Abstract
BACKGROUND: Patients with brain tumor are in risk of depression or depressive symptoms, but the estimated prevalence varies between studies. The aim of this study is to get a proper summarized estimate of depression prevalence in brain tumor patients. METHODS: Literature search on Pubmed, PsycINFO, and Cochrane library from January 1981 through October 2016. The prevalence of depression or depressive symptoms in brain tumor patients was estimated by screening scales and analyzed using stratified meta-analysis and subgroup analysis. The prevalence of depression level or symptoms during the follow-up periods was detected by secondary analysis. RESULTS: Among the 37 studies included in this meta-analysis, 25 used a cross-sectional design and 12 used longitudinal study. The pooled prevalence was 21.7% (971/4518 individuals, 95 % confidence interval (CI) 18.2%-25.2%) for overall sample. Lower prevalence was detected in studies with sample size ≥100 than <100, lower grade tumor than high grade tumor, studies using clinician-rated depression scales than self-rated or non-depression-specific ones, and in patients from UK, Germany and Italy than USA. After analyzing 6 longitudinal studies, prevalence of depression remained no change in the follow-up periods. No significant differences were observed between study designs and tumor types. CONCLUSIONS: The estimated prevalence of depression or depressive symptoms among brain tumor patients was 21.7%, affected by depression assessment type, sample size, tumor grade and country. Diagnosis and treatment of co-morbid depression in brain tumor patients need to be addressed in future studies for better life quality and oncology management.