Frailty predicts short-term incidence of future falls among British community-dwelling older people: a prospective cohort study nested within a randomised controlled trial

Abstract

Background: Although population-based studies have shown frailty predicted future falls, their follow-up periods were one year or longer and short-term fall risks associated with frailty are unknown.

Methods: A prospective cohort study nested within a randomised controlled trial was conducted to examine associations between frailty and short-term incident future falls among community-dwelling older people. Two hundred forty eight community-dwelling people > =65 years without history of > =three falls and allocated to a usual care arm of exercise intervention trial were prospectively monitored for falls over 24 weeks. Frailty index (FI) was constructed from 40 deficits at baseline. The future fall risks according to frailty status was examined using logistic regression models.

Results: Of 248 participants, 46 were classified as frail and 57 had one or more falls during follow-up. Both each 0.01 increase in FI and frailty defined as FI > =0.25 were significantly associated with higher risks of future falls in multivariate logistic regression models adjusted for age, gender and history of two falls in the previous year (odds ratio (OR) = 1.05, 95 % confidence interval (95 % CI) = 1.02-1.07, p < 0.001; OR = 3.04, 95 % CI = 1.53-6.02, p = 0.001, respectively). Receiver operating characteristic (ROC) curve analysis showed FI predicted future falls with fair accuracy with area under ROC curve of 0.62 (95 % CI = 0.53-0.71, p < 0.01).

Conclusions: Frailty was a significant and independent predictor of short-term future falls among community-dwelling older people who had volunteered for a physical activity study. It is important for healthcare practitioners to recognise frailty as a risk factor of imminent future falling even in older people who appear to be ageing well.

Fig. 1

Receiver operating characteristic curve using frailty index as a predictor of future falls. ROC: Receiver operating characteristic. Area under the receiver operating characteristic curve = 0.62, 95 % confidence interval 0.53–0.71, p < 0.01