Systematic Review & Evidence Synthesis
This cluster covers the formal methodologies used to aggregate and synthesize existing research literature into a single, rigorous body of evidence, as distinct from primary research or narrative literature reviews. It includes systematic reviews, meta-analyses, and related synthesis methods, along with the protocols, search strategies, and quality-appraisal tools that give these methods their rigor. Two frameworks anchor this space. PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) is a reporting guideline — a checklist and flow-diagram structure — that specifies what a systematic review must document to be transparent and reproducible, covering everything from search strategy to study selection and risk-of-bias assessment. It is widely required or recommended by journals publishing this type of research. The Cochrane Handbook for Systematic Reviews of Interventions provides the underlying methodological guidance for how a systematic review should actually be conducted, particularly in health and medical research, covering protocol registration, search comprehensiveness, and methods for combining results statistically. For a research administrator, this cluster is relevant to supporting researchers who conduct or contribute to evidence syntheses: understanding what protocol registration involves, what a PRISMA-compliant submission requires from an editorial standpoint, and how systematic reviews differ in scope and rigor from other publication types. It connects to peer-review content, since evidence-synthesis manuscripts are often subject to specialized review criteria, and to publishing-fundamentals content for readers who need a baseline understanding of study types before engaging with synthesis methodology specifically.
Guides
Optimal Information Size (OIS) and GRADE Imprecision
Optimal information size (OIS) is the sample size a single adequately powered trial would need to detect the effect of interest — a power calculation applied at the review level. GRADE compares a meta-analysis’s accumulated sample size against OIS to decide whether to downgrade certainty for imprecision, independent of confidence-interval width.
SUCRA Treatment Rankings: What the Number Does (and Doesn’t) Tell You
SUCRA compresses a treatment’s entire ranking distribution into one 0-100% number. Here’s what it represents, how it differs from probability-of-being-best, and the mean-vs-distribution limitation methodologists emphasize.
The Downs and Black Checklist: Assessing Quality Across RCTs and Observational Studies
The Downs and Black checklist is one of the few validated tools that can appraise both randomised and non-randomised health intervention studies on a single 27-item scale, making it useful for systematic reviews that mix study designs.
Baujat Plots: Finding Which Studies Drive Heterogeneity in a Meta-Analysis
A Baujat plot cross-references each included study’s contribution to overall heterogeneity against its influence on the pooled meta-analysis estimate, flagging studies in the upper-right quadrant as the ones most worth a sensitivity check.
QUADAS-2: Assessing Risk of Bias in Diagnostic Accuracy Studies
QUADAS-2 assesses risk of bias and applicability across four domains — patient selection, index test, reference standard, and flow and timing — using signalling questions to reach each judgment.
P-Curve Analysis: Testing for Evidential Value vs. P-Hacking
P-curve analysis examines the distribution of significant p-values in a literature to test whether findings reflect genuine effects or selective reporting and p-hacking.
Evidence Synthesis Methods: Choosing a Review Type From the Question, Not the Label
Review labels are unreliable: 14 types in 2009, 48 in 2019, several used interchangeably. Specify search, appraisal, synthesis and analysis first, then pick the family and its reporting guideline. Covers scoping vs systematic, rapid review trade-offs, umbrella overlap, and real costs.
Mixed Methods Appraisal Tool (MMAT): The 2018 Checklist Explained
The MMAT lets reviewers appraise qualitative, quantitative, and mixed methods studies with one checklist instead of switching tools per design, and deliberately produces no single overall score.
What Is a Cochrane Review?
A Cochrane Review is a specific, branded output — Cochrane Review Group oversight, Handbook-compliant methods, and CDSR publication — not just any review calling itself “systematic.”
Title and Abstract Screening: Workflow, Pilot Calibration and Disagreement Rules
How review teams screen titles and abstracts against pre-specified criteria: piloting and calibrating the criteria first, when dual independent screening is required versus a faster liberal-accelerated pass, and the Cochrane decision rules for resolving reviewer disagreement.
Outcome Switching and Selective Outcome Reporting: Detection and Prevention
How to detect outcome switching by comparing a trial’s registered outcomes against its published results, what CONSORT 2010 and ICMJE require, and how trial teams and reviewers can catch and prevent it.
