“We Don’t Have to Do Things the Way They’ve Been Done Before”; Mixed-Method Evaluation of a National Grant Program Tackling Physical Inactivity through Sport

National strategies are needed to continue to promote the broader benefits of participating in sport and organised physical activity to reduce physical inactivity and related disease burden. This paper employs the RE-AIM framework to evaluate the impact of the federally funded $150 million Move it AUS program in engaging inactive people in sport and physical activity through the Participation (all ages) and Better Ageing (over 65 years) funding streams. A pragmatic, mixed-methods evaluation was conducted to understand the impact of the grant on both the participants, and the funded organisations. This included participant surveys, case studies, and qualitative interviews with funded program leaders. A total of 75% of participants in the Participation stream, and 65% in the Better Ageing stream, were classified as inactive. The largest changes in overall physical activity behaviour were seen among socioeconomically disadvantaged participants and culturally and linguistically diverse participants. Seven key insights were gained from the qualitative interviews: Clarity of who, Partnerships, Communication, Program delivery, Environmental impacts, Governance, and that Physical inactivity must be a priority. The Move It AUS program successfully engaged physically inactive participants. Additional work is needed to better engage inactive people that identify as culturally and linguistically diverse, Aboriginal and/or Torres Strait Islander and those that live in disadvantaged communities in sport and physical activities. Tangible actions from the seven key insights should be adopted into workforce capability planning for the sport sector to effectively engage physically inactive communities.


Introduction
Physical inactivity is a major public health and economic concern to global communities [1]. Despite the benefits of physical activity (PA) on population health outcomes (physical, mental, and social), improved community connectedness, and contribution to economic growth [2][3][4], limited evidence exists on population-level strategies to increase PA, particularly in communities most likely to be inactive [5]. In the 2018/19 Federal Government Budget, Sport Australia committed more than $150m to 'Drive national sports participation and PA initiatives to get more Australian's moving more often' through the launch of the Australian roadmap 'Sport 2030', and investment through the Move It AUS grant program [6]. Support for this national plan is evident, with funding in sport identified as one of the eight best investments to tackle the growing inactivity crisis, and sport as a tool to enable active communities has been recognised and endorsed in the Global Action Plan on Physical Activity (GAPPA) [4,7]. More recently, the World Health Organization released the Fair Play advocacy brief, which signposted the necessity for PA to be a priority for all involved stakeholders to reduce the equity gap in physical inactivity [2]. Improved and equitable access to appropriate PA and sport activities must become a priority across society to address growing health inequities, made particularly evident during the COVID-19 pandemic [2,8].
Effectively evaluating the impact of population-level physical activity programs is complex and challenging. A lack of integrated evaluations and weak intervention designs are common, and few evaluations are transparent in their protocol reporting and many fail to assess program reach, omitting key process information required to make a judgement of value and translation [9,10]. The aims of this paper are to present the process and outcomes of a national government-funded national sport grant program 'Move it AUS' for physically inactive Australians using the RE-AIM framework. The paper will report the outcomes of the grant program's Reach, Effectiveness, Adoption, Implementation, and Maintenance of the different components of the program in the real-world contexts of the program's delivery [11]. This evaluation of the Move It AUS funded program will contribute to the evidence base on what works and what does not for reducing physical inactivity within communities and provide key insights on enhancing capability within the sport and recreation sectors to suggest appropriate and inclusive opportunities to recruit new target groups to participate.

Funding Overview
In 2019, Sport Australia announced a $56 million investment into two funding streams, the Participation (all ages) and Better Ageing (BA, over 65 years) streams. Funding focused on engaging inactive target groups in organised sport and PA, including people living with a disability, Aboriginal and Torres Strait Islanders, women and girls, disadvantaged communities, individuals with (or at risk of) long-term conditions, and culturally and linguistically diverse (CALD) and older people.

Evaluation Design
The 'Theory of Change' guided the evaluation methods to inform how the program impacted both the capacity of the sport and PA sector (funded organisations) [12]. A logic model was created for both Participation and BA streams to ensure data collected could appropriately explain whether the program achieved these outcomes ( Table 1). The use of realistic evaluation methods using the RE-AIM framework in this study aimed to understand the reasons for a certain outcome and to provide practice-relevant evidence [13].
A quasi-experimental mixed method design used a pre-post survey, alongside qualitative data collection with funded program operational leaders. Participant characteristics and program designs for the two streams (Participation and BA) differ and have been analysed and reported separately.

