Fig. 1.

Fixed effects (with 95% confidence intervals) of sibling × proband interaction effects of years of education (left panel), childhood trauma (middle panel), and hazardous alcohol use (right panel) on current depressive and/or anxiety symptom severity in the sibling, while controlling for current symptomatology in the proband. Current depressive and/or anxiety symptom severity was measured as standardized and averaged overall IDS/BAI score. Estimates of simple effects (γ) were retrieved from linear mixed-effects models with a random intercept of ‘Family-ID’ to account for within-family clustering and indicate the (presence/absence and direction of) association between a risk factor and current symptomatology in the sibling for different values of that risk factor in the proband. Low years of education (left panel) was associated with more severe symptoms in the sibling when the proband had high years of education (M + 1∙SD: γ = − 0.05, SE = 0.02, p = 0.007); no associations were found when the proband had average (M: γ = − 0.02, SE = 0.01, p = 0.104) or low years of education (M−1∙SD: γ = 0.01, SE = 0.02, p = 0.781). Higher childhood trauma (middle panel) was associated with more severe symptoms in the sibling and the strength of this association increases for decreasing trauma levels in the proband (M + 1∙SD: γ = 0.03, SE = 0.01, p < 0.001; M: γ = 0.04, SE = 0.005, p < 0.001; M−1∙SD: γ = 0.05, SE = 0.01, p < 0.001). Hazardous alcohol use (right panel) was associated with more severe symptoms in the sibling when the proband was not a user (No: γ = 0.44, SE = 0.11, p < 0.001); no association was found when the proband was a user (Yes: γ = − 0.11, SE = 0.15, p = 0.473). M mean, SD standard deviation, SE standard error