Prevalence estimates of depression in AD vary greatly across studies, and a reliable result is crucial
for further interventions. A systematic review and meta-analysis was conducted to identify the prevalence
of depression in AD patients. We searched PubMed and Embase, followed by data extraction, quality
assessment, prevalence estimates and subgroup analyses. A total of 63 studies were included in the review. The prevalences
of depression were 12.7% (CI, 8.8-17.8) and 42% (CI, 38-45) according to the DSM criteria for major depression
and the specific criteria for dementia respectively. Subgroup analyses stratified by case identification showed that the
prevalences of population-based stud ies were 5% (CI: 2-15) and 35% (CI: 29-41) according to the DSM criteria and the
specific criteria respectively, while those of single source were 17% (CI: 11-25) and 43% (CI: 37-49). Subgroup analyses
stratified by MMSE score revealed that the prevalences of severe AD were 8% (CI: 5-15) and 48% (CI: 41-54) according
to the DSM criteria and the specific criteria respectively, while those of mild AD were 14% (CI: 8-24) and 40% (CI: 32-
47). In conclusion, the prevalence of depression according to the DSM criteria was lower than that of the specific criteria.
Prevalence estimates conducted in population-based studies were lower compared with those in single-source studies irrespective
of the screening tools. Patients with severe AD tended to have higher prevalence of depression according to the
specific criteria, while the trend was opposite according to the DSM criteria. Thus, different settings and diagnostic approaches
should be taken into account before estimates of depression and further interventions.