748Misclassification of interpregnancy interval attributable to miscarriages/induced abortions: quantifying its impact on preterm births. (2nd September 2021)

• Record Type: Journal Article
• Authors: Tessema, Gizachew; Håberg, Siri E; Pereira, Gavin; Magnus, Maria C
• Is Part Of: International journal of epidemiology. Volume 50(2021)Supplement 1
• Publication Date: 2021-09-02
• Subjects: Epidemiology -- Periodicals; 614.4
• DOI: 10.1093/ije/dyab168.656
• Languages: English
• ISSNs: 0300-5771

Abstract

Abstract: Background: A paucity of studies explored how intervening pregnancy events (miscarriages and induced abortions) contribute to interpregnancy interval (IPI) misclassification and whether this misclassification translates into bias in the estimated effect of IPI on adverse birth outcomes. Methods: We included identified three national registries (Birth Register, the General Practitioner, and Patient Register) to obtain information on registered pregnancies (n = 167, 012), with at least two consecutive pregnancies per mother, occurring between 2008 and 2016 in Norway. We computed two IPIs, one that accounted only two consecutive pregnancies ending after 20 weeks of gestation ( standard IPI definition) and the other that accounted two consecutive pregnancies with various outcomes including miscarriages/induced abortions ( corrected IPI definition). We independently assessed the association between the two IPIs and preterm births (PTB). Log binomial regression analysis was conducted. Relative risks (aOR) and 95% confidence interval (CI) were estimated. Results: In the study, 4.7% of births were PTB. While 11.7% of pregnancies occurred <6 months in the corrected IPI definition, the proportion reduced to 4.4% in the standard IPI definition, with over 60% of IPIs were misclassified. After adjusting for confounders, there was a modest overestimation of risk of PTB for pregnancies following short IPI (<6 months) (RR = 1.76, 95% CI: 1.60-1.95) for the standard IPI definition as compared to the risk of PTB (RR = 1.52, 95% CI: 1.41-1.65) following short IPI in the corrected IPI definition. Conclusions: There was a significant misclassification bias associated with miscarriages/induced abortions which should not be ignored when assessing the effect of IPI on PTB.