Risk of early birth by body mass index in a propensity score--matched sample: A retrospective cohort study

Objective: Evaluate the risk of preterm (<37 weeks) or early term birth (37 or 38 weeks) by body mass index (BMI) in a propensity score--matched sample. Design: Retrospective cohort analysis. Setting: California, USA. Population: Singleton live births from 2011--2017. Methods: Propensity scores were calculated for BMI groups using maternal factors. A referent sample of women with a BMI between 18.5 and <25.0 kg/m(2) was selected using exact propensity score matching. Risk ratios for preterm and early term birth were calculated. Main outcome measures: Early birth. Results: Women with a BMI <18.5 kg/m(2) were at elevated risk of birth of 28-31 weeks (relative risk [RR] 1.2, 95% CI 1.1-1.4), 32-36 weeks (RR 1.3, 95% CI 1.2-1.3), and 37 or 38 weeks (RR 1.1, 95% CI 1.1-1.1). Women with BMI =25.0 kg/m(2) were at 1.2-1.4-times higher risk of a birth <28 weeks and were at reduced risk of a birth between 32 and 36 weeks (RR 0.8-0.9) and birth during the 37th or 38th week (RR 0.9). Conclusion: Women with a BMI <18.5 kg/m(2) were at elevated risk of a preterm or early term birth. Women with BMI =25.0 kg/m(2) were at elevated risk of a birth <28 weeks. Propensity score--matched women with BMI =30.0 kg/m(2) were at decreased risk of a spontaneous preterm birth with intact membranes between 32 and 36 weeks, supporting the complexity of BMI as a risk factor for preterm birth.