Abstract
OBJECTIVES: The objective of our study was to compare advantages and limitations of two generic Quality of Life questionnaires administered in older inpatients. DESIGN: Two validated generic health-related Quality of Life instruments : the MOS Short-Form 36 (9 dimensions, 36 items) and the Duke Health Profile (6 dimensions, 4 dysfunctions, 17 items) were administered to inpatients over 65 years. SETTING AND PARTICIPANTS: The sample was drawn from the CliniQualVie program that assessed systematically Quality of Life among hospitalized inpatients (18-79 years) in 10 medical and surgical wards at Nancy University Hospital. RESULTS: The two self-administered questionnaires were completed by 701 patients over 65 years at admission (mean age 71 +/- 4, 63% men). The proportion of patients who completed all items were 72.5% for the Duke and 66.9% for the SF-36 (p < .001). The Duke's internal consistency was low as compared with the SF-36, but other psychometric properties were comparable. Good correlations (Spearman) were observed between the two questionnaires for physical health (0.59, p < .0001), mental health (0.68, p < .0001) and health perception (0.56, p < .0001) scores. Low correlations were observed for the social score. CONCLUSIONS: This is the first study to our knowledge to assess the interest of using the Duke Health Profile in a general elderly inpatients population as compared with the SF-36 questionnaire. Although these two questionnaires have four comparable dimensions, they differ in their content and psychometric properties. The Duke questionnaire, due to its better completion rate and despite some psychometric limitations may be useful in this population, particularly in the more frail patients.