How are patient populations characterized in studies investigating depression in advanced cancer? Results from a systematic literature review.

Context: Prevalence rates of depression in patients with advanced cancer vary considerably. This may be because of heterogeneous samples and use of different assessment methods. Adequate sample descriptions and consistent use of measures are needed to be able to generalize research findings and apply them to clinical practice.

Objectives: Our objective was twofold: First, to investigate which clinically important variables were used to describe the samples in studies of depression in patients with advanced cancer; and second, to examine the methods used for assessing and classifying depression in these studies.

Methods: PubMed, PsycINFO, Embase, and CINAHL were searched combining search term groups representing "depression," "palliative care," and "advanced cancer" covering 2007-2011. Titles and abstracts were screened, and relevant full-text articles were evaluated independently by two authors. Information on 32 predefined variables on cancer disease, treatment, sociodemographics, depression-related factors, and assessment methods was extracted from the articles.

Results: After removing duplicates, 916 citations were screened of which 59 articles were retained. Age, gender, and stage of the cancer disease were the most frequently reported variables. Depression-related variables were rarely reported, for example, antidepressant use (17%) and previous depressive episodes (12%). Only 25% of the studies assessed and classified depression according to a validated diagnostic system.

Conclusion: Current practice for describing sample characteristics and assessing depression varies greatly between studies. A more standardized practice is recommended to enhance the generalizability and utility of findings. Stakeholders are encouraged to work toward a common standard for sample descriptions.