STROBE-MR: Reporting Mendelian Randomisation Studies
STROBE-MR is the 20-item reporting-guideline extension to STROBE for Mendelian randomisation studies — covering genetic instrument selection, the three core instrumental-variable assumptions, and pleiotropy-robust sensitivity analyses.
RECORD and RECORD-PE: Reporting Studies Using Routinely Collected Health Data
How to report observational studies built on claims, registry, or EHR data using the RECORD checklist, and when the RECORD-PE pharmacoepidemiology extension applies on top of it.
PRESS 2015: Peer Reviewing a Systematic Review Search Strategy
The PRESS 2015 checklist’s six elements for peer-reviewing a systematic review search strategy before it runs, how it differs from PRISMA-S, and who should do the review.
IPD Meta-Analysis: One-Stage vs Two-Stage, and PRISMA-IPD
Individual participant data (IPD) meta-analysis pools raw participant-level records instead of published summaries. How one-stage and two-stage synthesis differ, what data-sharing and harmonisation actually involve, and what PRISMA-IPD requires beyond standard PRISMA 2020 reporting.
ROBIS: Assessing Risk of Bias in a Systematic Review
ROBIS assesses whether a systematic review itself was conducted rigorously: its three phases, four domains, and how it differs from AMSTAR 2.
GRADE Summary of Findings Table: Building and Interpreting One
How to build a GRADE Summary of Findings table with GRADEpro GDT, read its seven columns, and avoid confusing it with an Evidence-to-Decision framework.
TREND Statement: Reporting Non-Randomised Behavioural Interventions
The TREND statement’s 22-item checklist for reporting nonrandomised evaluations of behavioural and public-health interventions, and how it differs from CONSORT.
Mixed-Methods Systematic Review: Convergent Integrated vs Convergent Segregated
JBI’s current mixed-methods systematic review methodology narrows the old three-design typology to one: convergent, split into convergent segregated (synthesize quantitative and qualitative evidence separately, then integrate the findings) and convergent integrated (qualitize or quantitize one data type so both can be pooled into a single synthesis).
Dose-Response Meta-Analysis: Modelling a Trend Across Exposure Levels
How dose-response meta-analysis pools a study’s several exposure-category risk estimates into one trend or curve, using the Greenland-Longnecker method, restricted cubic splines, and one- vs two-stage models.
Core Outcome Sets and the COMET Database
What a core outcome set (COS) is, how to search the COMET Database before designing a trial or review’s outcomes, and what COS-STAD and COS-STAR require of a trustworthy COS.
GRADE Evidence-to-Decision (EtD) Frameworks: From Certainty to Recommendation
The GRADE Evidence-to-Decision framework is the structured step between a certainty-of-evidence rating and an actual strong-versus-conditional recommendation, walking through problem priority, effects, values, balance, resources, equity, acceptability and feasibility.
Systematic Review Data Extraction Form: Fields, Piloting, and Dual Extraction
Cochrane’s four-step process for designing a systematic review data extraction form, what fields it should capture, the MECIR piloting requirement, and why dual extraction is mandatory for outcome data but only ‘highly desirable’ for study characteristics.
CINAHL Search Strategy: Subject Headings and Nursing-Specific Filters
How to build a CINAHL search: CINAHL Headings vs. MeSH, the Explode and Major Concept toggles, MH/MM field codes, the Exclude MEDLINE Records checkbox, and the EBP/publication-type limiters that matter for a nursing or allied-health systematic review.
Qualitative Evidence Synthesis: Choosing a Method and Reporting It with ENTREQ
How to choose among thematic synthesis, framework synthesis and JBI meta-aggregation for a qualitative evidence synthesis, and how to report it with the 21-item ENTREQ statement.
Subgroup and Sensitivity Analysis in Meta-Analysis: Pre-Specification and Interpretation
How subgroup analysis (does the effect differ across pre-defined groups) and sensitivity analysis (is the pooled result robust to analytic choices) differ, why pre-specification determines credibility, and how to report both honestly.
