Interrater reliability of a Pilates movement-based classification system

SummaryObjectiveTo determine the interrater reliability for identification of a specific movement pattern using a Pilates Classification system.MethodVideos of 5 subjects performing specific movement tasks were sent to raters trained in the DMA-CP classification system.ResultsNinety-six raters completed the survey. Interrater reliability for the detection of a directional bias was excellent (Pi = 0.92, and Kfree = 0.89). Interrater reliability for classifying an individual into a specific subgroup was moderate (Pi = 0.64, Kfree = 0.55) however raters who had completed levels 1–4 of the DMA-CP training and reported using the assessment daily demonstrated excellent reliability (Pi = 0.89 and Kfree = 0.87).ConclusionThe reliability of the classification system demonstrated almost perfect agreement in determining the existence of a specific movement pattern and classifying into a subgroup for experienced raters. There was a trend for greater reliability associated with increased levels of training and experience of the raters.