[A comparative psychometric study of the Spanish versions with 6, 17, and 21 items of the Hamilton Depression Rating Scale].

Background and Objective: Former Spanish studies on the Hamilton Depression Rating Scale (HDRS) have focused on the 17 items version and have been conducted in depressive in-patient series. This study compares the psychometric properties of the Spanish version of HDRS with 6, 17 and 21 items as assessed in depressive patients in ambulatory care.
Patients and Method: A prospective, observational and multicenter study was conducted in clinically stable or unstable depressed patients. The discriminative validity, reliability (internal consistency, test-retest and inter-raters) and sensitivity to change were assessed.
Results: One hundred and sixty eight patients from 15 psychiatric facilities were included. The 6, 17 and 21 items versions of HDRS showed appropriate discriminative validity (HDRS-Clinical Global Impression: p < 0.0001); internal consistency (Cronbach *: HDRS-6 = 0.6; HDRS-17 and 21 >= 0.7); test-retest reliability (intraclass correlation coefficient [ICC]: HDRS-6, 17 and 21 >= 0.9); inter-raters reliability (ICC: HDRS-6, 17 and 21 >= 0.9) and sensitivity to change (effect size >= 1.5 for HDRS-6, 17 and 21).
Conclusions: Spanish versions of HDRS with 6, 17 and 21 items show similar psychometric properties. HDRS-6 could be useful in ambulatory and primary care.