Estimating Pain and Disability in Virtual Patients with Low Back Pain: The Contribution of Nonverbal Behaviors

Abstract

It is well recognized that chronic low back pain (cLBP) can be estimated from nonverbal pain behaviors. However, only a few studies examined how clinicians rely on those when estimating specific outcomes, such as pain intensity and pain disability. Therefore, the present study examines (1) if facial expressions and guarding behaviors (including speed of the movement and lifting strategy) contribute to the prediction of pain intensity and disability in patients with cLBP; and (2) if these pain behaviors have been given the same importance according to the outcome. Twenty-five experienced clinicians and thirty-one novice clinicians were asked to estimate low back pain intensity and disability from a realistic virtual character performing a lifting lowering task. The studied pain behaviors were manipulated across different conditions. Pain intensity and disability were judged higher when the character moved more slowly and displayed painful facial expression. Speed of the movement and facial expressions explained a greater portion of variance when related to pain intensity assessment than to pain disability assessment. Results also showed a significant interaction between the lifting strategy, the speed of the movement and facial expressions, but only when estimating the character’s pain-related intensity. Novice clinicians rated pain disability higher than experienced clinicians did. Although pain-related concepts, pain intensity and related disability are not estimated through the same pain behaviors by clinicians. Clinical experience does not contribute to clinical judgments through the use of nonverbal pain behaviors when estimating pain outcomes but contributes to pain disability rating overall.