Cesarean section and the odds of stillbirth, miscarriage, or ectopic pregnancy—a multinomial logistic regression analysis.

<p>Cohort 1982–2010 (<i>n</i> = 832,996 women).</p>a<p>Number of events of the outcome of interest for each mode of delivery in parentheses. Outcome refers to a composite outcome whereby the odds of pregnancy failure (stillbirth, miscarriage, or ectopic pregnancy) were compared simultaneously using multinomial logistic regression models. The reference group was live birth (<i>n</i> = 514,191). Women with no recorded second pregnancy event (<i>n</i> = 236,425) were not included in this analysis.</p>b<p>Model 1: adjusted for maternal age; maternal origin; previous stillbirth, miscarriage, or ectopic pregnancy; mother's marital status; birth year; and measures of socio-economic status including mother's educational attainment and mother's and father's gross income.</p>c<p>Model 2: adjusted for Model 1 plus medical complications in the first live birth including delivery type (singleton versus twins or more), diabetes, gestational diabetes, placental abruption, placenta praevia, and hypertensive disorders (including eclampsia and preeclampsia).</p>d<p>Model 3: adjusted for Model 2 plus gestational age and birth weight.</p><p>*Where the number of events was less than ten for maternally requested cesarean, these were combined with the elective cesarean group for analyses.