Differences in older people's responses to CIDI's depression screening and diagnostic questions may point to age-related bias

Abstract

BackgroundPsychiatric surveys based on the Composite International Diagnostic Interview (CIDI) report very low rates of affective disorder in older people, perhaps because CIDI's long, convoluted screening questions present a special challenge to aged respondents. We have shown previously that inconsistencies in responses to CIDI's two screening questions about dysphoria and anhedonia rose in frequency with age. By contrast, responses to the short, simple K-10 and GHQ-12 mental health scales showed much less change over the lifespan. As a check on age-related bias, we now compare responses to CIDI's depression screening questions with responses to subsequent, simpler questions about other depressive symptoms. MethodsSecondary analysis of an Australian national survey in which CIDI was administered to 10,641 adults by trained lay interviewers. ResultsRates of positive responses to both CIDI screening questions fell from 11% in age-group 18–34years to 3% in respondents aged ≥65years. Responses to simpler CIDI items about other depressive symptoms and help-seeking behaviour showed much less change with age. ConclusionOlder respondents may deny symptoms when subjected to complex batteries of questions concerning severity, time frame, attribution and consequences. CIDI diagnoses need to be validated across the full age spectrum. LimitationsResidents of aged care facilities and those with low cognitive scores were excluded.