The Marchand Institute produces high-quality systematic reviews and meta-analyses at unusual volume. Our reviews are prospectively registered with PROSPERO, reported to PRISMA standards, and appraised with the Cochrane Risk of Bias 2 (RoB 2) tool for randomized trials. This page sets out exactly how we work, so that reviewers, editors, journalists and prospective collaborators can judge the process rather than take our word for it.
Scale of Output
| Indexed works | 137 (97 peer-reviewed articles, 35 conference presentations, 4 book chapters and textbooks, 1 patent) |
|---|---|
| PROSPERO protocol registrations | 65 under the Marchand Institute as organizational affiliation; 41 with Dr. Marchand as named contact |
| Citations | 1,200+ (Google Scholar, verified August 2026); 702 (Scopus); 641 (Web of Science Core Collection) |
| h-index | 18 (Google Scholar); 15 (Scopus); 15 (Web of Science) |
| i10-index | 26 (Google Scholar) |
| Typical annual output | 10–12 peer-reviewed articles plus 3–5 conference presentations |
For a comparison of scale: an independent, unfunded institute sustaining double-digit peer-reviewed meta-analyses per year is unusual. We are able to do this because evidence synthesis is the Institute’s core activity rather than a side project, and because we maintain a standing, trained research team rather than assembling one per study.
How We Choose a Question
We prioritise questions where a surgical or clinical decision is being made routinely on contested or outdated evidence, where randomized trials exist but have never been pooled, or where a previous review is materially out of date. We deliberately favour questions that change what a surgeon does on Monday morning over questions that are merely publishable.
Because we accept no industry funding, no sponsor has any influence over which questions we ask, which comparators we include, or whether we publish a null or unfavourable result.
Protocol Registration
Review protocols are registered prospectively with PROSPERO, the international prospective register of systematic reviews maintained by the Centre for Reviews and Dissemination at the University of York, before screening begins. Registration fixes the question, the eligibility criteria, the primary and secondary outcomes and the planned analysis in advance, which is the principal defence against outcome-switching and post-hoc analysis. Where the final review departs from the registered protocol, the deviation and its reason are reported in the manuscript.
Searching
We search PubMed / MEDLINE, Cochrane CENTRAL, Scopus and Web of Science from inception to a stated cut-off date, supplemented by hand-searching of reference lists and, where relevant, trial registries and conference proceedings. Search strategies are reported in full so that they can be reproduced. Searches are re-run before submission where a long interval has elapsed.
Screening and Data Extraction
Records are screened by two reviewers working independently at both title/abstract and full-text stages, with disagreements resolved by discussion or by a third reviewer. Data extraction is likewise performed in duplicate onto a piloted extraction form. Study selection is reported in a PRISMA flow diagram giving records identified, duplicates removed, records screened, full texts assessed, and exclusions with reasons.
Risk of Bias and Certainty
Randomized controlled trials are appraised with the Cochrane Risk of Bias 2 (RoB 2) tool across its five domains: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Non-randomized comparative studies are appraised with tools appropriate to the design, such as ROBINS-I or the Newcastle-Ottawa Scale. Where the evidence base warrants it, we apply a trustworthiness assessment to screen for implausible or irreproducible data before inclusion.
Certainty of evidence is characterised in the discussion, and we state plainly when pooled estimates are too imprecise or too heterogeneous to support a clinical recommendation.
Synthesis
We pool using random-effects models as the default, given expected clinical and methodological heterogeneity in surgical literature. Dichotomous outcomes are reported as risk ratios or odds ratios with 95% confidence intervals; continuous outcomes as mean differences or standardised mean differences. Heterogeneity is quantified with I² and Cochran’s Q. Where the data allow, we run subgroup and sensitivity analyses specified in the protocol, and we assess small-study effects and publication bias.
We also use methods suited to specific problems: time-stratified meta-analysis, where outcomes are compared only at matched follow-up intervals so that apparent differences are not artefacts of measurement timing; network meta-analysis, for comparing multiple devices or techniques that have never been tested head to head; and single-arm meta-analysis, where no comparator literature exists.
Reporting
All reviews are reported in accordance with the PRISMA statement, including the flow diagram, the full search strategy, the risk-of-bias summary, forest plots for each pooled outcome, and an explicit statement of limitations. Conference abstracts arising from a review are cross-referenced to the full publication.
Openness
We publish open access wherever the journal and our resources allow, and a substantial share of our catalogue is freely available through PubMed Central. Every item in our publications database carries its DOI and, where assigned, its PubMed identifier and PMCID, so that any claim we make can be traced to source. New work is announced through our publications RSS feed.
We will provide extracted datasets, search strategies or risk-of-bias assessments on reasonable request from other researchers.
Independence
The Marchand Institute is a 501(c)(3) tax-exempt non-profit (EIN 42-2857240) and accepts no pharmaceutical or medical device industry funding. Investigators declare no conflicts of interest and hold no financial relationships with industry. Read the full funding and independence statement.
Collaborate With Us
We take on external collaborators on systematic reviews and meta-analyses, including clinicians, residents, fellows and students. If you have a question that meets the criteria above, see Propose a Collaboration. You can also meet the research team or read about our clinical trials programme.