Original ArticleValidity of the Brief Pain Inventory for Use in Documenting the Outcomes of Patients With Noncancer PainKeller, San PhD*; Bann, Carla M. PhD†; Dodd, Sheri L. MSc‡; Schein, Jeff DrPH, MPH‡; Mendoza, Tito R. PhD§; Cleeland, Charles S. PhD§ Author Information From the *American Institutes for Research, Chapel Hill, NC; †RTI International, Research Triangle Park, NC; ‡Janssen Pharmaceutica, Inc., Titusville, NJ; and §M. D. Anderson Pain Research Group, Houston, TX. This work was supported by a contract with Janssen Pharmaceutica, Inc. Reprints: San Keller, PhD, American Institutes for Research, 1501 E. Franklin Street, Suite 200, Chapel Hill, NC 27514-2825 (e-mail: [email protected]). Received for publication June 9, 2002; revised January 11, 2003; second revision March 15, 2003; accepted March 17, 2003. The Clinical Journal of Pain: September/October 2004 - Volume 20 - Issue 5 - p 309-318 Buy Abstract Objectives: The Brief Pain Inventory (BPI) is a short, self-administered questionnaire that was developed for use in cancer patients. While most empirical research with the BPI has been in pain of that etiology, the questionnaire is increasingly evident in published studies of patients with non-cancer pain. The current research addresses the need for formal evaluation of the reliability and validity of the BPI for use in non-cancer pain patients. Methods: Approximately 250 patients with arthritis or low back pain (LBP) self-administered a number of generic and condition-specific health status measures (including the BPI) in the clinic of their primary care provider at 2 time points: the initial clinic visit and the first visit following treatment. Results: The reliability of BPI data collected from non-cancer pain patients was comparable to that reported in the literature for cancer patients and sufficient for group-level analyses (coefficient alphas were greater than 0.70). The factor structure of the BPI was replicated in this sample and the relationship of the BPI to generic measures of pain was strong. The BPI exhibited similar relationships to general and condition-specific measures of health as did a generic pain scale (SF-36 Bodily Pain). Finally, the BPI discriminated among levels of condition severity and was sensitive to change in condition over time in arthritis and LBP patients. Discussion: Results support the validity of the BPI as a measure of pain in patients without cancer and, in particular, as a measure of pain for arthritis and LBP patients. © 2004 Lippincott Williams & Wilkins, Inc.