Exploring influences on evaluation practice: a case study of a national physical activity programme
International Journal of Behavioral Nutrition and Physical Activity volume 18, Article number: 31 (2021)
Interventions to improve physical activity behaviour are a core part of public health policy and practice. It is essential that we evaluate these interventions and use the evidence to inform decisions to improve population health. Evaluation of ‘real-world’ interventions provide an opportunity to generate practice-relevant evidence, however these interventions are difficult to evaluate. Various guidelines have been developed to facilitate evaluation, but evidence about their effectiveness in practice is limited. To explore influences on evaluation practice in an applied context, we conducted a case study of Sport England’s ‘Get Healthy Get Active’ (GHGA) programme. This was a national programme that funded 33 projects that were delivered and evaluated across England. The programme was chosen as it was designed to generate evidence on the role of sport in increasing physical activity and improving health. The study aimed to explore and appraise whether strategies intended to facilitate project evaluation, including funder requirements to use a standardised evaluation framework and specific data collection methods, were effective in generating evidence that enabled the programme to meet its aims.
We applied a collective case study design involving 35 semi-structured interviews, and documentary analysis of multiple sources of evidence from 23 physical activity projects funded by GHGA. We applied thematic and framework analysis. We developed a logic model and mapped actual outcomes against intended outcomes. A narrative synthesis is provided. We discuss implications for the effective commissioning and evaluation of public health interventions.
We identified five main themes of influences on evaluation practices that can act as barriers and facilitators to good practice: programme and project design; evaluation design; partnerships; resources; and organisational structures and systems. These influences are context-specific and operate through a complex set of interactions.
Developing a better understanding of how influences on evaluation practice can act as facilitators or barriers is vital to help close current gaps in the evidence-based practice cycle. Critically, organisational structures and systems are needed to facilitate collaborative decision making; integration of projects and evaluation across partners organisations; transfer of knowldege and insights between stakeholders; and more rapid feedback and dissemination.
Interventions to increase physical activity are a core part of public health policy and practice [1,2,3,4], yet the complexity of public health interventions, which are often multi-component and multi-sectoral, inevitably leads to complexity in terms of their implementation and evaluation [5, 6]. Nevertheless, it is essential that we understand if and how these interventions are effective and act upon this evidence if we are to meet targets for increasing physical activity at the population level, including the World Health Organization Global Action Plan target for a 15% reduction in physical inactivity by 2030 .
Evidence-based public health aims to ensure that decisions and interventions are based on sound evidence to safeguard and improve the health of the population. Appropriate evaluation is central to the generation of this evidence [7,8,9,10]. One of the key challenges is to generate practice-relevant evidence, where external validity and adoption into routine practice may be more likely [10,11,12]. Evaluation of ‘real-world’ interventions, implemented as part of normal service delivery or in practice-based settings rather than in a research environment, provides an opportunity to address this challenge. However, this type of evaluation requires careful selection of approaches that are appropriate and feasible within real-world contexts [13,14,15].
Much progress has been made within the field of public health evaluation in the last two decades, and we have a better understanding of the challenges. Examples include limitations in expertise, capacity, and resources within normal service delivery to conduct evaluation, too much focus on operational objectives and outputs, and barriers to knowledge translation [7, 16,17,18,19]. As our understanding of the challenges to evaluation has developed, so too has the guidance available. This includes guidance on methodological approaches, such as theory-based or realist evaluation [20, 21], recommendations for good practice [8, 14, 16, 22,23,24], and specific frameworks to facilitate systematic evaluation [25,26,27]. The application of frameworks and logic models are now commonly recommended to guide the evaluation and reporting of physical activity interventions. However, our own systematic review of evaluation frameworks showed limited use and/or reporting of frameworks in evaluation studies of physical activity interventions . The reasons for this remain unclear.
Further to the concerns regarding the limited use of frameworks, additional gaps remain in our understanding of how to improve evaluation. Previous reviews of health promotion programmes have highlighted a need for a greater consideration of programme theory , investment and planning for evaluation , and a need for multi-level strategies that involve multiple stakeholders [7, 16, 19]. Collaboration with independent experts in evaluation, such as through research-practice partnerships, is recommended as an approach to improve the quality of evaluation, build capacity for evaluation [7, 16, 18, 19, 22], and improve the use of evidence to inform programme development . However, our understanding of the effectiveness of these strategies in practice remains limited [12, 19, 30, 31].
There is a need for research to develop a better understanding of how different factors interact to influence evaluation practice . Lack of insight into these influences may lead to variability in the quality of evaluation and reporting, which limits the generation and use of critical evidence to inform interventions and decisions to improve population health.
In this study, we report the findings of a case study of Sport England’s ‘Get Healthy Get Active’ (GHGA) programme  to explore evaluation practices, and influences on practice, in an applied context. Sport England is the agency in England with primary responsibility for developing grassroots sports and increasing physical activity across England . The GHGA programme was chosen as our case study as it was specifically designed to build an evidence base for the role of sport in increasing physical activity, improving health and reducing health inequalities ; evaluation was therefore a key element of the programme. The GHGA programme exemplifies multi-sectoral and multi-component approaches within public health . We explored the relationships between organisational structures and processes, and evaluation practice. Although we focus on a national programme to increase physical activity, the aim was to produce research findings that were applicable to other health-promotion interventions, particularly those operating in multi-sectoral public health contexts.
To identify the logic of the programme and explore the relationships between programme and project aims.
To explore influences on evaluation practices, including requirements to use a standardised evaluation framework and specific data collection methods.
To appraise whether the programme was effective in generating high quality generalisable evidence that enabled it to meet its aims.
To formulate and discuss implications for the effective commissioning and evaluation of public health interventions.
The GHGA Programme
Through the GHGA programme Sport England funded 33 physical activity projects, 31 projects within two funding rounds and two invited projects, which were delivered between 2013 and 2018 to communities and population groups across England. For clarity, we refer to the GHGA intervention as “the programme” and local, funded interventions as “projects”. Projects were developed, implemented and evaluated in partnership with Local Authorities, charities, Clinical Commissioning Groups and evaluation partners.
