A model to assess the costs and consequences of changes in diet and nutrition from potential population-wide policies

Abstract

INTRODUCTION: The prevalence of diabetes continues to increase - more than 38 million people in the US now have diabetes and 84 million have prediabetes. Because many new cases of incident diabetes may be attributed to suboptimal dietary quality, novel programs and policies to encourage healthy eating choices represent promising population-level approaches to reduce the number of new cases of diabetes.

METHODS: To help estimate the potential impact of such programs and policies, we created the Microsimulation of Nutrition, Diabetes, and Cardiovascular Disease (MONDAC), a model to estimate the impact of simulated population-level dietary changes on downstream outcomes: body mass index, diabetes incidence, cardiovascular disease (CVD) incidence, all-cause mortality, quality-adjusted life years, direct medical costs, and cost-effectiveness.

RESULTS: We used 24-hour recall data from the National Health and Nutrition Examination Survey to categorize food and beverage consumption into 51 mutually exclusive categories to understand the effects of dietary changes. We simulated the energy intake and dietary quality effects that result from increasing, decreasing, or reallocating intake of these 51 food categories. Reductions in calories induce weight loss via an energy balance model. Weight loss and improvements in dietary quality drive annual reductions in diabetes and CVD risk. Mortality was modeled using a lifetable approach, and direct medical care costs were applied using estimates from the literature. We cross-validated MONDAC with existing models to assess reliability of estimates. We provide an example simulation for MONDAC, modeling a reduction in sugar-sweetened beverage consumption at the national level in the US.

CONCLUSION: MONDAC provides a flexible approach to policy analysis to allow the user to simulate various food-related policies.