Reproductive
Birth spacing recommendations: a one-size-fits-all approach may not be appropriate Julie M. Petersen* Julie Petersen Mahsa Yazdy Anne Marie Darling Martha Werler National Birth Defects Prevention Study
Background: Guidelines strongly advise people who recently gave birth to wait 6 months before attempting the next pregnancy and consider interpregnancy intervals (IPI) of ≥18 months as optimal. We sought to evaluate whether age modifies associations between IPI and preterm birth.
Methods: We utilized data from the National Birth Defects Prevention Study (United States, 1997–2011) restricted to multiparas controls with livebirths in the two most recent pregnancies. IPI between the prior birth and the study pregnancy was categorized as <6, 6–11, 12–17, 18–23 (reference), 24–59, or ≥60 months. The study pregnancy was considered preterm if delivered at <37 0/7 weeks’ gestation. We stratified by age at the prior birth, categorized as <25 (n=2484), 25–29 (n=1626), or ≥30 (n=1209) years. We estimated risk ratios (RR) between IPI and preterm birth using Poisson regression, adjusting for measured potential confounders. We conducted a multiple quantitative bias analysis to adjust for exposure-outcome dependent misclassification and selection bias. We computed e-values for any lower bounds of the RR 95% confidence intervals (LBCI) that were >1.0 after multiple bias adjustment.
Results: Preterm birth risk was highest with <6 months IPI (covariate-adjusted RR point estimates ≥1.3 in all age groups; 25–29 years had the strongest association). Preterm birth risk was lowest with 6–17 months IPI among ≥30 years. Associations tended to move downward but these patterns remained after adjustment for multiple biases, with the greatest attenuation among 25–29 years, although the LBCI for <6 months IPI remained >1.0 (e-value=2.4).
Conclusions: People ≥30 years may benefit from shorter IPI than currently recommended, possibly to avoid increased risks associated with advanced age (e.g., reduced fecundability, chronic health conditions, pregnancy complications). Residual confounding is unlikely to explain the <6 months IPI-preterm birth association among people 25–29 years.
