Development of PRO-PMs assessing symptoms following completion of curative-intent chemotherapy.

Abstract

171 Background: Few patient-reported outcome performance measures (PRO-PMs) have been validated for the cancer population. The testing that has occurred mostly focuses on advanced cancers despite the fact that the majority of people with cancer are diagnosed with earlier stage disease. We developed and tested PRO-PMs to assess quality of life, pain and fatigue in adult patients completing curative-intent chemotherapy for breast, colon and non-small cell lung cancers. Our goal is to develop measures that target symptoms that impact entry into the survivorship phase. Methods: We recruited 20 diverse test sites from the Michigan Oncology Quality Collaborative (MOQC) and the Alliance of Dedicated Cancer Centers (ADCC). Test sites enrolled patients, administered surveys, and collected clinical and demographic data. A Technical Expert Panel and the Patient and Caregiver Council selected PROMs and provided testing guidance. We assessed data collection feasibility and clinician/staff/patient burden throughout the testing process. Results: PROMIS instruments were selected due to psychometric testing in the target population, public availability and acceptability to patients and test sites. 1,753 patients were enrolled between 10/1/19 – 3/31/21. The COVID public health emergency disrupted testing and resulted in lower than expected enrolled patients/completed surveys; however, adaptations led to expansions in survey administration methods. Preliminary practice-level performance results from 10 sites show variation across sites for pain interference (mean = 50.5, SD = 2.8, with a range of 44.6—54.6) and fatigue (mean 49.2, SD = 2.8, with a range of 44.6--54.3). Some test sites reported PRO implementation to be burdensome; however, most patients evaluated did not find survey completion to be burdensome. Conclusions: Next steps include testing risk adjustment variables/model, creating adjusted performance scores, reliability and validity testing. Despite the consensus goal of PRO-PMs in oncology, barriers to implementation persist and important methodologic barriers exist (e.g., ability to achieve sufficient sample size in an oncology practice; defining the most appropriate numerator calculation that reports the desired quality concept and is appropriate for the PROM being used; analytic best practices for PRO-PM adjustment/testing). This project is contributing to the knowledge base as we seek to ensure that PRO-PMs provide meaningful, actionable, patient-centered quality data with benefits that outweigh the burden of implementation.