According to the latest data, most countries, territories, and regions worldwide are “off track” to meet the World Health Organization’s target of achieving a relative 15% reduction in physical inactivity by 2030.1,2 Simple interventions, although part of the solution, are insufficient to achieve meaningful impacts.3,4 Rather, it is now well recognized that systems approaches involving coordinated multisectoral actions are needed for addressing complex, persistent, and multicausal problems, such as physical inactivity.2
In Australia, a national project named the “Australian Systems Approaches to Physical Activity (ASAPa)” was undertaken to support the practical implementation of systems approaches for promoting physical activity at the population level. ASAPa spanned the period from 2018 to 20223,5,6 and has evolved into a new phase of systems thinking, with research ongoing under the new project name ASAPa+. ASAPa+ is also national in scope but will initially focus on the State of New South Wales (NSW), the most populous jurisdiction in Australia, where the research team and contributing stakeholders are based.
ASAPa+, like its predecessor (ASAPa), aims to further advance systems approaches to physical activity promotion and to move from theoretical discussion to real-world application. Specifically, whereas ASAPa was located toward the conceptual end of the continuum presented by Bellew et al,3 ASAPa+ moves toward the advanced quantitative end of the continuum that will involve dynamic simulation modeling. System dynamics models (SDMs) are quantified, computer-based representations of complex systems that draw on the best available evidence to allow policy experiments to be conducted to see how a system will behave under different conditions and scenarios before they are implemented in the real world.7 They act as decision support tools to inform more efficient investment of resources.8
Stakeholder engagement can span the different stages of the model-building process but should ideally begin early with conceptual mapping of the problem to produce a shared mental model of the causal pathways contributing to the issue. This is to ensure that the scope of the ultimate model reflects key policy and planning considerations and takes account of the mechanisms by which potential interventions will affect outcomes.8 The participatory process can also enhance stakeholders’ knowledge and understanding of the complex system and its dynamics.
Promoting physical activity needs the joint efforts of relevant stakeholders from multiple sectors.9 Multisectoral action relies on evidence that addresses the priorities of both health and other sectors. Previous literature on the development of SDMs regarding physical activity has found few SDMs that examine the effects of interventions on nonhealth outcomes, and those that do mainly focus on the active transport domain of physical activity.10 Our proposed SDM will address this gap by incorporating multiple domains of physical activity and simulating a range of nonhealth-related outcomes, such as social (eg, social capital), environmental (eg, greenhouse gas emissions and air pollution), and economic (eg, health care cost savings) cobenefits. Through the development and ultimate adoption of this tool for decision making, we aim to enhance cross-sectoral support and boost political will for more coordinated and comprehensive, multisectoral action to increase population physical activity.
Although ASAPa+ is at an early stage of its 5-year term and the development of an SDM is the ultimate goal of this project, in this report, we share our progress so far and the preliminary results of the conceptual mapping phase, describe our future plans, and offer some recommendations for researchers exploring the use of systems approaches for physical activity.
Methods
The development of an SDM typically starts with mapping out at a conceptual level the influences and connections that contribute to the dynamic problem at hand (in this case, physical activity).11 We held a system mapping workshop with diverse stakeholders to develop an initial version of the conceptual maps, followed by interviews with diverse stakeholders to gain additional feedback and input to refine these.
Systems Mapping Workshop
In May 2023, we held a half-day systems mapping workshop in NSW, Australia. Nineteen participants were invited based on the relevance of their roles and sectors to the issue of physical inactivity at the local level. Participants were known to the research team or their colleagues and were selected purposefully to represent the relevant sectors broadly based on the International Society for Physical Activity and Health’s “8 Investments for Physical Activity.”9 Overall, participants represented stakeholders from the education, transportation, urban planning, health, and sport and recreation sectors as well as from research and nongovernment organizations.
Participants were provided with an overview of systems thinking and systems mapping before being convened into 2 groups to identify systemic determinants (ie, drivers) of physical activity and the interactive connections between drivers in adults and young people (ie, children and adolescents). Two skilled and experienced facilitators (Bauman and Freebairn) ensured fair and meaningful participation and the open deliberation of diverse and sometimes conflicting views. Initial hard-copy maps from the workshops were digitally transformed and refined using Kumu systems mapping software12 and then shared with all workshop participants to solicit feedback. Fifteen participants provided suggestions for improvement, such as adding, modifying, or deleting variables and/or links between them, and overall feedback on the presentation of the maps.
Stakeholder Interviews
Following the development of the initial systems maps, we conducted separate 30-minute interviews with 10 additional government policymakers to solicit feedback for further refining the systems maps. The interviewees were senior-level representatives from across the education, transport, urban planning, sport, and treasury sectors. Our existing senior government contacts from the health sector identified and facilitated connections with these interviewees. Interviewees were asked to focus on parts of the map most relevant to their sectors as well as provide overall feedback on the map.
