OBJECTIVESTo determine whether immune phenotypes associated with immunosenescence are predictive of frailty and mortality within 1-year in elderly nursing home residents.DESIGNCross sectional study of frailty; prospective cohort study of mortality.SETTINGThirty-two nursing homes in four Canadian cities between September 2009 and October 2011.PARTICIPANTSNursing home residents aged 65 and older (N = 1,072, median age 86, 72% female).MEASUREMENTSAfter enrollment, peripheral blood mononuclear cells were obtained and analyzed using flow cytometry for CD4(+) and CD8(+) T-cell subsets (naïve, memory (central, effector, terminally differentiated, senescent), and regulatory T-cells) and cytomegalovirus (CMV)-reactive CD4(+) and CD8(+) T-cells. Multilevel linear regression analysis was performed to determine the relationship between immune phenotypes and frailty; frailty was measured at the time of enrollment using the Frailty Index. A Cox proportional hazards model was used to determine the relationship between immune phenotypes and time to death (within 1 year).RESULTSMean Frailty Index was 0.44 ± 0.13. Multilevel regression analysis showed that higher percentages of naïve CD4(+) T-cells (P = .001) and effector memory CD8(+) T-cells (P = .02) were associated with a lower mean Frailty Index, whereas a higher percentage of CD8(+) central memory T-cells was associated with a higher mean Frailty Index score (P = .02). One hundred fifty one (14%) members of the cohort died within 1 year. Multivariable analysis showed a significant negative multiplicative interaction between age and percentage of CMV-reactive CD4(+) T-cells (hazard ratio = 0.87, 95% confidence interval = 0.79-0.96). No other significant factors were identified.CONCLUSIONImmune phenotypes found to be predictive of frailty and mortality in this study can help further understanding of immunosenescence and may provide a rationale for future intervention studies designed to modulate immunity.