ, Stockings EA, Ferrari AJ, Whiteford HA, Patton GA, et al.
2 The population cost-effectiveness of delivering universal and indicated school-based
3 interventions to prevent the onset of major depression among youth in australia.
4 Epidemiology and Psychiatric Sciences. 2017;26:545.
5 20. Merry SN, Stasiak K, Shepherd M, Frampton C, Fleming T, Lucassen MF. The
6 effectiveness of sparx, a computerised self help intervention for adolescents seeking help for
7 depression: Randomised controlled non-inferiority trial. BMJ. 2012;344:e2598.
8 21. Perry Y, Werner-Seidler A, Calear A, Mackinnon A, King C, Scott J, et al. Preventing
9 depression in final year secondary students: School-based randomized controlled trial.
10 Journal of Medical Internet Research. 2017;19:e369.
11 22. Higgins MC, Weiner J, Young L. Implementation teams: A new lever for
12 organizational change. Journal of Organizational Behavior. 2012;33:366.
13 23. Clarke DJ, Godfrey M, Hawkins R, Sadler E, Harding G, Forster A, et al.
14 Implementing a training intervention to support caregivers after stroke: A process evaluation
15 examining the initiation and embedding of programme change. Implement Sci. 2013;8:96.
16 24. Curran GM, Bauer M, Mittman B, Pyne JM, Stetler C. Effectiveness-implementation
17 hybrid designs: Combining elements of clinical effectiveness and implementation research to
18 enhance public health impact. Med Care. 2012;50:217.
19 25. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health
20 promotion interventions: The re-aim framework. American Journal of Public Health.
21 1999;89:1322.
22 26. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering
23 implementation of health services research findings into practice: A consolidated framework
24 for advancing implementation science. Implementation Science. 2009;4:50.
Page 37 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 37
1 27. Nilsen P. Making sense of implementation theories, models and frameworks.
2 Implementation Science. 2015;10:53.
3 28. Kirk MA, Kelley C, Yankey N, Birken SA, Abadie B, Damschroder L. A systematic
4 review of the use of the consolidated framework for implementation research. Implement Sci.
5 2016;11:72.
6 29. Leeman J, Wiecha JL, Vu M, Blitstein JL, Allgood S, Lee S, et al. School health
7 implementation tools: A mixed methods evaluation of factors influencing their use.
8 Implement Sci. 2018;13:48.
9 30. Murray E, Treweek S, Pope C, MacFarlane A, Ballini L, Dowrick C, et al.
10 Normalisation process theory: A framework for developing, evaluating and implementing
11 complex interventions. BMC Medicine. 2010;8:63.
12 31. Glasgow RE, Harden SM, Gaglio B, Rabin B, Smith ML, Porter GC, et al. Re-aim
13 planning and evaluation framework: Adapting to new science and practice with a 20-year
14 review. Frontiers in Public Health. 2019;7.
15 32. Proctor EK, Landsverk J, Aarons G, Chambers D, Glisson C, Mittman B.
16 Implementation research in mental health services: An emerging science with conceptual,
17 methodological, and training challenges. Administration and Policy in Mental Health.
18 2009;36:24.
19 33. Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, et al. Outcomes
20 for implementation research: Conceptual distinctions, measurement challenges, and research
21 agenda. Administration and Policy in Mental Health and Mental Health Services Research.
22 2011;38:65.
23 34. Calear AL, Christensen H, Mackinnon A, Griffiths KM, O'Kearney R. The
24 youthmood project: A cluster randomized controlled trial of an online cognitive behavioral
25 program with adolescents. Journal of Consulting and Clinical Psychology. 2009;77:1021.
Page 38 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 38
1 35. Dix KL, Green MJ, Tzoumakis S, Dean K, Harris F, Carr VJ, et al. The survey of
2 school promotion of emotional and social health (sspesh): A brief measure of the
3 implementation of whole-school mental health promotion. School Mental Health.
4 2019;11:294.
5 36. Fernandez ME, Walker TJ, Weiner BJ, Calo WA, Liang S, Risendal B, et al.
6 Developing measures to assess constructs from the inner setting domain of the consolidated
7 framework for implementation research. Implement Sci. 2018;13:52.
8 37. Deschesnes M, Trudeau F, Kébé M. Psychometric properties of a scale focusing on
9 perceived attributes of a health promoting school approach. Can J Public Health.
10 2009;100:389.
11 38. Aarons GA, Ehrhart MG, Farahnak LR. The implementation leadership scale (ils):
12 Development of a brief measure of unit level implementation leadership. Implement Sci.
13 2014;9:45.
14 39. Finch TL, Girling M, May CR, Mair FS, Murray E, Treweek S, et al. Nomad:
15 Implementation measure based on normalization process theory 2015 [Available from:
16 http://www.normalizationprocess.org.
17 40. Finch TL, Girling M, May CR, Mair FS, Murray E, Treweek S, et al. Improving the
18 normalization of complex interventions: Part 2 - validation of the nomad instrument for
19 assessing implementation work based on normalization process theory (npt). BMC Med Res
20 Methodol. 2018;18:135.
21 41. Weiner BJ, Lewis CC, Stanick C, Powell BJ, Dorsey CN, Clary AS, et al.
22 Psychometric assessment of three newly developed implementation outcome measures.
23 Implement Sci. 2017;12:108.
Page 39 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 39
1 42. Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, et al. Outcomes
2 for implementation research: Conceptual distinctions, measurement challenges, and research
3 agenda. Adm Policy Ment Health. 2011;38:65.
4 43. Salyers MP, Rollins AL, McGuire AB, Gearhart T. Barriers and facilitators in
5 implementing illness management and recovery for consumers with severe mental illness:
6 Trainee perspectives. Adm Policy Ment Health. 2009;36:102.
7 44. Perry Y, Calear AL, Mackinnon A, Batterham PJ, Licinio J, King C, et al. Trial for
8 the prevention of depression (tripod) in final-year secondary students: Study protocol for a
9 cluster randomised controlled trial. Trials. 2015;16:451.
10 45. Punukollu M, Burns C, Marques M. Effectiveness of a pilot school-based intervention
11 on improving scottish students’ mental health: A mixed methods evaluation. J Youth
12 Adolesc. 2019:1.
13 46. Outhwaite LA, Gulliford A, Pitchford NJ. A new methodological approach for
14 evaluating the impact of educational intervention implementation on learning outcomes. Int J
15 Res Meth Educ. 2019:1.
16 47. Poland BD. Transcription quality. Handbook of interview research: Context and
17 method. 2002:629.
18 48. Poland BD. Transcription quality as an aspect of rigor in qualitative research.
19 Qualitative inquiry. 1995;1:290.
20 49. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol.
21 2006;3:77.
22 50. Braun V, Clarke V. Reflecting on reflexive thematic analysis. Qual Res Sport Exerc
23 Health. 2019;11:589.
24 51. Noble H, Smith J. Issues of validity and reliability in qualitative research. Evid-Based
25 Nurs. 2015;18:34.
Page 40 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 40
1 52. Tashakkori A, Teddlie C. Handbook of mixed methods in social & behavioral
2 research. Thousand Oaks, CA: Sage; 2003.
3 53. Hargreaves JR, Goodman C, Davey C, Willey BA, Avan BI, Schellenberg JR.
4 Measuring implementation strength: Lessons from the evaluation of public health strategies
5 in low- and middle-income settings. Health Policy Plan. 2016;31:860.
6 54. Victora CG, Black RE, Boerma JT, Bryce J. Measuring impact in the millennium
7 development goal era and beyond: A new approach to large-scale effectiveness evaluations.
8 Lancet. 2011;377:85.
9 55. Ebert DD, Cuijpers P. It is time to invest in the prevention of depression. JAMA Netw
10 Open. 2018;1:e180335.
Page 41 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
School Contextual CharacteristicsSociodemographic factors (school size,
socio economic status, area)
Intervention Characteristics (SPARX)
Standardised: Delivered by app over 6 weeks
Technical issues, evidence strength and quality, relative advantage
Student-Level Outcomes
Primary outcome: Lower levels of depression
Secondary outcomes: Lower levels of anxiety, distress,
insomnia
Additional outcomes:Improved quality of life, wellbeing
Lower levels of suicidality, self-harm
Improved academic performance on standardised tests
Lower levels of health service use
Implementation Outcomes
School-level fidelity to the implementation strategy
Reach (e.g., intervention received by students, including number of modules completed and time spent using SPARX)
School-level uptake of the FPS
Staff and student acceptability/appropriateness
Implementation Strategy
App download in school time, with support from
research staff
Intervention completion, with support from school
staff
Reminders and support provided by school staff to
complete intervention
Therapeutic Components
PsychoeducationCognitive restructuringBehavioural activationRelaxation strategies
Practical problem solvingEmotion regulation and management
Individual CharacteristicsSchool staff (e.g., skills, motivation, self-
efficacy, buy in, knowledge about the intervention, leadership)
Study facilitators (e.g., skills, confidence)
School Organisational CharacteristicsSchool culture, buy in and support for the
intervention and study, support from leadership, ability of school to support delivery (including
technological infrastructure, counsellor availability)
Individual CharacteristicsStudents (e.g., mental health
history, intervention expectations)
Page 42 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Research Team
Facilitators
Teachers Counsellors
Year 8 Students
Training
Other staff
Delivery Support
Direct trainingSupport to access interventionDelivery support
School Implementation Team
Page 43 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 1
Additional File 1
Semi-Structured Interview Guides
School Staff INTRODUCTION: Thank you for agreeing to take part in this interview. I appreciate your time. I am interested in hearing your thoughts and experiences of the Future Proofing program in your school. For this interview, the FP program includes the intervention app (SPARX) and everything else that was done to support the program in your school. I hope to use this information to work out what works in different schools and how to best provide support. I will ask you a few questions about your experiences. This is chance for you to reflect and offer feedback on how we could improve. The interview will last up to one hour. It will be recorded, and your responses will be transcribed to ensure we accurately reflect your words. All personal information will be removed so that your answers are non-identifiable and confidential. Nothing you personally say will be shared with anyone on the FP team that you have been liaising with or with other staff at your school. You can choose to not answer any questions. Do you have any questions before we get started? General Opening Questions:
1. How did you first get involved in teaching/counselling/psychology?
2. Could you tell me about the core values and ethos of your school?
3. To the best of your knowledge, what mental health programs or initiatives has your
school implemented to care for students besides the FP program?
Specific Questions:
4. Overall, how would you describe your experience with the FP program in your school?
5. What was your role in delivering or supporting the FP program in your school?
a. Leader/driver, decision-maker, consent/recruitment, peripheral support
CFIR DOMAIN: INDIVIDUAL CHARACTERISTICS Personal attributes (e.g., motivations, expectations, capacity, values)
6. Why were you motivated to take on this role?
a. How did you come on board with the program?
7. Why do you think your school decided to support this program?
a. To what extent do you think there was a need for a mental health program like FP in
you school (Tension for change)?
8. What were your expectations about the FP program prior to coming on board?
a. How were your expectations met (or not met)? What went differently to what
you expected? What surprised you?
b. What were your expectations about the apps?
Knowledge and beliefs about the intervention 9. How much do you know about SPARX, the intervention? E.g., principles, rationale,
content?
Page 44 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 2
Self-efficacy 10. How confident were you in your ability to lead/support/make decisions about the FP
program? a. What personal qualities do you think helped you to do that?
CFIR DOMAIN: OUTER SETTING Relative Advantage
11. How did SPARX compare to other mental health programs that your school has supported or delivered?
CFIR DOMAIN: INNER SETTING School contextual factors – Barriers and Facilitators
12. What factors influenced your ability to deliver/support the FP program in your school? a. What else helped or hindered your ability? b. Enquire into school-specific factors and program-specific factors e.g., What
parts of the program were easy for you to do or support (e.g., the consent process, organising rooms, weekly team meetings)? OR What aspects of your school made it difficult for you to support the program?
c. What factors got in the way of students completing the mental health app/s? d. Were you (or others) able to overcome any barriers? How did you manage to do
this? School contextual factors – Leadership
13. Tell me about the involvement of your school executive in the FP program. a. What role did this person have (e.g., principal/vice principal)? Did you meet
regularly with this person or provide updates to them? 14. How do you think the school executive saw this initiative?