The programme provided an opportunity to explore evaluation practices, and to appraise whether strategies intended to facilitate project evaluation were effective. Sport England put in place several funding requirements to support evaluation. All projects were required to engage an independent evaluation partner, either an academic organisation or consultant. Projects were also required to use validated evaluation tools. This included the use of the Standard Evaluation Framework for physical activity interventions (SEF)  to guide project evaluation, the Single Item Physical Activity Measure , a validated tool to screen participants for eligibility for physical activity interventions, and the International Physical Activity Questionnaire (IPAQ)  to measure physical activity at baseline and follow-up.
We applied a collective case study design , using documentary analysis and semi-structured interviews, to conduct an in-depth analysis of multiple sources of evidence from a range of physical activity projects funded by GHGA. Ethical approval was received from the University of East Anglia Faculty of Medicine and Health Sciences Reseach Ethics Committee (REF: 201718–133).
Sampling and data collection for the documentary analysis
Agreement to conduct the research was gained from Sport England. We conducted initial screening of documents provided by Sport England or published on their website, such as the “Project Summaries”, to develop an overview of projects and to identify the lead organisation for each project. Each of the organisations responsible for the 31 projects in the two funding rounds were contacted and asked to share the final project evaluation report along with documents related to the funding application and intervention planning if available. Contact was initially made by email and then by telephone up to three times. All documents were given a unique code to de-identify them prior to importing them into NVivo 12 Pro for analysis.
Sampling and data collection for the semi-structured interviews
For the interviews, we applied purposive sampling to select stakeholders who were involved in the development, delivery or evaluation of the GHGA programme and projects. This included stakeholders with a role in the national programme and the project lead of each organisation who had shared an evaluation report. We applied snowball sampling to identify additional stakeholders, such as evaluation partners and project facilitators. Each stakeholder was contacted up to three times via email or telephone and invited to participate in an interview. We continued sampling until we were confident that the sample was representative of projects across the two funding rounds, and different types of lead organisation, evaluation partnership, and stakeholder role. All participants provided written consent prior to participating in the interview.
We used semi-structured interviews to ensure we obtained data in relation to the objectives yet allow flexibility that may elicit richer data. An interview guide was developed to facilitate practitioner reflection and allow clarification of findings from the documentary analysis. The guide was piloted with one practitioner, however using semi-structured interviews allowed us to be responsive to emerging findings and refine the questions throughout the data collection period in an iterative approach. The guide consisted of 13 open ended questions that explored practitioners’ experiences of the evaluation process, influences on evaluation, barriers and facilitators, and dissemination activities (provided in Additional file 1).
The interview guide was sent to participants in advance to provide them with prompts for reflection prior to the interview. Interviews were conducted face-to-face, by Skype or telephone. One participant communicated their responses via email. Interviews were conducted by the lead author (JF) between May and December 2019 and lasted an average of 46 min (range 25–86 min). Interviews were audio recorded and transcribed verbatim. All transcripts were sent to participants to check and provide the opportunity to add additional comments or clarification. Transcripts were given a unique numerical identifier to de-identify them before being imported into NVivo12 Pro.
Analysis of documents and interview data
To understand the programme aims and logic (objective one) we analysed Sport England’s organisational documentation related to programme design, funding and monitoring, to develop a logic model and pathway diagram. These were refined through interviews and consultation with key stakeholders at Sport England to ensure that our interpretation and representation of the programme was accurate.
To address objectives two and three we applied Framework Analysis [38, 39]. We combined deductive (a priori) and inductive (emergent) approaches to conduct thematic analysis of the documents and interview data. Initial categories and codes were identified a priori. These included codes related to the use and reporting of the SEF criteria, the single-item physical activity measure and the IPAQ. The SEF provides a structured framework to support project design, evaluation and reporting; the 52 criteria included in the SEF are intended to provide guidance on the information required to undertake a comprehensive and robust evaluation . The criteria are grouped into seven sections (Table 1). We used these criteria as codes to guide data extraction and anaylsis, and provide a systematic approach to summarise the projects and their evaluation. Other codes identified a priori were informed by our interview guide and research objectives, for example influences on evaluation design, barriers and facilitators, and dissemination. Through repeated reading and familiarization with the data emergent codes were added, for example reference to additional evaluation methods such as logic models and case studies. The codes were reviewed and organised into categories and sub-themes (by JF) to develop the coding framework and were iterated and agreed with all authors.
We extracted data from NVivo12 Pro into a final analytical framework matrix to systematically synthesise the data by cases and codes. Using the framework we analysed themes by individual cases (funded projects), across different data sources (documents and interviews), and across the whole data set (representing the programme). To explore how evaluation practices had been applied and documented, and to identify influencing factors, we combined data from the documentary anaysis with data from the interviews.
The findings are presented as a narrative synthesis. Firstly, we present the programme’s aim and logic, and then describe how these compare to project aims and characteristics (objective 1). We then present key themes identified as influences on evaluation practices (objective 2). To appraise whether the programme aim of generating evidence had been met (objective 3), we summarise the reported outputs and outcomes from the project and programme evaluation, and map these against the intended outcomes. Finally, we formulate and discuss implications for effective commissioning and evaluation of health promotion interventions (objective 4) within the discussion.
The case study sample
In addition to the programme-level documents provided by Sport England, representatives from 23 out of 31 (74%) projects shared documents, including the final evaluation reports. These documents formed our sample for the documentary analysis. Lead organisations of two projects declined to share reports, and the leads of the remaining projects did not respond, of which two organisations were known to be no longer in operation.
Thirty-five stakeholders participated in an interview, including stakeholders with a role in the development, management or evaluation of the national programme (n = 5), and stakeholders with a role in the design, delivery and/or evaluation of one or more local projects (n = 31). Some stakeholders had held more than one position with differing roles in the programme and projects. The interview sample was representative of 16 different projects; six from the first funding round and 10 from the second round.
Objective one: to identify the logic of the programme and explore the relationships between programme and project aims
The rationale for the programme and its evaluation is shown in a logic model (Fig. 1). A pathway diagram (Fig. 2) shows the contextual factors influencing the programme. The programme was described as a response to a review commissioned by Sport England that highlighted the limited evidence base for the role of sport in tackling inactivity , and to government strategies that sought to increase participation in sport and physical activity among the least active adults [40, 41]. Stakeholders involved in the programme’s design highlighted the desire to build evidence that could support the commissioning of sport interventions to improve physical activity and health. One programme-level stakeholder explained:
“The reason why we did it the way we did it, was because of the lack of the evidence base … so when somebody else does a systematic review we are hoping that there will be at least 33 papers that will come up, if not more, to help answer that question in future”. (stakeholder 1)
Table 2 summarises the aims and key characteristics of the projects. Whilst the primary aim of all projects aligned to the programme aims, projects also reported various secondary aims and objectives. Projects were delivered by a range of organisations and cross-sector partnerships in a range of locations and settings to diverse population groups. Several included multiple components and/or delivery pathways.
