[Validity and reliability of the Duke-UNC-11 questionnaire of functional social support]

Abstract

Objective: The aim of this study is to analyse the validity and reliability of the functional social support questionnaire, Duke-UNC-11.

Design: Descriptive. Crossover study.

Setting: Urban health centre.

Patients: 656 patients were interviewed in their homes. 60 had the questionnaire repeated (30 self-filled and 30 using an interviewer) an average of 6 days later.

Measurements and main results: The intraclass correlation coefficients of the 11 items in the Duke-UNC-11 were above 0.50, for both self-filled and interviewer questionnaires; the ones on the scale were 0.92 and 0.80, respectively. The factorial analysis separated two sub-scales, confidential support (7 items) and affective support (4 items). Low social support was significantly associated to: being over 40, widowed or divorced, living alone, over-user, worse subjective health, greater chronic morbidity, mental health disorder and family dysfunction. The multiple linear regression equation managed to explain 30% of the variability of social support, in which family function (family APGAR) explained 23.5%, education 3.3%, perception of internal health control 2%, mental health 1.2% and perception of susceptibility to/seriousness of illness 0.3%.

Conclusions: The questionnaire Duke-UNC-11 is valid and reliable.