Readiness for implementation
15. How would you describe the readiness of your school to take on the FP program? a. What was the preparation phase like for you/other key staff?
IMPLEMENTATION METRICS Appropriateness and Acceptability
16. How does the FP program align with your school’s approach toward student mental health?
17. How suited is the FP program to address depression prevention in your school? 18. How well does an app delivering therapy work to prevent depression in your students? 19. How do you feel about using technology to deliver psychological interventions in your
school? a. What worked well about this and what did not? b. Do you think SPARX met the student’s mental health needs? In what way?
20. What were the by-products of the intervention that you didn’t expect? a. What parts of the program do/do not suit your school?
**ADDITIONAL QUESTIONS FOR SCHOOL COUNSELLORS ONLY 21. What was your experience with responding to students identified as needing follow-up
during the FP program? a. Was there anything that could have been done to make this experience better for
you and the student? 22. What do you think of the FP risk protocol?
Page 45 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 3
KEY LEARNING AND FUTURE IMPLEMENTATION 23. What parts of the FP program, if any, could be made more attractive to your school to
increase uptake or support? a. In what ways could we make these parts more attractive?
24. To what extent do you think that the FP program could be integrated into your school’s existing mental health strategy?
25. What advice would you give another school contemplating taking on the FP program? 26. What’s been your major learning or insight from your school’s involvement in the FP
program? Closing
27. Is there anything else you would like to add?
Page 46 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 4
Volunteer Research Assistants (Study Facilitators) INTRODUCTION: Thank you for offering to take part in this interview. I appreciate your time. I am interested in your thoughts about and experiences with supporting the FP program as a Research Assistant Volunteer. I hope to use this information to work out what works in different schools and how we can help volunteers to best provide support. I will ask you a few questions about your experiences. This is chance for you to reflect and offer feedback on how we could improve. The interview will last up to one hour. It will be recorded and your responses will be transcribed to ensure we accurately reflect your words. All personal information will be removed so that your answers are non-identifiable and confidential. Nothing you personally say will be shared with the volunteer team or the Black Dog Institute more generally, so it will not affect your relationship with them. You can choose to not answer any questions. Do you have any questions before we get started? General Opening Questions:
1. What schools have you visited so far? 2. Roughly how many school visits have you attended?
Specific Questions: CFIR DOMAIN: INDIVIDUAL CHARACTERISTICS Personal attributes (e.g., motivation and expectations)
3. Why were you motivated you to volunteer? 4. What were your expectations coming on board with the FP program?
a. How were your expectations met (or not met)? b. What went differently to what you expected? What surprised you?
Perceptions of Training
5. What are your overall impressions of the training you received? a. How helpful and appropriate was the training? b. What parts were the most useful? c. What was missing from the training?
CFIR DOMAIN: INNER SETTING School visits – Barriers/Facilitators
6. What was your experience with completing the school visits? 7. Did anything unexpected happen during the school visits?
a. If so, what? b. How did you manage this? c. How well do you think you managed this?
8. In general, how was the FP program received by school staff during visits? 9. To what extent was it useful having a volunteer partner/s at the visits?
a. Were there any other benefits/disadvantages? School contextual factors - Barriers/Facilitators
10. What factors influenced your ability to assist students and teachers at the school-visits? a. Were these common across all the visits or specific to a particular visit? b. What aspects of the school made it difficult to support the program?
Page 47 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 5
c. What else helped or hindered your ability? (e.g., research platform/app specific factors)
d. How did you overcome these barriers? e. Were there any noticeable differences between different schools? If so, what were
they? KEY LEARNING AND FUTURE IMPLEMENTATION
11. What advice would you give a school contemplating taking on the FP program? 12. What advice would you give to others who are thinking about volunteering? 13. What parts of the FP program could be made more attractive to volunteers to increase
uptake or support? 14. What’s been your major learning or insight from taking part in the volunteer program?
Closing
15. Is there anything else you would like to add?
Page 48 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
1
Standards for Reporting Implementation Studies: the StaRI checklist for completion
Checklist itemReported on page # Implementation Strategy
Reported on page # Intervention
“Implementation strategy” refers to how the intervention was implemented
“Intervention” refers to the healthcare or public health intervention that is being implemented.
Title and abstractTitle 1 1-2 Identification as an implementation study, and description of the methodology in the title and/or keywords
Abstract 2 2-3 Identification as an implementation study, including a description of the implementation strategy to be tested, the evidence-based intervention being implemented, and defining the key implementation and health outcomes.
IntroductionIntroduction 3 5 Description of the problem, challenge or deficiency in healthcare or public health that the intervention being implemented aims
to address.Rationale 4 9-11 The scientific background and rationale for the
implementation strategy (including any underpinning theory/framework/model, how it is expected to achieve
its effects and any pilot work).
8 The scientific background and rationale for the intervention being implemented (including evidence
about its effectiveness and how it is expected to achieve its effects).
Aims and objectives
5 9 The aims of the study, differentiating between implementation objectives and any intervention objectives.
Methods: descriptionDesign 6 11 The design and key features of the evaluation, (cross referencing to any appropriate methodology reporting standards) and any
changes to study protocol, with reasonsContext 7 11-13 The context in which the intervention was implemented. (Consider social, economic, policy, healthcare, organisational barriers
and facilitators that might influence implementation elsewhere).Targeted
‘sites’8 8 The characteristics of the targeted ‘site(s)’ (e.g
locations/personnel/resources etc.) for implementation and any eligibility criteria.
8 The population targeted by the intervention and any eligibility criteria.
Description 9 17 A description of the implementation strategy 8 A description of the intervention
Page 49 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
2
Sub-groups 10 N/A Any sub-groups recruited for additional research tasks, and/or nested studies are described
Methods: evaluationOutcomes 11 18 Defined pre-specified primary and other outcome(s) of
the implementation strategy, and how they were assessed. Document any pre-determined targets
N/A Defined pre-specified primary and other outcome(s) of the intervention (if assessed), and how they were assessed. Document any pre-determined targets
Process evaluation
12 14 Process evaluation objectives and outcomes related to the mechanism by which the strategy is expected to work
Economic evaluation
13 N/A Methods for resource use, costs, economic outcomes and analysis for the implementation strategy
N/A Methods for resource use, costs, economic outcomes and analysis for the intervention
Sample size 14 7 Rationale for sample sizes (including sample size calculations, budgetary constraints, practical considerations, data saturation, as appropriate)
Analysis 15 26-28 Methods of analysis (with reasons for that choice)
Sub-group analyses
16 26-28 Any a priori sub-group analyses (e.g. between different sites in a multicentre study, different clinical or demographic populations), and sub-groups recruited to specific nested research tasks
ResultsCharacteristics 17 N/A Proportion recruited and characteristics of the recipient
population for the implementation strategyN/A Proportion recruited and characteristics (if appropriate)
of the recipient population for the interventionOutcomes 18 N/A Primary and other outcome(s) of the implementation
strategyN/A Primary and other outcome(s) of the Intervention (if
assessed)Process
outcomes19 N/A Process data related to the implementation strategy mapped to the mechanism by which the strategy is expected to work
Economic evaluation
20 N/A Resource use, costs, economic outcomes and analysis for the implementation strategy
N/A Resource use, costs, economic outcomes and analysis for the intervention
Sub-group analyses
21 N/A Representativeness and outcomes of subgroups including those recruited to specific research tasks
Page 50 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
3
Fidelity/ adaptation
22 N/A Fidelity to implementation strategy as planned and adaptation to suit context and preferences
N/A Fidelity to delivering the core components of intervention (where measured)
Contextual changes
23 N/A Contextual changes (if any) which may have affected outcomes
Harms 24 N/A All important harms or unintended effects in each group
DiscussionStructured discussion
25 28-30 Summary of findings, strengths and limitations, comparisons with other studies, conclusions and implications
Implications 26 N/A Discussion of policy, practice and/or research implications of the implementation strategy (specifically
including scalability)
N/A Discussion of policy, practice and/or research implications of the intervention (specifically including
sustainability)General
Statements 27 31-33 Include statement(s) on regulatory approvals (including, as appropriate, ethical approval, confidential use of routine data, governance approval), trial/study registration (availability of protocol), funding and conflicts of interest
Page 51 of 50
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review onlyProtocol for the Process Evaluation of a Complex
Intervention Delivered in Schools to Prevent Adolescent Depression: The Future Proofing Study
Journal: BMJ Open
Manuscript ID bmjopen-2020-042133.R2
Article Type: Protocol
Date Submitted by the Author: 26-Nov-2020
Complete List of Authors: Beames, Joanne; University of New South Wales, Black Dog InstituteLingam, Raghu; University of New South Wales, School of Women’s and Children’s HealthBoydell, Katherine; University of New South Wales, Black Dog InstituteCalear, Alison L.; Australian National University, Centre for Mental Health ResearchTorok, Michelle; University of New South Wales, Black Dog InstituteMaston, Kate; University of New South Wales, Black Dog InstituteZbukvic, Isabel; Orygen The National Centre of Excellence in Youth Mental HealthHuckvale, Kit; University of New South Wales, Black Dog InstituteBatterham, Philip; Australian National University, Centre for Mental Health ResearchChristensen, Helen; University of New South Wales, Black Dog InstituteWerner-Seidler, Aliza ; University of New South Wales, Black Dog Institute
<b>Primary Subject Heading</b>: Public health
Secondary Subject Heading: Mental health
Keywords: Child & adolescent psychiatry < PSYCHIATRY, Depression & mood disorders < PSYCHIATRY, PREVENTIVE MEDICINE
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on N
ovember 15, 2021 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2020-042133 on 12 January 2021. D
ownloaded from
For peer review onlyI, the Submitting Author has the right to grant and does grant on behalf of all authors of the Work (as defined in the below author licence), an exclusive licence and/or a non-exclusive licence for contributions from authors who are: i) UK Crown employees; ii) where BMJ has agreed a CC-BY licence shall apply, and/or iii) in accordance with the terms applicable for US Federal Government officers or employees acting as part of their official duties; on a worldwide, perpetual, irrevocable, royalty-free basis to BMJ Publishing Group Ltd (“BMJ”) its licensees and where the relevant Journal is co-owned by BMJ to the co-owners of the Journal, to publish the Work in this journal and any other BMJ products and to exploit all rights, as set out in our licence.
The Submitting Author accepts and understands that any supply made under these terms is made by BMJ to the Submitting Author unless you are acting as an employee on behalf of your employer or a postgraduate student of an affiliated institution which is paying any applicable article publishing charge (“APC”) for Open Access articles. Where the Submitting Author wishes to make the Work available on an Open Access basis (and intends to pay the relevant APC), the terms of reuse of such Open Access shall be governed by a Creative Commons licence – details of these licences and which Creative Commons licence will apply to this Work are set out in our licence referred to above.
Other than as permitted in any relevant BMJ Author’s Self Archiving Policies, I confirm this Work has not been accepted for publication elsewhere, is not being considered for publication elsewhere and does not duplicate material already published. I confirm all authors consent to publication of this Work and authorise the granting of this licence.
Page 1 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 1
1 Protocol for the Process Evaluation of a Complex Intervention Delivered in Schools to
2 Prevent Adolescent Depression: The Future Proofing Study
3
4 Joanne R. Beames1 [email protected]
5 Raghu Lingam2 [email protected]
6 Katherine Boydell1 [email protected]
7 Alison L. Calear3 [email protected]
8 Michelle Torok1 [email protected]
9 Kate Maston1 [email protected]
10 Isabel Zbukvic4 [email protected]
11 Kit Huckvale1 [email protected]
12 Philip J. Batterham3 [email protected]
13 Helen Christensen1 [email protected]
14 Aliza Werner-Seidler1* [email protected]
15 Affiliations:
16 1 Black Dog Institute, University of New South Wales, Sydney, NSW, Australia.
