This study reports on research to assess the accuracy of the pain intensity numeric rating scale (NRS) in evaluating the clinical importance of patients' pain. The Joint Commission on Accreditation of Health Care Organizations (JCAHO) requires that hospitals and clinics assess patients for pain, but does not mandate how that should be done. The NRS, the most widely used tool for assessing pain, requires patients to rate their pain from 0-10, with 10 being the most severe. However, the NRS was never validated as a screening test, although its use is mandated by some health care systems, including the VA system.
Two hundred and seventy five patients were enrolled in this study in 2005–2006 at a general internal medicine clinic at the University of North Carolina. Participants completed a face-to-face interview after seeing their physicians; afterwards, health data were abstracted from their electronic medical records. During the interview, patients were also administered the Brief Pain Inventory interference scale (BPI) as a means of obtaining a core outcome measurement for their pain.
Results indicated that, compared to the BPI, the accuracy of the NRS in screening for pain was only fair. Approximately one-third of patients that scored as having pain-related functional interference on the BPI scored a 0 (i.e., no pain) on the NRS. Additionally, 28 percent of patients who had reported pain as the primary reason for their visit received an NRS score of 0. The authors conclude that the ability of the NRS to capture patients' pain-related suffering (its sensitivity as a screening test) is only modest. Given the already stretched time and resources of many primary care physicians, the modest sensitivity and accuracy of this test raises the question of whether administering it widely is justified, given the lack of empirical evidence supporting its benefits in improving patient outcomes. At the least, the authors suggest, researchers should gather further data to determine whether pain screening actually improves patient outcomes, and whether the accuracy and effectiveness of the NRS can be improved, before its use becomes even more widespread.