The frailty index in Europeans: association with age and mortality

Objective: to construct sex-specific FIs from a large sample of Europeans and study their associations with age and mortality. Design: longitudinal population-based survey. Setting: the Survey of Health, Ageing and Retirement in Europe (SHARE, ). Subjects: a total of 16,217 females and 13,688 males aged >= 50 from wave 1 (2004-05). Mortality data were collected between 2005 and 2006 (mean follow-up: 2.4 years). Methods: regression curve estimations between age and an FI constructed as per the standard procedure. Logistic regressions were used to assess the relative effects of age and the FI towards mortality. Results: in both sexes, there was a significant non-linear association between age and the FI (females: quadratic R-2 = 0.20, P < 0.001; males: quadratic R-2 = 0.14, P < 0.001). Overall, the FI was a much stronger predictor of mortality than age, even after adjusting for the latter (females: age-adjusted OR 100.5, 95% confidence interval (CI): 46.3-218.2, P < 0.001; males: age-adjusted OR 221.1, 95% CI: 106.7-458.4, P < 0.001). Conclusion: the FI had the expected properties in this large sample of Europeans.