Editor’s note: This is part of a periodic series of Health Affairs Blog posts discussing the Culture of Health. In 2014 the Robert Wood Johnson Foundation announced its Culture of Health initiative, which promotes health, well-being, and equity. The initiative identifies roles for individuals, communities, commercial entities, and public policy that extend beyond the reach of medical care into sectors not traditionally associated with health. Health Affairs is planning a theme issue in November 2016 that will explore various aspects of the Culture of Health.
The Culture of Health Action Framework, developed recently by the Robert Wood Johnson Foundation (RWJF), focuses on improving the health and wellbeing of all Americans by supporting mobilization for collective change. The framework specifically focuses on four areas:
- Making health a shared value,
- Fostering cross-sector collaboration to improve well-being,
- Creating healthier, more equitable communities, and
- Strengthening the integration of health services and systems.
Moving forward in any of these areas—and their integration—to improve health requires new conceptual ideas as well as novel ways of organizing knowledge, combining insights, and developing solutions.
The main challenge to realizing this vision is determining how best to address the complexity of making all the moving parts work together in tractable ways. In creating a Culture of Health, we need to be able to know, in a timely fashion, what works and what does not work to improve health, why a given strategy succeeded or failed, and what changes are necessary to make meaningful progress.
Can Systems Science Help Advance The Culture Of Health Action Framework?
Systems science is an interdisciplinary field focused on the study of complex systems that are usually characterized by dynamic interdependencies and interactions among system components and outcomes. By combining systems thinking with advanced computational modeling techniques such as system dynamics, agent-based modeling, discrete-event simulation, and network analysis, systems science has the potential to provide insights about the complex connections and interaction among multiple determinants of health; thus, as evidenced by examples described in this post, systems science can play an important role in shaping and informing the implementation of the Culture of Health Action Framework.
Systems science emphasizes a holistic approach toward solving the complex population health challenges that are typically addressed by scientists, policymakers, and practitioners in fragmented ways. For example, it is clear that being physically active is an effective way to prevent obesity; however, an intervention that aims to increase physical activity may not be fully effective if factors such as the walkability or safety of a neighborhood are not considered in its design and implementation. In addition, systems science represents a promising approach to conducting counterfactual studies to assess the intended (or unintended) consequences of an intervention (or a set of alternative interventions) on population health outcomes.
For example, interventions that promote healthy eating may include incentivizing the retail sale of fresh produce, levying taxes on less healthy foods, offering nutrition education courses, or implementing mass media nutritional campaigns; a systems science approach can be useful to evaluate the likely impact of these interventions while taking into account factors such as the socioeconomic and cultural characteristics of a given community and the level of collaboration across sectors. This knowledge can in turn be used in innovative ways to create an environment that fosters collective action to improve health.
Examples Of How Systems Science Can Support A Culture Of Health
We believe that systems science has the potential to facilitate the adoption of the Culture of Health Action Framework by providing clarity around priority areas, encouraging dialogue about a broader view of health, providing citizens and organizations with actionable information to support cross-sector collaboration, and helping communities visualize feasible pathways for improvement and change. Below we describe four recent applications of systems science that are well aligned with the goals of the Culture of Health Action Framework.
Healthy food consumption is driven by many different factors such as food prices, income, personal preferences, and social norms. These factors may vary greatly across different neighborhood contexts. Li and colleagues used an evidence-based, agent-based model together with local data to analyze how nutrition education and the promotion of healthy food consumption through a mass media campaign would impact the consumption of fruits and vegetables across New York City neighborhoods.
Their study showed how modest improvements in social norms regarding health beliefs and taste preferences could substantially increase the proportion of the population consuming two or more servings of fruits and vegetables across different neighborhoods. However, Li and colleagues’ model also predicts that there would be wide variation in the impact of the proposed intervention across different neighborhoods, suggesting the need to take into account multiple interacting factors when designing the intensity of the intervention if the main goal is to achieve equity and meaningfully improve health.
To engage stakeholders and foster cross-sector collaboration, Loyo and colleagues developed an interactive system dynamics model to align efforts in cardiovascular disease prevention in Austin, Texas. By simulating a range of prevention scenarios and analyzing the simulated results, Austin health leaders and other stakeholders were able to identify several priority areas—improving the quality of primary care, considering social marketing against smoking, and reducing air pollution—that had not received enough attention before the modeling exercises. Insights from the system dynamics modeling work were used to align efforts and reassess priorities across different organizations.