Results
More than 100 variables and their interrelated connections were featured in the 2 systems maps (1 for adults, another for young people), showing the complexity of the physical activity system and the potential drivers contained therein. Five settings emerged from the systems map of adults’ physical activity—social and community, policy, built environment and transportation, health care, and workplace (Figure 1; online version is accessible via https://kumu.io/SusanLuo/active-adults)— and 4 for the systems map of young people’s physical activity—family, school, transportation, and community and environment (Figure 2; online version is accessible via https://kumu.io/SusanLuo/active-kids). The 2 maps shared similarities, such as emphasizing the potential drivers within transportation, community, and built environment sectors; however, the young people’s map had an additional focus on the school setting and the adults’ map on the workplace and health care settings. The participants who jointly created the young people’s map also identified how different factors contributed to various domains of physical activity, including leisure time, transportation-related (eg, active travel to/from school and nonschool-related active travel), and school-based (eg, physical education, school sports/events, and active learning and sports) physical activity.
Systems map of physical activity among adults (18y and older), from the figure notes which follow. The text and arrows in blue were modified based on the feedback from all workshop participants. The text and arrows in purple were modified based on the feedback from the stakeholder interviewees. The ovals represent different settings. Starting from top left and going clockwise, they are: ‘Social and community’, ‘Policy’, ‘Built environment and transportation’, ‘Healthcare’, and ‘Workplace’. For details including text and arrow colors, see the online version at https://kumu.io/SusanLuo/active-adults.
Citation: Journal of Physical Activity and Health 2024; 10.1123/jpah.2024-0359
Systems map of physical activity among young people (5-18 y old), from the figure notes which follow. The text and arrows in blue were modified based on the feedback from all workshop participants. The text and arrows in purple were modified based on the feedback from the stakeholder interviewees. The bold text in red indicates different physical activity domains. The ovals represent different settings. Starting from the top and going clockwise, they are: ‘Family’, ‘School’, ‘Transportation’, and ‘Community and environment’. For details including text and arrow colors, see the online version at https://kumu.io/SusanLuo/active-kids.
Citation: Journal of Physical Activity and Health 2024; 10.1123/jpah.2024-0359
Connections between factors were identified and modified throughout the process. An example of such a connection is the influence of parental awareness of physical activity on parenting style/family norms, which supports the independent mobility of children and eventually leads to their ability to access leisure-time physical activity opportunities on their own and engage in active travel to school (Figure 2).
The interviewees validated the existing configuration and depiction of factors within the maps and offered additional suggestions regarding missing or underrepresented variables and connections. The systems maps were further modified to take account of their suggestions (see texts and arrows in purple in Figures 1 and 2; for further details, see Kumu maps for adults https://kumu.io/SusanLuo/active-adults and young people https://kumu.io/SusanLuo/active-kids). For example, representatives from the education sector emphasized the importance of reflecting key pressure points that affect a school’s ability to maintain the delivery of high-quality physical education. These included teacher workload and staffing availability as well as ongoing physical education teacher training/in-service training, both of which were added as factors to the maps with new resulting connections to other existing factors in the map.
Discussion
This report describes the systems mapping work done so far as part of a collaborative and initial exploration into the interacting influences that affect population physical activity in NSW, Australia, and which serves as a conceptual starting point toward the creation of the SDMs. We consider the participatory codesign process involving researchers, practitioners, community leaders, and policymakers as a key strength of this study. In particular, engaging stakeholders across sectors at different stages of the project (during and after the mapping workshop) in various formats (ie, group discussion, individual written feedback, and one-on-one interviews) has allowed diverse opinions and ways of thinking to be considered and incorporated, which, in return, increases the validity of the SDM structure. Engaging potential end users from an early stage of the project also increases their trust in the model-building process and, therefore, the likelihood that the final SDM will be used by the stakeholders.8
Due to the participatory nature of their development, the systems maps are inherently subjective. As the variables and related connections were derived from the conversations during the workshop, they represent the perspectives of the participants, not the absolute “truth” or an exhaustive representation of all relevant influences on physical activity in NSW. The maps reflect local knowledge and are not necessarily generalizable to other settings and contexts. Even so, the broad elements identified in our maps are consistent with many of the elements reflected in other systems maps developed for physical activity4,13 and may be used as a base for future systems maps. For example, existing systems maps tend to converge around different layers of the socioecological model (eg, interpersonal factors, settings, environments, and policy). A key difference is that we chose not to incorporate intrapersonal variables (eg, motivation and self-efficacy) as our aim was to map systems-level upstream drivers of physical activity (and, particularly, modifiable social and environmental factors that could be tested and addressed using the ultimate SDM). We also developed 2 separate maps for adults and young people, which enabled differences in settings and factors to be more explicitly drawn out for the different population groups.