The pathway diagram (Fig. 2) shows changes in organisational structures and strategies, as well as organisational learning [34, 34, 43], which influenced programme processes and practices across the two funding rounds. A key factor was the shift to Local Authority Health and Well-being Boards and Clinical Commissioning Groups being made accountable for Public Health commissioning in England from 2013, which informed an additional funding requirement for projects to address local needs and gain approval from Local Health and Well-being Boards in Round Two; a change which is reflected in the target populations and objectives of those projects.
Objective Two: Influences on Evaluation Practices.
We identified five main themes describing factors that influenced evaluation practices: (1) programme and project design; (2) evaluation design; (3) partnerships; (4) resources; and (5) organisational structures and systems. Examples of how various factors within these themes can act as barriers or facilitators to evaluation are shown in Table 3, and explored further below. The data highlighted the complex inter-connections between influences, and how many influences can act as both facilitators and barriers depending on the project characteristics and context.
Programme and project design
Evaluation was shaped by the programme and project design. The choice and use of evaluation and data collection methods within projects was determined by programme and project objectives and outcomes of interest. However, these also needed to be adapted to the contexts and characteristics of the projects. Within this theme we identified four sub-themes of important influences on evaluation: timescales, participant demographics, settings, and implementation.
Timescales were seen as a barrier to data collection and to formative work. For example, short lead-in times impacted participant recruitment, ability to pilot evaluation methods, and to develop and embed data collection systems. Stakeholders noted that it took time to build relationships with delivery partners and to recruit participants. Timescales related to funding, project conclusion and outcome review were also felt to be a barrier to project sustainability. For example, stakeholders commented:
“the main thing was that lead in time, and I think the second thing is that it takes time to set up the project especially in these hard to reach communities and I think you can't underestimate how much time it takes to build those relationships with the participants, community groups, with the referrers…so it is how we can move away from that two to three years funding cycle, with the reality that it probably takes a year to two years to build relationships in the community and then you are taking that intervention away.” (stakeholder 15)
“I think there was sometimes a lack of time to actually pilot test some of the data collection instruments and processes because the projects are under pressure to start delivering as quickly as possible. And if we had had that time we might have maybe done things differently or refined things before we actually started to ensure it all went smoothly.” (stakeholder 21)
Participant demographics also influenced the outcomes of interest and how data were collected. Stakeholders described the importance of adapting data collection methods, project design and activities, to facilitate recruitment and data collection with specific demographic groups.
Project locations, settings and contexts, including resource availability and accessibility for participants, further impacted recruitment, implementation and response rates. The need for flexibility and adaptability was a recurring theme. This was linked to changes to projects during implementation, such as: staffing and promotional material; adding or tailoring activities and engagement opportunities; and refining eligibility criteria or referal processes. Flexibility in both project and evaluation implementation were described as essential to facilitate data collection, whilst also being a potential barrier to the generalisability of outcomes.
Evaluation design was shaped primarily by the requirements to use standardised data collection tools and a standard evaluation framework. In addition to these required elements, projects reported on a wide range of study designs, evaluation methods, and data collection tools, as shown in Table 4. As one stakeholder explained:
“There was a big influence there in terms of consistency across the projects across the country … Sport England were a big influence in terms of the IPAQ and the things that they were asking for, but we also had the additional secondary questions that we added into the evaluation that were very much around what do we need locally to evidence that this works … I know that a lot of the academic studies included a process evaluation, but that wasn't a direct output that Sport England were expecting, or they didn't dictate that.” (stakeholder 6)
To illustrate how the application and reporting of required and optional evaluation methods influenced the evaluation in practice these elements are discussed below.
Use of standardised tools
Sport England recommended using the Single Item Measure  to identify inactive participants for eligibility. Sixteen projects reported using this tool. Two projects did not refer to any screening tool, whilst four mentioned using alternative screening tools (Table 4). There was variability in how eligibility criteria were applied, and in the use made of the Single Item Measure; for example four projects used it to assess changes in physical activity over time. Stakeholders reflected on differences in how eligibity criteria and screening tools were applied as a challenge to recruitment and comparability across projects.
Projects were also required to use the IPAQ to collect baseline and follow-up measures. Twenty-two projects reported using IPAQ-short form or IPAQ-E (developed for older people), whilst one project had agreement to use an alternative tool, the Scottish Physical Activity Questionnaire (SPAQ). Sport England also recommended using a single question to assess sport participation; which ten projects referred to.
The use of standardised tools in real-world settings and with specific demographic groups was identified as a key challenge. In particular, stakeholders emphasised the negative effect of data collection burden on recruitment and response rates, and in turn on generalisability. For example, stakeholders described the following challenges in using the IPAQ:
“One of the biggest challenges is taking validated questions and looking at the practicality of implementing them in the community.” (stakeholder 15)
“They were a fairly lengthy questionnaire for the type of people we were working with and it led to a real reduction in numbers. The evaluation led to the reduction in numbers. The reduction in numbers was because of the way the evaluation was working but to make the evaluation effective we needed more people. So it was a bit of a vicious circle.” (stakeholder 19)
Use and reporting of the standard evaluation framework
The purpose of including the use of the essential SEF criteria as a funding requirement was to facilitate standardised evaluation and reporting. According to one programme-level stakeholder its strength was in the guidance on reporting contextual factors that would allow Sport England to “understand what works, for who and how; or what doesn’t.” (stakeholder 1).
Eleven (48%) of the evaluation reports, specifically stated that the evaluation was guided by the SEF. Eleven reports did not refer to any evaluation framework, and one referred to the RE-AIM framework  as guiding the evaluation.
