[Validity and reliability of the family Apgar family function test]

Abstract

Objective: The aim of this study is to analyse the validity and reliability of the Apgar family questionnaire on family function.

Design: A crossover and descriptive study.

Setting: Urban health centre.

Patients: 656 patients of the Zaidín-Sur health centre in Granada were interviewed at home. The questionnaire was repeated an average of 6 days later to 60 of the patients (30 with an interviewer and 30 filling in their own).

Measurements and main results: The within-class correlation coefficients for the 5 items in the Apgar-family we over 0.55, both for self-filled and interviewer questionnaires; the scale coefficients were 0.86 and 0.81, respectively. Item-scale correlation varied between 0.61 and 0.71. Crombach's alpha was 0.84 and none of the items, when taken out of the scale, increased the alpha. The factorial analysis isolated only one factor. 16% of the sample had dysfunctional Apgar-family: the average score was 8.4. After adjustment (multiple logistic regression), the following were significantly linked to family dysfunction: lack of social support; being widowed or separated/divorced; greater perception of susceptibility to, or seriousness of, illness; being an over-user of the health service; and being a woman.

Conclusions: The Apgar-family questionnaire on family function is valid and reliable.