We have identified several limitations that need to be noted. First, although the maps reflect the collective thinking and discussions during the workshop and stakeholder interviews, this is only a subset of all possible knowledge, and key variables may still be missing. Second, more dynamic elements, such as feedback loops and delays, will need to be added and further explored in subsequent stages of the project. Third, these maps are intended for the urban NSW context and may not be relevant to other settings, such as rural and remote areas in Australia.
In terms of future plans, we will continue to work with stakeholders from different sectors on identifying the priority intervention areas and leverage points as well as outcomes of interest (ie, health and nonhealth), continue to refine the maps or develop submaps for selected intervention areas, and parameterize the systems maps with data from various sources. While the process of data parameterization is subject to the availability of high-quality surveillance data for physical activity and its antecedents, and data on the cobenefits (eg, social, environmental, and economic gains) of relevant interventions, involving diverse stakeholders in the coproduction of the SDM can enhance the pool of data available while also alerting them to gaps that they may be in a position to help address. We also plan to facilitate bidirectional learning sessions wherein we can present versions of the SDM to policymakers as a means of socializing them to the tool and its potential application in decision making as well as to gain feedback to further refine the model. We aim to use reflective evaluation practices (including stakeholder feedback) to investigate whether the participatory modeling process has achieved its aims of building systems literacy among stakeholders, fostering new stakeholder connections to strengthen coordinated and collaborative action, and increasing confidence and capacity among policymakers to use the SDM as a decision-making tool for addressing population physical inactivity.14
Conclusions
Progress toward reducing population physical inactivity has been slow and suboptimal worldwide, with NSW being no exception. Systems approaches are regarded as critical for making meaningful progress. To support the practical implementation of systems approaches for physical activity in NSW, the ASAPa+ project aims to develop SDMs using participatory processes. We consider such participatory processes to be necessary elements for robust systems insights that can lead to more informed decision making for improving population-level physical activity. The SDMs will be unique in modeling the broader societal impacts of physical activity interventions to help foster more coordinated and comprehensive multisectoral action. In this report, we have shared our rationale for using participatory methods for SDM development, our progress and results so far from conceptual mapping, and our future plans in the hope that it will provide guidance to other researchers exploring this field of systems approaches.
Acknowledgments
We thank all the participants and stakeholders who got involved in the workshop and interview stage of the current project. Funding Source: This work was supported by a grant from Australia National Health and Medical Research Council, NSW Ministry of Health.
References
- 1.↑
Strain T, Flaxman S, Guthold R, et al. National, regional, and global trends in insufficient physical activity among adults from 2000 to 2022: a pooled analysis of 507 population-based surveys with 5·7 million participants. Lancet Glob Health. 2024;10:150. doi:
- 2.↑
World Health Organization. Global Action Plan on Physical Activity 2018-2030: More Active People for a Healthier World. 2018.
- 3.↑
Bellew W, Smith BJ, Nau T, Lee K, Reece L, Bauman A. Whole of systems approaches to physical activity policy and practice in Australia: The ASAPa project overview and initial systems map. J Phys Act Health. 2020;17(1):68–73. doi:
- 4.↑
Rutter H, Cavill N, Bauman A, Bull F. Systems approaches to global and national physical activity plans. Bull World Health Organ. 2019;97(2):162–165. doi:
- 5.↑
Bellew W, Nau T, Smith BJ, Ding M, Bauman A. Systems approaches to physical activity: new tools and resources. J Phys Act Health. 2022;19(10):645. doi:
- 6.↑
The Australian Prevention Partnerships Centre: Summative Project Report. https://preventioncentre.org.au/research-projects/employing-physical-activity-to-prevent-chronic-disease/]
- 7.↑
Freebairn L, Atkinson JA, Osgood ND, Kelly PM, McDonnell G, Rychetnik L. Turning conceptual systems maps into dynamic simulation models: An Australian case study for diabetes in pregnancy. PLoS One. 2019;14(6):e0218875. doi:
- 8.↑
Freebairn L, Rychetnik L, Atkinson JA, et al. Knowledge mobilisation for policy development: implementing systems approaches through participatory dynamic simulation modelling. Health Res Policy Syst. 2017;15(1):83. doi:
- 9.↑
International Society of Physical Activity and Health. Eight investments that work for physical activity. 2020. https://ispah.org/resources/key-resources/8-investments/
- 10.↑
Nau T, Bauman A, Smith BJ, Bellew W. A scoping review of systems approaches for increasing physical activity in populations. Health Res Policy Syst. 2022;20(1):104. doi:
- 11.↑
Barbrook-Johnson P, Penn AS.System dynamics. In: Barbrook-Johnson P, Penn AS, eds. Systems Mapping: How to Build and Use Causal Models of Systems. Springer International Publishing; 2022:113–128.
- 13.↑
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. doi:
- 14.↑
Lee GY, Hickie IB, Occhipinti JA, et al. Participatory systems modelling for youth mental health: an evaluation study applying a comprehensive multi-scale framework. Int J Environ Res Public Health. 2022;19(7):4015. doi:
