Medicine (RMH) - Research Publications

Permanent URI for this collection

http://hdl.handle.net/11343/193

Filters

2020 1
Reset filters

Settings
Statistics
Citations
Author: Darvall, Jai × Author: Lim, Wen × Author: Story, David × End date: 2020 × Start date: 2020 × Type: Journal Article ×

Search Results

Now showing 1 - 1 of 1
  • Item
    Thumbnail Image
    Development of a Frailty Index from Routine Hospital Data in Perioperative and Critical Care
    Darvall, JN ; Greentree, K ; Loth, J ; Bose, T ; De Silva, A ; Braat, S ; Lim, WK ; Story, DA (WILEY, 2020-12)
    BACKGROUND/OBJECTIVES: Frailty is common in surgical and intensive care unit (ICU) populations, yet it is not routinely measured. Frailty indices are able to quantify this condition across a range of health deficits. We aimed to develop a frailty index (FI) from routinely collected hospital data in a surgical and ICU population. DESIGN: Prospective observational single-center cohort study. SETTING: Tertiary referral metropolitan Australian hospital. PARTICIPANTS: A total of 336 individuals aged 65 and older undergoing surgery or aged 50 and older admitted to the ICU. MEASUREMENTS: Routine admission health data were used to derive an FI comprising 36 health deficits. We examined the FI correlation with existing frailty tools (Clinical Frailty Scale [CFS] and Edmonton Frail Scale [EFS]) and assessed its predictive ability for negative outcomes including 30-day mortality. RESULTS: Median FI was .17 (interquartile range [IQR]) = .10-.24) for ICU patients and .17 (IQR = .11-.25) for surgical patients; maximum FI was .58, and 25% (95% confidence interval [CI] = 10.4-29.6) of patients overall were diagnosed with frailty (FI score ≥.25). Correlation was strong between the FI and the EFS: ρ = .76 (95% CI = .70-.83) for ICU patients and .71 (95% CI = .64-.78) for surgical patients, and the CFS was .77 (95% CI = .70-.84) for ICU patients and .72 (95% CI = .65-.79) for surgical patients. The FI had good discriminative ability for prediction of 30-day mortality in ICU patients (multivariate odds ratio for each increase in FI of .1 = 2.04 [95% CI = 1.19-3.48]), comparable with the performance of the Acute Physiology and Chronic Health Evaluation III score (ICU patients) and the Portsmouth Physiological and Operative Severity Score for the Enumeration of Mortality and Morbidity score (surgical patients). CONCLUSION: It is feasible to construct an FI from hospital admission data in a cohort of critically ill and surgical patients.