Abstract
Background: Frailty is an important concept in the acute health care system due to its association with poor health outcomes such as mortality. This study aimed to describe the prevalence of frailty as assessed using the Clinical Frailty Scale (CFS) and the predictive ability of CFS for prolonged length of stay (LOS) (defined as ≥ 14 days) and 12-month mortality in general medicine patients, assessed by the geriatric evaluation and management (GEM) liaison service in a hospital.
Methods: This retrospective study involved patients admitted to general medicine and screened by the GEM liaison team. Patients aged ≥80 years, 65-79 with geriatric syndromes (falls or dementia) or Aboriginal and Torres Straits Islander aged ≥ 50 years were eligible for the study. Premorbid frailty was assessed by the GEM liaison registered nurse using the Clinical Frailty Scale (CFS). Multivariable analysis was undertaken.
Results: 394 patients were included in the study. Frailty prevalence was 80.5% with 20.6% classified as severely frail. In a multivariable analysis that adjusted for age, gender and co-morbidity, the CFS was a significant predictor of prolonged LOS (≥14 days) for those mild-moderately frail (OR= 2.96, 95% CI= 1.50-5.87, p= 0.002) but not for the severely frail (OR= 1.30, 95%CI = 0.55-3.05, p= 0.546). For 12-month mortality, higher odds of mortality were seen for the severely frail patients (OR= 9.46, 95%CI= 3.95-22.66, p< .001) but not for those who were mild-moderately frail (OR= 1.56, 95%CI= 0.72-3.40, p= 0.265).
Conclusions: Frailty was highly prevalent in this sample. CFS demonstrated significant prognostic value for mortality and prolonged length of stay.