Medically assisted reproduction and mental health in adolescence: evidence from the UK Millennium Cohort Study.

Abstract

The number and proportion of children conceived through medically assisted reproduction (MAR) is steadily increasing yet the evidence on their mental health in adolescence is inconclusive. Two main mechanisms with opposite effects can explain differences in mental health outcomes by conception mode: while more advantaged parental characteristics could positively influence it, higher parental stress could have a negative influence. Linear and logistic estimations on a longitudinal population-based birth cohort study of 9,897 individuals to investigate whether adolescents conceived through MAR are more likely than naturally conceived (NC) children to experience mental health problems at age 17, as reported by adolescents themselves and their parents. We test whether this association is confounded and/or mediated by parental background characteristics collected when the cohort member was around 9 months old (maternal age, maternal education level, ethnicity, income quintile), family structure variables measured in adolescence (number of siblings in the household at age 15, parental household structure at age 14) or maternal distress at age 14. Children conceived naturally and through MAR self-reported similar mental health outcomes. The only differences between MAR and NC adolescents are in the parental reports, with parents who conceived through MAR reporting their children had 3.82 (95% CI: 1.140 to 11.54) and 2.35 (95% CI: 1.145 to 4.838) higher odds of falling within the high category of SDQ total difficulties and emotional symptoms scales, respectively. The results did not change on adjustment for mediators, such as maternal distress, number of siblings in the household and parental household structure. The results reveal a lack of or small differences in MAR adolescents' mental health outcomes compared to children who were conceived naturally. While the results based on the parental reports could suggest that MAR adolescents are at higher risk of suffering from mental health problems, the differences are small and not supported by adolescents' own reports. The difference between MAR and NC adolescent's parental report might reflect differences in parental concern, their relationship or closeness and can help to reconcile the mixed findings of previous studies.