CONSORT-PRO and SPIRIT-PRO: Reporting Patient-Reported Outcomes in Trials
SPIRIT-PRO governs how a trial protocol specifies patient-reported outcomes; CONSORT-PRO governs how the completed trial reports them. Both extensions exist because base CONSORT/SPIRIT left PRO-specific gaps — most consequentially, how missing PRO data gets handled.
Trial Sequential Analysis: Monitoring Boundaries for Cumulative Meta-Analysis
How Trial Sequential Analysis adapts group sequential monitoring boundaries and required information size to cumulative meta-analysis, and what the Cochrane Handbook actually says about using it.
SAGER Guidelines: Reporting Sex and Gender in Research
The SAGER guidelines (Sex and Gender Equity in Research) are EASE’s 2016 editorial reporting standard for distinguishing and disaggregating sex and gender in research manuscripts, section by section.
SPIDER Framework for Qualitative Evidence Searching
The SPIDER framework (Sample, Phenomenon of Interest, Design, Evaluation, Research type) adapts PICO for qualitative evidence synthesis. A worked conversion example and the Methley et al. (2014) sensitivity/specificity evidence versus PICO.
JBI Critical Appraisal Checklists: Choosing the Right Instrument
JBI publishes 13 separate critical appraisal checklists, one per study design. This guide maps each checklist to its study design and shows how to choose the right one.
MINIMAR: The Reporting Checklist for Clinical AI and Machine Learning Models
MINIMAR (MINimum Information for Medical AI Reporting) is a 2020 JAMIA checklist covering the minimum information a clinical AI/ML model’s reporting needs: study population, patient demographics, model architecture and model evaluation.
Meta-Ethnography: Noblit and Hare’s Seven Phases and eMERGe Reporting
Noblit and Hare’s seven-phase meta-ethnography method for synthesizing qualitative research, the reciprocal translation and refutational synthesis concepts that drive it, and the eMERGe reporting guidance.
AGREE II: Appraising a Clinical Practice Guideline
AGREE II is the validated instrument for appraising a clinical practice guideline’s methodological quality — 23 items across six domains, distinct from reporting guidelines like PRISMA and CONSORT.
Inverse-Variance Weighting in Meta-Analysis: DerSimonian-Laird, REML and Hartung-Knapp
How random-effects meta-analysis weights are actually computed: the DerSimonian-Laird vs. REML choice for estimating between-study variance, and why the Hartung-Knapp adjustment corrects the confidence interval.
Cochrane Handbook Chapter Guide: Which Chapter Answers Your Review Question
A chapter-by-chapter navigational map of the Cochrane Handbook for Systematic Reviews of Interventions: which of its 26+ chapters covers searching, risk of bias, meta-analysis, GRADE, and more.
PsycINFO Search Strategy: Thesaurus Terms, Fields, and Limits
How to build a PsycINFO search strategy: the APA Thesaurus of Psychological Index Terms, structured fields like Methodology and Age Groups, platform differences across EBSCOhost/Ovid/ProQuest/APA PsycNet, and which limits to use with care.
Integrative Review: Whittemore and Knafl’s Five Stages
How to conduct an integrative review using Whittemore and Knafl’s five-stage methodology: problem identification, literature search, data evaluation, data analysis, and presentation, for synthesizing methodologically diverse sources.
Meta-Regression in Meta-Analysis: Covariates, the Ecological Fallacy, and the 10-Studies Rule
Meta-regression tests whether a study-level covariate explains heterogeneity in a meta-analysis — how it works, why it risks the ecological fallacy, and the 10-studies-per-covariate rule of thumb.
ROBINS-E: Risk of Bias in Non-Randomised Studies of Exposures
ROBINS-E’s seven bias domains for a non-randomised exposure study: the target-experiment framing, how it diverges from ROBINS-I (exposure vs intervention), and when to use it instead.
Standardized Mean Difference (SMD) and Hedges’ g in Meta-Analysis
Cochrane’s SMD is defined as Hedges’ g, not raw Cohen’s d. The formula, the small-sample correction (J), a worked example, and when to use SMD instead of the mean difference (MD) when pooling continuous outcomes in a meta-analysis.
