The extent and distribution of inequalities in childhood mortality by cause of death according to parental socioeconomic positions: A birth cohort study in South Korea

It has been shown that childhood mortality is affected by parental socioeconomic positions; in this article, we investigate the extent and distribution of inequalities across major causes of childhood death. We built a retrospective birth cohort using individually linked national birth and death records in South Korea. 1,329,540 children were followed up to exact age eight from 1995 to 1996 and total observed person-years were 10,594,168.18. Causes of death were identified from death records while parental education, occupation and birth characteristics were identified from birth records. Survival analysis was performed according to parental socioeconomic positions. Cox proportional hazard analysis was done according to parental education and occupation with adjustment of birth characteristics such as sex, parental age, gestational age, birth weight, multiple birth, the number of total births, and previous death of children. Cumulative incidence of mortality by age was obtained through a competing-risk method in each cause according to maternal education. From these results, distribution of inequalities across major causes of death was calculated. In total, 7018 deaths occurred during the eight years and mortality rate was 66.24 per 100,000 person-years. External cause was the most common cause of death followed by congenital malformations, nervous system diseases, perinatal diseases, cancer, respiratory, cardiovascular, infectious and gastrointestinal diseases. For all-cause mortality, hazard ratios (HR) were 1.98 (95% CI: 1.83-2.13) for paternal education, 1.90 (1.75-2.07) for maternal education, 1.40 (1.33-1.47) for paternal occupation and 2.33(1.98-2.73) for maternal occupation (between middle school graduation or lower and university or more for education, between manual and non-manual for occupation). Mortality differentials were found in every cause of death. External cause, respiratory, cardiovascular and infectious diseases showed larger HR than all-cause mortality: 2.20 (1.90-2.56), 2.87 (2.02-4.08), 2.50 (1.67-3.75) and 2.12 (1.43-3.15) respectively according to maternal education. On the contrary, congenital malformations and cancer had smaller HR than all-cause mortality: 1.49 (1.22-1.82) and 1.43 (1.00-2.05) respectively according to maternal education. In all-cause mortality and most of the causes, cumulative incidence of mortality increased rapidly until one or two years after birth and then slowed down. But in external cause and cancer, cumulative incidence of mortality accumulated at a constant pace. Thus, inequalities in these causes of death consistently widened. External cause was the leading cause of overall inequalities and its proportion was 36-42% followed by congenital malformations, respiratory diseases etc. We conclude that there were inequalities of childhood mortality in every major cause of death. External cause was the leading cause of both all-cause mortality and overall inequalities. Public health interventions to reduce inequalities are necessary and external cause should be primarily considered.