Design and validation of a questionnaire for the detection of major depression in elderly patients.

Abstract

BACKGROUND AND OBJECTIVE: The aim of this study was to design and validate a scale to detect major depressive disorders in elderly individuals in primary care (Detection of Depression in the Elderly Scale [DDES]). METHODS: We performed an observational and cross-sectional study for the validation of a scale, administered by means of a personal interview, in 259 patients aged 65 years old or older. Available instruments were reviewed and the questions were designed. Subsequently, a first pilot study was performed. In a second pilot study the reproducibility of the instrument was analyzed. The gold standard was the result of a standardized psychiatric interview performed by psychiatrists (DSM-IV criteria and SCAN interviews). RESULTS: The intraclass correlation coefficients corresponding to the test-retest and inter-rater reliability were 0.858 (95% confidence interval [CI], 0.634-0.946) and 0.908 (95% CI, 0.726-0.969) respectively. Two hundred sixteen subjects underwent an assessment, in which primary care and psychiatric evaluations were blinded. Major depression was diagnosed in 81 patients (37.5%; 95% CI, 31.1-44.4). The internal consistency of the DDES was good (Cronbach's alpha = 0.79). Exploratory factorial analysis revealed an 8-component structure (55.8% of explained variance). A cutoff score of 15 or more for the DDES showed sensitivity of 90.1% (95% CI, 80.95-95.33), specificity of 74.8% (95% CI, 66.48-81.71) and a likelihood ratio (+) of 3.58 (95% CI, 2.65-4.83). CONCLUSIONS: The DDES is a clinically useful instrument for the detection of major depression in elderly patients in primary care.