Reporting of the SEF criteria was variable. Tables 5 and 6 summarise which projects reportedon the criteria related to programme details and participant demographics. All projects gave a detailed description of their aims and objectives, recruitment methods, location and setting, and reported on age and gender. Those that targeted specific population groups described these in detail. Quality assurance mechanisms, potential unintended consequences, and costs were reported on by fewer projects. The rationale for the intervention, relevant policy context and health needs assessment were not always differentiated. The SEF recommends the use of a logic model, yet just five reports (22%) provided this.
All projects reported on the timing of data collection at baseline and follow-up. Whilst there was some variation in how impact data were reported, all projects reported on change in self-reported physical activity across time points. Seven (30%) projects reported a comparison of outcomes between intervention and control groups or across demographic, disease-risk, referral or service pathway sub-samples. Details of statistical tests used to analyse physical activity measures and the rationale for their use were reported fully, whilstsixteen (70%) projects reported on limitations and generalisability and ten (44%) reported on how findings were disseminated.
The SEF provides more limited guidance on process evaluation (Table 1). Participant numbers were reported variably based on attendance at at least one session, completion of a 10 or 12 week course, or registration at one-off events or online. One project provided a flow diagram of participant numbers with reasons for drop out. Fourteen (61%) projects combined exit survey and interview data to report on participant satisfaction. Nineteen (83%) projects reported on plans for sustainability. One project included this as a research objective to explore features that may lead to sustainable delivery models. Five (22%) projects described how the delivery model had been developed with sustainability in mind.
Use and reporting of optional evaluation components
Table 4 shows that projects included a range of additional self-report surveys. Nineteen (83%) of the projects conducted interviews and/or focus groups to provide additional understanding and insights about how the projects worked and were received. The choice and use of these methods was influenced by project level stakeholders’ priorities and expertise, but also limitations in the required tools to generate evidence in relation to evaluation objectives.
Several stakeholders reflected on the value of qualitative methods to answer questions about the project, for example:
“there's certain cohorts of people we work with where it’s really hard to collect robust evaluation and actually it's the qualitative that matters and the process. I'd like to see a lot more investment in process evaluation because I think at the moment at this time of system changes, so much transformation going on in the health system, and it’s the processes that are important.” (stakeholder 6)
“I think for us some of the most important information came from the qualitative side.” (stakeholder 15)
Twelve projects provided a separate section or report described as either a process or qualitative evaluation. There was variability in how qualitative methods were applied, analysed and reported. For example, some simply mentioned thematic analysis, whilst others provided details of the coding and method of reporting. Four projects combined different data sources to explore project impementation and contextual factors, whilst eight reported on data as case studies of individual participants, organisations or delivery pathways.
Resources, including staff, time, funding, equipment and facilities, were a major influence on evaluation as shown in Table 3. In particular, the availability and use of resources illustrates how the context and characteristics of each project can affect how factors interact and can act as both facilitators and barriers. For example staffing was essential for data collection and evaluation, and depended on the roles, responsibilities and capacity of partners, which in turn were dependent on organisational staffing structures, funding levels and time-scales. Stakeholders from some projects regarded the level of funding as enabling a more rigorous evaluation process than is often possible within real-world interventions, whilst stakeholders from other projects highlighted limited funding as a barrier to their ability to resource the evaluation.
Partnerships shaped the nature of project evaluations. All projects were required to have an independent evaluation partner, and were developed and implemented through working with a range of delivery and funding partners. Evaluation partners were central to the evaluation design. Whilst some stakeholders reflected on differing objectives, priorities and understanding between research and practice as potential sources of tension, most highlighted access to expertise, and in some cases access to additional resources for evaluation as a benefit.
Variation in the responsibilities, priorities and capacities of staff employed by delivery organisations and evaluation partners was thought to have impacted the evaluation design and process. Delivery staff were seen as essential to recruitment and managing data collection. Defining responsibilities, communication, and training were seen as vital to build capacity,and to get buy-in to the evaluation process. As shown in Table 3, the nature of the relationships and history of the partnerships were key influences. For example, close relationships and local partnerships enabled regular communication, and facilitated relationship building and sustainable partnerships, whereas arms-length relationships were described as barriers to successful partnerships and evaluation.
Organisational structures, systems and processes
We identified seven sub-themes of influences related to organisational structures, systems and processes: funding systems; staffing structures; systems for communication, monitoring and oversight; processes for capacity building and knowledge exchange; data management systems; wider external influences; and organisational culture and embeddedness of evaluation (Table 3).
Several of these factors are inter-connected, and also underpin factors identifed within the other main themes. For example, whilst defining roles and responsibilities early in the project was essential to successful partnership working and evaluation, this was dependent on appropriate funding and staffing structures. High staff turnover was mentioned as a challenge to evaluation in nine of the reports, and by eighteen of the stakeholders interviewed. Stakeholders felt this was linked to short funding cycles and contracts, and to have negatively influenced continuity, the capacity for evaluation and dissemination. In particular, stakeholders felt that delays in staff recruitment added to the challenges associated with short lead in times; and early departure of staff influenced dissemination and use of evidence. Having a central co-ordinator who could act as a conduit between partner organisations was seen as critical to successful project evaluation in several cases.
As shown in Table 3, various structures and systems that can act as facilitators to evaluation were identified. Examples include: steering groups and service level agreements to enable regular and formal communication and oversight; training and knowledge exchange to build capacity; and data management systems and processes to integrate evaluation within normal service delivery. Stakeholders reflected on the potential for efficiencies from integrated systems and processes, but also on the considerable time and resource implications of developing these and the difficulties in implementing them across multiple project partners and/or components.
A key underpinning theme was the importance of systems to facilitate monitoring, oversight and communication throughout the project planning, implementation and evaluation cycle. However stakeholders reflections on their experiences of these were variable. For example, service level agreements were seen as critical to agreeing and defining responsibilities in some projects, and as limiting flexibility in others. Many stakeholders reflected on the value of networking and knowledge exchange events facilitated by the funding agency, whilst others commented on a lack of such oportunities as a limitation:
“We found the workshops that they held, … actually to get the GHGA projects in a room together was really useful and because you could share the issues that you were having and people understood and you could share ideas and realize how people have overcome them.” (stakeholder 24)
“They were really good at that side of things, they would bring us in and then different projects would speak each time on different topic areas that we would cover in workshop scenarios, that was really good. They did that really well … I think Sport England could make a lot more of the network than they do in terms of avoiding that duplication of effort and resources.” (stakeholder 6)
“I never had a chance to talk to anyone else who was doing any of the other evaluations so there was never that kind of network and support which I think it might have been quite useful to have had.” (stakeholder 28)
Variability in communication and involvement of stakeholders in networking across different projects appears to have limited the opportunity for a more consistent approach to wider scale knowledge exchange and use of evidence. Some stakeholders also identified a need for organisational structures that enabled forward planning and closer working with local services to ensure that evaluation and evidence generation met future commissioning requirements.
