The present study examined psychometric properties of facial expressions of pain. A diverse sample of 129 people suffering from shoulder pain underwent a battery of active and passive range-of-motion tests to their affected and unaffected limbs. The same tests were repeated on a second occasion. Participants rated the maximum pain induced by each test on three self-report scales. Facial actions were measured with the Facial Action Coding System. Several facial actions discriminated painful from non-painful movements; however, brow-lowering, orbit tightening, levator contraction and eye closing appeared to constitute a distinct, unitary action. An index of pain expression based on these actions demonstrated test-retest reliability and concurrent validity with self-reports of pain. The findings support the concept of a core pain expression with desirable psychometric properties. They are also consistent with the suggestion of individual differences in pain expressiveness. Reasons for varying reports of relations between pain expression and self-reports in previous studies are discussed.

Background and Purpose. The Roland-Morris Questionnaire (RMQ) is a self-administered disability measure in which greater levels of disability are reflected by higher numbers on a 24-point scale. The RMQ has been shown to yield reliable measurements, which are valid for inferring the level of disability, and to be sensitive to change over time for groups of patients with low back pain. Little is known about the usefulness of this instrument in aiding decision making regarding individual patients. The purpose of this study was to determine the minimum level of detectable change when the RMQ is applied to individual patients. Subjects. The study sample consisted of 60 outpatients with low back pain. Methods. The RMQ was administered at the subjects' initial visit and again 4 to 6 weeks later. Conditional standard errors of measurement (CSEMs) were computed for initial and follow-up RMQ scores, and these values were used to estimate the minimum level of detectable change. Results. Minimum levels of detectable change at the 90% confidence level varied from 4 to 5 RMQ points. Conclusion and Discussion. The magnitude of CSEMs is sufficiently small to detect change in patients with initial scores in the central portion of the scale (4–20 RMQ points); however, the magnitude is too large to detect improvement in patients with scores of less than 4 and deterioration in patients who have scores greater than 20.

Pain is typically assessed by patient self-report. Self-reported pain, however, is difficult to interpret and may be impaired or in some circumstances (i.e., young children and the severely ill) not even possible. To circumvent these problems behavioral scientists have identified reliable and valid facial indicators of pain. Hitherto, these methods have required manual measurement by highly skilled human observers. In this paper we explore an approach for automatically recognizing acute pain without the need for human observers. Specifically, our study was restricted to automatically detecting pain in adult patients with rotator cuff injuries. The system employed video input of the patients as they moved their affected and unaffected shoulder. Two types of ground truth were considered. Sequence-level ground truth consisted of Likert-type ratings by skilled observers. Frame-level ground truth was calculated from presence/absence and intensity of facial actions previously associated with pain. Active appearance models (AAM) were used to decouple shape and appearance in the digitized face images. Support vector machines (SVM) were compared for several representations from the AAM and of ground truth of varying granularity. We explored two questions pertinent to the construction, design and development of automatic pain detection systems. First, at what level (i.e., sequence- or frame-level) should datasets be labeled in order to obtain satisfactory automatic pain detection performance? Second, how important is it, at both levels of labeling, that we non-rigidly register the face?

Background and Purpose. This study compared the ability of the Roland-Morris (RM), Oswestry (OSW), and Jan van Breemen Institute (JVB) pain and function questionnaires to detect change over time. Subjects. The sample consisted of 88 patients with mechanical low back pain who were referred by physicians to the outpatient physical therapy department of a teaching hospital. Methods. Questionnaires were completed by the subjects at their initial visit and 4 to 6 weeks later. Clinically important change was estimated by having the subject and the clinician independently complete two rating scales. Sensitivity to change was assessed using receiver operating characterisic (ROC) curve analysis. Results. The ROC curve areas for the RM (0.79), OSW (0.78), and JVB pain (0.79) questionnaires were significantly greater than for the JVB function questionnaire (0.66). Blank and multiple responses per item were present on approximately 20% of the OSW questionnaires and 14% of the JVB questionnaires. Words rather than checks were evident on 3% of the RM questionnaires. Conclusion and Discussion. Based on the latter finding, we believe the RM questionnaire may be the preferred instrument for assessing change over time in patients with low back pain.