Age of onset and cumulative risk of mental disorders: a cross-national analysis of population surveys from 29 countries.
Level 3 - non-randomized controlled study
Cross-national analysis of coordinated population-based cross-sectional surveys
PubMed 37531964 · doi:10.1016/S2215-0366(23)00193-1
What was done
Data were analyzed from 156,331 adult respondents (aged 18 years or older; 85,308 female and 71,023 male) across 32 face-to-face community surveys in 29 countries (12 low- and middle-income, 17 high-income) in the World Mental Health surveys conducted between 2001 and 2022. The structured WHO Composite International Diagnostic Interview was used to determine age of onset, lifetime prevalence, and morbid risk up to age 75 years for 13 DSM-IV mental disorders by sex. Analyses used survey weighting and jackknife repeated replications to adjust for selection probability and geographical clustering.
What was found
Lifetime prevalence of any mental disorder was 28.6% (95% CI 27.9-29.2) for males and 29.8% (95% CI 29.2-30.3) for females. Morbid risk by age 75 years was 46.4% (95% CI 44.9-47.8) in males and 53.1% (95% CI 51.9-54.3) in females. Conditional probabilities of first onset peaked at approximately age 15 years, with a median age of onset of 19 years (IQR 14-32) for males and 20 years (IQR 12-36) for females. The most prevalent conditions were alcohol use disorder and major depressive disorder for males, and major depressive disorder and specific phobia for females.
Why it matters
This study provides updated global benchmarks indicating that approximately half of the general population will experience a diagnosable mental disorder across their lifetime. It underscores that intervention and early detection frameworks must prioritize late childhood, adolescence, and young adulthood.
Limits
The analysis was limited to 13 DSM-IV mental disorders and did not assess ethnicity. Diagnoses and ages of onset depended on retrospective recall by respondents aged 18 and older, making findings vulnerable to recall inaccuracies, telescoping, and survival bias.