Getting the questions right: an example of loss of validity during transfer of a brief screening approach for depression in the elderly

Screening for depression in the elderly has been advocated to improve detection and management. This article summarises the trend towards briefer screening instruments, and the integration of mental health screening with other assessments. The study aimed to validate a single question depression screen which has previously shown adequate sensitivity and specificity in a new context: a multi-faceted assessment instrument used by nurse practitioners within a community sample of over 75 year olds. The GMS-AGECAT computerised interview assessment was used as a gold standard to determine the accuracy of the depression question in this new setting. Three hundred and twenty-eight patients were screened by their own nurse practitioners, of whom 100 consenting patients underwent a further interview with a research assistant using the GMS-AGECAT. The prevalence of depression was 30%, the sensitivity of the question was 67%, and its specificity 60% (compared with 88% and 71% previously). Responses indicating disability and loneliness were more closely correlated with depression than the depression screen itself. Relevant factors may include: the derivation of the question, the effect of a different sample, altered reliability when used by multiple interviewers, differing patient expectations, and the wording and context of the question within the multi-faceted screening instrument. Depression screening questions need repeated validation when used in different contexts. Patient and staff expectations may influence how screening instruments are used in practice in a way that may also alter reliability. Further studies are needed to establish the causes of loss of validity when screening approaches are used in new settings. Copyright © 2000 John Wiley & Sons, Ltd.