Identifying sincerity of effort based on the combined predictive ability of multiple grip strength tests

摘要

Study Design: Retrospective Cohort. Introduction: Detecting sincerity of effort (SOE) of grip strength remains a frustrating and elusive task for hand therapists because there are no valid, reliable, or widely accepted assessments for identifying feigned effort. Some therapists use various combinations of different SOE tests in an attempt to identify feigned effort, but there is lack of evidence to support this practice. Purpose: The present study examined the ability of a combination of three grip strength tests commonly used in the clinic to detect SOE: the five rung grip test, rapid exchange grip test, and coefficient of variation. A secondary purpose was to compare the predictive ability between the logistic and linear regression models. Methods: Healthy participants (n = 146) performed the three SOE tests exerting both maximal and submaximal efforts. We compared the ability of two regression models, the logistic and linear models, to predict sincere versus insincere efforts. Results: Combining the three tests predicted SOE better than each test alone. Yet, the full logistic model, which was the best predictor of SOE, explained only 42% of variance and correctly classified only 58% of the efforts. Conclusions: Our findings do not support the clinical practice of combining these three tests to detect SOE. Level of Evidence: Not applicable.