Contents
On this page
- Quick answer
- Overview
- Case Study 1: Explaining High Heterogeneity
- Case Study 2: Meeting MECIR Standards Before Cochrane Submission
- Case Study 3: Bringing a Stalled Review Back on Schedule
- Case Study 4: Preparing an Evidence Synthesis Aligned with IQWiG Methods
- Lessons Across the Case Studies
- Conclusion
- Frequently asked questions
- Sources
InsightsCase Study
Case study
Meta-Analysis Case Studies: Four Routes from Research Question to Robust Evidence Synthesis
Four illustrative scenarios: high heterogeneity, a MECIR methods audit, a stalled review and a synthesis aligned with IQWiG General Methods 8.0.
- Meta-analysis
- Case study
- IQWiG
- Scenarios
- 4, illustrative
- Heterogeneity
- REML, HKSJ, prediction interval
- Searches
- Rerun within 12 months
- IQWiG, 3 trials
- Bayesian random-effects
At a glanceCase Study
Explore meta-analysis supportDiscuss your project
Quick answer
- High heterogeneity is answered with a better random-effects analysis, a prediction interval and cautious, pre-specified subgroup tests.
- A Cochrane review should trace every analysis decision to Chapter 10 or a MECIR standard.
- For very few trials, IQWiG General Methods 8.0 sets a Bayesian random-effects approach.
Overview
Each meta-analysis case study below describes a common challenge, the approach a review team would take and the likely result. The four scenarios cover heterogeneity, Cochrane conduct standards, the end-to-end review workflow and evidence synthesis for German HTA.
Disclosure: these are illustrative, anonymised scenarios based on typical research challenges. They do not describe identifiable clients or real results, and they are not a promise of acceptance by any journal or HTA body.
Case Study 1: Explaining High Heterogeneity
Scenario 1 profile
| Profile | Detail |
|---|---|
| Scenario profile | Academic cardiology research group |
| Topic | Heterogeneity in meta-analysis |
| Service focus | Statistical analysis and interpretation |
Challenge. A pooled analysis of 14 trials shows an I² of 78%, and peer reviewers question whether pooling is justified.
Approach.
- Re-run the random-effects model with a REML estimator and Hartung-Knapp-Sidik-Jonkman confidence intervals1.
- Add τ² and a 95% prediction interval to the results12.
- Run the two pre-specified subgroup analysis comparisons with a formal test for subgroup differences (MECIR C67)5.
- Run a sensitivity analysis excluding trials at high risk of bias.
Illustrative outcome. A pre-specified dose subgroup may account for part of the heterogeneity. Because subgroup comparisons are observational, the authors would interpret this cautiously (MECIR C68) and report the average effect alongside the prediction interval, which answers the reviewers’ question directly15.
Case Study 2: Meeting MECIR Standards Before Cochrane Submission
Scenario 2 profile
| Profile | Detail |
|---|---|
| Scenario profile | Review team preparing a Cochrane intervention review |
| Topic | Cochrane Handbook Chapter 10 and MECIR |
| Service focus | Methods audit |
Challenge. An internal check finds that the effect measure for one outcome is not justified and that two subgroup analyses were not pre-specified.
Approach.
- Map every analysis decision to Chapter 10 and the current MECIR standards15.
- Justify the effect measure for each outcome type and confirm that pooling is meaningful (C62).
- Label the two subgroup analyses as post hoc and interpret them cautiously (C67, C68).
- Document deviations from the protocol.
Illustrative outcome. With each decision traceable to a Handbook section or MECIR standard, the team can answer methodological queries from Cochrane editors quickly and consistently.
Planning a meta-analysis?
Pubrica’s evidence synthesis team supports protocol and PROSPERO registration, searching, screening, extraction, risk of bias, statistical synthesis, GRADE and PRISMA 2020 reporting. Review authors keep ownership of and accountability for the review, and no provider can guarantee acceptance.
Case Study 3: Bringing a Stalled Review Back on Schedule
Scenario 3 profile
| Profile | Detail |
|---|---|
| Scenario profile | Hospital research department |
| Topic | Meta-analysis workflow |
| Service focus | Project management and data synthesis |
Challenge. Fourteen months after PROSPERO registration, screening is incomplete and the search is out of date.
Approach.
- Rerun the search strategy with an information specialist, since searches must be within 12 months of publication (MECIR C37, mandatory)5.
- Organise duplicate screening and extraction with clear deadlines.
- Complete data synthesis, GRADE assessment and summary of findings tables.
