ABSTRACT:
Objective:
To evaluate the validity and reliability of the Johns Hopkins Fall Risk Assessment Tool (JH-FRAT), which assesses the risk of falls in hospitalized inpatients.
Method:
Study with 297 patients at a hospital in São Paulo, using retrospective data form 2014. Validity was assessed by accuracy (sensitivity, specificity, positive predictive value - PPV and negative predictive value - NPV) and discriminant analysis (comparison of patients with and without falls in relation to the scale items and comparison of previous risk situations in relation to the injury). The χ2 test and Fisher’s exact test were used. Reliability was assessed by reproducibility between methods and interobserver test-retest comparison in a subsample of 60 patients. We used the Kappa, quadratic weighted Kappa and PABAK statistics.
Results:
Sensitivity was 97.0%, specificity was 6%, PPV was 36.2% and NPV was 90.6%. Five of the eight items of the scale and the overall classification showed risk discrimination capability (p < 0.050). The risk of previous situations did not discriminate the injury resulting from the falls (p = 0.557). Reproducibility between methods was substantial (PABAK = 0.71). The interobserver reproducibility ranged between items (PABAK 0.25 to 1.00) and was substantial to the overall risk classification (PABAK = 0.71).
Conclusion:
JH-FRAT showed validity and reliability expected of a screening tool for risk of falls, and it can contribute to the implementation of fall management strategies in hospitals.
Keywords:
Data accuracy; Reproducibility of results; Reliability and validity; Nursing methodology research; Accidental falls; Quality of health care