17 2 Population Child Health Clinical Research Group, School of Women’s and Children’s
18 Health, University of New South Wales, NSW, Australia.
19 3 Centre for Mental Health Research, The Australian National University, Canberra, ACT,
20 Australia.
21 4 Orygen, Parkville, Victoria, Australia; Centre for Youth Mental Health, The University of
22 Melbourne, Parkville, Victoria, Australia
23 *Corresponding author:
24 Dr Aliza Werner-Seidler, Black Dog Institute, University of New South Wales, Sydney,
25 NSW, Australia. [email protected]; +61 2 9382 3808
Page 2 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 2
1 Abstract
2 Introduction: Process evaluations provide insight into how interventions are delivered across
3 varying contexts and why interventions work in some contexts and not in others. This
4 manuscript outlines the protocol for a process evaluation embedded in a cluster randomised
5 trial of a digital depression prevention intervention delivered to secondary school students
6 (The Future Proofing Study). The purpose is to describe the methods that will be used to
7 capture process evaluation data within this trial.
8 Methods and Analysis: Using a hybrid type 1 design, a mixed methods approach will be
9 used with data collected in the intervention arm of the Future Proofing Study. Data collection
10 methods will include semi-structured interviews with school staff and study facilitators,
11 automatically collected intervention usage data, and participant questionnaires (completed by
12 school staff, school counsellors, study facilitators, and students). Information will be
13 collected about: i) how the intervention was implemented in schools, including fidelity; ii)
14 school contextual factors and their association with intervention reach, uptake and
15 acceptability; iii) how school staff, study facilitators, and students and responded to
16 delivering or completing the intervention. How these factors relate to trial effectiveness
17 outcomes will also be assessed. Overall synthesis of the data will provide school cluster-level
18 and individual-level process outcomes.
19 Ethics and Dissemination: Ethics approval was obtained from the University of New South
20 Wales Human Research Ethics Committee (HC180836; 21st January 2019) and the NSW
21 Government State Education Research Applications Process (SERAP 2019201; 19th August
22 2019). Results will be submitted for publication in peer reviewed journals and discussed at
23 conferences. Our process evaluation will contextualise the trial findings with respect to how
24 the intervention may have worked in some schools but not in others. This evaluation will
25 inform the development of a model for rolling out digital interventions for the prevention of
Page 3 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 3
1 mental illness in schools.
2 Trial Registration: The randomised controlled trial has been prospectively registered:
3 ACTRN12619000855123. Registered 31 May 2019,
4 https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=377664&isReview=true
5 Keywords: Process evaluation, complex intervention, mixed methods, adolescent mental
6 health, depression, prevention, randomised controlled trial, implementation.
Page 4 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 4
1 Strengths and Limitations of this Study
2 The methodology of this embedded process evaluation is underpinned by
3 implementation frameworks and logic modelling.
4 Flexible and pragmatic quantitative and qualitative data collection methods will be
5 used to balance research rigor with feasibility within the school delivery context.
6 Process data from a range of key stakeholders will be collected, including school
7 staff, study facilitators, and students.
8 To minimise burden on schools, fidelity data from teachers and in-depth qualitative
9 data from students will not be collected.
10 The methodology and study processes were pilot tested to ensure appropriateness
11 within the school context.
Page 5 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 5
1 Introduction
2 Embedded in randomised controlled trials, process evaluations provide insight into
3 reasons why interventions work in some contexts and not in others. These process evaluations
4 can help to demystify the ‘black box’ of complex intervention trials by taking into account
5 contextual factors, differences in ways the intervention is delivered, and adaptations made for
6 intervention delivery into a particular system (1, 2). Contextual factors typically include
7 features of the organisation or broader environment that influence the delivery of the
8 intervention (e.g., leadership, engagement, culture, political landscape). Considering the
9 contribution of contextual factors is necessary to aid interpretation of trial outcomes,
10 maximise the knowledge gained from trials, identify optimal delivery processes across
11 different settings, and inform broader dissemination efforts. In recognition of this need for
12 process evaluations, the UK’s Medical Research Council (MRC) set out a framework that
13 emphasises the value of these evaluations in order to capture both contextual and
14 implementation factors associated with complex interventions (3).
15 In line with best-practice recommendations that study protocols that pre-specify
16 methods and approaches should be published to maintain research integrity (4), we describe a
17 protocol for a mixed methods process evaluation embedded within a cluster randomised
18 controlled trial (cRCT) known as the Future Proofing Study (FPS). The FPS is a large school-
19 based trial examining whether depression can be prevented using cognitive behaviour therapy
20 (CBT) delivered by smartphone application (5).
21 There has been an increase in the availability of digital mental health programs which
22 show promise in addressing the significant disease burden associated with depression (6).
23 Depression often first emerges during adolescence (7) and treatment alone cannot adequately
24 reduce this burden (8). Accumulating evidence indicates that early adolescence is the ideal
25 developmental window during which to intervene because it captures young people before
Page 6 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 6
1 the incidence of depression increases exponentially around the age of 16-17 years (9, 10).
2 Therefore, increased efforts are being directed toward prevention approaches that are
3 developed specifically for delivery among adolescents.
4 While there is evidence supporting the value and effectiveness of prevention
5 programs in schools (11, 12), their uptake has been significantly limited by low levels of
6 help-seeking (13) as well as practical constraints in cost and scalability. A lack of
7 understanding about how to deliver these interventions sustainably at scale, and engage those
8 who stand to benefit, have hampered the translation of prevention approaches into the
9 community.
10 Two ways to overcome the barriers of cost and scalability of implementing adolescent
11 depression interventions are:
12 Delivering universal prevention programs in schools, and
13 Using technology to deliver programs, which tends to be lower cost than traditional
14 face-to-face methods (14).
15 Working with schools to deliver universal prevention programs, to every student,
16 regardless of their risk, circumstance, or symptom profile, not only dramatically increases the
17 potential reach of interventions, it also means that these programs can eventually be
18 integrated into the curriculum and delivered to every student, making this approach a
19 sustainable implementation strategy for widescale delivery and dissemination. Additionally,
20 working with schools to deliver such programs reduces the need for young people to actively
21 seek professional help. This is critically important - despite having a significant need, fewer
22 than 80% of young people with mental illness seek help and receive the services they need
23 (15, 16). The scaffolding provided by schools can be leveraged to deliver prevention
24 programs to all students, which fits with a trend for schools to be designated the first point of
25 contact to support youth mental health problems when they first emerge (9). Process
Page 7 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 7
1 evaluations of face-to-face mental health programs have been documented, some of which
2 have been delivered in school settings (e.g., 17). However, we were unable to find any
3 process evaluation descriptions in the literature that evaluate digital mental health programs
4 in school settings. Conducting process evaluations specific to digital methods of delivery are
5 important because the contextual barriers and facilitators, as well as concerns around fidelity
6 of the intervention, are likely to have unique characteristics. For example, facilitator or
7 training in intervention delivery will be different in supporting the use of an automated
8 program relative to a face-to-face program. This process evaluation is an initial step towards
9 addressing that gap.
10 Technology offers a promising way to deliver mental health interventions to the
11 community. Digital mental health interventions offer two key advantages over face-to-face
12 approaches – first, they are cheaper to access and more cost effective (18, 19), and second,
13 they can be used to reach people across vast geographic areas. The latter is particularly
14 important given Australia’s geography, where remoteness and low population density meant
15 that people in regional and rural areas often do not have the same level of access to mental
16 health services as people in metropolitan areas. With more young people than ever using
17 smartphones, mental health intervention delivered online or through applications represents
18 an exciting avenue for reaching adolescents.
19 The Future Proofing Study (FPS)
20 The trial that is the subject of this process evaluation is the FPS. The FPS addresses
21 barriers of reach, cost, and scalability by delivering a depression prevention program via a
22 smartphone app to Year 8 secondary school students aged 12-13 years. This study is being
23 conducted with approximately 200 Australian schools (up to 10,000 participants), of which
24 half will be allocated to the intervention condition. The primary outcome is symptoms of
25 depression. Secondary outcomes include anxiety, psychological distress and insomnia, among
Page 8 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 8
1 others. Symptom outcomes will be assessed at baseline, post-intervention, six months
2 (primary outcome only), and then annually for five years. The primary endpoint is 12 months
3 following baseline. More details are available from the Australian New Zealand Clinical
4 Trials Registry (ANZCTRN12619000855123) and protocol paper (5).
5 The intervention being delivered is known as SPARX, a CBT-based program
6 incorporating gamification principles. The development and initial evaluation of SPARX has
7 been documented in detail previously (20). SPARX has been tested in an Australian sample
8 of secondary students (delivered via computer) and shown to prevent depression in the lead
9 up to final school exams (21). The gamified intervention teaches young people about the
10 relationship between thoughts, feelings, and behaviour. Skills learnt through SPARX include
11 emotion identification, emotion regulation, behavioural activation (being active), recognising
12 and challenging unhelpful thoughts, and practical problem solving. SPARX consists of seven
13 20-minute modules. The intervention is fully automated, and the therapeutic components are
14 standardised. See the trial protocol for full intervention details (5).
15 School Engagement and Recruitment
16 The FPS requires multiple levels of approval, beginning with state department and
17 independent school body approval (NSW Department of Education; Catholic school
18 dioceses), followed by individual school engagement. An engagement strategy for this study
19 involves sending electronic communication material to all schools across New South Wales
20 and in other Australian capital cities, targeting school principals and wellbeing staff to invite
21 them to participate in this study opportunity. Schools are invited to submit expressions of
22 interest and are subsequently followed up by the research team over the phone to explain
23 more about the study. The schools for which the FPS is a good fit (as determined by the
24 school) are then signed onto the study, with support from the principal, the school counsellor,
25 and at least one other staff member (typically a teacher). In line with best practice in
Page 9 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 9
1 implementation science (22), this group of 2-3 staff members (typically not including the
2 principal) will form the school-specific ‘study implementation team’. After signing on to the
3 study, several webinars are scheduled throughout the lead up to the study start date, so that
4 school staff and parents can listen to a 15-minute study overview from the trial manager and
5 have their questions answered.
6 Preparation
7 In preparation for in-class assessment sessions, study facilitators (volunteer research
8 assistants) are recruited to support the study. The purpose of these facilitators is to attend
9 schools to introduce the study to students, and ensure the technology is functioning so
10 students can download the SPARX app and complete the baseline (and post-assessment)
11 questionnaires. All facilitators go through an interview and screening process prior to
12 selection, then attend a half day face-to-face training session before supporting the study in
13 schools. This training provides an overview of the study and detailed information about their
14 roles within schools, including a step-by-step guide to running the sessions. Discussion and
15 practice are core components of the training.
16 Aims and Objectives
17 We adapted the MRC framework for complex interventions to focus on effectiveness of
18 an evidence-based intervention within a specific context of delivery (i.e., schools). Overall,
19 the objective of this process evaluation is to understand how SPARX is implemented and
20 delivered in schools, and to identify systematic differences and variation in delivery.
21 Specifically, the aims are:
22 1. To evaluate the reach (including completion), uptake and acceptability of the
23 intervention (school- and student-level).
24 2. To understand the contribution of contextual factors (e.g., characteristics of the
25 outer/inner setting, intervention, individuals) on:
Page 10 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 10
1 o School-level fidelity to the implementation strategy. For example, different
2 schools will likely provide different levels of study support based on available
3 resourcing.
4 o Implementation outcomes (intervention reach, uptake, acceptability), as
5 assessed from the perspectives of school staff, teachers, and students. For
6 example, young people’s openness to receiving mental health material via an
7 app will likely impact intervention acceptability and completion.