Objective 3: appraisal of whether the programme was effective in generating high quality generalisable evidence that enabled it to meet its aims
Figure 3 provides a summary of project and programme outputs mapped against the intended outcomes included in the logic model (Fig. 1). Two separate evaluation consultancies were commissioned to produce summary reports from Round One and Round Two respectively. At the time of writing, only the reports following Round One were available [34, 43]; these reported numbers of participants engaged in the programme, changes in numbers of participants identified as active or inactive, and case studies of individual projects. Stakeholders at programme and project levels acknowledged the challenges of pooling large data sets from multi-component, multi-sectoral projects due to diverse project designs, settings and participant demographics, and variability in response rates, secondary outcomes, and in how outcome measures were analysed and reported:
“It was good to specify a measure to get the consistency across all the programmes, I guess the quality of that data collection probably varied quite a lot across different projects, depending on who did the data collection and how it was done.” (stakeholder 21)
One programme level stakeholder commented on the need to accept flexibility in how projects applied the specified requirements but that this:
“created a number of challenges at programme level, when you try to pull it all together.” (stakeholder 1)
Programme level stakeholders reported that findings had informed the development of resources to support project and service design and evaluation [44,45,46], and that several project reports had been included in subsequent reviews of practice [47, 48]. In total nine projects disseminated findings through published articles in academic journals, eleven through publicly available reports, and nine through conference presentations. Five stakeholders mentioned plans for publishing articles, but identified a lack of time or time lag between end of project and publication as a challenge.
Project level stakeholders felt the need for knowledge exchange activities and reporting methods that were more appropriate to a wider audience, including local stakeholders and commissioners. Stakeholders involved in projects that had been showcased through best practice projects and conferences saw it as an important way of valueing the project and disseminating findings. Other stakeholders, who had not been involved seemed less aware of dissemination activities beyond what they were doing locally, and were keen to know more about how findings from across the programme were being shared. For example, stakeholders commented:
“I think it is a constant frustration that I have, that there is a huge amount of knowledge that gets built up and then never gets shared.” (stakeholder 31)
“I don't think out of all those projects across the whole network, that was really shared with people. So I think we got to hear more about it because we were part of it. I think where they have done one or two things more recently where they do try and bring people back together where they are all working on similar types of project and I think that's really valuable but I still think they can do a lot more to then share that with the wider network.” (stakeholder 30)
Whilst there was limited understanding amongst some project level stakeholders of how the reports were received, used or shared at the programme level, many described project evaluation as influencing practices, project sustainability or partnerships locally. One programme-level stakeholder commented on learning and capacity building remaining at a project or person level, and fragmentation of projects across multiple organisations, limiting the ability to influence at scale.
The GHGA programme included physical activity projects with a wide range of secondary aims, partnerships, participant groups, settings, and project and evaluation designs. Despite the variability in projects, we identified common influences on evaluation practices that act as facilitators or barriers depending on the context and how they interact within a project. Multiple factors influence programme implementation and evaluation in real-world interventions [16, 19]. This is especially true in multi-sectoral and multi-component programmes such as GHGA. This makes gauging the role of any one factor difficult. Accordingly, our findings highlight the importance of understanding the interactions between influences on evaluation practices and, in particular, the implications for commissioning and evaluation of interventions. Whilst our focus is on physical activity interventions, the findings are applicable to other interventions, particularly those operating in multi-agency public health contexts.
A frequent criticism of real world evaluation has been that evaluation is approached as an “add on” to intervention design and implementation, and that insufficient attention is given to evaluation during intervention planning [7, 16]. Previous studies of health promotion programmes have also identified barriers such as limited investment for evaluation, and differing value placed on evaluation by stakeholders [7, 8, 61, 62]. Within the GHGA programme these barriers were largely overcome by the specification of evaluation as a funding requirement at the outset of the programme. Our study showed the vital role that commissioners play in influencing evaluation practice through resourcing and demands for evaluation, and more critically, in providing appropriate guidance and support, and how they value different forms of evidence.
Stakeholders’ understanding of what counts as evidence, and their use of appropriate evaluation methods, are recognised challenges of conducting real-world evaluation [8, 63,64,65,66]. Evaluation in an applied context often requires a balance to be found between scientific rigour and pragmatism, internal and external validity, and standardisation and adaptability [8, 22]. It can be a challenge to balance differing stakeholder priorities for evidence. The value of combining systematic and flexible approaches [67,68,69], and applying theory based approaches [20, 21, 70] to evaluate the variability within complex interventions is well recognised. Standardised requirements for evaluation of funded projects can facilitate a systematic approach to evaluation and improve the consistency of reporting. This may be particularly important within multi-project programmes like GHGA, which are designed and funded nationally but delivered and evaluated through local projects. We have previously argued that appropriate use of an evaluation framework to guide evaluation and reporting can improve the quality of an evaluation study . Use of a framework can also facilitate identification and agreement of evaluation objectives and methods between stakeholders . Logic models are commonly recommended to identify objectives, inputs, contextual factors and outcomes to help explain an intervention’s theory or rationale [22, 24, 72, 73]; their use is also recommended in the SEF . Qualitative or mixed methods are also advocated to help explain quantitative findings, and generate evidence about project implementation, programme theory or causal mechanisms [14, 24, 29, 69]. Despite putting in place specific evaluation requirements, there was considerable variation in how important evaluation components were applied and reported. Components that were reported in detail, such as project descriptions and participant demographics, reflected the more detailed guidance of these components in the evaluation framework applied. Gaps in the evaluation reports highlighted limitations in the guidance provided in the SEF and the field generally on important evaluation components, and limited the ability to compare or generalise findings across projects. Further guidance or training is needed to improve the evaluation and reporting of specific components, in particular qualitative methods, process evaluation, economic evaluation, logic models, and data analysis. We argue that specifying evaluation requirements alone is insufficient. The context-specific nature of influences within diverse projects makes it more critical to implement processes that facilitate collaborative decision making to select, agree and apply the most appropriate methods to generate the evidence required and valued, rather than specifying standardised data collection across heterogenous projects.
