Interpregnancy Interval and Birth Outcomes: A Propensity Matching Study in the California Population

Introduction Previous studies that used traditional multivariable and sibling matched analyses to investigate interpregnancy interval (IPI) and birth outcomes have reached mixed conclusions about a minimum recommended IPI, raising concerns about confounding. Our objective was to isolate the contribution of interpregnancy interval to the risk for adverse birth outcomes using propensity score matching. Methods For this retrospective cohort study, data were drawn from a California Department of Health Care Access and Information database with linked vital records and hospital discharge records (2007–2012). We compared short IPIs of < 6, 6–11, and 12–17 months to a referent IPI of 18–23 months using 1:1 exact propensity score matching on 13 maternal sociodemographic and clinical factors. We used logistic regression to calculate the odds of preterm birth, early-term birth, and small for gestational age (SGA). Results Of 144,733 women, 73.6% had IPIs < 18 months, 5.5% delivered preterm, 27.0% delivered early-term, and 6.0% had SGA infants. In the propensity matched sample (n = 83,788), odds of preterm birth were increased among women with IPI < 6 and 6–11 months (OR 1.89, 95% CI 1.71–2.0; OR 1.22, 95% CI 1.13–1.31, respectively) and not with IPI 12–17 months (OR 1.01, 95% CI 0.94–1.09); a similar pattern emerged for early-term birth. The odds of SGA were slightly elevated only for intervals < 6 months (OR 1.10, 95% CI 1.00–1.20, p < .05). Discussion This study demonstrates a dose response association between short IPI and adverse birth outcomes, with no increased risk beyond 12 months. Findings suggest that longer IPI recommendations may be overly proscriptive.