8 3. To examine the impact of school-level variation (in implementation fidelity and
9 outcomes) on clinical effectiveness outcomes at the school- and student-level. School-
10 level clinical effectiveness is defined as changes in clinical outcomes (e.g., self-
11 reported depression) for different schools. The differing ways in which schools
12 support and deliver the intervention will inevitably impact its effectiveness, and this
13 evaluation will assess these differences (3, 23). For example, degree of support from
14 senior school leaders who would be expected to understand that participating in study
15 activities is a priority for the school will likely impact effectiveness at the cluster
16 level.
17 This process evaluation has been designed to capture important information from
18 teachers, school staff and students at both the school and individual level, which may
19 ultimately impact the effectiveness of the intervention on clinical mental health outcomes for
20 students. Findings will provide insight into factors which support and/or hinder the
21 implementation of digital universal mental health programs in school settings. Knowledge
22 gained from this process evaluation will help to inform the development of a model and guide
23 for how to best deliver digital mental health programs to young people in schools.
24
25
Page 11 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 11
1 Methods and Analysis
2 Design
3 This study uses a hybrid type 1 approach (24), with a focus on implementation
4 process factors and outcomes in the context of an effectiveness trial. The evaluation is guided
5 by the CFIR and RE-AIM framework (25, 26). In keeping with Nilsen’s categorisation of
6 implementation theories, models, and frameworks, these frameworks help us to understand
7 different parts of the implementation process (27). The CFIR will be used to identify
8 barriers and facilitators to intervention implementation and effectiveness (27). The CFIR is
9 a widely used deterministic theoretical framework which has been applied in multiple settings
10 (28), including schools (e.g., 29). The CFIR identifies five major domains, including (i) the
11 outer setting, which includes the social, political and economic context that the organisation
12 in which implementation is occurring exists; (ii) the inner setting, which includes features and
13 characteristics of the organisation such as leadership and relative priority; (iii) the
14 characteristics of individuals, which includes organisational staff knowledge and attitudes
15 about the intervention and their role and identification within the wider organisation; (iv) the
16 characteristics of the intervention itself and; (v) implementation processes, which includes the
17 ways that the intervention will be delivered in a given context (including fidelity to the
18 implementation strategy). Normalisation Process Theory will be used to provide additional
19 insights into implementation processes (NPT; 30). NPT aims to identify and explain the
20 implementation of new interventions and how they become integrated into routine care.
21 The RE-AIM framework will be used to evaluate the implementation outcomes,
22 including intervention reach, uptake, and acceptability/appropriateness (25, 31). These
23 outcomes also map onto the framework of implementation outcomes proposed by Proctor and
24 colleagues (32, 33). Reach refers to the proportion of eligible participants who opened, used,
25 and completed the intervention, as well as the proportion of students from the entire cohort of
Page 12 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 12
1 eligible students in intervention schools who consented to participate. Uptake refers to the
2 proportion of schools that were onboarded to the study (intervention and control arms), and
3 of school staff who were willing to support the delivery of the intervention (intervention arm
4 only). Reach and uptake also incorporate the representativeness of the sample (school-level
5 and individual-level). Acceptability and appropriateness refer to the perceived agreeableness
6 or fit of the intervention. This evaluation will incorporate how the barriers and facilitators
7 identified through the CFIR impacted the implementation outcomes, and, in turn, how these
8 implementation outcomes impacted effectiveness outcomes.
9
10 Logic Model
11 Following MRC guidance (3), the research team developed a logic model for the FPS
12 process evaluation in a series of participatory workshops. The logic model was prospectively
13 informed by key CFIR constructs identified in previous literature as being important in
14 school-based studies, but also feasible and appropriate to measure within the school context
15 in the FPS. The key constructs included the outer setting, inner setting, individual
16 characteristics, and intervention characteristics. The logic model (Figure 1) was developed to
17 consider the key factors (mapped to the CFIR) that would potentially impact implementation
18 of the intervention (mapped to RE-AIM), as well as its clinical effectiveness. The process
19 evaluation method, including the selection of dependent variables and design of surveys and
20 semi-structured interview guides, were derived from this logic model. See Table 1 for
21 CFIR/RE-AIM domains, key research questions, process data, and data that will be collected
22 within each domain.
23 A pilot study of 8 schools conducted in 2019 was used to assess the suitability of the
24 planned implementation strategy and process evaluation methods. We retrospectively applied
25 our logic model to collected data to evaluate whether our methodology captured relevant
Page 13 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 13
1 constructs, and sufficient variation in these constructs. We integrated learnings from this pilot
2 study using a dynamic feedback process to strengthen our methodology for the full-scale FPS
3 phase. Some minor changes were made following this pilot which relate to the CFIR,
4 including a greater emphasis on sub-constructs such as school climate (inner setting).
5
6 <INSERT FIGURE 1 HERE>
7
Page 14 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 14
1 Table 1 – Process evaluation details including process data, outcome data, data type and source
CFIR and RE-AIM Constructs Research Aim Process or Outcome Data Data Type and Source
Outer SettingSchool contextual characteristics
What was the broad context of the schools in which the SPARX intervention was delivered? (Aim 2)
School socio-economic indexSchool location (metropolitan/regional)
Publicly available information (ICSEA, GPS)
Inner SettingSchool organisational characteristics
What were the characteristics of the delivery environments (schools)? (Aim 2)
What were the barriers and facilitators that affected buy-in, delivery, and student uptake? (Aim 2)
School size, type, composition, fundingSchool culture
Implementation climate, relative priority, competing demands, leadership, school counsellor availability and level of support, networks and communication, school culture and climate, school readiness for implementation
Publicly available information (school size, type, funding)SSPESH (assesses school culture) Interviews with school staff
Implementation Climate measures (staff and students)Relative Priority measure Competing Demands measureInterviews with school staffChecklists completed by trial managerOther administrative data including number of staff allocated to assist with delivery and consent process, level of communication with the research team
School leadership How supportive of delivering SPARX were school principals, deputy principals and executives? (Aim 2)
Level of support and buy in from school leaders
Interviews with school staff
Individual CharacteristicsSchool staff What were the characteristics (including attitudes,
beliefs, traits) of school staff supporting the delivery of the intervention? (Aim 2)
How supportive of delivery were school staff who were involved on-the-ground? (Aim 2)
Age, gender, current employment, role etc
Leadership, skills, motivations, expectations, self-efficacy, expectations, time available, knowledge and beliefs about the intervention, study buy in
Demographics questionnaire
Implementation Leadership ScaleInterviews with school staff and study facilitators
Study facilitators How well did study facilitators attending schools support the delivery of the intervention? (Aim 2)
Age, employment Skills, self-efficacy or confidence, motivations and expectations
Demographics questionnaireSelf-Confidence measure Interviews with study facilitators
Page 15 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 15
Students What were the characteristics of young people that affected intervention uptake and effectiveness? (Aim 2)
History of mental illness Reported by Year 8 students as part of the online FPS survey
Intervention CharacteristicsSPARX Were there any barriers to intervention use? (Aim 2
and 3)
What do staff think about the efficacy and advantage of using the intervention? (Aim 2)
Technical issues
Evidence strength and quality, relative advantage
Logs of technical issues sent through schools, parents and participantsIT data pertaining to technical problemsInformal feedback provided by schools and research staff attending schools
Relative Advantages measureAnticipated Benefits measureInterviews with school staff
Implementation Processes
Normalisation and integration How did school staff perceive the implementation processes? (Aim 2)
Coherence, cognitive participation, and collective action
NoMAD
Fidelity to the implementation strategy
To what extent was the intervention implemented as planned? (Aim 2)
School delivery of the FP program, including changes to the plan
Completed implementation checklists, emails and feedback formsInterviews with school staff
Implementation Outcomes Reach What was the extent to which those who were eligible
to receive SPARX used it? (Aim 1)Proportion of eligible participants who consented to participate; proportion who opened, used and completed the SPARX interventionRepresentativeness of the student sample
Administrative data about consent. Digital analytic data including usage (app downloads, installs, opens), completion rate (number of modules completed) and time spent using SPARXReported by Year 8 students as part of the online FPS survey
Uptake How many eligible schools participated in the study? Within those schools, how many staff supported the delivery of SPARX? (Aim 1)
Proportion of eligible schools that were onboarded to the study; proportion of school staff (in intervention schools) who supported SPARX Representativeness of the sample
Administrative data
Publicly available information about schools and self-report demographic data from school staff
Acceptability/appropriateness How satisfied were participants with the intervention? (Aim 1)
Acceptability/appropriateness of the intervention, expectations
Acceptability/appropriateness of the Future Proofing program (including SPARX)
Reported by Year 8 students as part of the online FPS surveyInformal conversations and feedback provided by Year 8 studentsImplementation Appropriateness measureInterviews with school study staff
Page 16 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 16
How satisfied were school staff with supporting the intervention the FP program (including SPARX)? (Aim 1)
Informal conversations and feedback provided by school staff
Across DomainsHow might the relationship between the intervention, the staff supporting the program, and context of each school shape variation in outcomes (implementation strength metric)? (Aim 2 and 3)How might the school-level variation (in implementation fidelity and outcomes) affect clinical effectiveness outcomes (e.g., self-reported depression)? (Aim 3)What key lessons emerge from this study that can be generalised to the implementation of digital mental health programs in schools more broadly?
1 Note. The process data and outcomes are mapped onto Figure 2. CFIR = Consolidated Framework for Intervention Research; RE-AIM = Reach, Effectiveness, Adoption, Implementation, 2 Maintenance; Aim = Aims outlined in the Aims and Objectives section; SSPESH = Survey of School Promotion of Emotional and Social Health; NoMAD = Normalisation Measure 3 Development questionnaire.
Page 17 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 17
1 School Implementation Strategy
2 The school implementation strategy was developed by the study authors. The authors
3 drew on their experience in school-based intervention delivery and integrated feedback from
4 teachers and school staff from several completed school-based trials that delivered digital
5 interventions to school students (21, 34). Stakeholder consultation specifically for this study
6 involved discussions with the Department of Education, consultation with several school
7 parent committees and consultation with both youth and parent Lived Experience Advisory
8 Panels. This strategy was also refined following the first pilot wave involving eight
9 intervention schools.
10 The school ‘study implementation teams’ are principally responsible for the
11 implementation of the intervention and liaising with the research team. As described earlier,
12 these teams typically incorporate at least one classroom teacher and one school counsellor to
13 assist with the intervention delivery. Study facilitators also support the delivery of the
14 SPARX intervention by attending schools for the first school session.
15 Implementation Strategy. During the active intervention phase, schools allocate a
16 minimum of 4 x 20-minute school class sessions during which students complete the SPARX
17 intervention. The additional three sessions may be completed either in class if permitted by
18 the schools, or in the students’ own time. The implementation strategy developed by the
19 project team and stakeholders to support the completion of the SPARX intervention
20 comprises:
21 Standardised facilitator training delivered face-to-face over a half-day
22 Study facilitators are present at schools to support students in downloading the
23 SPARX app and completing the baseline assessments
24 School implementation team provided with a SPARX user-guide and information
25 booklet they can refer to during the sessions
Page 18 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 18
1 Schools provide students with weekly verbal reminders in home classrooms to use the
2 app regularly
3 Schools publish brief information about the study and mental health tips in the school
4 weekly newsletter
5 Schools liaise with research team weekly to troubleshoot problems
6 This is the strategy being outlined to schools by the research team and adherence to this
7 strategy will be assessed. Whether or not schools schedule more than the four mandated in-
8 class sessions for intervention completion is flexible and can be adapted to suit the
9 preferences of the school. Student participants receive a $20AUD voucher after the
10 intervention period to cover any phone or data related costs incurred. See Figure 2 for details
11 of training and delivery structure.
