Health and Nutrition in the Context of Developing Countries

This dissertation examines three different aspects of population health and nutrition in low- and middle-income countries. Chapter 1 is methodological in nature and investigates the implications of using self-reported anthropometrics as proxies for measured ones. I analyze misreporting patterns of height and weight (and resulting body mass index, or BMI) in China, India, Russia, and South Africa, and find heterogeneity of reporting patterns both between these countries and high-income countries, as well as heterogeneity among these countries themselves. Adjustments of self-reported heights and weights are investigated, and the use of measured, self-reported, and adjusted BMI are compared in various applications. Chapters 2 and 3 study more substantive topics. Less-developed countries have traditionally dealt with issues of stunting, wasting, and underweight, and most resources have been put toward rectifying these. But as these countries continue to work on problems of under-nutrition, over-nutrition has been rising rapidly at rates not historically witnessed before, resulting in a double burden of malnutrition. Chapter 2 examines the population-level double burden by analyzing within- and between- country nutritional disparities among older adults in China, Ghana, India, Mexico, Russia, and South Africa. Using both country-specific and pooled partial proportional odds models, I analyze patterns of BMI categories along various socioeconomic dimensions. I conclude that economic development and the nutrition transition are intertwining processes, and progression through these processes results in shifts of the population segments affected by the two nutritional extremes. In Chapter 3, I use a birth cohort study from Guatemala to study changes in nutritional status over the life course, the double burden at an individual level. With an analysis of transitions, multiple regressions, and structural equation modeling, I find that while early anthropometrics are generally not associated with adulthood BMI, there are direct relationships between childhood nutritional status and growth with some chronic disease indicators, such as triglycerides and fasting blood glucose. Furthermore, these relationships are not mediated by BMI. When taken together, these three chapters have both significant research and policy implications for population health in the developing context