PURPOSE: The patient-reported outcomes (PRO) field is replete with instruments measuring the same concepts, but using different questions, response scales, and scoring methods. The large number of different instruments hinders meta-analyses and comparisons of outcomes across studies assessing disease burden, clinical effectiveness, and health systems performance. To overcome these obstacles, methodology from educational testing has recently been adopted to develop common, concept-based metrics (eg, depression, anxiety, fatigue, physical function, and pain interference), allowing results from one PRO measure to be expressed in the metric of another measure using crosswalk tables. Transforming PRO scores to a common metric can potentially bridge gaps between databases and enable broader understanding of patient-centered outcomes. The objectives of this workshop are to educate attendees regarding the methodology, data and statistical requirements, application, and challenges associated with developing and using crosswalks.
DESCRIPTION: The session will include: a) an outline of the standard statistical assumptions and data requirements for developing crosswalks and how psychometric approaches differ from those used in health economics, b) an overview of the most common designs and analytic methods for developing crosswalks, c) specific examples of projects to develop crosswalks including the National Institutes of Health-funded PROsetta Stone project, which has linked Patient-Reported Outcomes Measurement Information System (PROMIS®) measures and other widely-used instruments, d) introduction to web applications (eg, common-metrics.org) that can be used by researchers and practitioners to put data from different instruments on a common scale using established crosswalks, and e) a discussion of the challenges associated with developing crosswalks in one language/culture and generalizing the results to others. Participants will provide feedback on the strengths, limitations and potential applications of the crosswalk methodology and prioritize domains and instruments for development of new crosswalks.