Baseline statistics of the parents and their offspring included in our study.

Maternal variables are divided into mothers with normotensive (control) pregnancies relative to those who experienced pregnancy-induced hypertension (PIH) or preeclampsia. Values are means (± SD) except where indicated otherwise. P-values from t-tests between normotensive and PIH/preeclampsia cohorts were all between P = 4.867e−14 and PP = 0.135). All data are based on records of the Danish National Patient Registry, spanning the years 1979–2007. Apgar5 score is an evaluation of the neonate’s health five minutes after birth on a scale from 0 to 10, where 10 indicates best possible health. Woman with PIH were on average slightly older, had more previous spontaneous and induced abortions, higher family income, and higher BMI compared to the rest of the sample. These variables were all included in the Cox regressions so their effects were partialled out. Because BMI was only recorded from 2003 onwards, it was not included as a covariate in this analysis but was included in a smaller analysis, of which the results were consistent in direction and magnitude to those in Table S1, suggesting that BMI is not a major explanatory variable.