Search

Objective: The present study aimed to resolve inconsistency in reported prevalence of affective symptoms by dementia stage. Methods/Design: We conducted a meta‐analysis of studies with data on dementia stage and prevalence of depression, anxiety, or apathy assessed using validated tools. We performed random‐effects meta‐analysis and subgroup analysis on symptom prevalence by dementia stage, according to CDR. Results: The meta‐analysis included 5897 people with dementia from 20 studies. Prevalence rates of depression in mild, moderate, and severe dementia were 38% (95% CI 32%–45%), 41% (95% CI 33%–49%), and 37% (95% CI 17%–56%) respectively. The corresponding prevalence for anxiety was 38% (95% CI 31%–45%), 41% (95% CI 31%–52%), and 37% (95% CI −8%–82%); and 54% (95% CI 45%–62%), 59% (95% CI 44%–73%), and 43% (95% CI 10%–75%) for apathy. The prevalence of depression, anxiety, and apathy did not differ with regard to dementia stage and type. The prevalence of depression in Alzheimer's disease (AD) was significantly lower when it was assessed using diagnostic criteria compared to screening tools. The prevalence of depression in AD was lowest in America, while anxiety in vascular dementia was higher in Europe than Asia. Conclusions: Depression, anxiety, and apathy symptoms are highly prevalent across dementia stages. There is no evidence of any changes in prevalence of affective symptom as the illness progresses. Evaluation methods and cultural difference may explain some of the variance, suggesting further investigation of factors that may influence the report of symptoms, such as carer psychosocial characteristics, and more cross‐cultural studies are needed. Key points: Random‐effects meta‐analysis showed no evidence of any changes in prevalence of depression, anxiety, and apathy as dementia progresses in general dementia, AD and VaD populationThe variance of prevalence in depression and anxiety may be attributable to symptom evaluation method and region of study conductedFuture studies could examine the pattern and management of affective symptoms in relation to carer psychosocial characteristics and across culture [ABSTRACT FROM AUTHOR]