Background: We recently completed a randomised controlled trial in Goa India in which we observed a pattern of discordance with our two primary outcome measures; the Beck Depression Inventory (BDI-II) classified patients as moderately severe at the end of treatment, whilst the Patient Health Questionnaire (PHQ-9) classified these same patients as being only mildly depressed. The aim of this study is to explore whether the disparity between these two measures is seen in other settings. Method: The relationship between BDI-II and PHQ-9 scores was compared between the Indian trial and two other studies (from United Kingdom and United States) that administered both measures to patients. Linear regression was used to quantify the non-concordance between the two measures across studies. Patients were classified by severity category on the BDI-II and PHQ-9, respectively, and relationship assessed using chi-square test. We further quantified the proportion assigned a higher category on the BDI-II than the PHQ-9 and assessed the difference in prevalence between studies using a test of proportions. Results: Correlations between PHQ-9 and BDI-II were high and similar across studies (India: r=0.79; UK: r=0.87; US: r=0.77). Regression coefficients were similar across studies, but the predicted BDI-II mean score was significantly higher in the India study (24.3) compared to the US (20.5) or UK (20.8) studies. India participants had poorer outcomes on the BDI-II than the PHQ-9 and this difference was significant relative to both the UK (prevalence difference (PD): -15.9%; p Conclusions: The BDI-II and PHQ-9 measures are highly correlated, but the BDI-II tends to assign high severity scores in an Indian sample compared to UK/US samples. Where it is necessary to read items to patients, it seems likely that the PHQ-9 is a more accurate measure given that the BDI-II is longer and more complex.