Evaluation partnerships were a strong influence on evaluation. Many of the benefits of partnership working that we identified in this study, such as access to expertise, capacity building, and efficiencies from shared resources or integrated systems were also found in other studies [7, 12, 16, 19]. We also suggest that partnerships can bring greater opportunities for evaluation to be tailored to the needs of individual projects and stakeholders, and to enable a more flexible and innovative evaluation approach. However, the effectiveness of partnerships were dependent on the nature of the relationships, the embeddedness and continuity of partnerships, and on organisational structures and systems. In line with other studies, we also found partnerships to be context specific, and changeable . For funders and partners to initiate and embed processes and systems that facilitate partnerships and that retain benefits of partnership working beyond a projects lifetime, it is essential that we develop a better understanding of the influences of, and on, partnership working.
Our appraisal of the extent to which the programme had generated evidence to achieve its aims (Fig. 3) identified several resources and publications resulting from the programme, but showed that dissemination and use of evidence remains a challenge. At this stage, questions remain as to how useful local project evaluation has been in addressing the programme aim to build an evidence-base that would inform scale up of effective interventions or translation to other settings. The programme sits within a system of evolving national and local policies, strategies and priorities, and knowledge base (Fig. 2). Our findings highlight the importance of rapid feedback to ensure that evidence and insights are disseminated and used to inform policy and practice. Further, we show the importance of thinking forward to the next cycle of project planning and funding to ensure that relevant evidence is generated and used beyond the project. Systems that enable collaboration in the early stages of evaluation planning to identify and agree types of evidence needed and stakeholder engagement throughout the project lifespan are essential. In additition, systems are needed that minimise time lags between project end and dissemination and facilitate knowledge transfer between and beyond projects and partners. The role of research partners is critical in bringing practice-relevant studies to publication , and reviewers and editors also have a role in this. Our study showed that funders and practitioners have a vital role in facilitating and contributing to knowledge-exchange activities. Multi-sectoral and multi-component projects, particularly where projects and evaluation are locally designed and implemented, need appropriate processes and systems to facilitate flows of information between all stakeholders. Without this, fragmentation of projects can lead to fragmentation of learning across organisations and individual stakeholders. In line with other studies [16, 18, 19], we show that cross-sector partnerships and networks appear to offer opportunites to improve knowledge-management and dissemination. Further research is needed to understand their value and how these can be implemented and embeded to help close current gaps in the evidence-based practice cycle.
Our findings have highlighted the important influences of differing stakeholder demands for evaluation, and resources for evaluation, in shaping the design and implementation of intervention evaluation. More critically, it showed the important influence of the underpinning organisational structures and systems, and the complex interactions between influences that act as facilitators or barriers to good practice, even when measures to address known challenges are put in place. Previous studies have identified a need for multi-level strategies to improve evaluation and for more research to understand these [16, 19]; this study supports this view. We argue that stakeholders need to work together to understand, develop and implement systems to enable: (i) collaborative decision making; (ii) synergies between data needed for project delivery, participant engagement, accountability, research and evaluation; and (iii) timely knowledge transfer and dissemination. It is vital to improve our understanding of how influences interact to facilitate or limit good practice within evaluation. This will enable structures and systems to be developed and implemented that capitalise on factors acting as facilitators and that address barriers, and help to ensure that effective interventions are adopted, and that ineffective interventions or unnecessary research are avoided.
Strengths and limitations
A key strength of this study is that we combined data from multiple sources, including evaluation reports and documents from 23 physical activity projects and from the programme as a whole, and data from 35 stakeholder interviews. A further strength is our use of a rigorous and transparent methodology to extract and analyse the data. The logic model that we imputed from the documents was based on the programme aims, objectives and intended outputs reported, and implied outcomes, and was further refined through consultation and interviews with key stakeholders.
There are several limitations of the study. Time lags between end of project delivery and publication mean that our appraisal of the evidence generated could not include the final programme summary evaluation that has been commissioned, and we may have missed additional publications from individual projects. The retrospective nature of the study limited the use of a more ethnographic approach. This may also have contributed to a lower response rate from project organisations and our ability to obtain documents related to project planning and the funding application. This time line also limited our ability to adopt a more collaborative approach to agree the theory of the programme as represented on the logic model.
We identified multiple influences on evaluation practice that can act as barriers and facilitators to good practice. These influences are context-specific and operate through a complex set of interactions. It is vital that commissioners, researchers and practitioners engaged in intervention evaluation or with an interest in improving evaluation and the generation of high-quality evidence, develop a better understanding of these influences and implement appropriate systems and processes to support good practice. Critically, organisational structures, systems and processes are needed to: (i) build and retain individual and organisational capacity for evaluation; (ii) enable collaborative and flexible decision making to identify and agree the most appropriate evaluation objectives, methods and types of evidence; and (iii) improve the transfer of knowledge and insights between stakeholders. This is critical to close current gaps in the evidence-based practice cycle, and ensure that relevant evidence is generated and used in a timely manner.
Availability of data and materials
Documents used to support the findings of this study are publicly available. Other dataset(s) used and analysed during the current study are not publicly available due to them containing information that could compromise research participant consent and anonymity. Data sets are available from the corresponding author on reasonable request, and subject to permission from Sport England.
- BMI :
Body mass index
- CCG :
Clinical commissioning group
- GP :
Get healthy get active
International physical activity questionnaire
- PA :
- RCT :
Randomised controlled trial
- RE-AIM :
Reach, effectiveness, adoption, implementation, maintenance framework
Standard evaluation framework for physical activity interventions
Scottish physical activity questionnaire
- WHO :
World health Organization
World Health Organization. Global action plan on physical activity 2018–2030: more active people for a healthier world. Geneva: World Health Organization; 2018.
Public Health England. Everybody Active, Every Day: An evidence-based approach to physical activity. London: Public Health England; 2014.
Public Health England. A Guide to Community-Centred Approaches to Health and Well-being Full Report. London: Public Health England; 2015. Contract No.: 2014711
Sport England. Towards an Active Nation. London: Sport England; 2016.
Datta J, Petticrew M. Challenges to evaluating complex interventions: a content analysis of published papers. BMC Public Health. 2013;13(1):568.