12
13 <INSERT FIGURE 2 HERE>
14
15 Data Collection Methods and Participant Groups for Process Evaluation
16 There are three participant groups taking part in the process evaluation: Year 8
17 students, school staff members (e.g., teaching and counselling staff), and study facilitators
18 (see Table 2). School staff members will be members who are responsible for leading the
19 delivery of the study in their school or will be teaching staff who have a supporting role (e.g.,
20 home room teachers who provide reminders to students to complete the intervention).
21 Four types of data will be collected and triangulated: self-report questionnaire data,
22 digital analytic data, administrative data, and qualitative interview data. Self-report
23 questionnaire data specific to the process evaluation will be collected from staff and study
24 facilitators using online survey software (Qualtrics) programmed by one of the authors (JB).
25 These questionnaires assess demographic information, school organisational characteristics,
Page 19 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 19
1 individual characteristics, intervention characteristics, implementation processes and
2 implementation outcomes. Where no available published questionnaires were identified as
3 being suitable for this process evaluation, we adapted existing standardised measures or
4 developed our own items (details below). Self-report questionnaire data from Year 8 students
5 about intervention use and feedback will be collected from an online survey (details
6 described in 5). Digital analytic data about intervention use will be captured by the purpose-
7 built Black Dog Institute research platform, which is being used for the broader FPS.
8 Administrative data will be collected through a range of sources, including communications
9 with the research team. Qualitative data will be collected using semi-structure interviews. All
10 data collected will be from intervention schools only as no intervention is implemented in
11 control schools. Student and school staff data will be collected immediately after the
12 intervention period has been completed (i.e., after the six-week intervention stage); facilitator
13 data will be collected both before and after the intervention period.
14
15 Table 2. Summary of data forms (and collection point) provided by each of the participant
16 groups
Participant group Questionnaire Individual interview Digital analyticsYear 8 students ✓ (post-intervention) ✓ (ongoing)School staff ✓ (post-intervention) ✓ (post-intervention)Facilitators ✓ (before first school
visit and after final post-intervention visit)
✓ (after final school post-intervention visit)
17
18 Implementation Predictors
19 School Organisational Characteristics (Inner Setting).
20 Publicly Available Information. Publicly available information will be collected
21 about school contextual characteristics, including socio-economic level, size, location, type,
22 and funding.
Page 20 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 20
1 General School Culture. School staff will complete the Survey of School Promotion
2 of Emotional and Social Health (SSPESH; 35) and a measure of organisational culture (36).
3 The SSPESH assesses a school’s capacity to promote social and emotional wellbeing, and
4 contains four subscales: Positive School Community, Student Social and Emotional Learning,
5 Engaging Families, and Supporting Students Experiencing Mental Health Difficulties. Items
6 are rated on a 4-point Likert scale from 0 (not yet in place) to 3 (completely in place) and
7 preliminary investigations support the scale structure and criterion-related validity (35).
8 We adapted a 9-item questionnaire about general culture within a health care setting
9 to use within the school setting (36). Items are rated on a 5-point Likert scale from 1 (strongly
10 agree) to 5 (strongly disagree). Lower scores indicate a more positive working culture, which
11 includes transparency, productive working relationships, and receptivity to feedback.
12 Previous investigation has shown that this measure has good internal consistency, although
13 overlaps somewhat with other sub-constructs within the inner setting (e.g., learning climate;
14 35).
15 Implementation Climate. School staff will complete a measure of implementation
16 climate that we adapted for use in the school context (36). Items are rated on a 5-point Likert
17 scale from 1 (strongly agree) to 5 (strongly disagree). Lower scores indicate increased staff
18 receptivity to the program and support within the school, including rewards and recognition.
19 This measure has acceptable internal consistency and good discriminant validity (36).
20 Relative Priority. School staff will complete one adapted item from the School
21 Contextual Barriers subscale of the Perceived Attributes of the Healthy Schools Approach
22 Scale (37). The item is rated on a 5-point scale Likert scale from 1 (strongly agree) to 5
23 (strongly disagree). Higher scores indicate that other activities did not interfere with
24 implementation of the FP program.
25 Competing Demands. Staff will complete 3-items, developed specifically for this
Page 21 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 21
1 study, which assess how much time they allocated, and desired to allocate, to the FP program
2 relative to other competing workload demands on visual analogue scales (anchor by 0 = no
3 work time, 100% = all of my work time). Higher scores indicate increased allocation of time
4 and prioritisation of the FP program.
5 Individual Characteristics.
6 Leadership. School staff will complete the Implementation Leadership Scale (ILS;
7 38). The ILS is a 12-item scale that assesses leadership behaviours that support
8 implementation of evidence-based practices. The ILS contains four subscales, and two will be
9 included in the current study (Knowledge Leadership and Supportive Leadership). These
10 scales assess the degree to which the staff member was knowledgeable about and offered
11 support to the program. Items are rated on a 5-point Likert scale from 0 (very much so) to 4
12 (not at all) and will be re-coded such that higher scores indicate more effective
13 implementation leadership behaviours. This measure has excellent internal validity,
14 convergent validity, and discriminant validity (38).
15 Self-Efficacy and Confidence. Study facilitators will rate their level of confidence in
16 performing 23 different tasks during school-visits on visual analogue scales, which are based
17 on the training program they completed (anchor by 0 = not at all confident, 100 = very
18 confident). Higher scores indicate higher levels of confidence. A similar questionnaire will be
19 repeated following their school-visits, along with four short answer questions that assess
20 experiences with supporting staff and students during school-visits.
21 Intervention Characteristics.
22 Relative Advantage and Evidence Strength and Quality. School staff will complete
23 the 2-item Relative Advantages subscale and one item from the Anticipated Benefits subscale
24 of the Perceived Attributes of the Healthy Schools Approach Scale (37). Adapted for the
25 current study, items are rated on a 5-point scale Likert scale from 1 (strongly agree) to 5
Page 22 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 22
1 (strongly disagree). Scores will be re-coded such that higher scores indicate greater perceived
2 advantage of the FP program compared to others and greater perceived impacts on mental
3 health, respectively.
4 Implementation Processes.
5 Normalisation and Integration into Routine Practice. School staff will complete
6 NPT’s accompanying tool, the Normalisation MeAsure Development questionnaire
7 (NoMAD; 39). The NoMAD is a 23-item measure that assesses how professionals involved
8 in the implementation of a complex intervention perceive implementation processes. It is a
9 flexible measure that can be altered to more accurately describe the adoption of new
10 interventions at the provider level. Ten items of the NoMAD will be used in this study to
11 assess intervention buy-in from school staff, and how staff members incorporated the
12 initiative into their standard work responsibilities. These items are grouped into three
13 categories: coherence (i.e., making sense of an intervention), cognitive participation (i.e.,
14 working with others to support an intervention), and collective action (i.e., the type of work
15 that people do to support an intervention). Items are rated on a 5-point Likert scale from 1
16 (strongly agree) to 5 (strongly disagree). Initial validation demonstrated that the NoMAD has
17 good face validity, construct validity, and internal consistency (40).
18 Fidelity to the Implementation Strategy. Fidelity will be captured through
19 implementation checklists, communications between schools and research staff, and
20 interviews with school staff.
21 Implementation Outcomes
22 Reach. SPARX app usage data from students will allow for the assessment of app use
23 (downloads, installs, and opens), completion (number of modules completed, out of a total of
24 seven), and time spent using SPARX. Administrative data about the number of students in
25 intervention schools with consent to participate will also be used as an indicator of reach.
Page 23 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 23
1 Self-report data about individual characteristics (e.g., gender, mental health history) will be
2 used to gauge the representativeness of the students in the intervention schools.
3 Uptake. Administrative data about the proportion of schools that were onboarded to
4 the study and the proportion of teachers who supported the intervention will be collected to
5 provide an index of uptake by schools. Representativeness of the sample will be informed by
6 publicly available information about school characteristics (e.g., socio-economic status,
7 location) and self-report data about school staff characteristics (e.g., role, gender).
8 Appropriateness. School staff will complete the Intervention Appropriateness
9 Measure (IAM; 41). The IAM is a pragmatic 4-item measure of the perceived fit, relevance,
10 or compatibility of an evidence-based practice for a context, person, or problem (42). Items
11 have been adapted for this study and are rated on a 5-point Likert scale from 1 (strongly
12 agree) to 5 (strongly disagree). Scores will be re-coded such that higher scores indicate higher
13 levels of perceived appropriateness. The IAM demonstrated strong psychometric properties
14 in previous research (41). School staff will also complete one adapted item from the Agency
15 Leadership Support subscale of the Barriers and Facilitators to Implementing Survey (43).
16 The item is rated on a 5-point Likert scale from 1 (strongly agree) to 5 (strongly disagree) and
17 will be re-coded such that higher scores indicate greater compatibility of the FP program
18 within a particular school.
19 Acceptability. Year 8 students will complete an 11-item feedback questionnaire
20 about SPARX. The questionnaire assesses three domains, including: (i) reasons for non-
21 adherence; (ii) intervention acceptability; and (iii) skills learnt from the intervention. Items
22 will be quantified individually. This questionnaire has previously been used to assess the
23 acceptability of the SPARX program in the school context (44).
24 Individual Interviews
25 Interview guides for school staff and study facilitators were derived from the logic
Page 24 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 24
1 model, CFIR online resources (e.g., https://cfirguide.org/evaluation-design/qualitative-data/),
2 and the broader literature investigating the delivery of interventions in school settings (45, 46;
3 see Additional File 1 for interview guides). Interviews will provide information about both
4 implementation predictors and outcomes. For school staff, questions focus on: motivations
5 and expectations about the intervention and study processes more broadly; knowledge and
6 beliefs about the intervention; relative advantages of the intervention; self-efficacy; barriers
7 and facilitators affecting the delivery of the intervention, including fidelity; appropriateness
8 and acceptability of the intervention; and recommendations for future implementation.
9 School counsellors will also be asked questions about their experiences of managing high risk
10 participants within the study and compatibility with existing workload. For study facilitators,
11 questions focus on: motivations and expectations about their role; confidence and
12 competence in supporting the delivery of SPARX in schools; the quality of their training; and
13 perceptions about their ability to support the study as required.
14 The interview guides allow for flexibility in questioning and diversion in responses.
15 Questions will primarily be open-ended, with specific prompts and follow-up questions being
16 used as necessary to encourage respondents to elaborate on their ideas and provide examples.
17 Patient and Public Involvement
18 The Future Proofing Study was developed with key stakeholders including school
19 personnel, school counsellors, parents, adolescents, and individuals with a lived experience of
20 mental illness. All aspects of the study design, appropriateness of outcomes measures and
21 consent procedures were developed in consultation with these stakeholder groups. The
22 current process evaluation involves consultation with school staff members to understand
23 their experience. The SPARX intervention itself was codesigned with young people. All
24 study results will be shared directly with participants and their schools through lay summaries
25 and infographics.
Page 25 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 25
1 Procedure
2 The FPS has three waves of delivery (October-December 2020; April-July 2021; July-
3 September 2021). Process evaluation data will be collected at each wave. All school students
4 will complete relevant survey questions at the same time as completing primary measures for
5 the FPS. All study facilitators will be asked to provide informed consent for their data to be
6 used for research purposes following the compulsory face-to-face training session they attend
7 with the research team at the Black Dog Institute. Data provided by facilitators will be from
8 two surveys, one completed immediately following the training and another completed after
9 their final school visit. All school staff from intervention schools (on average, three from
10 each school, estimated number of intervention schools = 100) who were directly involved in
11 the study will be invited to complete one survey following the six-week SPARX intervention
12 period. All surveys will be completed online.
13 After completing their respective online questionnaires, all school staff and study
14 facilitators will be given the option to participate in a 60-minute individual interview
15 (completed either in person or remotely). Purposive sampling will be used to capture a range
16 of diverse school settings and experiences. Face-to-face interviews will be held in a quiet
17 room on school grounds or at the Black Dog Institute. Virtual interviews, which may be
18 required due to COVID-19 restrictions, will also be conducted. All interviews will be audio-
19 recorded and transcribed verbatim. All contact with study facilitators and school staff,
20 including the semi-structured interviews, will be made by research staff who have had no
21 previous contact with them during the trial. This independence minimises the risk of bias and
22 demand effects.
