Since the introduction of specified diagnostic criteria for mental disorders in the
1970s, there has been a rapid expansion in the number of large-scale mental health
surveys providing population estimates of the combined prevalence of common mental
disorders (most commonly involving mood, anxiety and substance use disorders). In
this study we undertake a systematic review and meta-analysis of this literature.
We applied an optimized search strategy across the Medline, PsycINFO, EMBASE and PubMed
databases, supplemented by hand searching to identify relevant surveys. We identified
174 surveys across 63 countries providing period prevalence estimates (155 surveys)
and lifetime prevalence estimates (85 surveys). Random effects meta-analysis was undertaken
on logit-transformed prevalence rates to calculate pooled prevalence estimates, stratified
according to methodological and substantive groupings.
Pooling across all studies, approximately 1 in 5 respondents (17.6%, 95% confidence
interval:16.3-18.9%) were identified as meeting criteria for a common mental disorder
during the 12-months preceding assessment; 29.2% (25.9-32.6%) of respondents were
identified as having experienced a common mental disorder at some time during their
lifetimes. A consistent gender effect in the prevalence of common mental disorder
was evident; women having higher rates of mood (7.3%:4.0%) and anxiety (8.7%:4.3%)
disorders during the previous 12 months and men having higher rates of substance use
disorders (2.0%:7.5%), with a similar pattern for lifetime prevalence. There was also
evidence of consistent regional variation in the prevalence of common mental disorder.
Countries within North and South East Asia in particular displayed consistently lower
one-year and lifetime prevalence estimates than other regions. One-year prevalence
rates were also low among Sub-Saharan-Africa, whereas English speaking counties returned
the highest lifetime prevalence estimates.
Despite a substantial degree of inter-survey heterogeneity in the meta-analysis, the
findings confirm that common mental disorders are highly prevalent globally, affecting
people across all regions of the world. This research provides an important resource
for modelling population needs based on global regional estimates of mental disorder.
The reasons for regional variation in mental disorder require further investigation.