The effects of compulsory health insurance on birth outcomes: evidence from China's UEBMI scheme

Background Despite extensive research concerning the impact of health insurance on the advancement of infant health in developed countries, few studies have adjusted their results for potential confounding due to adverse selection in insurance coverage, wherein those who anticipate a need for health services tend to be the ones that acquire insurance. The presence of compulsory health insurance in China, such as the Urban Employee Basic Medical Insurance (UEBMI) scheme may provide an opportunity to estimate the effect of health insurance on infant health, by reducing the endogeneity problem into insurance due to the adverse selection. The objective is to assess the relationship between UEBMI and infant health outcomes in one sizeable municipal-level obstetrics hospital in Shanghai, East China. Methods Medical records data from the Shanghai First Maternity and Infant Hospital from January 1, 2013 to April 30, 2019 were used to form an analysis dataset of 160,429 live births which was comprised of Shanghai residents with UEBMI coverage (n = 101,153) and women without any insurance coverage (n = 59,276). A propensity score matching approach using conjoint quantile regression and probit regression models was used to eliminate latent endogeneity of UEBMI coverage in order to garner robust results. Further analysis stratified by maternal migrant status was conducted to further assess the sensitivity of the findings to distinct patient subgroups. Results The UEBMI scheme was shown to be associated with improvements in infant birth outcomes. The scheme was associated with: an increase in birth weight of about 30 g (p < 0.001, 95% CI 23.908-35.295). This finding was evident in other five different birth outcomes (premature birth, low birth weight, very low birth weight, low Apgar score, and an abnormal health condition at birth). After stratifying by migrant status, the UEBMI was shown to have a greater effect on migrants compared to local residents of Shanghai. Conclusions Our findings suggest that health insurance coverage for pregnant women, especially for migrants, has the potential to significantly and directly improve infant health outcomes. Further research is required to determine whether these findings can be replicated for other Chinese jurisdictions.