Considerations for multi-centre conditioned pain modulation (CPM) research; an investigation of the inter-rater reliability, level of agreement and confounders for the Achilles tendon and Triceps Surae

Objective: This study aimed to investigate the inter-rater reliability of the conditioned pain modulation (CPM) effect. Methods: The reliability between two examiners assessing the CPM effect via pressure pain thresholds and induced using the cold pressor test of 28 healthy volunteers at the mid-portion Achilles tendon (AT) and Triceps Surae musculotendinous junction was performed. Reliability was calculated using intraclass correlation coefficient (ICC). Confounders were assessed using multivariable generalised estimating equations (GEES). Bias in the level of agreement was assumed if the confidence intervals (Cis) of the mean difference in Bland-Altman plots did not cross the line of equality. Results: The inter-rater reliability of the CPM effect was poor to moderate in the AT (ICC 95% CI =0.00-0.66) and Triceps Surae (ICC 95% CI =0.00-0.69). However, when accounting for confounders within the GEE, there were no differences between testers and Bland-Altman plots reported good agreement between testers. Habitual completion of running-related physical activity was a confounder for both the AT parallel-paradigm (p=0.017) and sequential-paradigm (p=0.029). Testing order was a confounder for the AT (p=0.023) and Triceps Surae (p=0.014) parallel-paradigm. Conclusion: This study suggests the CPM effect may be site specific (i.e. differences between the AT and Triceps Surae exist). In addition, differences in the reliability between examiners are likely due to the influence of confounders and not examiner technique and therefore appropriate analysis should be used in research investigating the CPM effect.

Automated summary

This study found that the inter-rater reliability of the CPM effect varied at the Achilles tendon and Triceps Surae musculotendinous junction, but after considering confounders, there were no significant differences between testers.