Revista médica de Chile
versión impresa ISSN 0034-9887
PUSCHEL, Klaus et al. Design and validation of an instrument to assess families at risk for health problems. Rev. méd. Chile [online]. 2012, vol.140, n.4, pp. 417-425. ISSN 0034-9887. http://dx.doi.org/10.4067/S0034-98872012000400001.
Background: There is a paucity of screening instruments with a high clinical predictive value to identify families at risk and therefore, develop focused interventions in primary care. Aim: To develop an easy to apply screening instrument with a high clinical predictive value to identify families with a higher health vulnerability. Material and Methods: In the first stage of the study an instrument with a high content validity was designed through a review of existent instruments, qualitative interviews with families and expert opinions following a Delphi approach of three rounds. In the second stage, concurrent validity was tested through a comparative analysis between the pilot instrument and a family clinical interview conducted to 300families randomly selected from a population registered at a primary care clinic in Santiago. The sampling was blocked based on the presence of diabetes, depression, child asthma, behavioral disorders, presence of an older person or the lack of previous conditions among family members. The third stage, was directed to test the clinical predictive validity of the instrument by comparing the baseline vulnerability obtained by the instrument and the change in clinical status and health related quality of life perceptions of the family members after nine months of follow-up. Results: The final SALUFAM instrument included 13 items and had a high internal consistency (Cronbach's alpha: 0.821), high test re-test reproducibility (Pearson correlation: 0.84) and a high clinical predictive value for clinical deterioration (Odds ratio: 1.826; 95% confidence intervals: 1.101-3.029). Conclusions: SALUFAM instrument is applicable, replicable, has a high content validity, concurrent validity and clinical predictive value.
Palabras llave : Family characteristics; Healthcare surveys; Predictive value of tests.