Levels of Evidence: The OCEBM 2011 Table, GRADE, and Where the Pyramid Breaks
A levels-of-evidence hierarchy ranks study design; GRADE rates certainty in one effect estimate for one outcome. This guide sets out what the OCEBM 2011 table actually says, what changed from the 2009 version, the published thresholds at which GRADE upgrades observational evidence or downgrades randomised trials, and what the methodological critique of the evidence pyramid actually argues.
Nominal Group Technique: A Four-Stage Session Script
A runnable nominal group technique session script: the four stages with worked timings, the facilitation rules that carry the method, how the voting and aggregation actually work, what the published evidence says genuinely varies, and a direct comparison with Delphi on cost, anonymity and speed.
AMSTAR 2: The 16-Item Tool for Appraising Systematic Reviews
AMSTAR 2 is the standard tool for appraising the methodological quality of a systematic review. Here is what its 16 items assess and how the confidence rating is calculated.
Realist Review: What It Is and the RAMESES Reporting Standards
A realist review asks why and for whom an intervention works, not just whether it does. Here is the method, context-mechanism-outcome configurations, and the RAMESES reporting standards.
MOOSE Checklist: Reporting Meta-Analyses of Observational Studies
MOOSE is the reporting guideline for meta-analyses of observational studies in epidemiology. Here is what its 35 items require and when to use it instead of PRISMA.
SQUIRE 2.0: The Reporting Guideline for Quality Improvement Studies
SQUIRE 2.0 is the reporting guideline for healthcare quality improvement studies. Here is what its 18 items require, how it differs from SQUIRE 1.0, and when to use it instead of CONSORT.
PICO Framework: A Question-Formulation Guide for Systematic Review Authors
What the PICO framework is for, the four elements and the comparator/control distinction authors most often botch, the full variant family (PICOT, PICOS, PECO, PEO, SPIDER, SPICE, ECLIPSE, CIMO, PCC) and when to use each, how PICO becomes a database search strategy and a PROSPERO/PRISMA-ready protocol, and a fill-in template.
Systematic Review vs Scoping Review: PRISMA Guide
How to tell a systematic review from a scoping review before you commit to one: differing objectives, PRISMA 2020 versus PRISMA-ScR reporting, Cochrane versus JBI frameworks, PROSPERO registration, and whether risk-of-bias appraisal is required for your question.
CHEERS 2022 Checklist: Reporting Economic Evaluations
What the CHEERS 2022 checklist requires when reporting a health economic evaluation, how it differs from CHEERS 2013, and how it relates to CONSORT, STROBE, and PRISMA.
TIDieR Checklist: Describing an Intervention So It Can Be Replicated
The TIDieR checklist is a 12-item reporting guideline for describing interventions in enough detail that another researcher can replicate them. Full checklist, worked guidance, and how it relates to CONSORT, SPIRIT, and PRISMA.
Narrative Synthesis and SWiM: When Meta-Analysis Isn’t Possible
How to synthesize systematic review findings without meta-analysis: the SWiM reporting guideline, common narrative synthesis methods, and when to choose this approach over pooling.
Risk of Bias Assessment: RoB 2, ROBINS-I and the Traffic-Light Plot
A practical guide to the RoB 2 tool for randomized trials and ROBINS-I for non-randomized studies: the domains, judgement categories, and how to build a traffic-light plot with robvis.
Choosing a Critical Appraisal Tool for Your Study Design
A decision table matching study design to the right critical appraisal tool (CASP, AMSTAR 2, Newcastle-Ottawa Scale, RoB 2, ROBINS-I, QUADAS-2), plus a step-by-step appraisal procedure and a direct AMSTAR-vs-CASP-vs-Newcastle-Ottawa comparison.
The Newcastle-Ottawa Scale (NOS): How to Score Study Quality
How the Newcastle-Ottawa Scale (NOS) star system works for cohort and case-control studies: the three domains, a worked scoring example, good/fair/poor interpretation, and how NOS compares to ROBINS-I and GRADE.
