A comparison between uni- and multidimensional frailty measures: prevalence, functional status, and relationships...

Over the years, a plethora of frailty assessment tools has been developed. These instruments can be basically grouped into two types of conceptualizations - unidimensional, based on the physical-biological dimension - and multidimensional, based on the connections among the physical, psychological, and social domains. At present, studies on the comparison between uni- and multidimensional frailty measures are limited.

The aims of this paper were: 1) to compare the prevalence of frailty obtained using a uni- and a multidimensional measure; 2) to analyze differences in the functional status among individuals captured as frail or robust by the two measures; and 3) to investigate relations between the two frailty measures and disability.

Two hundred and sixty-seven community-dwelling older adults (73.4±6 years old, 59.9% of women) participated in this cross-sectional study. The Cardiovascular Health Study (CHS) index and the Tilburg Frailty Indicator (TFI) were used to measure frailty in a uni- and multidimensional way, respectively. The International Physical Activity Questionnaire, the Center of Epidemiologic Studies Depression scale, and the Loneliness Scale were administered to evaluate the functional status. Disability was assessed using the Groningen Activity Restriction Scale. Data were treated with descriptive statistics, one-way analysis of variance, correlations, and receiver operating characteristic analyses through the evaluation of the areas under the curve.

Results showed that frailty prevalence rate is strictly dependent on the index used (CHS =12.7%; TFI =44.6%). Furthermore, frail individuals presented differences in terms of functional status in all the domains. Frailty measures were significantly correlated with each other (r=0.483), and with disability (CHS: r=0.423; TFI: r=0.475). Finally, the area under the curve of the TFI (0.833) for disability was higher with respect to the one of CHS (0.770).

Data reported here confirm that different instruments capture different frail individuals. Clinicians and researchers have to consider the different abilities of the two measures to detect frail individuals.