Age- and sex-specific associations between risk scores for schizophrenia and self-reported health in the general population

PURPOSE: The health correlates of polygenic risk (PRS-SCZ) and exposome (ES-SCZ) scores for schizophrenia may vary depending on age and sex. We aimed to examine age- and sex-specific associations of PRS-SCZ and ES-SCZ with self-reported health in the general population. METHODS: Participants were from the population-based Netherlands Mental Health Survey and Incidence Study–2 (NEMESIS-2). Mental and physical health were measured with the 36-item Short Form Survey 4 times between 2007 and 2018. The PRS-SCZ and ES-SCZ were respectively calculated from common genetic variants and exposures (cannabis use, winter birth, hearing impairment, and five childhood adversity categories). Moderation by age and sex was examined in linear mixed models. RESULTS: For PRS-SCZ and ES-SCZ analyses, we included 3099 and 6264 participants, respectively (age range 18–65 years; 55.7–56.1% female). Age and sex did not interact with PRS-SCZ. Age moderated the association between ES-SCZ and mental (interaction: p = 0.02) and physical health (p = 0.0007): at age 18, + 1.00 of ES-SCZ was associated with − 0.10 of mental health and − 0.08 of physical health, whereas at age 65, it was associated with − 0.21 and − 0.23, respectively (all units in standard deviations). Sex moderated the association between ES-SCZ and physical health (p < .0001): + 1.00 of ES-SCZ was associated with − 0.19 of physical health among female and − 0.11 among male individuals. CONCLUSION: There were larger associations between higher ES-SCZ and poorer health among female and older individuals. Accounting for these interactions may increase ES-SCZ precision and help uncover populational determinants of environmental influences on health. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00127-022-02346-3.