The Effect of Changing Education Structure Across Cohorts on Excess Body Weight and Diabetes in A Developing Country: A Longitudinal and Causal Inference Analysis

Abstract

Diabetes is one of the most rapidly growing chronic diseases in the world. The increase in body mass index (BMI) at the population level is the primary driver of this disease, which has contributed to a twofold increase in the global burden of non-communicable diseases (NCDs) and deaths globally from 1990 to 2017. Low and Middle-Income Countries (LMICs) are the most affected by the global rise in obesity and diabetes. Without sufficient and effective prevention strategies, the growth of obesity and diabetes poses severe challenges to sustainable development in LMICs.

LMICs are undergoing a rapid nutritional transition (NT), indicated by an increase in the consumption of processed food containing high levels of fat and sugar, and ultra processed food. These dietary shifts have been linked to rising trends in obesity and type 2 diabetes in LMICs. The rapid NT is occurring concurrently with economic development and changes in sociocultural trends within LMICs populations, creating a unique timing and pace of transition, and susceptibility to obesity and diabetes that is likely to be different from high-income countries.

In this thesis, I provide a different perspective on understanding obesity and diabetes patterns, focusing on a macro-level demographic perspective centred on the interaction between the education transition and nutritional theories through investigation of cohort and individual experiences. This study aims to understand how cohort changes in educational structure are shaping the distribution of high BMI and diabetes over time in an LMIC context during the nutritional transition. Finally, this study identifies the population education transition effect on excess body weight in a LMIC undergoing dynamic macro-level changes. This study is conducted in the Indonesian population as a case study of LMICs.

Using longitudinal analysis, this study finds different BMI trajectories over the life span by sex, with women having a higher likelihood of obesity at mid-adulthood. Educational patterning for the risk of high BMI is observed across birth cohorts, with rapid increases among lower educated people and younger birth cohorts over time. This study identifies the possible shifting effect of education in future, from increasing the risk of obesity to protecting against obesity, led by highly educated women.

BMI trajectories over adulthood strongly predicts the probability of diabetes both in men and women. We identified that rapid decline in BMI over age is linked to a higher risk of diabetes among women, suggesting a long-term effect of undiagnosed diabetes. Greater educational attainment is associated with an increase in the likelihood of diabetes in men, but a higher education level is a protective factor from diabetes in women.

Using a quasi-experimental design, this study finds the population education transition effect on excess body weight with stronger positive causal effect over time. Heterogeneity in the effect of schooling is identified, with a consistent positive effect in men but not in women. Dietary patterns and physical activity link education to the risk of excess body weight. However, I argue that the present finding of a positive effect of education on obesity is likely a temporal effect that can possibly be shifted to negative with changes in the population-level environment. The combination of changing food environments, improving population knowledge on risks of excess weight, and continued educational expansion may prevent the rising excess weight and anticipate future new health risks in the population. Show more