Revisiting the Long-Term Health Effect of China’s 1959-1961 Famine: An Instrumental Variable Approach

Hongwei Xu, University of Michigan

Demographers have tested the provocative fetal-origin hypothesis by using famine as a natural experiment. Empirical identification of casual effects of prenatal famine exposure on later-life health is challenging. Existing evidence supporting the fetal-origin hypothesis from studies of famine is limited partly because of data limitation and/or methodological flaws. In fact, the disruption effect of prenatal exposure to famine is often overshadowed by a selection effect when cohort comparisons are carried out in population-based samples. Capitalizing on the biomarker data from the 2011 China Health and Retirement Longitudinal Study (CHARLS), this study proposes to adopt an instrumental variable strategy to identify long-term effects of prenatal exposure to China’s 1959-61 famine on later-life cardiovascular and metabolic disease risks. Exploratory analysis using the conventional cohort comparison strategy reveals little significant disruption effect, except that for LDL cholesterol. Significant selection effect is present in cardiovascular disease risk as measured by resting pulse.