Under-Five Cause- and Age-Specific Mortality in the US: An Adaptation of the Log-Quadratic Model

Francisco Villavicencio , Johns Hopkins University
Tyler McCormick, University of Washington, Seattle
Jamie Perin, Johns Hopkins Univerisity
Li Liu, Johns Hopkins University

Young children under five years of age are at elevated overall mortality risk with respect to adults, which is closely related to specific causes such as pneumonia and premature birth. The mortality pattern from birth to age five is also expected to vary by cause. However, to date the causes of child death have only been described for broad age categories such as neonates (0-27 days), infants (1-11 months) and children age 12 to 59 months (or among older children) and existing methods have not comprehensively estimated under-five cause- and age-specific mortality (U5-CASM). Previous research has attempted to quantify U5-CASM in China by extending the log-quadratic model originally proposed by Wilmoth and colleagues to estimate all-cause mortality. However, this study is limited by the quality of the Chinese sample registration data, which is likely suboptimal compared to that of many well-functioning vital registration (VR) systems. In this context, our present research has two main goals. First, to investigate U5-CASM patterns in the United States using a high age resolution, with especial focus on the US context, some of its main causes of death such as congenital malformations and preterm births, and the potential effects of the opioid crisis. Second, to validate the extension of the log-quadratic model for U5-CASM using high quality VR data from the US.

See extended abstract

 Presented in Session 29. Estimation Methods