Petticrew M, Rehfuess E, Noyes J, Higgins JP, Mayhew A, Pantoja T, et al. Synthesizing evidence on complex interventions: how meta-analytical, qualitative, and mixed-method approaches can contribute. J Clin Epidemiol. 2013;66(11):1230–43.
Brug J, Tak NI, Te Velde SJ. Evaluation of nationwide health promotion campaigns in the Netherlands: an exploration of practices, wishes and opportunities. Health Promot Int. 2011;26(2):244–54.
Li V, Carter SM, Rychetnik L. Evidence valued and used by health promotion practitioners. Health Educ Res. 2015;2:193.
Learmonth AM. Utilizing research in practice and generating evidence from practice. Health Educ Res. 2000;15(6):743–56.
Brownson RC, Roux AVD, Swartz K. Commentary: generating rigorous evidence for public health: the need for new thinking to improve research and practice. Annu Rev Public Health. 2014;35:1–7.
Glasgow RE, Emmons KM. How can we increase translation of research into practice? Types of evidence needed. Annu Rev Public Health. 2007;28(1):413–33.
Harden SM, Johnson SB, Almeida FA, Estabrooks PA. Improving physical activity program adoption using integrated research-practice partnerships: an effectiveness-implementation trial. Transl Behav Med. 2017;7(1):28–38.
Northridge ME, Metcalf SS. Enhancing implementation science by applying best principles of systems science. Health Res Policy Syst. 2016;14(1):74.
Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. Br Med J. 2008;337:a1655.
Craig P, Cooper C, Gunnell D, Haw S, Lawson K, Macintyre S, et al. Using natural experiments to evaluate population health interventions: new Medical Research Council guidance. J Epidemiol Community Health. 2012;66(12):1182–6.
Lobo R, Petrich M, Burns SK. Supporting health promotion practitioners to undertake evaluation for program development. BMC Public Health. 2014;14:1315.
Schneider CH, Milat AJ, Moore G. Barriers and facilitators to evaluation of health policies and programs: policymaker and researcher perspectives. Eval Program Plann. 2016;58:208–15.
Francis LJ, Smith BJ. Toward best practice in evaluation: a study of Australian health promotion agencies. Health Promot Pract. 2015;16(5):715–23.
Schwarzman J, Bauman A, Gabbe B, Rissel C, Shilton T, Smith B. Organizational determinants of evaluation practice in Australian prevention agencies. Health Educ Res. 2018;33(3):243–55.
Pawson R, Tilley N. Realistic Evaluation. London: Sage Publications Ltd; 1997.
Brousselle A, Buregeya J-M. Theory-based evaluations: framing the existence of a new theory in evaluation and the rise of the 5th generation. Evaluation. 2018;24(2):153–68.
Pettman TL, Armstrong R, Doyle J, Burford B, Anderson LM, Hillgrove T, et al. Strengthening evaluation to capture the breadth of public health practice: ideal vs. real. J Public Health. 2012;34(1):151–5.
Smith RD, Petticrew M. Public health evaluation in the twenty-first century: time to see the wood as well as the trees. J Public Health. 2010;32(1):2–7.
Moore G, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process Evaluation of Complex Interventions. Oxford: MRC Population Health Research Unit; 2015.
Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health promotion interventions: the RE-AIM framework. Am J Public Health. 1999;89(9):1322–7.
Cavill N, Roberts K, Rutter H. Standard evaluation framework for physical activity interventions. Oxford: National Obesity Observatory; 2012.
Centers for Disease Control and Prevention (CDC). Framework for program evaluation in public health. 1999;48(RR-11).
Fynn JF, Hardeman W, Milton K, Murphy J, Jones A. A systematic review of the use and reporting of evaluation frameworks within evaluations of physical activity interventions. Int J Behav Nutr Phys Act. 2020;17(1):1–17.
Jolley G. Evaluating complex community-based health promotion: addressing the challenges. Eval Program Plann. 2014;45:71–81.
Denford S, Lakshman R, Callaghan M, Abraham C. Improving public health evaluation: a qualitative investigation of practitioners' needs. BMC Public Health. 2018;18(1):190.
Bowen S, Zwi AB. Pathways to “evidence-informed” policy and practice: a framework for action. PLoS Med. 2005;2.
Sport England. New £5 million drive to help tackle inactivity. [Internet] London: Sport England; 2014. updated 23 October 2014; Available from https://www.sportengland.org/news/new-5-million-drive-to-help-tackle-inactivity. Accessed 4 Jan 2021.
Sport England. Why we're here. London: Sport England; 2021. [Internet] updated 2021. Available from: https://www.sportengland.org/why-were-here#buildinganactivenation-5887. Accessed 8 Feb 2021.
Cavill N, Adams E, Gardner S, Ruane S. Tackling inactivity what we know: key insights from our get healthy get active pilots. London: Sport England; 2016.
Milton K, Bull F, Bauman A. Reliability and validity testing of a single-item physical activity measure. Br J Sports Med. 2011;45(3):203–8.
Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–95.
Crowe S, Cresswell K, Robertson A, Huby G, Avery A, Sheikh A. The case study approach. BMC Med Res Methodol. 2011;11(1):100.
Richie J, Spencer L. Qualitative data analysis for applied policy research. In: Bryman A, Burgess B, editors. Analysing qualitative data. London: Routledge; 1994.
Gale NK, Heath G, Cameron E, Rashid S, Redwood S. Using the framework method for the analysis of qualitative data in multi-disciplinary health research. BMC Med Res Methodol. 2013;13(1):1–8.
Department of Health. Start active, stay active: a report on physical activity from the four home countries’ chief medical officers. London: HM Government; 2011.
England S. Creating a sporting habit for life: a new youth sport strategy. London: DCMS; 2012.
Cavill N, Richardson D, Foster C. Improving health through participation in sport: a review of research and practice. British Heart Foundation Health Promotion Research Group University of Oxford 2012.
England S. Tackling inactivity. London: Sport England; 2016.
Sport England. Tackling Inactivity: The Design Principles 2016 [Available from: https://sportengland-production-files.s3.eu-west-2.amazonaws.com/s3fs-public/tackling-inactivity-design-principles.pdf.
