Frailty is characterized by increased vulnerability to stressors that poses the older subject at risk of adverse health-related outcomes, including hospitalization, disability and mortality. Early identification of community-dwelling frail older subjects is important in order to implement preventive strategies against negative health-related outcomes, in particular disability. Validated brief screening tools are needed to detect frail community-dwelling elders.

The aim of the present study, promoted by the European Union Geriatric Medicine Society (EUGMS) working group on “Frailty in older persons”, is to determine the agreement between the Gérontopôle Frailty Screening Tool (GFST) (administered by the general practitioner) and the Fried’ criteria for frailty phenotype as reference measure (administered by a blinded assessor). The study is performed in older primary care patients in nine European countries after translation of the GFST into eight languages.

The sample (n=109 older patients,) included 37.6%, 56.9%, and 5.5% robust, pre-frail or frail, and disabled individuals, respectively. The GFST showed a sensitivity of 71.0%, a specificity of 70.2%, a positive predictive value of 75.9% and a negative predictive value of 64.7% at the identification of non-disabled frail elders. The positive and negative likelihood ratios were 2.38 and 0.41, respectively. In logistic regression models only slow gait speed (odds ratio [OR]: 19.65, 95% confidence interval [95% CI]: 4.69–82.35) and mobility issues (OR: 18.04, 95% CI: 3.11–104.78) were significantly associated with the condition of frailty in the absence of disability.

Our findings demonstrate an overall moderate agreement between the GFST and the frailty phenotype.