In this study, the association of non-institutional births with neonatal and infant mortality in Japan was investigated as well as sociodemographic characteristics of non-institutional births.

The Vital Statistics data in Japan (the birth data from 2012 to 2021 and mortality data from 2012 to 2022) were used. Births were classified into three types based on the place of birth and birth attendant: institutional births, non-institutional births with a physician or a midwife, and non-institutional births without a physician or a midwife. Modified Poisson regression was used in order to investigate the association between the type of birth and neonatal and infant mortality and the association between sociodemographic characteristics and the type of birth.

A total of 9,422,942 births were used in the analysis. The results of regression analysis investigating an association between the type of birth and neonatal and infant mortality showed that non-institutional births were positively associated with neonatal and infant mortality regardless of the attendance of a physician or a midwife. Furthermore, the results of regression analysis investigating predictors of non-institutional births showed that factors such as non-urban regions, older maternal age groups, and unmarried status of mothers were positively associated with the two types of non-institutional births, and non-Japanese mother was positively associated with non-institutional births without a physician or a midwife.

Non-institutional births were a predictor of neonatal and infant mortality regardless of the attendance of a physician or midwife, and some sociodemographic characteristics have been proven to be predictors of non-institutional births.