Mental illness quietly accounts for one of the largest shares of non-fatal disease burden worldwide, yet its scale is routinely underestimated because it rarely appears in mortality statistics. This landmark dataset — spanning 33 years, 204 countries, and 12 distinct disorders — offers the most granular picture yet of how that burden has shifted, with implications for healthcare systems, insurance structures, and workforce productivity globally.
The GBD 2023 mental health analysis applied Bayesian meta-regression to epidemiological data drawn from a systematic literature review, estimating prevalence, incidence, and disability-adjusted life-years (DALYs) for conditions including major depressive disorder, anxiety disorders, bipolar disorder, schizophrenia, ADHD, autism spectrum disorders, anorexia nervosa, bulimia nervosa, and several others. Because most mental disorders do not directly cause death, the primary burden metric is years lived with disability (YLDs). Mortality was modeled for anorexia nervosa — the eating disorder with the highest case fatality rate — using a Cause of Death Ensemble model to capture years of life lost. Results were stratified by sex, age group, and Socio-demographic Index (SDI) quintile, enabling comparison between high-income and low-to-middle-income contexts across all 21 GBD super-regions from 1990 through 2023.
What distinguishes this iteration from earlier GBD mental health analyses is its capacity to surface trajectory, not just prevalence snapshot. The 33-year time series allows detection of accelerating trends — particularly in younger cohorts and among women — that cross-sectional surveys obscure. The SDI stratification is especially valuable: it tests whether rising mental disorder prevalence is a phenomenon of affluence (better diagnosis, reporting) or genuine incidence increase across economic contexts. A key limitation is that GBD estimates remain constrained by the quality and geographic coverage of primary data; low-SDI regions are systematically underrepresented in the literature, which likely means burden in those areas is undercounted. This is confirmatory and expansionary work rather than paradigm-shifting — the broad contours were known — but the scale and rigor make it an essential benchmark for any serious discussion of global mental health investment.