Prevalence of depression: Comparisons of different depression definitions in population-based samples of older adults

Background Depression prevalence in older adults varies largely across studies, which probably reflects methodological rather than true differences. This study aims to explore whether and to what extent the prevalence of depression varies when using different diagnostic criteria and rating scales, and various samples of older adults. Methods A population-based sample of 3353 individuals aged 60–104 years from the Swedish National Study on Aging and Care in Kungsholmen (SNAC-K) were examined in 2001–2004. Point prevalence of depression was estimated by: 1) diagnostic criteria, ICD-10 and DSM-IV-TR/DSM-5; 2) rating scales, MADRS and GDS-15; and 3) self-report. Depression prevalence in sub-samples by dementia status, living place, and socio-demographics were compared. Results The prevalence of any depression (including all severity grades) was 4.2% (moderate/severe: 1.6%) for ICD‐10 and 9.3% (major: 2.1%) for DSM-IV-TR; 10.6% for MADRS and 9.2% for GDS-15; and 9.1% for self-report. Depression prevalence was lower in the dementia-free sample as compared to the total population. Furthermore, having poor physical function, or not having a partner were independently associated with higher depression prevalence, across most of the depression definitions. Limitations The response rate was 73.3% and this may have resulted in an underestimation of depression. Conclusion Depression prevalence was similar across all depression definitions except for ICD-10, showing much lower figures. However, independent of the definition used, depression prevalence varies greatly by dementia status, physical functioning, and marital status. These findings may be useful for clinicians when assessing depression in older adults and for researchers when exploring and comparing depression prevalence across studies.