Electronic Patient-Reported Outcome Monitoring to Improve Quality of Life After Joint Replacement

Key Points Question Does patient-reported outcome measure (PROM)–based remote monitoring after hip and knee replacement surgery improve health outcomes compared with the standard of care? Findings This multicenter randomized clinical trial included 3697 patients with primary hip replacement and 3110 with knee replacement across 9 German hospitals. The intervention group had a statistically significantly higher mean increase from baseline to 12 months after surgery in health-related quality of life score, fatigue score, and depression score compared with the control group. Meaning This randomized clinical trial found that postoperative PROM monitoring, including alerts for critical recovery, led to small improvements in several health dimensions (health-related quality of life and fatigue after hip and knee replacement and depression after hip replacement).

All patients who signed the consent forms to participate in the "PROMoting Quality" trial n=7,827 Patients who received surgery n=7,204 Randomized patients n=6,878  a PROM-scores have different score ranges: -0.661-1 for the EQ-5D-5L, 0-100 for the EQ-VAS, 0-100 for the       a The score ranges for the PROMs are: -0.661-1.0 for the EQ-5D-5L, 0-100 for the EQ-VAS, 0-100 for the HOOS-PS and

Final
V et al.JAMA Network Open.

eFigure 4 .
Count of Knee Replacement Patients Whose PROM-Scores Initiated Alerts per Intervention Time/Times a a the total number of patients with critical values (alerts) were 157, 386 and 461 for month one, three and six respectively.The total number of patients who did not have a critical value (no alert) were 1329, 1319 and 1285 respectively.
V et al.JAMA Network Open.
Absolute Intervention Alert Thresholds for Each PROM-Score Descriptive Statistics of the Compliance Corrected Study Population eTable 2.a the absolute alert thresholds were set through a Delphi panel(Kuklinski et al. 2020)b the relative change to a worse score initiated an alert for all PROM-scores with "Yes".This table is based on the table published in the study protocol of the PROMoting Quality trial (Kuklinski et al. 2020) © 2023 Steinbeck V et al.JAMA Network Open.eTable 3. © 2023 Steinbeck V et al.JAMA Network Open.

procedure Hip replacement patients Knee replacement patients month month 1 month 3 month 6 month 1 month 3 month 6 number of patients in the intervention group
Descriptive Statistics of the Admission, Discharge and 12-Month Follow-up PROM-Scores Readmission and Reoperation Descriptive Statistics in the Intervention Versus the Control Group Separated per Joint Replacement Type Information on Critical Values and Reactions to PROM-Based Critical Value Alerts From Study Assistants and Patients -Critical Values and Alert Reactions for thefor the PROMIS-depression.Higher values in the EQ-VAS and EQ-5D-5L indicate better health levels, whereas lower values in H/KOOS-PS, PROMIS-fatigue, and depression indicate better health (less health impairment).©2023 Steinbeck V et al.JAMA Network Open.eTable 4.

number of those with critical values (% of critical value patients) call
Count of Hip Replacement Patients Whose PROM-Scores Initiated Alerts per Intervention Time/Times a a the total number of patients with critical values (alerts) were 157, 454 and 424 for month one, three and six respectively.The total number of patients who did not have a critical value (no alert) were 1615, 1099 and 974 respectively.
Therapy Changes in the Intervention Group of Patients Who Visited a Physician and Were Called Versus Not Called Classification of Notes by Study Nurses of Patients Who Were Not Called Following an Alert (n= 560) Difference in the PROM-Score-Change From Baseline to 12 Months Post-Surgery, Between Intervention and Control Group (Compliance Corrected Study Population) eTable 7.b Reaction by refers to the month in which a change in therapy was reported.This row specified the next measurement time after the call taking place/not taking place, as the reaction based on the call could have only be visible in the data after the call at the next measurement time.eFigure 5. Other © 2023 Steinbeck V et al.JAMA Network Open.eTable 8.
Mixed Effects Models With PROM-Score Changes as Outcomes, Controlling for Gender, Age, Mobilization and Admission PROM Scores as Fixed Effects and Hospital as Random Intercept (Intention to Treat Study Population) Mixed Effects Models With PROM-Score Changes as Outcomes, Controlling for Gender, Age, Mobilization and Admission PROM Scores as Fixed Effects and Hospital as Random Intercept (Compliance Corrected Study Population) Mixed Effects Models With PROM-Score Change Z-Scores as Outcomes, Controlling for Gender, Age, Mobilization and Admission PROM Scores as Fixed Effects and Hospital as Random Intercept (Intention to Treat Study Population) Mixed Effects Models With PROM-Score Change Z-Scores as Outcomes, Controlling for Gender, Age, Mobilization and Admission PROM Scores as Fixed Effects and Hospital as Random Intercept (Compliance Corrected Study Population) for the PROMIS-fatigue and 41-79.4 for the PROMIS-depression.Due to the low score range of the EQ-5D-5L, three digits after the decimal point are reported b The mean difference in points refers to the difference between intervention and control group in terms of their mean change in the respective PROM-score © 2023 Steinbeck V et al.JAMA Network Open.eTable 9. © 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.eTable 11. © 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.eTable 12. © 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.© 2023 Steinbeck V et al.JAMA Network Open.