The identification of frail older adults in primary care: comparing the accuracy of five simple instruments

Abstract

Background: many instruments are available to identify frail older adults who may benefit from geriatric interventions. Most of those instruments are time-consuming and difficult to use in primary care. Objective: to select a valid instrument to identify frail older adults in primary care, five simple instruments were compared. Methods: instruments included clinical judgement of the general practitioner, prescription of multiple medications, the Groningen frailty indicator (GFI), PRISMA-7 and the self-rated health of the older adult. Fried's frailty criteria and a clinical judgement by a multidisciplinary expert panel were used as reference standards. Data were used from the cross-sectional Dutch Identification of Frail Elderly Study consisting of 102 people aged 65 and over from a primary care practice in Amsterdam. In this study, frail older adults were oversampled. We estimated the accuracy of each instrument by calculating the area under the ROC curve. The agreement between the instruments and the reference standards was determined by kappa. Results: frailty prevalence rates in this sample ranged from 11.6 to 36.4%. The accuracy of the instruments ranged from poor (AUC = 0.64) to good (AUC = 0.85). Conclusion: PRISMA-7 was the best of the five instruments with good accuracy. Further research is needed to establish the predictive validity and clinical utility of the simple instruments used in this study.