A revised short version of the depression questionnaire.

Background: The Depression Questionnaire (QD) is an instrument designed to measure depression in an inpatient-rehabilitation setting. This study proposes a reduced version of the original 24-item QD, to make it better suited to the characteristics of patients in this setting.

Methods: The study population consisted of 801 individuals admitted to cardiac rehabilitation (564 males and 237 females) who had validly completed the original version of the QD. Data were analyzed first through confirmatory factor analysis and then using logistic models.

Results: Factor analysis of QD-24 items form, confirmed the plausibility of the monofactorial solution, both for males and females. Factor analysis combined with clinical observation suggested the possibility of reducing the QD to 17 items. Two further items were eliminated through use of simple logistic models. We thus obtained a shortened 15-item version of the QD (QD-R) that maintains a good index of separation (0.80), that is, good capacity to discriminate between individuals with different levels of depression, and a good item-trait interaction (chi(2)153=165.37, P>0.05). A second confirmatory factor analysis applied to the 15-item monofactorial model confirmed the validity of this shortened version of the instrument. The QD-R scores significantly correlated with metres walked during the 6 minute walking test (r=-0.316; P<0.01).

Conclusion: Combined use of confirmatory factor analysis and simple logistic models together with observations drawn from clinical experience constitutes a valid method for shortening a questionnaire while at the same time maintaining, if not improving, its psychometric properties. The QD-R, seems adequate to how much is required in rehabilitation to point out the outcome, in case of variation of depression, easily to fill in by the debilitated individual or elderly, mainly because some item excessively influenced by the pathology or the hospital context were removed.