Adoption patterns of an electronic patient-reported outcomes tool in oncology.
e19127 Background: Patient-reported outcomes (PROs) are critical for the delivery of patient-centric care. Previous research has shown that the use of PROs can impact clinical outcomes in Oncology. Despite these benefits, PROs have not been widely implemented as part of standard of care. Significant barriers exist for its adoption including lack of integration between PRO tools and electronic health record systems (EHRs). We developed MSK Engage, a home-grown PRO tool integrated with the EHR and embedded within MSK's patient portal (MyMSK). It was launched in late 2015 and it is currently used enterprise-wide for both standard of care and research purposes. The purpose of this study was to describe the usage patterns of PROs in a real-world setting. Methods: We conducted a retrospective analysis of patients that have been assigned at least one MSK Engage survey instrument between 2015 and 2019. Patient demographic information was extracted from the EHR system. Usage patterns including the number of survey instruments assigned and completed were extracted from MSK Engage. Survey completion rates were stratified by patient demographic information and number of assigned surveys. We performed univariate and multivariate analyses to identify factors associated with survey completion rates. Results: During the study period, 57 survey instruments were implemented, with the majority being used for standard of care (41 for standard of care only, 9 for research protocols, 6 for standard of care and research, and one for a quality improvement initiative). In total, 246,638 patients were assigned 557,559 surveys. Of those, 146,505 patients completed a total of 265,174 surveys, with a median completion time of 3 minutes. Among the patients that were assigned a survey, 73.7% were women. Women in our sample were older than men, with median age 52 and 47 years old, respectively. Men were more likely to complete the assigned surveys with 61.2% completion rate among men and 43.3% among women (p < 0.001). On multivariate analysis, men were less likely to miss surveys than women (odds ratio [OR], 0.49; 95% CI, 0.48-0.49). We also identified a significant correlation between the survey completion rate and number of assigned surveys, demonstrating that assigning a large number of surveys for each patient can lead to a reduced survey response. Conclusions: Providers have quickly adopted MSK Engage demonstrating that it is feasible to incorporate PROs as part of standard of care. Patients are willing to answer PROs; however, it is important to mindful when assigning surveys to avoid negative effects of survey burden.