Defining a patient-reported outcome measure (PROM) selection process: What criteria should be considered when choosing a PROM for routine symptom assessment in clinical practice?

187 Background: PROMs capture the patient’s perspective on their health status and thus improve person-centred care. The Edmonton Symptom Assessment System has been the standard PROM used for all patients in Ontario. Recently, Cancer Care Ontario has chosen to expand the use of PROMs by adding cancer specific tools. The objective of this work was to develop a process to systematically evaluate and select PROMs for routine use in clinical practice. Methods: The development of the PROM selection process included several steps: 1) a literature scan to identify existing PROM selection guidelines and considerations; 2) development of clinically relevant criteria by a panel of methodological experts and patient advisors; and 3) use of a case study to test the process and identify gaps. In the case study, disease-site experts and patients were given the opportunity to review the PROMs and provide feedback. Results: The literature scan highlighted six resources endorsing three common considerations for PROM selection: symptom coverage, usability, and psychometric properties. The expert panel further developed criteria for each consideration. Symptom coverage criteria include percentage of prevalent, bothersome and expert endorsed symptoms addressed by the PROM. Usability criteria include: reference time-frame, scale, time to complete, scoring, plain language and translations. Psychometric criteria are: internal consistency, reliability, responsiveness, discrimination ability, meaningful change and translation validity. Each consideration is scored as weak, average or good based on how well the criteria are met. A summary matrix illustrates the overall assessment. In the case study, the selection process identified two top-performing PROMs; disease-specific patient focus groups were held to inform the final decision. Conclusions: This rigorous evaluation of candidate measures resulted in the selection of a PROM that was accepted by methodological experts, clinical advisors and patients. The case study demonstrated the value of disease-specific patient input on PROM selection.