Frailty prevalence and 10-year survival in community-dwelling older adults: results from the ESTHER cohort study

The frailty index (FI), defined by a deficit accumulation approach, has emerged as a promising concept in gerontological research, but applications have been mostly restricted to populations from Canada and the United States aged 65 years or older. Baseline data from the German ESTHER cohort study (N 9,886; age 50-75; mean follow-up 8.7 years) were used to create a FI through a deficit accumulation approach. For estimation of frailty prevalence, we used cut-points for the FI to define three categories (non-frail 0 to a parts per thousand currency sign0.20; pre-frail > 0.20 to < 0.45; frail a parts per thousand yen0.45). We assessed variation of the FI by age and sex: 10-year survival according to baseline FI was assessed by Kaplan-Meier curves and bivariate and multivariate Cox proportional hazard models. Cubic splines were used to assess sex-specific dose-response associations. Prevalence of frailty was 9.2 and 10.5 % in women and men, respectively. Age-specific prevalence of frailty ranged from 4.6 % in 50-54 year old participants to 17.0 % in 70-75 year old participants. Below 60 years of age, men had a higher FI than women. However, the FI showed a stronger increase with age among women (3.1 % per year) than among men (1.7 % per year) and was higher among women than men in older age groups. Adjusted hazard ratios (95 % confidence intervals) for all-cause mortality were 1.08 (0.84-1.39), 1.32 (1.05-1.66), 1.77 (1.41-2.22), and 2.60 (2.11-3.20) for the 2nd, 3rd, 4th, and 5th quintile of the FI compared to 1st quintile, respectively. There was a strong dose-response relationship between the FI and total mortality among both men and women and both younger (< 65 years) and older subjects. We found sex differences in the FI and its increase with age, along with a consistent strong association of the FI with mortality in both sexes, even for age group 50-64.