We read with a particular interest the study by Cuenca-Martínez et al,1 which found neural mobilization intervention having at least a moderate clinical benefit on pain intensity and disability in people with different musculoskeletal disorders and on mechanosensitivity in asymptomatic people. These conclusions were obtained by conducting an overview (ie, review of reviews). The authors performed 3 meta-meta-analyses (MMAs), one for each clinical outcome, in which the effect estimate was the standardized mean difference (SMD).
Overviews are second-order evidence synthesis studies, as their unit of analysis is systematic reviews (SRs) with or without MA. These studies have challenges, including the potential overlap of primary studies included in SRs.2 When 1 or more primary studies are included in 2 or more SRs, the results and conclusions of the overview may be biased as study data might be counted more than one time, leading to a biased and over-precisely pooled effect size. The study by Cuenca-Martínez et al1 appears to be one such case.