Choosing and Using Patient-Reported Outcome Measures in Clinical Practice

Abstract The increasing use of patient-reported outcome (PRO) measures is forcing clinicians and health care systems to decide which to select and how to incorporate them into their records and clinical workflows. This overview addresses three topics related to these concerns. First, a literature review summarizes key psychometric and practical factors (such as reliability, responsiveness, computer adaptive testing, and interpretability) in choosing PROs for clinical practice. Second, three clinical decision support (CDS) issues are highlighted: gathering PROs, electronic health record impact on providers, and incorporating PROs into CDS design and implementation. Lastly, the salience of cross-cutting domains as well as nine key pragmatic decisions are reviewed. Cross-cutting domains are those that are relevant across most medical and mental health conditions, such as the SPADE symptom pentad (sleep problems, pain, anxiety, depression, and low energy/fatigue) and physical functioning. The nine pragmatic decisions include: 1) generic vs. disease-specific scales; 2) single- vs. multi-domain scales; 3) universal scales vs. user-choice selection; 4) number of domains to measure; 5) prioritization of domains when multiple domains are assessed; 6) action thresholds; 7) clinical purpose (screening vs. monitoring); as well as the 8) frequency and 9) logistical aspects of PRO administration.