abstract
- BACKGROUND: In psychiatric epidemiology, symptoms are often assessed retrospectively. This raises concerns about the accuracy of the information recalled. In this study, we sought to examine the level of agreement between survey items assessing recent and more remote depressive episodes. METHODS: Data from the Canadian National Population Health Survey (NPHS) were used. The NPHS is a prospective study following a representative cohort of household residents sampled in 1994 and 1995. Every 2 years, participants are administered the Composite International Diagnostic Interview Short Form for Major Depression (CIDI-SFMD). The 2004 NPHS interview also included items asking about past episodes of depression and diagnoses of depression done by health professionals. We used cross-tabulation and logistic regression to explore the relationship between these responses. RESULTS: Approximately, 90% of respondents with CIDI-SFMD-defined major depressive episodes in the year preceding the 2004 interview also reported lifetime episodes or professional diagnoses of depression in 2004. However, responses to the 2004 lifetime items corresponded less closely to CIDI-SFMD results from the same individuals earlier in the longitudinal survey. Only 40.8% of respondents having the most recently identified episode in 1994 subsequently affirmed a past episode of depression in 2004. CONCLUSIONS: Reporting of depressive episodes diminishes with time, suggesting that retrospective assessment of such episodes may be vulnerable to inaccuracy.