Social determinants of mid-life mental health decline: a life-course analysis from the 1970 British Cohort Study.

Mental well-being has been shown to follow a U-shaped pattern across the life course in cohorts studied to date, with psychological distress worsening as people enter their forties. The reasons for this decline remain a puzzle. This study examined the social determinants associated with this phenomenon and their relative importance as predictors of mid-life mental health. We used data from the 1970 British Cohort Study, aged 26, 34, and 42 (N = 6992, 51.5% female). Mental distress was measured with the nine-item Malaise Inventory. We contrasted a group of those whose mental health declined with those exhibiting stable mental health. Random forest (RF) and logistic regression models explored whether decline was predicted by socio-economic and family factors, physical health, and health behaviours (at ages 26, 34, or birth). Social class at birth (variable importance [VI] = 0.027) and income quintile at 34 (VI = 0.027) were the most important predictors in RF analysis. Regression confirmed that working-class background (odds ratio [OR] = 1.53 [1.27-1.85]) and being in the highest-income quintile at 34 (OR = 0.68 [0.53-0.88]) strongly predicted the mental health decline. The predicted probability of decline was two and a half times higher for those with an unfavourable, compared to a favourable, combination of social determinants (28% vs 11%). Importantly, a working-class background was not associated with mental health decline between ages 26 and 34. Early-life social class and income were among the strongest predictors of mid-life psychological distress. Decline was concentrated among those exposed to economic disadvantage, suggesting that this group should be a particular focus of mid-life mental health prevention and support.