Who we are
My Meta-Analysis supports researchers who need rigorous evidence synthesis and statistical analysis for publication. The work centers on meta-analysis and systematic review, including network meta-analysis, Bayesian meta-analysis and meta-regression, and extends to the surrounding tasks that determine whether a paper is accepted: protocol development, search strategy, data extraction, risk-of-bias and certainty assessment, manuscript preparation and responses to peer review.
Medicine and the health sciences are the flagship area, because that is where the methods of evidence synthesis are most developed. The same methods are used in psychology, education, management, economics, the social sciences, ecology and other fields, with the study designs, effect measures and conventions that each discipline expects.
How we approach the work
Evidence synthesis fails in predictable ways: a question that is too vague to answer, a search that missed relevant studies, pooling of studies that do not ask the same question, analyses chosen after the results were seen, and reports that leave the reader unable to judge what was done. Our approach is designed around those failure points.
- Method before software. The design, the model and the analysis plan are chosen for the question and the studies available, and are written down before results are seen.
- Published standards. Reviews are planned and reported against recognized guidance, including the Cochrane Handbook, PRISMA 2020 and its extensions, MOOSE and GRADE.
- Honest scope. Not every method suits every field or every dataset. Where pooling is not justified, we say so and propose another design.
- Reproducibility. Analyses are documented so that a second person can repeat them.
- Clear responsibility. The researchers remain the authors and are accountable for their work. Support is described openly, as set out in the research integrity and authorship statement.
What makes an evidence synthesis trustworthy
A reader judging a review or meta-analysis can ask eight questions, and they describe what we try to deliver. Was the question specific enough to be answered, and was it stated before the work began? Was there a protocol, and is it available? Was the search comprehensive and documented, so that it could be repeated? Were studies selected by stated criteria and not by convenience? Was the risk of bias in each study assessed with an appropriate tool and taken into account in the conclusions? Were the studies combined by methods that suit the data, with heterogeneity examined and not ignored? Is the report complete enough for another researcher to judge what was done and to reproduce it? And do the conclusions stay within what the evidence supports, with the certainty of that evidence stated?
These are not abstract ideals. They are the points at which reviews are most often criticized by editors and peer reviewers, and the points that determine whether a synthesis is useful to the people who rely on it. A project that is planned around them is also more likely to be published, although no outcome can be promised.
Standards we work to
- Cochrane Handbook
- The main reference for the methods of systematic reviews of interventions, including meta-analysis, risk of bias and certainty of evidence.
- PRISMA 2020
- The reporting guideline for systematic reviews, with a checklist and flow diagram and extensions for scoping, diagnostic accuracy, network and individual participant data reviews.
- MOOSE
- The reporting guideline for meta-analyses of observational studies.
- GRADE
- A structured approach to rating the certainty of a body of evidence and the strength of recommendations.
- PROSPERO and protocol registration
- Registration of review protocols, which limits selective reporting by fixing the plan before the results are seen.
- EQUATOR Network
- A directory of reporting guidelines for different study designs, used to find the one a journal expects.
- ICMJE recommendations
- The reference on authorship, acknowledgment and the conduct of publication, summarized in our research integrity statement.
Where a journal or a discipline uses a different standard, the journal's instructions and the field's conventions take precedence, and the plan is adapted to them.
What we offer
The core services are described on their own pages: meta-analysis, systematic review, network meta-analysis, Bayesian meta-analysis, meta-regression, statistical analysis and publication support. The methods section explains each approach and when it is appropriate, and the resource library contains guides to the concepts that arise most often. The process from first enquiry to submission is described in detail.
Disciplines
The service is organized around eleven research areas: medical and health sciences, life and biological sciences, psychology and behavioral sciences, management and business, economics and econometrics, education, social sciences, environment and sustainability, agriculture and food sciences, sports and exercise sciences and engineering and technology. Each area page explains which study designs and methods are common in the field and which are not, because meta-analysis is well established in some disciplines and rare in others.
People and credentials
[OWNER VERIFICATION REQUIRED: describe the people who do the work, their roles and verified qualifications. Do not add names, degrees, affiliations or experience that cannot be substantiated.]
What we do not do
We provide research and evidence-synthesis support. We do not provide clinical advice or advice about the care of individual patients, and the results of a review are not patient-specific guidance (see the disclaimer). We do not fabricate or alter data, select results to reach a conclusion, or write references that cannot be verified. We do not guarantee acceptance of a paper by any journal. These commitments are set out further in the research integrity statement.
Working with us
If you would like to discuss a project, the contact page has a short form that asks for the question, the study type, the target journal and the deadline. The process page explains what happens next.