Fig. 2

Classification of included articles and clinical practices according to the assessment of reproducibility. Detailed Legend: ^aThe sum of clinical practices with consistent (n = 28) and inconsistent (n = 35) effect estimates between original and reproduction attempts does not sum to 66 due to three practices that could not be categorized as their single reproduction attempt was in progress [38, 42, 43]. ^bPractices wherein all reproduction attempts demonstrated similar effect estimates (e.g., all lack of efficacy). ^cPractices wherein effect estimates from each reproduction attempt differed from the previous attempt. ^dEach box represents the way in which the reproduction attempt changed the results of the original study (e.g., efficacy/harm represents practices wherein the original study demonstrated efficacy but reproduction attempt demonstrated harm)