RESUMO
BACKGROUND: Frailty predisposes individuals to a variety of complications. However, there is no consensus on the definition of frailty. OBJECTIVES: To examine whether various frailty measures are equivalent in identifying the same individuals as being frail and whether the measures also predict similar outcomes. DESIGN, SETTING AND PARTICIPANTS: The Helsinki Businessmen Study cohort, which is a long-term observational study of men born in 1919-1934, was used as the population. We investigated these men by their postal questionnaire responses in 2000 and 2005. The mean age of the men (N=480) was 73 years at the start of follow-up. MEASUREMENTS: We compared two phenotypic frailty measures, the Helsinki Businessmen Study measure (HBS), the modified Women's Health Initiative Observational Study (WHI-OS), and the Frailty Index (FI) comprising 20 items. All three measurements were applied to Helsinki Businessmen Study cohort data collected via simple postal questionnaire from 480 men. We investigated how effectively these three measures distinguished between the not frail, prefrail, and frail individuals, and predicted mortality, falls, weight change, and health-related quality of life (HRQoL, 15D instrument) during a 5-year follow-up. RESULTS: The HBS and the modified WHI-OS identified 35 persons (7.3%) each as frail but their respective sets comprised different groupings of individuals that partly overlapped. The FI identified 86 persons (17.9%) as frail. One-hundred-and-two (21.3%) men were classified as frail by at least one of the measures. All three measures significantly predicted higher mortality, higher number of fallers, and lower HRQoL for frail participants. None of the measures showed different results for weight change between the frailty groups or frailty stages. CONCLUSIONS: All three measures identified somewhat different sets of participants as frail. They all predicted increased mortality, falls and reduced HRQoL for the frail groups.