CASP Checklists: Which One and How to Apply It
A decision table for picking the right CASP checklist by study design (RCT, cohort, qualitative, systematic review and more), how the shared validity/results/applicability structure works, and how CASP compares to JBI, Newcastle-Ottawa, and GRADE.
ARRIVE Guidelines for Animal Research Reporting: The Full 2.0 Checklist
The complete ARRIVE 2.0 checklist for reporting animal research: all 21 items across the Essential 10 and Recommended Set, journal requirements, and how it compares to CONSORT, STROBE, and PRISMA.
The Delphi Method: Building Expert Consensus in Rounds
The Delphi method builds expert consensus through anonymous, iterated rounds of rating and feedback. This guide covers round structure, panel selection, how consensus is measured, stopping rules, and modified/real-time variants, with a worked example.
Searching Grey Literature for a Systematic Review
How to search grey literature for a systematic review: which sources to check (theses, trial registries, conference abstracts, preprints, government/NGO reports), how to translate a search strategy where Boolean syntax isn’t supported, and how to document the search under PRISMA-S so it satisfies PRISMA 2020 reporting.
Fixed-Effect vs. Random-Effects Meta-Analysis: How They Differ and When to Use Each
A side-by-side comparison of the fixed-effect and random-effects meta-analysis models — their assumptions, weighting, and confidence-interval behavior — with one dataset pooled both ways to show exactly how and why the results diverge.
How to Conduct a Scoping Review: A Step-by-Step JBI/PRISMA-ScR Guide
A step-by-step walkthrough of the JBI nine-step scoping review framework, from PCC question and eligibility criteria through charting the data and PRISMA-ScR reporting, plus when a scoping review is the wrong choice.
How to Write a PICOT Question
How to turn a vague research topic into a searchable PICOT question, with an element table, five fully worked conversions (including two nursing-specific scenarios), and their Boolean search translations.
How to Detect and Assess Publication Bias
How to detect and assess publication bias in a systematic review or meta-analysis: reading a funnel plot, running Egger’s test, applying trim-and-fill, and correctly interpreting the result.
How to Register a Systematic Review on PROSPERO
A step-by-step walkthrough of registering a systematic review on PROSPERO: when to register, what the form requires, submitting the record, and amending it later.
How to Read a Forest Plot: A Step-by-Step Guide
A step-by-step guide to reading a forest plot: the study rows, confidence-interval whiskers, weights, the diamond, and the line of no effect, walked through with a worked example.
Heterogeneity in Meta-Analysis: I², τ², and Prediction Intervals
How to define, quantify, and investigate heterogeneity in meta-analysis — Cochran’s Q, I², tau-squared, prediction intervals, fixed vs. random-effects models, and PRISMA 2020 reporting requirements.
Boolean Search Operators for Literature Searching
How AND, OR, NOT, nesting, truncation, proximity operators, and field tags/controlled vocabulary combine to build a reproducible database search strategy for systematic and scoping reviews, with a worked PICO example and PRISMA-S reporting guidance.
The EQUATOR Network: Which Reporting Guideline Does Your Study Need?
A study-design routing table for the EQUATOR Network’s reporting guideline library — CONSORT, PRISMA, STROBE, SPIRIT, STARD, TRIPOD, ARRIVE, SQUIRE, CARE, CHEERS, COREQ/SRQR — with links to CASRAI’s own guideline coverage.
How to Peer Review a Systematic Review or Meta-Analysis
Peer reviewing a systematic review or meta-analysis means checking PRISMA 2020 reporting compliance, search-strategy reproducibility, and risk-of-bias appraisal rigor — not the same checklist used for an original-research manuscript.
Systematic Literature Review Tools: A Comparison Guide
A comparison of the software used across the systematic review pipeline — Covidence, DistillerSR, EPPI-Reviewer, Rayyan, ASReview, RevMan, and Abstrackr — by workflow stage, cost, and best fit.
PRISMA: Full Form, Checklist and Flow Diagram
PRISMA — Preferred Reporting Items for Systematic Reviews and Meta-Analyses — is a reporting checklist, not a methodology. This guide covers the full 27-item PRISMA 2020 checklist, the flow diagram, how to use PRISMA in practice, and its main extensions.
