ArticleSingle vs composite measures of pain intensity: Relative sensitivity for detecting treatment effectsJensen, Mark P.a,*; Hu, Xiaojunb; Potts, Susan L.b; Gould, Errol M.bAuthor Information Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. aDepartment of Rehabilitation Medicine, University of Washington, Seattle, WA, USA bEndo Pharmaceuticals, Chadds Ford, PA, USA *Corresponding author. Address: Department of Rehabilitation Medicine, University of Washington, Box 359612, Harborview Medical Center, 325 Ninth Avenue, Seattle, WA 98104, USA. Tel.: +1 206 543 3185; fax: +1 206 685 3244. E-mail address:[email protected] Article history: Received 12 July 2012; Received in revised form 16 November 2012; Accepted 12 December 2012. Pain: April 2013 - Volume 154 - Issue 4 - p 534-538 doi: 10.1016/j.pain.2012.12.017 Buy Metrics Abstract Summary A single-item recall rating of pain intensity is as valid as composite scores made up of up to 9 items for detecting changes in pain. Assay sensitivity remains a significant issue in pain clinical trials. One possible method for increasing assay sensitivity for detecting changes in pain intensity is to increase the reliability of pain intensity assessment by increasing the number of intensity ratings obtained, and combining these ratings into composite scores. The current study performed secondary analyses from a published clinical trial to test this possibility. The reliability and assay sensitivity pain intensity scores made up of 1 to 9 24-hour pain intensity recall ratings were compared. Although the reliability of the outcome measures improved as the number of items increased, this increase in reliability was not associated with an increase in assay sensitivity. A single 24-hour recall rating was about as valid (sensitive) for detecting treatment effects as composite scores made up of 2 to 9 different ratings. If this finding replicates in other pain populations, it has significant implications for the design and conduct of pain clinical trials. Specifically, it suggests the possibility that assessment burden (and associated costs and problems related to missing data) might be greatly reduced by specifying a single recall rating as the primary outcome variable. Research is needed to explore this possibility further. © 2013 Lippincott Williams & Wilkins, Inc.