H Smythe1. 1. Department of Medicine, University of Toronto, ON, Canada.
Abstract
OBJECTIVE: To minimize interobserver variation in tender point examination, a training program was developed that focused on learning to deliver 4 kg force. METHODS: A 3 step process was designed to provide both measurement and feedback, serving both didactic and research functions. There were 5 repetitions within each step. During the first and 3rd steps, the subjects pressed on the footplate of a dolorimeter with no immediate feedback about the force measured. In the 2nd "training" step, they were given immediate feedback about the force delivered, and were taught to watch the amount and pattern of blanching beneath and around the thumbnail. RESULTS: Twenty-nine observers participated in 30 studies. Initial values varied from 1.78 to 8.92 kg (means within observers), although the mean value was 4.05 kg (means of all observers). The training step reduced observer variation by 65%, with reduction of mean absolute error from 1.25 to 0.44 kg. This improvement was sustained in the 3rd step, in which no feedback was given. Naïve participants performed as well as experts, and there were no detectable sex differences. One observer was retested after an interval of 8 months, and performed no better during the the initial phase of the 2nd test than in the first. Each training session required 3 to 5 minutes. CONCLUSION: Specific training efficiently reduces interobserver variations in pressure used in tender point examinations. These skills should be periodically refreshed to prevent drift into error.
OBJECTIVE: To minimize interobserver variation in tender point examination, a training program was developed that focused on learning to deliver 4 kg force. METHODS: A 3 step process was designed to provide both measurement and feedback, serving both didactic and research functions. There were 5 repetitions within each step. During the first and 3rd steps, the subjects pressed on the footplate of a dolorimeter with no immediate feedback about the force measured. In the 2nd "training" step, they were given immediate feedback about the force delivered, and were taught to watch the amount and pattern of blanching beneath and around the thumbnail. RESULTS: Twenty-nine observers participated in 30 studies. Initial values varied from 1.78 to 8.92 kg (means within observers), although the mean value was 4.05 kg (means of all observers). The training step reduced observer variation by 65%, with reduction of mean absolute error from 1.25 to 0.44 kg. This improvement was sustained in the 3rd step, in which no feedback was given. Naïve participants performed as well as experts, and there were no detectable sex differences. One observer was retested after an interval of 8 months, and performed no better during the the initial phase of the 2nd test than in the first. Each training session required 3 to 5 minutes. CONCLUSION: Specific training efficiently reduces interobserver variations in pressure used in tender point examinations. These skills should be periodically refreshed to prevent drift into error.