Background Mental health problems tend to run in families, with studies showing transdiagnostic associations across generations. Nevertheless, if these associations were attributable to unmeasured familial (either environmental or genetic) factors that influence both generations, then treating the parental conditions would not break the intergenerational transmission. This study aims to investigate the associations between parental psychiatric conditions and offspring psychiatric, behavioral, and psychosocial outcomes, after controlling for unmeasured familial factors shared by offspring of monozygotic (MZ) twin parents (i.e., cousins). Methods and findings We conducted a children-of-MZ twins study that consisted of 15,603 individuals (born to 7,742 MZ twin parents) born in Sweden between 1970 and 2000, and followed them from their date of birth to the date of the outcome or December 31, 2020, when the offspring were between 21 and 51 years old. The exposures were whether the MZ twin parents were diagnosed with any psychiatric condition, any internalizing condition, or any externalizing condition. The outcomes included register-based psychiatric conditions, behavioral problems, suicide, and psychosocial problems in the offspring. We performed stratified Cox regression for time-to-event outcomes and conditional logistic regression for binary outcomes to compare offspring exposed to an MZ twin parent with psychiatric conditions against their unexposed cousins. We adjusted for the highest parental educational level, maternal and paternal age at childbirth, offspring birth year, offspring sex, and psychiatric diagnosis of the nontwin parent. Offspring of parents with any parental psychiatric condition, internalizing condition, or externalizing condition had significantly higher probabilities for all the psychiatric, behavioral, and psychosocial outcomes, with hazard ratios (HRs) ranging from 1.34 (95% confidence interval CI 1.21, 1.49; p < 0.001) to 2.53 (95% CI 1.96, 3.26; p < 0.001) for time-to-event outcomes and odds ratios ranging from 1.33 (95% CI 1.17, 1.52; p < 0.001) to 1.87 (95% CI 1.63, 2.14; p < 0.001) for binary outcomes. Although these associations attenuated when comparing differentially exposed cousins whose parents were MZ twins (20 out of 27 associations were no longer statistically significant within cousin pairs), associations between broad spectra remained statistically significant. Specifically, across the main analysis and several sensitivity analyses, statistically significant within-twin-family associations remained between any parental psychiatric condition and any offspring psychiatric condition (HR = 1.28, 95% CI 1.13, 1.44; p < 0.001), between parental internalizing conditions and any offspring psychiatric condition (HR = 1.26, 95% CI 1.09, 1.45; p = 0.002), and between parental externalizing conditions and any offspring psychiatric condition (HR = 1.27, 95% CI 1.08, 1.51; p = 0.005). The main limitations of this study were unmeasured confounders not shared by cousins, the lack of diagnostic data from primary care, and limited statistical power for some specific clustered outcomes. Conclusions Although the intergenerational transmission between parental psychiatric conditions and offspring psychiatric, behavioral, and psychosocial outcomes appeared partially attributable to unmeasured familial (environmental or genetic) factors that influenced both generations, there was also evidence of either nonshared factors or direct causal effects. If the latter, then treating parental psychiatric conditions would reduce the risk of psychiatric vulnerability in offspring.
Zhou et al. (Tue,) studied this question.