Auchincloss and colleagues constructed an agent-based model to analyze how economic segregation, food preferences, and the pricing of healthy foods could lead to income disparities in diet. The modeling results suggest that both favorable prices and healthy food preferences are needed to eliminate income disparities in diet. With this information in hand, the authors identified several potentially effective interventions, including subsidies for healthy foods, nutrition education, and financial incentives for stores selling healthy foods in low-income neighborhoods.
Milstein and colleagues built a system dynamics model to assess the impact of a wide range of national health policies—such as expanding health insurance coverage, improving health care quality, and increasing primary care capacity—on the health and health care costs of the entire United States population over 25 years. Their modeling results highlighted the importance of integrating high-quality, accessible health care with population-based interventions focusing on prevention to improve health and health equity and lower health care costs.
What Are The Challenges To Implementing A Systems Science Approach?
Despite its promise, there are obstacles to using systems science to advance a Culture of Health; these stem from what we would call technical and cultural barriers existing across different sectors. First, it is usually difficult to sustain collaborations among systems science modelers, policymakers, leaders, and other potential model users because of differences in training and divergent perspectives on how to approach systems science problems. For example, systems science modelers tend to focus on methodological rigor and technical feasibility when thinking about a specific health problem, whereas public health practitioners and policymakers may be more interested in finding pragmatic, real-world solutions to this same problem. These differences in problem conceptualization may lead to disagreement during the model development process, which can only be solved through novel ways to promote communication and foster inclusive engagement across disciplines, professions, and sectors.
Second, systems science modeling requires a broad range of data throughout the model development processes (i.e., conceptualization, parameterization, calibration, validation), given that its goal is to understand complex processes and solve vexing problems. This brings to the forefront issues related to data availability and quality. For example, it is clear that a person’s health behaviors (e.g., smoking, diet) are influenced by his/her social network; however, social network data is not commonly available and, as a result, systems science models oftentimes will include simplifying assumptions about the composition and structure of a social network. Also, systems science modelers usually need to combine data from different sources of varying levels of quality to construct a comprehensive model; this inevitably will raise flags for many. To address these issues, systems science modelers must recognize the limitations of the data sources they use, strive to obtain better data, and conduct rigorous model validation to foster trust.
Third, although systems science has been widely used in many other fields such as biological science, physics, and meteorology, it is still a relatively new approach to solving problems and finding solutions that may support collective action in health. The development of a systems science model may seem daunting to people with no prior training in systems engineering or computer programming. There are also misconceptions about the benefits and limitations of systems science modeling, which may lead to unrealistic expectations about what this analytical approach can accomplish. Some people think, for example, that a systems science model can be a crystal ball that predicts the future.
The value of a systems science model is indeed to allow for the assessment of answers to “what-if” questions and to draw insights from different simulated scenarios that can inform a process, pathway, or decision. A well-designed systems science model can be used to make plausible predictions of the future. However, these predictions will inevitably have a level of uncertainty. What is clear is that there is a need to expose professionals across multiple disciplines to systems science methodologies; curriculum development and training should be highly interdisciplinary if the goal is to breach barriers that hinder collaboration across professions and sectors in model building and problem solving.
Finally, funding for interdisciplinary research in general and systems science in particular is very limited. Funding organizations and academic institutions have the tendency to reward field-specific expertise whereas interdisciplinary approaches—such as systems science modeling—do not receive the same level of support. Government agencies as well as philanthropic and scientific advisory organizations are showing some interest in systems science work, but these efforts need to be more consistent and sustained.
Effectively Using Systems Science To Enhance Health
Despite all the challenges identified above, systems science approaches have the potential to bring people together across many sectors to use data, existing knowledge, and advanced modeling for collective action to improve health and wellbeing. The paths to achieving this vision are many and they are all full of dynamic interdependencies, just like other complex problems. The key is to recognize that engagement and collaboration across sectors are necessary. Systems science seems like a promising way to facilitate discussion and mobilization around how to build a Culture of Health.
Research work at the Center for Health Innovation is funded in part by grants from the Robert Wood Johnson Foundation.