Reproductive
Considering pregnancies as repeated versus isolated events: An empirical comparison of common approaches across selected perinatal outcomes Shalmali Bane* Shalmali Bane Suzan Carmichael Julia Simard Maya Mathur
Despite published guidance on how to address repeated pregnancies to the same individual, a variety of approaches are observed. While some of these approaches are supported by the chosen research question, others are consequences of constraints inherent to a given dataset (e.g., missing parity). We compared common cohort selection and analytic approaches used for epidemiological research.
Using vital records linked to hospital discharge records for singletons, we created four cohorts: (1) all births (2) randomly selected one birth per individual (3) first observed birth per individual (4) primiparous births. Sampling of births was not conditional on cluster. Study outcomes were severe maternal morbidity (SMM) and preeclampsia/eclampsia, and the independent variables were self-reported race/ethnicity (as a social factor) and systemic lupus erythematosus. We assessed the distribution of maternal characteristics, the prevalence of outcomes, overall and stratified by parity, and risk ratios (RRs). Among all births, we compared RRs from three analytic strategies: with standard inference, cluster-robust inference, and adjusting for parity.
Outcome prevalence was consistently lowest among all births and highest among primiparous births. RRs differed for study outcomes across all four cohorts, with the most pronounced differences between the primiparous-birth cohort and other cohorts. Robust inference minimally impacted the confidence bounds of estimates, compared to the standard inference, (e.g., lupus-SMM association: 4.01, 95% CI 3.54-4.55 vs. 4.01, 95% CI 3.53-4.56), while adjusting for parity slightly shifted estimates.
Researchers should consider the alignment between methods used, sampling strategy, and research question. This could include refining the research question to better match possible inference, considering alternative data sources, and acknowledging data limitations. If parity is an established effect modifier, stratified results should be presented.
