Agreement of Patient Responses Between Electronic- and Paper- Patient Reported Outcome Measures in Total Hip Arthroplasty.

Abstract

Total hip arthroplasty (THA) outcomes are commonly assessed using patient-reported outcome measures (PROMs). As electronic PROMs (ePROMs) gain popularity, it is essential to ensure equivalence to paper-based PROMs (pPROMs). This study aimed to evaluate the interchangeability of ePROMs and pPROMs in patients undergoing THA. A prospective cohort analysis of 96 THA patients who completed preoperative ePROMs and pPROMs was conducted. Electronic Oxford Hip Score (OHS) and EuroQol 5-Dimension 5-Level (EQ-5D-5L) were completed zero to 28 weeks prior to surgery, and pPROMs were completed on the day of surgery. Correlations were assessed with Spearman's rank correlation, and agreement and bias were evaluated using intraclass correlation coefficients (ICC) and Bland-Altman analysis. The effect of time between completion of ePROM and pPROM on score difference was assessed. Electronic scores were lower compared to the paper format (OHS: median 19.0 [interquartile range (IQR) 13.8 to 25.3] versus 19.0 [14.0 to 25.0]; difference -1.5; P = 0.005, EQ-5D-5L: median 0.575 [0.455 to 0.681] versus 0.597 [0.472 to 0.704]; difference -0.028; P = 0.05). These differences were below established minimal clinically important differences (MCIDs). Agreement was good for OHS (ICC 0.873, 95% CI [confidence interval] 0.805 to 0.917) and moderate for EQ-5D-5L (ICC 0.677, 95% CI 0.551 to 0.773). Minimal bias (OHS -1.3, LoA -9.3 to 6.7; EQ-5D-5L -0.025, LoA -0.26 to 0.20) was detected, with 14.6 and 2.1% of scores, respectively, exceeding MCID. The median time interval between ePROM and pPROM administration was six days (IQR 3 to 27 days) and had no effect on differences. However, reduced variability was seen when ePROMs were completed within one week of surgery. Electronic and paper-based OHS and EQ-5D-5L show good agreement and can generally be used interchangeably. However, caution is required for individual scores near clinically important thresholds. Given their efficiency and scalability, we recommend a full transition to ePROMs.