Ethics
The University of Sydney ethics committee granted ethics approval for this evaluation (2019/533 and 2020/250). Where required, written informed consent was attained prior to data collection. Surveys were expected to be distributed by the funded organisations to all registered participants before and after their participation in the funded program, or at 6 months post initial registration. Socio-demographic data were collected from participants including postcode, which was used to classify both socioeconomic status using the Socio-Economic Indexes for Areas (SEIFA) [14], and remoteness using the Accessibility and Remoteness Index of Australia (ARIA+) [15]. This survey data were used to inform on the reach, effectiveness, and maintenance aspects of programs in achieving the aims of the funding. The primary outcome of the survey data was meeting PA guidelines status, assessed using the validated single item measures for children 5-17 [16] and adults 18+ years old [17]. The definition for physically inactive were adults who were not completing 30 mins of PA on 5 or more days per week, and for children 60 mins of PA on 7 days per week [4]. Secondary outcomes, including organised sport participation, were aligned where possible with existing validated or accepted measures [18].

Qualitative Interviews (Organisational Outcomes)
All Move It AUS grant funded organisations were invited to participate in a 30-45 min qualitative interview (Appendix A). A purposive convenience sample of organisations was recruited, and a nominated leader from each organisation participated in the interview at the conclusion of the program delivery. Interview questions related to the RE-AIM framework and provided perspectives on the process and outcomes of program delivery (Appendix A). The interviews were conducted and recorded online using Zoom (ZoomVideo Communications Inc., San Jose, CA, USA, 2016).

Data Analysis
Participants' demographic characteristics were calculated using descriptive statistics, including frequencies and proportions. Logistic regression models were used to determine whether the pre/post timepoint was associated with meeting physical activity guidelines and sport participation (at least twice per week). Model 1 is unadjusted and model 2 adjusts for age, sex, language, remoteness, and socioeconomic status. All analyses were performed in SAS Enterprise Guide 9.4 (SAS Institute, Cary, NC, USA).
All audio recordings of the qualitative interviews were transcribed verbatim for analysis by a professional transcription company (Way With Words). Framework Analysis was deemed an appropriate approach to analyse qualitative data due to the systematic nature of the semi-structured interviews [19]. Interview transcriptions were analysed using the Framework Analysis approach in NVivo software (NVivo 12 Plus). Once familiarised with the transcripts, the research team conducted an iterative process that identified codes and sub-codes within the interviews to form a thematic scheme of the data ( Figure 1). The RE-AIM framework was applied to the coded themes to understand the areas in which the grant programs were successful, or could be improved, in achieving the aims of the funding.

Results
In total, 88 diverse organisations were funded to deliver activities from July 2019 to July 2020 (Participation stream, n = 62), or July 2019 to July 2020 (BA stream, n = 26). Due to the impact of the COVID-19 pandemic, 32 of the funded programs were provided an extension beyond the planned completion date in June 2020. The 62 organisations in the Participation stream engaged 495,528 people, with 43,638 participants participating across 26 BA funded programs. A small proportion of participants responded to the pre/post survey, with 3483 (0.8%) and 6687 (15.3%) of Participation and BA stream participants responding to the survey, respectively. The results for each funding stream are presented separately herein, using the RE-AIM framework.

Reach
Funded programs successfully reached target groups ( Table 2) and inactive populations. A total of 75% of participants in Participation programs and 77% of participants in BA programs did not meet PA guidelines at baseline. However, there was low representation in some key target groups within the data set of inactive participants, including Aboriginal and or Torres Strait Islanders (5% Participation, <1% Better Ageing), CALD participants (11% Participation, 11% BA), and those living in outer regional/remote communities (7.6% Participation, 12% BA).

Effectiveness
43% of participants in the Participation stream reported increases in PA behaviours. Participants were 19% (non-significant) more likely to meet guidelines at follow-up, compared to baseline (OR:1.19, 95% CI 0.93, 1.53) ( Table 3). Weekly minutes of PA increased in the Participation stream with an increase from 447.5 min per week of organised sport and PA, to 534.7 mins per week.
There was a decline in the number of participants achieving PA guidelines in the BA programs, with participants 35% less likely to meet guidelines at follow-up, compared to baseline (OR: 0.65, 95% CI 55, 0.76) ( Table 3). This was particularly evident for those in the lowest SEIFA categories and those speaking a language other than English at home (Table 3). However, once engaged in the program, older adults in the BA program that spoke a language other than English at home typically spent more time in the funded activity (115 min) than native English speakers (100 min). Of all the older adults engaged, 27% also reported significant improvements in their balance after participation in the funded program.
The qualitative data also evidenced that targeted approaches to engage and deliver appropriate activities to new target groups was in engagement. Appropriate program design was also necessary to effectively retain participants to achieve PA guidelines through continued participation. The organisations were at different levels of readiness, which impacted the adoption and integration of the program within the organisations' strategies for long term results. Our findings showed that organisations with existing internal buy-in from leaders were more likely be using the support from Sport Australia to scale-up an idea already in place, rather than scoping out a pilot to test the feasibility of a new product (Appendix B). The importance of a strong organisational commitment and the integration of positive internal communication supporting the funded activity was reported as critical to the successful adoption within organisations. Note: "All" column includes those who could not be classified as pre or post.

