#2528 Comparing patient-reported and performance-based physical function assessments in hemodialysis: insights from the CONVINCE trial

Abstract Background and Aims Physical function is a crucial outcome in nephrology. In alignment with the Patient-Reported Outcomes Measurement Information System (PROMIS) Initiative, physical function is characterized as the ability to execute activities involving physical actions, ranging from self-care to more complex activities, such as social activities. This definition of physical function spans various domains of the International Classification of Functioning, Health, and Disability (ICF). The assessment of physical function commonly employs two distinct approaches, patient-reported outcomes (PRO) and observed performance outcomes (PerfO). PRO and PerfO assessments have distinct advantages and limitations and were found to correlate only moderately, adding complexity to cross-instrument outcome comparisons and interpretation. Despite this, there is limited understanding of the factors contributing to the discordance between PRO and PerfO assessments in patients with kidney disease. In this study, we applied a previously established common T-score metric for patient-reported and performance-based physical function to compare patterns of associations with potential laboratory-based, psychosocial, sociodemographic, and health-related determinants in hemodialysis patients. Method Cross-sectional data from the baseline assessment of the CONVINCE trial (N = 1,132), a multinational randomized controlled trial comparing high-dose hemodiafiltration with high-flux hemodialysis, were analysed. To investigate the relationships between potential determinants and performance-based versus patient-reported physical function, we performed multiple linear regression analysis. Dependent variables were standardized T-scores derived from the PROMIS physical function short-form 4a (PRO) and the Physical Performance Test (PerfO). Both measures assess a generic concept of physical function. We compared the effect sizes of each potential determinant in the multiple regression model using partial f2 values, with thresholds of 0.02, 0.15, and 0.35 denoting small, medium, and large effects. Results Both, PerfO and PRO exhibited significant associations with a laboratory marker-based indicator of muscle mass (simplified creatinine index), albeit with relatively small effects (partial f² of 0.04 and 0.05, respectively). None of the other laboratory measures investigated in this study showed any relevant associations with physical functioning. Age demonstrated a negative association with physical function, with a larger effect size observed for performance-based (partial f² = 0.11) compared to patient-reported physical function (partial f² = 0.04). In contrast to performance-based physical function, patient-reported physical function displayed a stronger association with self-reported health domains, particularly pain interference and fatigue. Exploring the differences between patient-reported and performance-based T-scores at the individual level, we found that younger age and more cognitive/emotional fatigue were associated with lower patient-reported compared to performance-based physical function (small effect sizes). Conclusion PRO and PerfO assessments of physical function in hemodialysis patients had similar associations with the simplified creatinine index, suggesting comparable links to objective physical impairment. However, PRO-based physical function exhibited a notably stronger association with other patient-reported variables, suggesting that patient-reported constructs of physical function might represent more than just capability, but ‘capability free from symptomatic constraints’. This distinction should be taken into account when directly comparing or aggregating PRO and PerfO scores. A noteworthy aspect of our study is the utilization of a common T-score metric for both PRO and PerfO measures. This method enhances interpretability of results and facilitates comparisons across diverse assessment types, thereby offering valuable insights into the determinants of physical function in patients undergoing hemodialysis.