23
24 Data analysis Plan
25 Quantitative. Survey questionnaire data (from approximately 100 schools) will be
Page 26 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 26
1 exported into data analytic software for analysis. Descriptive statistics will be calculated for
2 all participant groups and will provide information about intervention use and its
3 acceptability (questionnaire data from Year 8 participants), differential implementation,
4 fidelity to the implementation strategy, school context factors within each school
5 (questionnaire including several short answer questions from school staff), and competence
6 and experience in schools from those facilitating the study (facilitator questionnaires).
7 Differences between school clusters will be assessed using Analysis of Variance methods
8 (ANOVA). Regression models will assess the on the effects of contextual factors on
9 implementation outcomes (e.g., reach, uptake, acceptability).
10 Qualitative. Interviews will be digitally audio-recorded and transcribed verbatim.
11 The transcripts will be checked for accuracy against the sound files as per best practice in
12 transcription (47, 48). Qualitative data will then be imported into NVivo to aid in data
13 management and analysis. Thematic analysis will be undertaken to identify, interpret and
14 report on the repeated patterns of meaning within the data, drawing from Braun and Clark’s
15 classic six phase model (49, 50). An iterative and reflexive approach will be used to analyse
16 the data, incorporating themes from the data together with topics covered in the interview
17 guide. Two coders will independently engage in a familiarisation phase before generating
18 codes and initial themes for a subset of the data. These codes and themes will be reviewed
19 and discussed by the two coders, with refinement occurring via an interactive process. A
20 senior qualitative analysist will also review the first-stage coding framework and scheme
21 before all transcripts are coded by the first coder. Refinement will continue to occur via an
22 interactive process until final codes and themes are realised and defined across the whole data
23 set. Research rigor will be enhanced by a team approach to analysis, reflexive field notes, and
24 prolonged engagement with the subject matter (51).
25 Triangulation of Qualitative and Quantitative Data. Triangulation involves the use
Page 27 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 27
1 of multiple approaches to address a research question. The combination of several approaches
2 increases confidence in the findings and provides a more comprehensive account of the
3 results than individual approaches would do alone (52). In this study, reliability, validity and
4 confidence will be maximised through cross verification and exploration of differences
5 between the outcomes of the various methods. This takes place in several ways:
6 Maximising validity in analysis of qualitative data within the research team using
7 techniques such as discussing coding, constant comparison, accounting for deviant
8 cases, and systematic coding;
9 Triangulation of school staff and research assistant interviews with results from the
10 questionnaires, exploring and accounting for differences;
11 Triangulation of self-report and interview data with publicly available information
12 relating to school contextual characteristics (e.g., school socio-economic level and
13 size) and school delivery of the program, including deviations to the implementation
14 strategy.
15 Mapping the perspectives of different stakeholders across the study (school staff,
16 study facilitators).
17 Additional Analyses. We will generate an ‘implementation strength’ metric for each
18 school and its relationship to primary and secondary trial outcomes. This is an emerging
19 evaluation approach, used mainly in low and middle-income countries. The approach aims to
20 understand the degree of implementation effort needed during intervention delivery to
21 achieve desired benefits (53, 54).
22 The implementation strength metric will provide funders and policy makers with an
23 objective measure to monitor effectiveness of implementation if it goes beyond this trial and
24 becomes a sustainable approach. The metric can be used to assess whether the approach to
25 implementation meets the minimum level required to prevent the onset of mental health
Page 28 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 28
1 problems in adolescents. The metric will be based on implementation inputs and contextual
2 factors informed by the intervention logic model and process evaluation frameworks (CFIR
3 and RE-AIM) (e.g., adoption from teachers and other staff in the school and fidelity to the
4 intervention strategy), and will be developed with the Future Proofing research team using
5 principal component analysis.
6
7 Ethics and Dissemination
8 Ethics approval was obtained from the University of New South Wales Human
9 Research Ethics Committee (HC180836; 21st January 2019) and the NSW Government State
10 Education Research Applications Process (SERAP 2019201; 19th August 2019). Results will
11 be submitted for publication in peer reviewed journals and discussed at conferences. Our
12 process evaluation will contextualise the trial findings with respect to how the intervention
13 may have worked in some schools but not in others. This evaluation will inform the
14 development of a model for rolling out digital interventions for the prevention of mental
15 illness in schools.
16
17 Discussion
18 This paper describes the design of a mixed methods process evaluation of a cluster
19 randomised controlled trial, the FPS. The FPS investigates the impact of a digital cognitive-
20 behavioural therapy intervention when delivered at scale in school settings. Digital mental
21 health programs have tremendous potential to prevent up to 22% of depression cases and,
22 when delivered at scale, could have population-level impacts (55). However, these programs
23 have not been translated into practice and policy because optimal ways to scale and deliver
24 these interventions are not yet well understood.
25 As an initial step to address this issue, the current process evaluation will attend to
Page 29 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 29
1 contextual and implementation factors that vary across schools and provide a lens through
2 which to interpret trial efficacy outcomes. We expect that results will provide a richly
3 detailed and nuanced understanding of the key factors involved in the effective delivery of
4 digital mental health programs across different schools. We expect that results will not only
5 contextualise our trial findings but will also be used as a model to guide the delivery of
6 school-based interventions that focus on preventing mental illness more broadly. Findings
7 from this process evaluation will indicate whether the approach used in the FPS trial is likely
8 to be sustainable in the school environment going forward and, if so, the threshold level of
9 support required in order to prevent depression and benefit student mental health.
10 The prospective publication of this protocol outlines our planned methodological
11 approach. It also serves as a road map for other researchers on a practical way of carrying out
12 process evaluations of complex interventions in the school setting. As is the case with the
13 delivery of interventions across different contexts, we acknowledge that our approach has
14 inbuilt flexibility to explore the data and make provisions for unexpected implementation
15 factors that arise.
16 Limitations and Strengths
17 There are several limitations to our process evaluation that warrant mention. First, we
18 are not including direct observation of teachers in their role supporting the delivery of
19 SPARX. While this would provide objective fidelity data, it requires resources beyond the
20 scope of this project and is not representative of how the program will be sustained following
21 the conclusion of the trial. Second, given the complexity of the study and high demand placed
22 on students (e.g., engaging with the study apps and completing the online surveys at multiple
23 time points over five years), we are not collecting in-depth qualitative data by way of
24 interviews. Instead, we collect information about students’ perceptions of the intervention
25 (e.g., acceptability) through short self-report questions in the online survey. Second, the
Page 30 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 30
1 qualitative interviewer is a member of the research team (but not the evaluation team). Care
2 will be taken to ensure that this staff member has no contact with schools prior to the
3 interview visit to minimise bias. However, there remains a risk that demand effects may
4 impact the information that is shared. Third, the process evaluation process (questionnaires
5 and interview) will undoubtedly add to the burden placed on school staff. Given that the FPS
6 is already placing a significant burden on the time of busy school staff, this additional
7 component might contribute to low levels of participation.
8 To the point of burden on schools, one of the strengths of the design is that we have
9 undertaken a pilot phase involving eight intervention schools and have been able to refine our
10 processes (e.g., introduce an incentive) for school staff members to participate. This process
11 evaluation also involves the combination of qualitative and quantitative methods which will
12 be triangulated to provide a coherent and comprehensive picture of the data. The use of the
13 ‘implementation strength’ metric represents a novel approach in this field, borrowed from the
14 low-middle income country implementation science sector. The inclusion of this approach
15 will provide important information to funders and policy makers following on from this trial,
16 indicating the level of implementation support required to prevent mental illness and improve
17 wellbeing of adolescent school students.
18 Notwithstanding the limitations raised above, this process evaluation will contribute
19 to the broader knowledge base and indicate how best to deliver digital mental health
20 prevention programs in school settings.
21
22
23
24
25
Page 31 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 31
1 Additional Files
2 Additional File 1 (Semi-Structured Interview Guides).docx
3
4 List of Abbreviations
5 FPS = Future Proofing Study
6 cRCT = Cluster Randomised Controlled Trial
7 CFIR = Consolidated Framework for Implementation Research
8 RE-AIM = Reach, Effectiveness, Adoption, Implementation, Maintenance framework
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Page 32 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 32
1 References
2 1. Oakley A, Strange V, Bonell C, Allen E, Stephenson J, Team RS. Process evaluation
3 in randomised controlled trials of complex interventions. BMJ. 2006;332:413.
4 2. Greenhalgh T, Papoutsi C. Studying complexity in health services research:
5 Desperately seeking an overdue paradigm shift. BMC Med. 2018;16:95.
6 3. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process
7 evaluation of complex interventions: Medical research council guidance. BMJ.
8 2015;350:h1258.
9 4. Grant A, Treweek S, Dreischulte T, Foy R, Guthrie B. Process evaluations for cluster-
10 randomised trials of complex interventions: A proposed framework for design and reporting.
11 Trials. 2013;14:15.
12 5. Werner-Seidler A, Huckvale K, Larsen ME, Calear AL, Maston K, Johnston L, et al.
13 A trial protocol for the effectiveness of digital interventions for preventing depression in
14 adolescents: The future proofing study. Trials. 2020;21:2.
15 6. Tal A, Torous J. The digital mental health revolution: Opportunities and risks.
16 Psychiatric Rehabilitation Journal. 2017;40:263.
17 7. Sawyer SM, Azzopardi PS, Wickremarathne D, Patton GC. The age of adolescence.
18 Lancet Child Adolesc Health. 2018;2:223.
19 8. Andrews G, Issakidis C, Sanderson K, Corry J, Lapsley H. Utilising survey data to
20 inform public policy: Comparison of the cost-effectiveness of treatment of ten mental
21 disorders. British Journal of Psychiatry. 2004;184:526.
22 9. Lawrence D, Johnson S, Hafekost J, de Hann KB, Sawyer M, Ainley J, et al. The
23 mental health of children and adolescents: Report on the second australian child and
24 adolescent survey of mental health and wellbeing. 2015.
25 10. Lee FS, Heimer H, Giedd JN, Lein ES, Šestan N, Weinberger DR, et al. Adolescent
Page 33 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 33
1 mental health—opportunity and obligation. Science. 2014;346:547.
2 11. Werner-Seidler A, Perry Y, Calear AL, Newby JM, Christensen H. School-based
3 depression and anxiety prevention programs for young people: A systematic review and
4 meta-analysis. Clin Psychol Rev. 2017;51:30.
5 12. van Zoonen K, Buntrock C, Ebert DD, Smit F, Reynolds CF, Beekman AT, et al.
6 Preventing the onset of major depressive disorder: A meta-analytic review of psychological
7 interventions. Int J Epidemiol. 2014;43:318.
8 13. Gulliver A, Griffiths KM, Christensen H. Perceived barriers and facilitators to mental
9 health help-seeking in young people: A systematic review. BMC Psychiatry. 2010;10:113.
10 14. Donker T, Blankers M, Hedman E, Ljotsson B, Petrie K, Christensen H. Economic
11 evaluations of internet interventions for mental health: A systematic review. Psychological
12 Medicine. 2015;45:3357.
13 15. Essau CA. Frequency and patterns of mental health services utilization among
14 adolescents with anxiety and depressive disorders. Depress Anxiety. 2005;22:130.
15 16. Rickwood DJ, Deane FP, Wilson CJ. When and how do young people seek
16 professional help for mental health problems. Med J Aust. 2007;187:S35.
17 17. Kösters MP, Chinapaw MJM, Zwaanswijk M, van der Wal MF, Utens EMWJ, Koot
18 HM. FRIENDS for Life: Implementation of an indicated prevention program targeting
19 childhood anxiety and depression in a naturalistic setting. Mental Health & Prevention.
