Overview

Abstract (English)

Background Most people will partner at some point during their lives. Yet little is known about the association between childhood behavior and patterns of long-term romantic partnering in adulthood. Methods In this population-based cohort study, behavioral ratings were prospectively obtained from teachers when children (n = 2,960) were aged 10–12 years – for inattention, hyperactivity, aggression–opposition, anxiety, and prosociality – and linked to their tax return records from age 18 to 35 years (1998–2015). We used group-based based trajectory modeling to estimate the probability of partnership (marriage/cohabitation) over time and multinomial logistic regression models to examine the association between childhood behavior and trajectory group membership. The child’s sex and family socioeconomic background were adjusted for. Results Five distinct trajectories of partnering were identified: early-partnered (n = 420, 14.4%), mid-partnered (n = 620, 21.3%), late-partnered (n = 570, 19.2%), early-partnered-separated (n = 460, 15.5%), and delayed-or-unpartnered (n = 890, 30.0%). Participants in the early-partnered-separated and delayed-or-unpartnered trajectories were more likely to have left high school without a diploma and to have lower earnings and higher welfare receipt from age 18 to 35 years. After adjustment for sex and family background, inattention and aggression–opposition were uniquely and additively associated with increased likelihood of following an early-partnered-separated trajectory, while inattention and anxiety were associated with an increased likelihood of following a delayed-or-unpartnered trajectory. Childhood prosocial behaviors were consistently associated with earlier and more sustained patterns of partnership. Conclusions Children with behavioral problems are more likely to separate or to be unpartnered across early adulthood. This may have consequences for their psychological health and wellbeing and that of their families.

Abstract (French)

Please note that abstracts only appear in the language of the publication and might not have a translation.