Identifying the loss of functional independence of older people residing in the community: Validation of the PRISMA-7 instrument in Brazil

Arch Gerontol Geriatr. 2018 Jan:74:62-67. doi: 10.1016/j.archger.2017.09.008. Epub 2017 Oct 7.

Abstract

The identification of the older people who are functionally dependant is essential for planning interventions with emphasis on secondary and tertiary prevention. Therefore, a simplified tool, able to identify these individuals for a more comprehensive evaluation, is required. However, the use of a measuring tool in a different socio-cultural context requires prior adaptation. This article aims to verify the psychometric properties of the PRISMA-7 in Brazilian context. This instrument was previously shown as able to identify older persons living in the community who have risk of functional loss in a Canadian study. In a sample of 1748 older persons, the internal consistency assessed by Cronbach's alpha showed a borderline value credited to the reduced number of questionnaire items. Factor analysis identified two well correlated factors, except for items 2 and 6. For criterion validity the SMAF scale (French acronym, Functional Autonomy Measurement System) was used as the gold standard. The analysis of the ROC curve indicated a sensitivity of 74.4% and specificity of 87.4% for the cut-off point of 4, while the Canadian version had a cut-off point of 3 for disability detection. Inter and intra-observer reliability, investigated by the Kappa Cohen were high and statistically significant. In conclusion, the results suggest that the validation process was adequate, and recommend the Brazilian version of PRISMA-7, to track older people with functional loss in the community.

Keywords: Aged; Case-finding; Functional autonomy; Risk factors; Screening; Transcultural adaptation (process).

Publication types

  • Validation Study

MeSH terms

  • Activities of Daily Living
  • Aged
  • Aged, 80 and over
  • Brazil
  • Canada
  • Disability Evaluation*
  • Factor Analysis, Statistical
  • Female
  • Humans
  • Male
  • Middle Aged
  • Psychometrics*
  • Surveys and Questionnaires