Sport England Evaluation framework: measurement and evaluation guidance tools and resources. London: Sport England Date Unknown. Available from http://evaluationframework.sportengland.org/. Accessed 9 Feb 2021.
Sport England. Applying behaviour change theories: real world examples from the get healthy get active Programme. London: Sport England; 2016.
Varney J, Lawson R, Williams T, Copeland R, Brannan M, Lane A, et al. Moving at scale: promising practice and practical guidance on evaluation of physical activity programmes in the UK. 2018.
Beedie C, Mann S, Copeland R, Domone S. Identifying what works for local physical inactivity interventions. 2014.
University of Hertfordshire Higher Education Corporation. Hertfordshire Evaluation Framework: Hertfordshire Sports Partnership; 2017 [Available from: https://sportinherts.org.uk/app/uploads/2018/05/Hertfordshire-Evaluation-Framework.pdf.
Mansfield L. Resourcefulness, reciprocity and reflexivity: the three Rs of partnership in sport for public health research. Int J Sport Policy Polit. 2016;8(4):713–29.
Mansfield L, Kay T, Anokye N, Fox-Rushby J. Community sport and the politics of aging: co-design and partnership approaches to understanding the embodied experiences of low-income older people. Front Sociol. 2019;4:5.
Anokye N, Mansfield L, Kay T, Sanghera S, Lewin A, Fox-Rushby J. The effectiveness and cost-effectiveness of a complex community sport intervention to increase physical activity: an interrupted time series design. BMJ Open. 2018;8(12):e024132.
Mansfield L, Anokye N, Fox-Rushby J, Kay T. The health and sport engagement (HASE) intervention and evaluation project: protocol for the design, outcome, process and economic evaluation of a complex community sport intervention to increase levels of physical activity. BMJ Open. 2015;5(10):e009276.
Burke A, Jones A, Hughes R, Player E. From evidence to practice: developing best practice guidelines for the delivery of activities to people living with moderate to advanced dementia using a pragmatic observational study. Dementia. 2020:1471301220957805.
Buffin J. The benefits of the Commuinity sports initiative: an evaluation of a 6 month pilot programme to enhance recovery among problematic drug and alcohol users. 2014.
Sanders GJ, Roe B, Knowles ZR, Kaehne A, Fairclough SJ. Using formative research with older adults to inform a community physical activity programme: get healthy, get active. Prim Health Care Res Dev. 2019;20(e60):1–10.
Sanders G. Get healthy get active: prevention is better than care. [Thesis] Omskirk: Edgehill University; 2018.
Adams EJ, Musson H, Watson A, Mason L. Bright spots, physical activity investments that work: workplace challenge. Br J Sports Med. 2018;52(16):1026–8.
Atchinson R, Frith G, Roden A, Copeland RJ, Reece LJ. Active for health Rotherham: be active to stay healthy. Br J Sports Med. 2019;53(16):1036–7.
Cavill N, Richardson D, Faghy M, Bussell C, Rutter H. Using system mapping to help plan and implement city-wide action to promote physical activity. J Public Health Res. 2020;9(3):1759.
Jolley GM, Lawless AP, Baum FE, Hurley CJ, Fry D. Building an evidence base for community health: a review of the quality of program evaluations. Aust Health Rev. 2007;31(4):603–10.
Habicht JP, Victora CG, Vaughan JP. Evaluation designs for adequacy, plausibility and probability of public health programme performance and impact. Int J Epidemiol. 1999;28(1):10–8.
Rychetnik L, Bauman A, Laws R, King L, Rissel C, Nutbeam D, et al. Translating research for evidence-based public health: key concepts and future directions. J Epidemiol Community Health. 2012;66(12):1187–92.
Rychetnik L, Frommer M, Hawe P, Shiell A. Criteria for evaluating evidence on public health interventions. J Epidemiol Community Health. 2002;56:119–27.
Milstein B, Wetterhall S. A framework featuring steps and standards for program evaluation. Health Promot Pract. 2000;1(3):221–8.
Koenig G. Realistic evaluation and case studies: stretching the potential. Evaluation. 2009;15(1):9–30.
Lipsey MW, Cordray DS. Evaluation methods for social intervention. Annu Rev Psychol. 2000;51(1):345–75.
Wagemakers A, Vaandrager L, Koelen MA, Saan H, Leeuwis C. Community health promotion: a framework to facilitate and evaluate supportive social environments for health. Eval Program Plann. 2010;33(4):428–35.
Bauman A, Nutbeam D. Planning and evaluating population interventions to reduce noncommunicable disease risk–reconciling complexity and scientific rigour. Public Health Res Pract. 2014;25(1):e2511402.
Stame N. Theory-based evaluation and types of complexity. Evaluation. 2004;10(1):58–76.
Wimbush E, Watson J. An evaluation framework for health promotion: theory. Qual Effectiveness Eval. 2000;6(3):301–21.
W K Kellogg Foundation. Logic Model Development Guide. Battle Creek Michigan: WK Kellogg Foundation; 2004.
Cavill N, Roberts K, Ells L. Evaluation of weight management, physical activity and dietary interventions: an introductory guide. Oxford: Public Health England; 2015.
This work was undertaken with the permission and support of Sport England. In particular we would like to thank Darcy Hare (Strategic Lead for Evaluation at Sport England) and Suzanne Gardner (National Partnerships Lead Health and Inactivity at Sport England) and all the participants who shared documents with us and participated in this study. We would also like to thank Charlotte Salter for her advice on qualitative research and framework analysis.
The work was undertaken by the Centre for Diet and Activity Research (CEDAR), a UKCRC Public Health Research Centre of Excellence. Funding from the British Heart Foundation, Cancer Research UK, Economic and Social Research Council, Medical Research Council, the National Institute for Health Research, and the Wellcome Trust, under the auspices of the UK Clinical Research Collaboration, is gratefully acknowledged. The funders had no role in any element of this research.
Ethics approval and consent to participate
Ethical approval was received from the University of East Anglia Faculty of Medicine and Health Sciences Reseach Ethics Committee (REF: 201718–133). Consent to particpate was received from Sport England and the lead organisations of all projects included in the case study.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Fynn, J.F., Hardeman, W., Milton, K. et al. Exploring influences on evaluation practice: a case study of a national physical activity programme. Int J Behav Nutr Phys Act 18, 31 (2021). https://doi.org/10.1186/s12966-021-01098-8