Abstract
This study investigated the association between parental educational level and infant mortality using data from Vital Statistics and Census in Japan. We used the Census data in 2020 and birth and mortality data from the Vital Statistics from 2018 to 2021 in Japan. Data linkage was conducted between birth data and the Census to link the educational level with parents for birth data and between the birth data and mortality data to identify births that resulted in infant mortality. Four educational levels were compared: "junior high school," "high school," "technical school or junior college," and "university." A multivariate logistic regression model was used to investigate an association between parental educational level and infant mortality using other risk factors as covariates. After the data linkage, data on 890,682 births were analyzed. The proportion of junior high school or high school graduates was higher among fathers and mothers for births with infant mortality compared with that among those for births without infant mortality; in contrast, the proportion of university graduates was lower for births with infant mortality than those without infant mortality. Regression analysis showed that mothers with junior high school or high school graduates were significantly and positively associated with infant mortality compared with those with university graduates. As a conclusion, lower educational level in mothers was positively associated with infant mortality, and it was shown that a difference in infant mortality depending on parental educational level existed in Japan.