Complex systems and participatory approaches to address maternal health disparities: Findings from a system dynamics group model building project in North Texas.

Focusing on non-Hispanic Black women (NHBW) in North Texas, this study employed participatory system dynamics modeling to explore three hypotheses: (1) stakeholders will conceptualize structural racism is a pervasive macrostructural force that exerts downstream impacts to shape and perpetuate maternal health disparities among NHBW; (2) stakeholders will identify key causal forces and leverage points that exist across levels of influence; and (3) stakeholders will identify complex interactions, in the form of circular causality, that are present among the key causal forces and leverage points that shape NHBW maternal health disparities. Nine participants engaged in a virtual system dynamics group model-building session that focused on eliciting key variables, behavior-over-time graphs (BOTGs), causal loop diagram (CLD), and targets for action. Participants identified 83 key variables. BOTGs included an average of 6.56 notations and time horizons that, on average, started in 1956. The CLD featured 11 reinforcing and seven balancing feedback loops. Eleven targets for action were identified. Structural racism was revealed as a pervasive macrostructural force that shaped maternal health outcomes among NHBW. Key causal forces and leverage points were identified across levels of influence. Finally, feedback loops within the CLD exhibited circular causality.

Trends in Energy Imbalance Gap and Body Weight Status in the Japanese Adult Population: A System Dynamics Approach.

BACKGROUND: The double burden of malnutrition is a growing public health problem in Japan. We estimated the dynamics of the energy imbalance gap (EIG) (average daily difference between energy intake and expenditure) to explain trends in the prevalence of underweight, overweight, and obese Japanese adults. METHODS: We used individual-level data on body height and weight from the National Health and Nutrition Surveys from 1975 to 2015. We calibrated a validated system dynamics model to estimate the EIG for Japanese adults aged 20 to 74 years by survey year, sex, and weight status classified by the body mass index (BMI). RESULTS: The overall EIG for men increased from 2.3 kcal/day in 1975 to 4.7 kcal/day in 1987 and then decreased to 2.3 kcal/day in 2015. The overall EIG for women consistently decreased from 4.3 kcal/day in 1975 to -0.5 kcal/day in 2015. By BMI class, the EIG for men with a BMI of <30 kg/m2 began to decrease around 1990, indicating a deceleration in the prevalence of overweight and obese men. The EIG consistently decreased for women with a BMI of <25 kg/m2 and reached negative values from the late 2000s to early 2010s, indicating a gradual decrease in the prevalence of overweight and obese women. CONCLUSIONS: The dynamics of the EIG were different across sex and weight groups. Public health interventions should target a further decrease in the EIG for normal-weight, overweight, and obese men and a stop in the decreasing trends of the EIG in underweight and normal-weight women.

Modeling US adult obesity trends: a system dynamics model for estimating energy imbalance gap.

OBJECTIVES: We present a system dynamics model that quantifies the energy imbalance gap responsible for the US adult obesity epidemic among gender and racial subpopulations. METHODS: We divided the adult population into gender-race/ethnicity subpopulations and body mass index (BMI) classes. We defined transition rates between classes as a function of metabolic dynamics of individuals within each class. We estimated energy intake in each BMI class within the past 4 decades as a multiplication of the equilibrium energy intake of individuals in that class. Through calibration, we estimated the energy gap multiplier for each gender-race-BMI group by matching simulated BMI distributions for each subpopulation against national data with maximum likelihood estimation. RESULTS: No subpopulation showed a negative or zero energy gap, suggesting that the obesity epidemic continues to worsen, albeit at a slower rate. In the past decade the epidemic has slowed for non-Hispanic Whites, is starting to slow for non-Hispanic Blacks, but continues to accelerate among Mexican Americans. CONCLUSIONS: The differential energy balance gap across subpopulations and over time suggests that interventions should be tailored to subpopulations' needs.

Trends in adults' energy imbalance gaps over two decades in Belgium using system dynamics modelling.

