MO323: Effects of Patient Frailty and Comorbidity on Duration of Hospitalisation as an Outcome Among Adults With Acute Kidney Injury

Abstract

Abstract BACKGROUND AND AIMS Acute kidney injury (AKI) is commonly associated with an adverse outcome in hospitalised patients. Frailty and comorbidity are risk factors for acute kidney injury. The aim of this study was to assess the strength of association between frailty and comorbidity as modifiers for duration of hospitalisation (LOS) as an outcome among inpatients with acute kidney injury in our institution. METHOD A retrospective observational study as part of a service assessment of adult inpatients, in a 1-month period, during the pre-pandemic phase was conducted to evaluate for changes needed to existing care pathways in our institution. AKI was identified with the help of the national algorithm endorsed by NHS England, which is incorporated in hospital reporting systems. Clinical frailty and comorbidity were estimated using the Rockwood Clinical Frailty Scale (CFS) and the Charlson comorbidity index (CCI). Pairwise correlations between LOS, CFS and CCI were estimated and tested for statistical significance using Bonferroni-adjusted significance levels. Ordinary least-squares linear regression was used to assess the prediction of LOS using CFS and CCI in separate models. RESULTS In 148 patients in our cohort, the mean age was 76.4 [95% confidence interval (95% CI) 74.1–78.8] years. Of these, 54% were male, 24.3% were diabetic, 29.7% had at least one criterion for prior vascular disease and 16.2% had a history of cancer. A total of 22.3% had a prior history of chronic kidney disease. In terms of severity at presentation, 67.6% had stage 1 AKI, 18.9% had stage 2, and 13.5% had stage 3 AKI. The mean CCI was 5.4 (95% CI 5.06–5.81) and the mean RFI was 4.6 (95% CI 4.3–5.0). 50% were both at least moderately frail (CFS ≥ 5) and had a CCI of >5. The mean duration of hospitalisation was 8.7 days. Spearman's correlation coefficient constant for CFS with LOS was 0.23 and was 0.11 with CCI, with only the correlation between LOS and CFS being statistically significant at the 5% level after Bonferroni correction. As expected, there was a statistically significant correlation between CCI and CFS (0.49). Beta coefficients for ordinary least-squares linear regression individual models with LOS as the outcome variable were 0.69 (95% CI 0.18–1.19; P < .007) for CFS and 0.27 (95% CI –0.16 to 0.70; P = .22) for CCI. CONCLUSION Compared with patient comorbidity, clinical frailty has a stronger and statistically significant association with the duration of hospitalisation among hospitalised adults with AKI. This study quantifies the magnitude of factors over and above those resulting from patient comorbidity that are contributory to frailty in adult inpatients with AKI. Further studies are required to improve understanding of how clinical frailty affects duration of hospitalisation in patients with AKI, to help inform future clinical care pathways and improve duration of hospitalisation as a clinical outcome.