- Prepare the PRISMA 2020 flow diagram and checklist3.
Illustrative outcome. With a named owner and deadline for each remaining stage of the meta-analysis workflow, the team can plan a realistic submission date with an up-to-date search and complete PRISMA reporting.
Case Study 4: Preparing an Evidence Synthesis Aligned with IQWiG Methods
Scenario 4 profile
| Profile | Detail |
|---|---|
| Scenario profile | Pharmaceutical company in Germany |
| Topic | IQWiG General Methods |
| Service focus | Meta-analysis for an AMNOG dossier |
Challenge. Only three randomised trials compare the new drug with the appropriate comparator, and a frequentist random-effects analysis gives a very wide confidence interval.
Approach.
- Restrict the analysis to the patient-relevant outcomes defined for the assessment4.
- Assess risk of bias at outcome level.
- Follow IQWiG General Methods 8.0 (section 10.3.7): a Bayesian random-effects meta-analysis with an informative prior for heterogeneity, checked against a qualitative summary of conclusive effects4.
- Run interaction tests for pre-specified subgroups.
Illustrative outcome. A synthesis aligned with IQWiG methods leaves less scope for methodological objections during the assessment, although the benefit rating remains IQWiG’s and the G-BA’s decision. See our IQWiG General Methods guide for the detail.
Lessons Across the Case Studies
The same principles recur in every scenario1345:
Table 1. Principles that recur across the four scenarios
| Principle | In practice |
|---|---|
| Pre-specify | Register the protocol and analysis plan before screening. |
| Describe heterogeneity fully | Report Q, I², τ² and a prediction interval. |
| Explain, do not just measure | Use pre-specified subgroups and sensitivity analyses, and interpret subgroups cautiously. |
| Match the audience | Follow MECIR for Cochrane and the General Methods for IQWiG. |
| Report transparently | Use PRISMA 2020 and document every deviation. |
Conclusion
Each meta-analysis case study shows that robust evidence synthesis depends on pre-specification, full reporting of heterogeneity and methods matched to the audience, from journals and Cochrane to IQWiG. For the standards behind these decisions, see our meta-analysis Q&A; for help with a similar problem, see meta-analysis services.
Pubrica team verdict
In each scenario the fix is the same: decide methods in advance, report heterogeneity fully and match the methods to the audience.
Frequently asked questions
Sources
Sources were checked on 13 November 2026. Requirements and guidance may change after that date; confirm the current version with the issuing body.
- 1Higgins, J. P. T., Thomas, J., Chandler, J., Cumpston, M., Li, T., Page, M. J., & Welch, V. A. (Eds.). (2024). Cochrane handbook for systematic reviews of interventions (Version 6.5). Cochrane. https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current
- 2Higgins, J. P. T., & Thompson, S. G. (2002). Quantifying heterogeneity in a meta-analysis. Statistics in Medicine, 21(11), 1539–1558. https://doi.org/10.1002/sim.1186
- 3Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., Chou, R., Glanville, J., Grimshaw, J. M., Hróbjartsson, A., Lalu, M. M., Li, T., Loder, E. W., Mayo-Wilson, E., McDonald, S., . . . Moher, D. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, Article n71. https://doi.org/10.1136/bmj.n71
- 4Institute for Quality and Efficiency in Health Care. (2025). General methods (Version 8.0). IQWiG. https://doi.org/10.60584/General-Methods_V8.0
- 5Higgins, J. P. T., Lasserson, T., Thomas, J., Flemyng, E., & Churchill, R. (2023). Methodological expectations of Cochrane intervention reviews (MECIR) (Version August 2023). Cochrane. https://www.cochrane.org/authors/handbooks-and-manuals/mecir-manual
meta-analysis case studyheterogeneitysubgroup analysisprediction intervalMECIR standardsPROSPERO registration
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation, so check its guide if it has one.
Dr. Charlie. (2026, November 13). Meta-Analysis Case Studies: Four Routes from Research Question to Robust Evidence Synthesis. Pubrica. https://pubrica.com/insights/case-study/meta-analysis-case-studies/
This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.
Next steps
Planning a meta-analysis?
Pubrica’s evidence synthesis team supports protocol and PROSPERO registration, searching, screening, extraction, risk of bias, statistical synthesis, GRADE and PRISMA 2020 reporting. Review authors keep ownership of and accountability for the review, and no provider can guarantee acceptance.
This article is general information for researchers, not legal, regulatory or funding advice. Check the current requirements of your journal, funder, ethics committee or regulator before you act on it.
More in Case Study