False positives associated with responder/non-responder analyses based on motor evoked potentials

Background: A trend in the non-invasive brain stimulation literature is to assess the outcome of an intervention using a responder analysis whereby participants are di- or trichotomised in order that they may be classified as either responders or non-responders. Objective: Examine the extent of the Type I error in motor evoked potential (MEP) data subjected to responder analyses. Methods: Seven sets of 30 MEPs were recorded from the first dorsal interosseous muscle in 52 healthy volunteers. Four classification techniques were used to classify the participants as responders or non-responders: (1) the two-step cluster analysis, (2) dichotomised thresholding, (3) relative method and (4) baseline variance method. Results: Despite the lack of any intervention, a significant number of participants were classified as responders (21–71%). Conclusion: This study highlights the very large Type I error associated with dichotomising continuous variables such as the TMS MEP