Implementation
The qualitative analysis of the interviews with program leads was synthesised into seven key insights (Appendix B):

1.
Clarity of who organisations aimed to reach was provided in the Move It AUS grant guidelines informed program design, recruitment, and delivery to overcome barriers specific to the nominated target groups.

2.
Partnerships were recognised as a mode of working synergistically to reach new audiences or provide new offerings designed to reducing physical inactivity through new target groups.

3.
Communication was redefined externally to emphasise the fun, social, and noncompetitive aspects of sport participation and internally to advocate for internal buy-in for the funded activity and new target audience.

4.
Program designs included a traditional or modified sport, the provision of educational or capacity building resources, or a multifaceted approach. High quality program deliverers and program flexibility were central to the effective implementation and adoption of funded programs. 5.
COVID-19 disruptions forced funded organisations to pivot online, which impacted reach and program delivery both positively and negatively. Although this time enabled organisations to reflect on improving key aspects of delivery, it also emphasised the importance social connections in project delivery. 6.
Governance from Sport Australia allowed organisations to try new approaches in recruiting target groups and legitimised internal commitment to these new strategies. 7.
Participation strategies to reduce physical inactivity were recognised as a priority across the sport ecosystem despite competing priorities for resourcing within funded organisations.

Maintenance
There was a 5% increase in participants in the Participation stream that was seen in those that "don't know" whether they will drop out of sport or PA after the program, with a reduction in the proportion of participants who had already dropped out by 8.7%. A total of 91% of participants in the BA stream reported that they were planning to continue their current sports and physical activities at the post time point. This suggests there may be an impact of retention in programs, despite the impact of COVID-19 on participation opportunities.
The sustainability of programs was a common concern within funded organisations (coded under "program delivery" and "governance") ( Figure 1). Most interviewees reported that an extended grant delivery time would be preferable to allow consideration for mechanisms for sustainability. Other solutions presented by organisations included effective partnerships and continued collection of data on the impact of the program to support future grant applications and strategic directions (Figure 1).

Discussion
This study aimed to evaluate how a national grant program reached and engaged inactive communities in sport and physical activity, and to understand the impact on the capability and capacity of the sport sector to meet the needs of inactive communities. This is the first time that a nationally funded grant program in Australia has specifically targeted inactive participants and results suggest that the clear focus and strategic direction of the grant program was successful in recruiting inactive people. Despite the impact of COVID-19, the programs successfully demonstrated how organised sport can reach inactive populations and investment in sport can achieve health outcomes for these populations [5].
Seven key insights were synthesised from the results, providing an improved understanding on what works and what does not when designing and implementing PA and sport initiatives for inactive populations. Investments in "sport and recreation for all" have been listed as one of the Eight Best Investments to reduce physical inactivity [3,7]. This evaluation will inform policies that may better support sporting organisations as health promotion, supporting on this directive.
A major barrier to reducing physical inactivity is the initial engagement and reach to recruit key inactive groups [5,20]. The targeted approach of the Move It AUS grants were based on findings reported by both GAPPA, and through Sport Australia's AusPlay data, which identified key inactive population groups [4,6]. Clarity enabled organisations to create a strategic focus and unified approach that guided all aspects of program delivery, helped identify key partners, and informed communication strategies. Funded organisations tried new communication strategies to reach new target groups that were like those found in other studies, including word of mouth, local marketing, and cross-promotion through partnerships [5,21]. Communication of sport as accessible to all, not just those that have experience in sporting activities is important and challenges perceived barriers to participation. However, methods to better engage primary target groups (including CALD, rural communities, and indigenous Australians) into organised PA and sport programs are still required [2].
Program design should begin with clearly defined target groups and include co-design where possible [5]. Challenging pre-conceived notions of sport as intimidating or out-ofreach through the development of beginner-friendly, non-competitive, and social options was reported as critical in engaging inactive participants. Core components of delivery should also include flexibility in delivery, embedded social opportunities, and skilled and qualified staff. The impact of a champion of the program within the organisation was emphasised by Ooms (et al., 2015) and signposts the importance of identifying and training skilled volunteers to deliver programs to ensure participant engagement and enjoyment [5]. Involvement in the funded PA programs presented an opportunity for community members to connect and create support networks. Sport and organised PA also have added benefits of group delivery, which have a positive impact on social and mental health across the lifespan [22]. Using sport as both a social connector and vehicle to achieve PA targets will be particularly relevant in recovering from the detrimental social, physical, and mental effects of experiencing various lockdowns and periods of social distancing due to COVID-19 [23,24].
There was a reduction in the proportion of participants achieving PA guidelines in the BA stream, most significantly observed in the most disadvantaged and CALD participants. Research has found that the effect of COVID-19 on population participation in PA was not equal, and further work is required to address the widening equity gap in PA, particularly in the return to sport and organised PA after COVID-19 [8,25]. However, our findings also reported that once engaged, CALD communities participating in BA programs engaged for an average of 15 mins longer each week in funded activities than English speakers. This evidences the feasibility of tailored programs as a gateway to achieve physical activity guidelines for inactive minority groups. Barriers to participation are greater for these minority groups, therefore, socioecological models for understanding participation rates may be used to better understand how to design and deliver effectively tailored organised sport and PA [26,27].
Recently released by World Health Organisation, the Fair Play advocacy brief has called for greater cross-sectoral collaboration to better engage and retain inactive target groups [2]. Our evaluation found that despite the complicated process of aligning strategic objectives organisations, partnerships were a critical factor to the success in reaching and delivering sport and PA programs to new target groups. Casey et al. (2011) made the case for long-term commitments in funding strategies and partnerships to provide sustainability, which was echoed from program providers concerned about resources required to maintain delivery [28]. Similarly, Staley et al. (2019) found that addressing inactivity through sport requires collaboration and support across multiple levels of the ecosystem [20]. Crosssectoral collaboration is instrumental in reaching specific inactive target groups and should be embedded in future initiatives to support sustained delivery and fair access to sport and PA opportunities across the lifespan [2,5,28,29].

