Service
Systematic review & meta-analysis
Most rejected systematic reviews are not rejected for the statistics. They are rejected because the search was not reproducible, the bias assessment was superficial, or the studies should never have been pooled.
A publishable systematic review has a protocol registered before screening begins, a search strategy written by database with full syntax so any reader can reproduce it, at least two independent reviewers screening and extracting with a documented disagreement process, a formal risk-of-bias assessment using the appropriate instrument, a pooling decision justified on clinical and methodological grounds rather than on an I² threshold, heterogeneity investigated rather than reported, an assessment of small-study effects, and a certainty rating using GRADE. Reviews are most often rejected because the search cannot be reproduced or because incompatible studies were pooled.
What this service includes
- Protocol & registration
- PRISMA-P protocol and PROSPERO registration completed before screening, which editors increasingly check.
- Reproducible search
- Search strategies built per database — MEDLINE, Embase, Cochrane CENTRAL, Scopus, Web of Science, CINAHL and grey literature — with full syntax and date of search reported.
- Dual screening & extraction
- Two independent reviewers, a documented adjudication process, kappa agreement, and a complete PRISMA flow diagram.
- Risk of bias
- RoB 2, ROBINS-I, ROBINS-E, Newcastle–Ottawa, QUADAS-2 and PROBAST as appropriate, with domain-level judgements and supporting quotations.
- Advanced pooling
- Random-effects models with REML and Hartung–Knapp adjustment, prediction intervals, meta-regression, influence and leave-one-out diagnostics, and small-study-effect assessment.
- Network & IPD meta-analysis
- Frequentist and Bayesian network meta-analysis with transitivity, inconsistency and node-splitting; one-stage and two-stage IPD models.
- Diagnostic accuracy synthesis
- Bivariate and HSROC models, joint pooling of sensitivity and specificity, and explicit handling of threshold effects.
- Certainty of evidence
- GRADE summary-of-findings tables and CINeMA for networks, so the conclusion states how much confidence is warranted.
What you receive
- Registered protocol (PROSPERO)
- Reproducible search strategies per database
- PRISMA flow diagram and screening log
- Risk-of-bias tables with judgements
- Forest, funnel and network plots
- GRADE summary-of-findings table
- Full manuscript draft
Independently verified before it reaches you
Every primary result is reproduced from the cleaned dataset by a second statistician before delivery, and you receive the analysis scripts so any number in the manuscript can be traced back to a line of code and a row of data.
Common questions
Related
Often needed alongside this.
Biostatistics & data analysis
Descriptive statistics through to advanced modelling — with assumptions tested, methods justified, and every number reproducible.
ExploreManuscript & publication support
Writing, scientific editing, reporting-guideline compliance, journal selection, submission and point-by-point reviewer responses.
ExploreResearch mentorship
Structured, project-based training that ends with you running your own research — not with you needing us again.
ExploreTell us about your project.
Book a free 15-minute consultation. Tell us the research question, the data you have, and the deadline — we will tell you honestly what is achievable.
Or call +20 100 163 8864 · Sunday–Thursday, 09:00–18:00 (GMT+2, Cairo)