The Influence of Political and Economic Characteristics on Infant Mortality and Under-Five Mortality Rates

This study intends to understand the influences of political, economic, and social factors as well as health services on infant mortality and under-five mortality in developing countries. This study in particular is interested in whether the political, economic, and social characteristics influence infant mortality and under-five mortality directly or indirectly through health service provision. Since few prior studies examined different domains of influences simultaneously and did not collect longitudinal data in this area of research, this study contributes to this research area by addressing these limitations. This study used a longitudinal ecological study design. Based on the theories of modernization and dependency theory, this study will collect 1980-2009 data of 72 developing countries. The data are from World Development Indicators, World Bank Governance Indicators, UN Commodity Trade Statistics Database, and World Health Statistics. Outcome variables are infant mortality and under-five mortality. Political characteristics include the six dimensions of the World Bank Governance Index: voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and control of corruption. The economic characteristics include GDP per capita, foreign debt, foreign investment, and commodity concentration. The social characteristics include percentage of women finished primary education, ethnic fragmentation, and percentage of people in rural areas. The health service characteristics include physician density, health spending/GDP, percentage of skilled birth attendant, percentage of infant immunization, and percentage of access to safe water The analyses will use mixed models to take into account the repeated measurements in this study. This study will also construct time-lag models to assess whether political, economic, social, and health service characteristics will influence the outcomes five years later. The mediating hypotheses will be examined by the Baron and Kenny (1986) method in which the coefficients of political, economic, and social factors may disappear once health service characteristics were added into the regression models.