Jack Homer

Simplified versions of three larger, previously published system dynamics policy simulation models were built and tested. The three models address (a) evolving use of a new medical technology; (b) the U.S. opioid epidemic; and (c) local healthcare system performance and financing. The smaller models produce key outputs and policy results that resemble those of the original larger models, although, of course, in a less thorough and detailed way. The purpose of this exercise was to determine whether simplification might be a broadly applicable strategy for modelers seeking to make their large policy models more transparent. If model simplification can lead to the same central policy findings as a larger version does, then it may serve as an additional tool for learning beyond descriptive documentation, experiential “flight simulators”, and loop dominance analysis of the large model.

Multiple studies, starting with Preston’s work in 1975, have suggested that gross domestic product per capita (GDPPC) is an important explanatory factor for understanding differentials in life expectancy at birth (LEB) in countries around the world. This proposition was tested in the present study using two-period cross-sectional regression across a large number of both advanced and developing countries and 16 socioeconomic factors, including GDPPC. The best-performing regression equations in the periods around 2000 and 2015 included four to six of these factors (government effectiveness, safe sanitation, poverty and contraception, plus, in the circa-2000 period, the Gini index and CO₂ emissions); perhaps surprisingly, these equations did not include GDPPC. The results were examined in greater detail for the world’s 15 most populous countries, helping to identify key drivers of LEB growth for each of these countries from circa 2000 to 2015. The fact that GDPPC drops out of the best equations calls into question the view that economic growth is the correct primary target for nations seeking to increase their average life expectancy.

In America and around the world, one’s chances for well-being depend on systems that are not yet built for everyone to thrive together. Knowing that life expectancy and life evaluation are far below their full potential, with stark injustices by race/ethnicity, we ask: how can the United States make a great stride toward multiracial well-being?. This study explores potential impacts of a federal plan for thriving people and places. We estimate the likely effects of 68 recommendations using ReThink Health’s Thriving Together Model (TTM), revised with new data and new features including a multisolving ratio that accounts for greater cost-effectiveness when a proposed action advances multiple goals at once. The TTM is a previously published system dynamics model that simulates changes over time when community assets (both funding and in-kind resources) are invested in four drivers of population well-being (i.e., Vital Conditions, Belonging and Civic Muscle, Fairness in System Design, and Urgent Services Capacity). All drivers work together through a dynamic structure that influences individual states of thriving, suffering, and life expectancy (overall and by race/ethnicity). The model specifies three reinforcing dynamics, including an “expanding the pie” loop that can increase available assets and improve all four well-being drivers over time. Results reveal a plausible scenario over 25 years in which thriving could rise 20 percentage points, suffering could drop 2.5 percentage points, and average life expectancy could grow by 2.6 years – all from equitable progress across racial/ethnic groups. Every subgroup improves, but the greatest gains would likely be among Black and Hispanic Americans. Sensitivity tests confirm that the model’s conclusions are robust across identified uncertainty ranges. The federal plan points the way toward a just transition for multiracial well-being. It does not require new appropriations or authorities: only the will and wherewithal to bring these recommendations to life.

US life expectancy now lags significantly behind the majority of high-income countries, having grown more slowly since 1980 for reasons that are not evident and have been debated. An exploratory system dynamics model is presented that reproduces the full pattern of US life expectancy from 1960 to the present. Multiple socioeconomic and behavioral factors help to explain the historical pattern, two of them apparently most responsible for the stagnation since 1980: the growth of obesity and the leveling off of growth in social spending. Some of the factors in the model are traced back to earlier causes, and obesity’s growth in particular is traced back to excess growth in private health care spending and its adverse effect on workers’ wages. The model’s base run does a good job of reproducing a variety of historical time series data going back to the 1960s, and counterfactual tests produce plausible results and clarify the model’s main themes. The model may thus be considered a reasonable starting point for more conclusive future modeling of US life expectancy.

US life expectancy now lags significantly behind the majority of high-income countries, having grown more slowly since 1980 for reasons that are not evident and have been debated. An exploratory system dynamics model is presented that reproduces the full pattern of US life expectancy from 1960 to the present. Multiple socioeconomic and behavioral factors help to explain the historical pattern, two of them apparently most responsible for the stagnation since 1980: the growth of obesity and the leveling off of growth in social spending. Some of the factors in the model are traced back to earlier causes, and obesity’s growth in particular is traced back to excess growth in private health care spending and its adverse effect on workers’ wages. The model’s base run does a good job of reproducing a variety of historical time series data going back to the 1960s, and counterfactual testing produces plausible results. The model may thus be considered a reasonable starting point for more conclusive future modeling of US life expectancy.

An existing simplified simulation model of global climate, population, economy, and governance is extended, through statistical and other data analysis, to include several measures of human well-being and population displacement. The revised model is used to explore (in testing to the year 2060) vicious cycles and causal cascades that, some have warned, could lead to acceleration of climate change in the coming decades. Model scenario testing addresses two uncertainties, namely the strength of the effect of climate change on governance erosion, and the strength of the “heat trap” effect from physical tipping points such as ice loss and permafrost thaw. This testing indicates that a strong heat trap effect could very well accelerate climate change, but that even a strong governance erosion effect is unlikely to do so. Governance erosion would hurt human well-being but is probably not much of a climate threat per se. These results are tentative but point to the possibility of formally modeling causal cascades beyond what the larger climate models have done to date.

Summary: Adolescent overweight and obesity (AdOWOB) in Europe has proven to be a persistent and complex problem, and appropriate systems methods may help in evaluating potential policy options. This paper describes the development of a system dynamics model of AdOWOB as part of the EU‐funded CO‐CREATE project. The model was developed using literature and data from the Health Behavior in School‐Aged Children (HBSC) study across 31 European countries. We identified 10 HBSC variables that were included as direct or indirect drivers of AdOWOB in the dynamic model, seven at the level of the individual, and three related to the social environment. The model was calibrated to 24 separate cases based on four gender and perceived wealth segments for each of the five CO‐CREATE countries (The Netherlands, Norway, Poland, Portugal, and the UK) and for Europe overall. Out of 10 possible intervention points tested, exercise, fruit, life dissatisfaction, school pressure, and skipping breakfast were identified as the top five most influential ones across the 24 cases. These model‐based priorities can be compared with the policy ideas suggested by the CO‐CREATE adolescents.

We present a practical guide and step-by-step flowchart for establishing uncertainty intervals for key model outcomes in a simulation model in the face of uncertain parameters. The process starts with Powell optimization to find a set of uncertain parameters (the optimum parameter set or OPS) that minimizes the model fitness error relative to historical data. Optimization also helps in refinement of parameter uncertainty ranges. Next, traditional Monte Carlo (TMC) randomization or Markov Chain Monte Carlo (MCMC) is used to create a sample of parameter sets that fit the reference behavior data nearly as well as the OPS. Under the TMC method, the entire parameter space is explored broadly with a large number of runs, and the results are sorted for selection of qualifying parameter sets (QPS) to ensure good fit and parameter distributions that are centrally located within the uncertainty ranges. In addition, the QPS outputs are graphed as sensitivity graphs or box-and-whisker plots for comparison with the historical data. Finally, alternative policies and scenarios are run against the OPS and all QPS, and uncertainty intervals are found for projected model outcomes. We illustrate the full parameter uncertainty approach with a (previously published) system dynamics model of the U.S. opioid epidemic, and demonstrate how it can enrich policy modeling results.