Mass Disasters and Global Child Health in the Demographic and Health Surveys

Abstract

Both natural disaster researchers and policymakers tend to assume that every disaster is unique, and therefore generalizations are not possible. But effect consistency and the appropriateness about generalizing have not been systematically examined. Accordingly, this dissertation tests if disasters have consequences on health and health behaviors in multiple low- and middle-income countries. Manuscript 1 asks, Can epidemiologic methods be fruitfully applied for disaster public health research using existing data? Manuscript 2 asks, What are the health consequences of disasters, and is there evidence that ‘all disasters are unique’ in their effects on child mortality, fever, and diarrheal disease? Manuscript 3 considers the question, Do disasters cause mothers to change their behavior when seeking treatment for child illness? We exploit the Demographic and Health Surveys (DHS) that, by chance, were in the field before and/or after a natural disaster. Analysis shows that disasters are associated with consistent declines in cases of fever and diarrheal disease, but this is not to say that disasters are health-promoting in their own rite. Disasters were not related to change in maternal treatment-seeking behaviors, nor were they related to differences in location where treatment is sought. These particularly severe disasters did not systematically impact maternal health behavior. This study provided evidence that disasters of various types have similar impact on health and behavior. A change in the perception about disasters and global child health may reveal opportunities for research and policy innovations that can reshape how disaster preparedness is administered globally.