20 2017:6,44.
21 18. Donker T, Blankers M, Hedman E, Ljótsson B, Petrie K, Christensen H. Economic
22 evaluations of internet interventions for mental health: A systematic review. Psychological
23 Medicine. 2015;45:3357.
24 19. Lee YY, Barendregt JJ, Stockings EA, Ferrari AJ, Whiteford HA, Patton GA, et al.
25 The population cost-effectiveness of delivering universal and indicated school-based
Page 34 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 34
1 interventions to prevent the onset of major depression among youth in australia.
2 Epidemiology and Psychiatric Sciences. 2017;26:545.
3 20. Merry SN, Stasiak K, Shepherd M, Frampton C, Fleming T, Lucassen MF. The
4 effectiveness of sparx, a computerised self help intervention for adolescents seeking help for
5 depression: Randomised controlled non-inferiority trial. BMJ. 2012;344:e2598.
6 21. Perry Y, Werner-Seidler A, Calear A, Mackinnon A, King C, Scott J, et al. Preventing
7 depression in final year secondary students: School-based randomized controlled trial.
8 Journal of Medical Internet Research. 2017;19:e369.
9 22. Higgins MC, Weiner J, Young L. Implementation teams: A new lever for
10 organizational change. Journal of Organizational Behavior. 2012;33:366.
11 23. Clarke DJ, Godfrey M, Hawkins R, Sadler E, Harding G, Forster A, et al.
12 Implementing a training intervention to support caregivers after stroke: A process evaluation
13 examining the initiation and embedding of programme change. Implement Sci. 2013;8:96.
14 24. Curran GM, Bauer M, Mittman B, Pyne JM, Stetler C. Effectiveness-implementation
15 hybrid designs: Combining elements of clinical effectiveness and implementation research to
16 enhance public health impact. Med Care. 2012;50:217.
17 25. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health
18 promotion interventions: The re-aim framework. American Journal of Public Health.
19 1999;89:1322.
20 26. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering
21 implementation of health services research findings into practice: A consolidated framework
22 for advancing implementation science. Implementation Science. 2009;4:50.
23 27. Nilsen P. Making sense of implementation theories, models and frameworks.
24 Implementation Science. 2015;10:53.
25 28. Kirk MA, Kelley C, Yankey N, Birken SA, Abadie B, Damschroder L. A systematic
Page 35 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 35
1 review of the use of the consolidated framework for implementation research. Implement Sci.
2 2016;11:72.
3 29. Leeman J, Wiecha JL, Vu M, Blitstein JL, Allgood S, Lee S, et al. School health
4 implementation tools: A mixed methods evaluation of factors influencing their use.
5 Implement Sci. 2018;13:48.
6 30. Murray E, Treweek S, Pope C, MacFarlane A, Ballini L, Dowrick C, et al.
7 Normalisation process theory: A framework for developing, evaluating and implementing
8 complex interventions. BMC Medicine. 2010;8:63.
9 31. Glasgow RE, Harden SM, Gaglio B, Rabin B, Smith ML, Porter GC, et al. Re-aim
10 planning and evaluation framework: Adapting to new science and practice with a 20-year
11 review. Frontiers in Public Health. 2019;7.
12 32. Proctor EK, Landsverk J, Aarons G, Chambers D, Glisson C, Mittman B.
13 Implementation research in mental health services: An emerging science with conceptual,
14 methodological, and training challenges. Administration and Policy in Mental Health.
15 2009;36:24.
16 33. Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, et al. Outcomes
17 for implementation research: Conceptual distinctions, measurement challenges, and research
18 agenda. Administration and Policy in Mental Health and Mental Health Services Research.
19 2011;38:65.
20 34. Calear AL, Christensen H, Mackinnon A, Griffiths KM, O'Kearney R. The
21 youthmood project: A cluster randomized controlled trial of an online cognitive behavioral
22 program with adolescents. Journal of Consulting and Clinical Psychology. 2009;77:1021.
23 35. Dix KL, Green MJ, Tzoumakis S, Dean K, Harris F, Carr VJ, et al. The survey of
24 school promotion of emotional and social health (sspesh): A brief measure of the
25 implementation of whole-school mental health promotion. School Mental Health.
Page 36 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 36
1 2019;11:294.
2 36. Fernandez ME, Walker TJ, Weiner BJ, Calo WA, Liang S, Risendal B, et al.
3 Developing measures to assess constructs from the inner setting domain of the consolidated
4 framework for implementation research. Implement Sci. 2018;13:52.
5 37. Deschesnes M, Trudeau F, Kébé M. Psychometric properties of a scale focusing on
6 perceived attributes of a health promoting school approach. Can J Public Health.
7 2009;100:389.
8 38. Aarons GA, Ehrhart MG, Farahnak LR. The implementation leadership scale (ils):
9 Development of a brief measure of unit level implementation leadership. Implement Sci.
10 2014;9:45.
11 39. Finch TL, Girling M, May CR, Mair FS, Murray E, Treweek S, et al. Nomad:
12 Implementation measure based on normalization process theory 2015 [Available from:
13 http://www.normalizationprocess.org.
14 40. Finch TL, Girling M, May CR, Mair FS, Murray E, Treweek S, et al. Improving the
15 normalization of complex interventions: Part 2 - validation of the nomad instrument for
16 assessing implementation work based on normalization process theory (npt). BMC Med Res
17 Methodol. 2018;18:135.
18 41. Weiner BJ, Lewis CC, Stanick C, Powell BJ, Dorsey CN, Clary AS, et al.
19 Psychometric assessment of three newly developed implementation outcome measures.
20 Implement Sci. 2017;12:108.
21 42. Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, et al. Outcomes
22 for implementation research: Conceptual distinctions, measurement challenges, and research
23 agenda. Adm Policy Ment Health. 2011;38:65.
24 43. Salyers MP, Rollins AL, McGuire AB, Gearhart T. Barriers and facilitators in
25 implementing illness management and recovery for consumers with severe mental illness:
Page 37 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 37
1 Trainee perspectives. Adm Policy Ment Health. 2009;36:102.
2 44. Perry Y, Calear AL, Mackinnon A, Batterham PJ, Licinio J, King C, et al. Trial for
3 the prevention of depression (tripod) in final-year secondary students: Study protocol for a
4 cluster randomised controlled trial. Trials. 2015;16:451.
5 45. Punukollu M, Burns C, Marques M. Effectiveness of a pilot school-based intervention
6 on improving scottish students’ mental health: A mixed methods evaluation. J Youth
7 Adolesc. 2019:1.
8 46. Outhwaite LA, Gulliford A, Pitchford NJ. A new methodological approach for
9 evaluating the impact of educational intervention implementation on learning outcomes. Int J
10 Res Meth Educ. 2019:1.
11 47. Poland BD. Transcription quality. Handbook of interview research: Context and
12 method. 2002:629.
13 48. Poland BD. Transcription quality as an aspect of rigor in qualitative research.
14 Qualitative inquiry. 1995;1:290.
15 49. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol.
16 2006;3:77.
17 50. Braun V, Clarke V. Reflecting on reflexive thematic analysis. Qual Res Sport Exerc
18 Health. 2019;11:589.
19 51. Noble H, Smith J. Issues of validity and reliability in qualitative research. Evid-Based
20 Nurs. 2015;18:34.
21 52. Tashakkori A, Teddlie C. Handbook of mixed methods in social & behavioral
22 research. Thousand Oaks, CA: Sage; 2003.
23 53. Hargreaves JR, Goodman C, Davey C, Willey BA, Avan BI, Schellenberg JR.
24 Measuring implementation strength: Lessons from the evaluation of public health strategies
25 in low- and middle-income settings. Health Policy Plan. 2016;31:860.
Page 38 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 38
1 54. Victora CG, Black RE, Boerma JT, Bryce J. Measuring impact in the millennium
2 development goal era and beyond: A new approach to large-scale effectiveness evaluations.
3 Lancet. 2011;377:85.
4 55. Ebert DD, Cuijpers P. It is time to invest in the prevention of depression. JAMA Netw
5 Open. 2018;1:e180335.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Page 39 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 39
1 Declarations
2 Authors’ Contributions
3 HC and AW-S conceived of the study and secured the funding. AW-S and JRB led the design
4 of the process evaluation, with input from all authors (including KH, IZ, PJB), and expert
5 guidance from RL and KB. AW-S drafted the manuscript, with assistance from JRB. ALC,
6 MT, KM, RL, KB and HC have a continuing role in monitoring the conduct and outcomes of
7 the process evaluation. All named authors contributed substantially to the approved final
8 manuscript.
9
10 Funding Statement
11 Funding for this project came from an NSW Ministry of Health Early-Mid Career Fellowship
12 awarded to AW-S, and a Black Dog Institute Post-Doctoral Fellowship awarded to JRB,
13 secured by HC. ALC is supported by NHMRC fellowships 1122544 and 1173146. PJB is
14 supported by NHMRC Fellowship 1158707. Funding for the randomised controlled trial
15 within which this process evaluation is embedded came from an NHMRC Project Grant
16 Awarded to HC GNT1120646. The funding bodies had no role in any aspect of the study
17 design or this manuscript.
18
19 Competing Interests Statement
20 The authors declare that they have no competing interests.
21
22 Acknowledgments
23 We would like to thank Professor Melanie Barwick for her advice on this manuscript,
24 together with all the facilitators from the Training Institute for Dissemination and
25 Implementation Research in Health (TIDIRH; Australia 2020) for their comments and
Page 40 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 40
1 feedback on this project.
2
3 Trial Status
4 Recruitment for the trial is underway. Data collection commences in October 2020 (delayed
5 from April 2020 due to COVID-19).
6
7 Consent to Participate
8 All schools, staff, parents, and participants must provide active consent to participate.
9
10 Consent for Publication
11 Not applicable.
12
13 Availability of Data and Materials
14 Not applicable.
15
16 Author information
17 Not applicable.
Page 41 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Embedded process evaluation protocol for trial to prevent depression in school students 41
1 Figure Legends
2 Figure 1. Logic model. The model shows that CFIR constructs including school context
3 characteristics, school organisational characteristics, and individual characteristics, will
4 influence how staff engage with the implementation strategy. The intervention itself, which
5 includes the core cognitive-behavioural therapeutic components, is conceptualised as
6 standardised across individuals because it is delivered digitally, follows a fixed schedule and
7 does not incorporate tailored content. The yellow input factors are expected to vary across
8 schools and individuals, thus influencing engagement and flexibility of the implementation
9 strategy and in turn, implementation outcomes and student-level outcomes. The logic model
10 and implementation plan were externally peer reviewed by an experienced and internationally
11 recognised implementation scientist outside the team within an implementation workshop.
