Background: AMA's Guides to the Evaluation of Permanent Impairment is used to rate loss of function and determine compensation and ability to work after injury or illness; however, there are few studies that evaluate reliability or construct validity.
Objectives: To evaluate the reliability of the fifth and sixth editions for back injury; to determine best methods for further study.
Results: Intra-class correlation coefficients within and between raters were relatively high. There was wider variability for individual cases. Impairment ratings were lower and correlated less well for the sixth edition, though confidence intervals overlapped.
Conclusions: The sixth edition may not be an improvement over the fifth. A research agenda should include investigations of reliability and construct validity for different body sites and organ systems along the entire rating scale and among different categories of raters.