Limitations
Funded organisations were responsible for disseminating the surveys. Some surveys were not able to be determined at either time point but are included in the aggregate total of respondents. Some organisations modified the data collection methods during program delivery due to unforeseen practicality implications such as language barriers and available resourcing.
Findings would be strengthened in future with information on maintenance and the long-term impact of the Move It AUS grants, particularly in the absence of COVID-19 implications. Participation in the qualitative surveys was voluntary, and the possibility for self-selection bias should be noted.

Conclusions
The strong engagement of inactive people in Move It AUS funded programs demonstrates the success and acceptability of targeted interventions in engaging inactive people to reduce physical inactivity and improve health for all. The sport sector is motivated and mobilised to be part of the solution to physical inactivity, and integration of the seven key insights from this study can inform future policies and opportunities supporting sporting programs for inactive populations in the future. Utilising grant programs to broaden the population engagement throughout sport and enhance the capability of the sport sector is one strategy for increasing population levels of PA and understanding the unique contribution sport makes to our local communities.  Conflicts of Interest: Sport Australia delivered a federally funded grant, Move It AUS, which was evaluated in this paper. This study was funded by Sport Australia to evaluate the impact of this grant. At the time of data collection, there were no other competing interests. At the time of manuscript submission, L.R. is an employee of Sport Australia, however, there are no known vested interests related to this research connected to this position. 12.
What is the target audience for your program and what is the reason for this?
One of the target audiences highlighted in Move It Aus grant applications, or simply inactive population of a specific age group? Explain why that choice was made?

13.
Were you successful in delivering the program to the target group? Funds, engagement of effective deliverers who engage with target market, staff, attitudes of participants

14.
What were three things that worked well and why? Participation rate, conversion to memberships, positive feedback

15.
What challenges or barriers (at least 3) did you come across while delivering the program? Did you overcome them? How will you overcome them in the future? Capacity building of the organization, staff recruitment, enhancement of the sporting area, targeted participation, how does the program fit within the organisational structure etc.

QUESTIONS INTERVIEWER PROMPTS
How does the funded program fit within your organisation?

18.
How does your funded program fit within your organisation? Is it a new program or scaling/alteration of existing program?

19.
Has this project influenced your wider organisation in any way?
If yes, how? Recognition, collaborations, motivation to improve, employment etc.
20. Tell me about the priorities of your organisation? What are your key performance outcomes and how was this program designed to impact these KPIs?
Increased membership, improved public perception of organisation, increased participation of target group etc.

How does your organisation tackle physical inactivity outside of this grant program?
Targeted approaches to increasing participation amongst inactive or disengaged members of public? Or not at all? Why not? Is this the first time this approach has been taken and why?
22. What are your key learnings for you and your organisation from this program?
Implementation issues, target audience difficulties, staff management of the program, how did you keep the participants engaged, how has it impacted your key KPIs and organisational outcomes 23. Does your organization intend to increase the reach of this program? How?
Capacity building -staff, volunteers, type of sports, frequency of program, means to increase participation rate, engagement, study the attitudes of target audience, technological support, collaboration etc.

Your funded program and organisation's role within the global approach to reducing physical inactivity
24. On a scale of 1-10, how important is tackling physical inactivity to your organisation? Self-driven research, funding programs for the inactive, evaluation of programs on improving PA outside of this current evaluation?