12 For details on assessment of these factors, see Table 1.
13
14 Figure 2. Details of implementation strategy training and delivery structure.
15
16
Page 42 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
School Contextual CharacteristicsSociodemographic factors (school size,
socio economic status, area)
Intervention Characteristics (SPARX)
Standardised: Delivered by app over 6 weeks
Technical issues, evidence strength and quality, relative advantage
Student-Level Outcomes
Primary outcome: Lower levels of depression
Secondary outcomes: Lower levels of anxiety, distress,
insomnia
Additional outcomes:Improved quality of life, wellbeing
Lower levels of suicidality, self-harm
Improved academic performance on standardised tests
Lower levels of health service use
Implementation Outcomes
School-level fidelity to the implementation strategy
Reach (e.g., intervention received by students, including number of modules completed and time spent using SPARX)
School-level uptake of the FPS
Staff and student acceptability/appropriateness
Implementation Strategy
App download in school time, with support from
research staff
Intervention completion, with support from school
staff
Reminders and support provided by school staff to
complete intervention
Therapeutic Components
PsychoeducationCognitive restructuringBehavioural activationRelaxation strategies
Practical problem solvingEmotion regulation and management
Individual CharacteristicsSchool staff (e.g., skills, motivation, self-
efficacy, buy in, knowledge about the intervention, leadership)
Study facilitators (e.g., skills, confidence)
School Organisational CharacteristicsSchool culture, buy in and support for the
intervention and study, support from leadership, ability of school to support delivery (including
technological infrastructure, counsellor availability)
Individual CharacteristicsStudents (e.g., mental health
history, intervention expectations)
Page 43 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Research Team
Facilitators
Teachers Counsellors
Year 8 Students
Training
Other staff
Delivery Support
Direct trainingSupport to access interventionDelivery support
School Implementation Team
Page 44 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 1
Additional File 1
Semi-Structured Interview Guides
School Staff INTRODUCTION: Thank you for agreeing to take part in this interview. I appreciate your time. I am interested in hearing your thoughts and experiences of the Future Proofing program in your school. For this interview, the FP program includes the intervention app (SPARX) and everything else that was done to support the program in your school. I hope to use this information to work out what works in different schools and how to best provide support. I will ask you a few questions about your experiences. This is chance for you to reflect and offer feedback on how we could improve. The interview will last up to one hour. It will be recorded, and your responses will be transcribed to ensure we accurately reflect your words. All personal information will be removed so that your answers are non-identifiable and confidential. Nothing you personally say will be shared with anyone on the FP team that you have been liaising with or with other staff at your school. You can choose to not answer any questions. Do you have any questions before we get started? General Opening Questions:
1. How did you first get involved in teaching/counselling/psychology?
2. Could you tell me about the core values and ethos of your school?
3. To the best of your knowledge, what mental health programs or initiatives has your
school implemented to care for students besides the FP program?
Specific Questions:
4. Overall, how would you describe your experience with the FP program in your school?
5. What was your role in delivering or supporting the FP program in your school?
a. Leader/driver, decision-maker, consent/recruitment, peripheral support
CFIR DOMAIN: INDIVIDUAL CHARACTERISTICS Personal attributes (e.g., motivations, expectations, capacity, values)
6. Why were you motivated to take on this role?
a. How did you come on board with the program?
7. Why do you think your school decided to support this program?
a. To what extent do you think there was a need for a mental health program like FP in
you school (Tension for change)?
8. What were your expectations about the FP program prior to coming on board?
a. How were your expectations met (or not met)? What went differently to what
you expected? What surprised you?
b. What were your expectations about the apps?
Knowledge and beliefs about the intervention 9. How much do you know about SPARX, the intervention? E.g., principles, rationale,
content?
Page 45 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 2
Self-efficacy 10. How confident were you in your ability to lead/support/make decisions about the FP
program? a. What personal qualities do you think helped you to do that?
CFIR DOMAIN: OUTER SETTING Relative Advantage
11. How did SPARX compare to other mental health programs that your school has supported or delivered?
CFIR DOMAIN: INNER SETTING School contextual factors – Barriers and Facilitators
12. What factors influenced your ability to deliver/support the FP program in your school? a. What else helped or hindered your ability? b. Enquire into school-specific factors and program-specific factors e.g., What
parts of the program were easy for you to do or support (e.g., the consent process, organising rooms, weekly team meetings)? OR What aspects of your school made it difficult for you to support the program?
c. What factors got in the way of students completing the mental health app/s? d. Were you (or others) able to overcome any barriers? How did you manage to do
this? School contextual factors – Leadership
13. Tell me about the involvement of your school executive in the FP program. a. What role did this person have (e.g., principal/vice principal)? Did you meet
regularly with this person or provide updates to them? 14. How do you think the school executive saw this initiative?
Readiness for implementation
15. How would you describe the readiness of your school to take on the FP program? a. What was the preparation phase like for you/other key staff?
IMPLEMENTATION METRICS Appropriateness and Acceptability
16. How does the FP program align with your school’s approach toward student mental health?
17. How suited is the FP program to address depression prevention in your school? 18. How well does an app delivering therapy work to prevent depression in your students? 19. How do you feel about using technology to deliver psychological interventions in your
school? a. What worked well about this and what did not? b. Do you think SPARX met the student’s mental health needs? In what way?
20. What were the by-products of the intervention that you didn’t expect? a. What parts of the program do/do not suit your school?
**ADDITIONAL QUESTIONS FOR SCHOOL COUNSELLORS ONLY 21. What was your experience with responding to students identified as needing follow-up
during the FP program? a. Was there anything that could have been done to make this experience better for
you and the student? 22. What do you think of the FP risk protocol?
Page 46 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 3
KEY LEARNING AND FUTURE IMPLEMENTATION 23. What parts of the FP program, if any, could be made more attractive to your school to
increase uptake or support? a. In what ways could we make these parts more attractive?
24. To what extent do you think that the FP program could be integrated into your school’s existing mental health strategy?
25. What advice would you give another school contemplating taking on the FP program? 26. What’s been your major learning or insight from your school’s involvement in the FP
program? Closing
27. Is there anything else you would like to add?
Page 47 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 4
Volunteer Research Assistants (Study Facilitators) INTRODUCTION: Thank you for offering to take part in this interview. I appreciate your time. I am interested in your thoughts about and experiences with supporting the FP program as a Research Assistant Volunteer. I hope to use this information to work out what works in different schools and how we can help volunteers to best provide support. I will ask you a few questions about your experiences. This is chance for you to reflect and offer feedback on how we could improve. The interview will last up to one hour. It will be recorded and your responses will be transcribed to ensure we accurately reflect your words. All personal information will be removed so that your answers are non-identifiable and confidential. Nothing you personally say will be shared with the volunteer team or the Black Dog Institute more generally, so it will not affect your relationship with them. You can choose to not answer any questions. Do you have any questions before we get started? General Opening Questions:
1. What schools have you visited so far? 2. Roughly how many school visits have you attended?
Specific Questions: CFIR DOMAIN: INDIVIDUAL CHARACTERISTICS Personal attributes (e.g., motivation and expectations)
3. Why were you motivated you to volunteer? 4. What were your expectations coming on board with the FP program?
a. How were your expectations met (or not met)? b. What went differently to what you expected? What surprised you?
Perceptions of Training
5. What are your overall impressions of the training you received? a. How helpful and appropriate was the training? b. What parts were the most useful? c. What was missing from the training?
CFIR DOMAIN: INNER SETTING School visits – Barriers/Facilitators
6. What was your experience with completing the school visits? 7. Did anything unexpected happen during the school visits?
a. If so, what? b. How did you manage this? c. How well do you think you managed this?
8. In general, how was the FP program received by school staff during visits? 9. To what extent was it useful having a volunteer partner/s at the visits?
a. Were there any other benefits/disadvantages? School contextual factors - Barriers/Facilitators
10. What factors influenced your ability to assist students and teachers at the school-visits? a. Were these common across all the visits or specific to a particular visit? b. What aspects of the school made it difficult to support the program?
Page 48 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
Semi-Structured Interview Guides (Version 2, 21/02/2020) HREC number: HC180836 5
c. What else helped or hindered your ability? (e.g., research platform/app specific factors)
d. How did you overcome these barriers? e. Were there any noticeable differences between different schools? If so, what were
they? KEY LEARNING AND FUTURE IMPLEMENTATION
11. What advice would you give a school contemplating taking on the FP program? 12. What advice would you give to others who are thinking about volunteering? 13. What parts of the FP program could be made more attractive to volunteers to increase
uptake or support? 14. What’s been your major learning or insight from taking part in the volunteer program?
Closing
15. Is there anything else you would like to add?
Page 49 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
1
Standards for Reporting Implementation Studies: the StaRI checklist for completion
Checklist itemReported on page # Implementation Strategy
Reported on page # Intervention
“Implementation strategy” refers to how the intervention was implemented
“Intervention” refers to the healthcare or public health intervention that is being implemented.
Title and abstractTitle 1 1-2 Identification as an implementation study, and description of the methodology in the title and/or keywords
Abstract 2 2-3 Identification as an implementation study, including a description of the implementation strategy to be tested, the evidence-based intervention being implemented, and defining the key implementation and health outcomes.
IntroductionIntroduction 3 5 Description of the problem, challenge or deficiency in healthcare or public health that the intervention being implemented aims
to address.Rationale 4 9-11 The scientific background and rationale for the
implementation strategy (including any underpinning theory/framework/model, how it is expected to achieve
its effects and any pilot work).
8 The scientific background and rationale for the intervention being implemented (including evidence
about its effectiveness and how it is expected to achieve its effects).
Aims and objectives
5 9 The aims of the study, differentiating between implementation objectives and any intervention objectives.
Methods: descriptionDesign 6 11 The design and key features of the evaluation, (cross referencing to any appropriate methodology reporting standards) and any
changes to study protocol, with reasonsContext 7 11-13 The context in which the intervention was implemented. (Consider social, economic, policy, healthcare, organisational barriers
and facilitators that might influence implementation elsewhere).Targeted
‘sites’8 8 The characteristics of the targeted ‘site(s)’ (e.g
locations/personnel/resources etc.) for implementation and any eligibility criteria.
8 The population targeted by the intervention and any eligibility criteria.
Description 9 17 A description of the implementation strategy 8 A description of the intervention
Page 50 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
2
Sub-groups 10 N/A Any sub-groups recruited for additional research tasks, and/or nested studies are described
Methods: evaluationOutcomes 11 18 Defined pre-specified primary and other outcome(s) of
the implementation strategy, and how they were assessed. Document any pre-determined targets
N/A Defined pre-specified primary and other outcome(s) of the intervention (if assessed), and how they were assessed. Document any pre-determined targets
Process evaluation
12 14 Process evaluation objectives and outcomes related to the mechanism by which the strategy is expected to work
Economic evaluation
13 N/A Methods for resource use, costs, economic outcomes and analysis for the implementation strategy
N/A Methods for resource use, costs, economic outcomes and analysis for the intervention
Sample size 14 7 Rationale for sample sizes (including sample size calculations, budgetary constraints, practical considerations, data saturation, as appropriate)
Analysis 15 26-28 Methods of analysis (with reasons for that choice)
Sub-group analyses
16 26-28 Any a priori sub-group analyses (e.g. between different sites in a multicentre study, different clinical or demographic populations), and sub-groups recruited to specific nested research tasks
ResultsCharacteristics 17 N/A Proportion recruited and characteristics of the recipient
population for the implementation strategyN/A Proportion recruited and characteristics (if appropriate)
of the recipient population for the interventionOutcomes 18 N/A Primary and other outcome(s) of the implementation
strategyN/A Primary and other outcome(s) of the Intervention (if
assessed)Process
outcomes19 N/A Process data related to the implementation strategy mapped to the mechanism by which the strategy is expected to work
Economic evaluation
20 N/A Resource use, costs, economic outcomes and analysis for the implementation strategy
N/A Resource use, costs, economic outcomes and analysis for the intervention
Sub-group analyses
21 N/A Representativeness and outcomes of subgroups including those recruited to specific research tasks
Page 51 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from
For peer review only
3
Fidelity/ adaptation
22 N/A Fidelity to implementation strategy as planned and adaptation to suit context and preferences
N/A Fidelity to delivering the core components of intervention (where measured)
Contextual changes
23 N/A Contextual changes (if any) which may have affected outcomes
Harms 24 N/A All important harms or unintended effects in each group
DiscussionStructured discussion
25 28-30 Summary of findings, strengths and limitations, comparisons with other studies, conclusions and implications
Implications 26 N/A Discussion of policy, practice and/or research implications of the implementation strategy (specifically
including scalability)
N/A Discussion of policy, practice and/or research implications of the intervention (specifically including
sustainability)General
Statements 27 31-33 Include statement(s) on regulatory approvals (including, as appropriate, ethical approval, confidential use of routine data, governance approval), trial/study registration (availability of protocol), funding and conflicts of interest
Page 52 of 51
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Novem
ber 15, 2021 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2020-042133 on 12 January 2021. Dow
nloaded from