Authors: Sean M. Murphy, Robert Rosenman, Daniel L. Friesner, Janet Raisor
Addresses: Accounting, Economics and Finance, West Texas A&M University, P.O. Box 60187, Canyon, TX 79016-0001, USA. ' School of Economic Sciences, Washington State University, P.O. Box 646210, Pullman, WA 99164-6210, USA. ' Department of Pharmacy Practice, North Dakota State University, P.O. Box 6050, Fargo, ND 58108-6050, USA. ' Rehabilitation Professional Services, St. Mary's Medical Center, 3700 Washington Avenue, Evansville, IN 47750, USA
Abstract: In this paper, we apply recent advances in discrete choice modelling to characterise what patient-specific characteristics influence intervention effectiveness when effectiveness is measured by a reduction in perceived pain. Our analysis goes beyond previous studies by examining whether misclassifying patient responses influences the outcomes of such analyses. Using data from several outpatient rehabilitation sites on patients with back injuries, we find that the scores are misclassified. We also find evidence suggesting that distributional misspecification may bias parameter and misclassification probability estimates.
Keywords: perceived back pain; misclassification; monotone rank estimator; MRE; perceived pain; discrete choice modelling; back injuries; distributional misspecification; health outcomes; healthcare.
International Journal of Behavioural and Healthcare Research, 2010 Vol.2 No.2, pp.123 - 135
Published online: 20 Dec 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article