“…The prevalence estimate of frailty syndrome The Italian Longitudinal Study on Aging, 2nd survey, 1995Aging, 2nd survey, -1996Aging, 2nd survey, and 3rd survey, 2000Aging, 2nd survey, -2001 ADL Activities of Daily Living, HR hazard ratios, CI confidence interval, IADL Instrumental Activities of Daily Living a HRs were adjusted for age, sex (coded 0 for women and 1 for men), and failure type as covariate in partially adjusted models b HRs were adjusted for age categories (coded 1 for 65-69, coded 2 for 70-74, coded 3 for 75-79, and coded 4 for 80-84), gender (coded 0 for women and 1 for men), education, pack-years {pack-years cigarettes [coded 0 for pack-years cigarettes = 0 (never smoking) and 1 for pack-years cigarettes ≥0.5]}, Instrumental Activities of Daily Living score, Mini Mental State Examination score, Charlson comorbidity index, serum albumin levels, in fully adjusted models for the nondemented population; MMSE and IADL scores were excluded from the models for the demented population in the ILSA was also similar to that of another Italian population-based study, the Invecchiare in Chianti Study (8.8%; Cesari et al 2006). However, several other population-based studies have applied the same definition of frailty syndrome of the CHS and the ILSA with very different prevalence estimates that varied from 6.9% to 20.0% (Fried et al 2001;Woods et al 2005;Ottenbacher et al 2005;Bandeen-Roche et al 2006;Gill et al 2006;Hirsch et al 2006;AvilaFunes et al 2008;Wong et al 2010). In these population-based studies, differences in the tools used in estimating the different components of the frailty syndrome, different races, and different age and gender distribution of the samples could be the source of this great variability.…”