Prediction of Adverse Health Outcomes in Older People Using a Frailty Index Based on Routine Primary Care Data

Abstract

BACKGROUND: A general frailty indicator could guide general practitioners (GPs) in directing their care efforts to the patients at highest risk. We investigated if a Frailty Index (FI) based on the routine health care data of GPs can predict the risk of adverse health outcomes in community-dwelling older people. METHODS: This was a retrospective cohort study with a 2-year follow-up period among all patients in an urban primary care center aged 60 and older: 1,679 patients (987 women [59%], median age, 73 years [interquartile range, 65-81]). For each patient, a baseline FI score was computed as the number of health deficits present divided by the total number of 36 deficits on the FI. Adverse health outcomes were defined as the first registered event of an emergency department (ED) or after-hours GP visit, nursing home admission, or death. RESULTS: In total, 508 outcome events occurred within the sample population. Kaplan-Meier survival curves were constructed according to FI tertiles. The tertiles were able to discriminate between patients with low, intermediate, and high risk for adverse health outcomes (p value < .001). With adjustments for age, consultation gap, and sex, a one deficit increase in the FI score was associated with an increased hazard for adverse health outcomes (hazard ratio, 1.166; 95% confidence interval [CI], 1.129-1.210) and a moderate predictive ability for adverse health outcomes (c-statistic, 0.702; 95% CI, 0.680-0.724). CONCLUSIONS: An FI based on International Classification of Primary Care (ICPC)-encoded routine health care data does predict the risk of adverse health outcomes in elderly population.