BACKGROUND: The energy imbalance gap (EIG) represents the average daily difference between energy intake and energy expenditure. The maintenance energy gap (MEG) captures the increased energy intake needed to maintain a higher average bodyweight compared with an initial distribution of bodyweight. This study quantified the dynamics of the EIG and MEG over time and across different genders/regions/BMI groups for Belgian adults. METHODS: A validated system dynamics model was adapted to estimate the trends/dynamics of the EIG among different subpopulations over two decades in Belgium. The model was calibrated using data from the six Belgian national Health Interview Surveys (1997, 2001, 2004, 2008, 2013, 2018). RESULTS: EIG was negative for all BMI groups among Belgian females in 2018, implying the start of a decrease in prevalence of overweight/obesity in this subpopulation. However, this was not the case among Belgian males. Flemish and Walloon males had positive EIGs across BMI groups in 2018, however, Brussels' males showed negative EIGs across BMI groups. Flemish and Brussels' females showed negative EIGs across all BMI groups in 2018, while Walloon females showed positive EIGs across almost all BMI groups. According to the MEG, Belgian men consumed (and expended) on average 59 kcal/day more in 2018 than in 1997 to maintain their heavier body weight. The MEG for Belgian women was 46 kcal/day in 2018, triple the MEG in 2004. CONCLUSIONS: The detailed heterogeneous trends of the EIG describe the obesity patterns for different subpopulations in Belgium and could be used to model the differential effects of specific nutrition policies targeting energy intake.

Three Decades of New Zealand Adults Obesity Trends: An Estimation of Energy Imbalance Gaps Using System Dynamics Modeling.

OBJECTIVE: The energy imbalance gap (EIG) captures the average daily excess energy intake and governs the speed of change in body mass. This study quantifies the dynamics of the EIG over time and across different ethnic, sex, and weight groups in New Zealand. METHODS: A novel method in system dynamics was used to estimate the trends/dynamics of the EIG among Pacific, Maori, Asian, and European/other men and women over the past 3 decades in New Zealand. The model was calibrated using data from the national New Zealand Health Survey (1988-2014/15). RESULTS: All ethnic/sex subpopulations except European/other women and Pacific men showed a drop in their EIGs starting in early to mid-2000. For European/other subpopulations, the EIG was positive in 2014/15 but lower than it was in previous years, meaning that the prevalence of obesity still increased but at a slower pace. For the Pacific subpopulation, increasing trends of the EIG across all BMI classes in 2014/15 implied that obesity prevalence for this subpopulation increased at a rate faster than before. Among Asian women, almost all BMI classes showed a negative EIG in 2012 to 2014/15, implying decreasing prevalence of obesity in this subpopulation. Maori populations with obesity showed a negative EIG in 2014/15. CONCLUSIONS: The detailed heterogeneous trends of the EIG explain the obesity patterns for different ethnic, sex, and BMI subgroups in New Zealand and suggest the need for customizing targets/policy interventions for different subpopulations.

The Additional Costs and Health Effects of a Patient Having Overweight or Obesity: A Computational Model.

OBJECTIVE: This paper estimates specific additional disease outcomes and costs that could be prevented by helping a patient go from an obesity or overweight category to a normal weight category at different ages. This information could help physicians, other health care workers, patients, and third-party payers determine how to prioritize weight reduction. METHODS: A computational Markov model was developed that represented the BMI status, chronic health states, health outcomes, and associated costs (from various perspectives) for an adult at different age points throughout his or her lifetime. RESULTS: Incremental costs were calculated for adult patients with obesity or overweight (vs. normal weight) at different starting ages. For example, for a metabolically healthy 20-year-old, having obesity (vs. normal weight) added lifetime third-party payer costs averaging $14,059 (95% range: $13,956-$14,163), productivity losses of $14,141 ($13,969-$14,312), and total societal costs of $28,020 ($27,751-$28,289); having overweight vs. normal weight added $5,055 ($4,967-$5,144), $5,358 ($5,199-$5,518), and $10,365 ($10,140-$10,590). For a metabolically healthy 50-year-old, having obesity added $15,925 ($15,831-$16,020), $20,120 ($19,887-$20,352), and $36,278 ($35,977-$36,579); having overweight added $5,866 ($5,779-$5,953), $10,205 ($9,980-$10,429), and $16,169 ($15,899-$16,438). CONCLUSIONS: Incremental lifetime costs of a patient with obesity or overweight (vs. normal weight) increased with the patient's age, peaked at age 50, and decreased with older ages. However, weight reduction even in older adults still yielded incremental cost savings.

Connecting micro dynamics and population distributions in system dynamics models.

Researchers use system dynamics models to capture the mean behavior of groups of indistinguishable population elements (e.g., people) aggregated in stock variables. Yet, many modeling problems require capturing the heterogeneity across elements with respect to some attribute(s) (e.g., body weight). This paper presents a new method to connect the micro-level dynamics associated with elements in a population with the macro-level population distribution along an attribute of interest without the need to explicitly model every element. We apply the proposed method to model the distribution of Body Mass Index and its changes over time in a sample population of American women obtained from the U.S. National Health and Nutrition Examination Survey. Comparing the results with those obtained from an individual-based model that captures the same phenomena shows that our proposed method delivers accurate results with less computation than the individual-based model.