Association of frailty with risk of incident hospital-treated infections in middle-aged and older adults: A large-scale prospective cohort study.

Abstract

Although frailty is associated with a range of adverse health outcomes, its association with the risk of hospital-treated infections is uncertain. A total of 416220 participants from the UK Biobank were included in this prospective cohort study. Fried phenotype was adopted to evaluate frailty, which included five aspects (gait speed, physical activity, grip strength, exhaustion, and weight). More than 800 infectious diseases were identified based on electronic health records. Cox proportional models were used to estimate the associations. During a median 12.3 years (IQR 11.4-13.2) of follow-up (4747345 person-years), there occurred 77988 (18.7%) hospital-treated infections cases. In the fully adjusted model, compared with participants with non-frail, the HRs (95% CIs) of those with pre-frail and frail for overall hospital-treated infections were 1.22 (1.20, 1.24) and 1.78 (1.72-1.84), respectively. The attributable risk proportion of pre-frail and frail were 18.03% and 43.82%. Similarly, compared to those without frailty, the HRs (95% CIs) of those with frailty for bacterial infections were 1.76 (1.70-1.83), for viral infections were 1.62 (1.44-1.82), and for fungal infections were 1.75 (1.47-2.08). No association was found between frailty and parasitic infections (HR: 1.17, 95% CI: 0.62-2.20). Frailty was significantly associated with a higher risk of hospital-treated infections, except for parasitic infections. Studies evaluating the effectiveness of implementing frailty assessments are needed to confirm our results.