Dr Amanda Siriwardana1,2,3, Anna Hoffman4, Prof Rachael Morton5, Dr Brendan Smyth1,4,5, Prof Mark Brown4,6
1The George Institute for Global Health, Sydney, Australia, 2Sydney Medical School, University of Sydney, Sydney, Australia, 3Department of Renal Medicine, Royal North Shore Hospital, Sydney, Australia, 4Department of Renal Medicine, St George Hospital, Sydney, Australia, 5NHMRC Clinical Trials Centre, Sydney, Australia, 6Faculty of Medicine, University of New South Wales, Sydney, Australia
Aim: To use anchor-based and distribution-based methods to estimate the minimal clinically important difference (MCID) for the European Quality of Life-5 Dimensions-5 Levels Questionnaire (EQ-5D-5L) utility index in dialysis patients.
Background: Health-related quality-of-life (HRQOL) is a critical patient-centred clinical and economic outcome when evaluating interventions for people managed with dialysis. The EQ-5D-5L is a commonly used HRQOL measure, however the MCID that constitutes a meaningful treatment effect for dialysis patients has not been established.
Methods: 6-monthly EQ-5D-5L measurements were collected from adult dialysis patients between April 2017 and November 2020 at a renal network in Sydney, Australia. EQ-Visual Analogue Scale and global symptom burden scores using the IPOS-Renal inventory were collected simultaneously and used as anchors. Longitudinal EQ-5D-5L utility was examined using a linear mixed effects model with random intercepts and random slopes, and EQ-5D-5L utility index MCID estimates were derived using several anchor-based and distribution-based methods.
Results: 352 patients with ≥1 EQ-5D-5L observation were included, constituting 1127 observations. Mean EQ-5D-5L utility at baseline was 0.719 (SD ±0.267), and mean EQ-5D-5L utility decreased over time by -0.017 per year (95% CI -0.029 to -0.006, p=0.004). Using cross-sectional anchor-based methods, MCID estimates ranged from 0.073 to 0.107. Using longitudinal anchor-based methods, MCID for improvement and deterioration ranged from 0.046 to 0.158 and 0.048 to 0.111, respectively. Using receiver operating characteristic curves, MCID for improvement and deterioration ranged from 0.037 to 0.122 and 0.063 to 0.074, respectively. MCID estimates from distribution-based methods were largely consistent with anchor-based estimates.
Conclusions: Anchor-based and distribution-based approaches provided EQ-5D-5L utility MCID estimates ranging from 0.034 to 0.158. These estimates can now be used in clinical trial power calculations for dialysis interventions.
Biography:
Dr Amanda Siriwardana is a Nephrologist at Royal North Shore Hospital and a PhD candidate through the George Institute for Global Health and University of Sydney. She has an MBBS (Hons1) and GCert Public Health, both from the University of New South Wales. Her current PhD research focuses on large observational cohorts exploring outcomes for older patients with advanced kidney disease.