School performance in offspring born to mothers with hyperemesis gravidarum

Hyperemesis gravidarum is characterized by severe nausea, vomiting, and potentially undernutrition during pregnancy. Despite affecting up to 3% of all pregnancies, little is known about the impact of hyperemesis gravidarum on the cognitive development of the child.

We aimed to compare the school performance of offspring exposed to hyperemesis gravidarum during pregnancy with their nonexposed counterparts and siblings.

We conducted a population-based cohort study including all liveborn children, born in Denmark between January 1, 1986, and December 31, 2005. We included 1,162,249 offspring, of which 12,720 (1.1%) had been exposed in utero to a maternal diagnosis of hyperemesis gravidarum. The 2 coprimary outcomes were mean grade point average in the final year of elementary school (ninth grade, 15-16 years) and a composite outcome of low school performance, defined as a mean grade point average below 4 or no registered mean grade point average. The difference in mean grade point average was evaluated using uni- and multivariable linear mixed models and differences in proportions of the binary outcomes were compared using uni- and multivariable logistic regression models.

Of the exposed offspring, 29.3% had low school performance compared to 25.7% of the nonexposed group. After adjustment for confounders (birth year, maternal age, multiple pregnancies, parity, maternal education, maternal smoking, and sex), we found that exposed offspring had increased odds of low school performance (odds ratio 1.19, 95% confidence interval 1.14-1.24). Furthermore, the exposed group had a lower mean grade point average of 6.55 compared with 6.78 in the nonexposed group, yielding an adjusted mean difference of -0.18 (-0.22 to -0.14). However, within-sibling analyses, comparing exposed offspring with their unexposed siblings (same mother), revealed no increased odds of low school performance (adjusted odds ratio 0.98 [0.92-1.04]) or difference in mean grade point average (-0.02 [-0.08 to 0.05]).

We found a small association between in utero exposure to hyperemesis gravidarum and low school performance. However, no difference in school performance was observed between exposed and unexposed siblings, indicating that the association could be attributed to genetics or other familial factors rather than hyperemesis gravidarum.