CORR® publishes systematic reviews, meta-analyses, and network meta-analyses. We do not consider selective reviews or book-chapter-style overview articles.
Authors of systematic reviews and meta-analyses must identify all meta-analyses and systematic overviews published in past 5 years. We will not accept meta-analyses or systematic overviews on the same topic published within five years unless the authors can demonstrate that the literature has substantially changed or that the earlier analyses were flawed. Meta-analyses or systematic reviews will not be accepted if the same (or largely the same) papers are used to arrive at similar conclusions as those in previous reviews. For more information, please see:
A review article designed to answer specific, testable questions by querying prespecified databases with specific search terms, that applies clear inclusion/exclusion criteria to determine which source articles can help answer those questions, assesses study quality of the source articles, and answers the research questions without pooling (averaging or combining) data from the source studies together.
In general, CORR® discourages systematic reviews of level-IV evidence, and systematic reviews of retrospective evidence (even retrospective-comparative studies) will receive particular scrutiny.
A systematic review as defined above that also pools (averages or mathematically combines) data from the source studies. Additional methodological steps must be taken to be sure that this is done rigorously, including assessing for heterogeneity (and accounting for it in the analysis performed), and assessing the population(s) of source studies for the possibility of positive-outcome bias (publication bias).
In general, CORR® permits meta-analysis only when the source studies are randomized (or high quality prospective, comparative trials), and only when the source studies have reasonably similar inclusion and exclusion criteria.
A meta-analysis, as defined above, that uses statistical approaches to compare treatments that were not compared directly in the original head-to-head trials (e.g., If there are randomized controlled trials comparing treatment A to treatment B, and treatment B to treatment C, a network meta-analysis will synthesize the direct comparisons of A to B and B to C, but also use mathematical approaches to indirectly compare A to C).
As with traditional meta-analyses, network meta-analyses should only be performed when the source studies are randomized (or high quality prospective, comparative trials), and only when the source studies have reasonably similar inclusion and exclusion criteria. Additional issues arise in these studies, as well, including transitivity, coherence, effect modifiers, and trustworthiness of treatment rankings. Network meta-analyses should should only be done in collaboration with an expert statistical team.