Meta-analytic Case Series — Pooling Case Series for Evidence Synthesis
Meta-analytic Pooling of Case Series Studies · Also known as: pooled case series, systematic review of case series, case series meta-analysis, aggregated case series
A meta-analytic case series is an evidence-synthesis design that systematically identifies, appraises, and statistically pools outcome data from multiple single-arm case series on a defined clinical condition or intervention. It occupies a middle tier of evidence — above individual case reports and unsystematic series, but below pooled randomized trials — and is particularly valuable when experimental designs are ethically or practically unavailable.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use a meta-analytic case series when: (1) the clinical question concerns a rare disease or intervention for which RCTs do not exist or are infeasible; (2) multiple small case series on the same topic have been published and a pooled estimate would improve precision; (3) the objective is to characterise complication rates, natural history, or procedural outcomes rather than to compare treatments. Do NOT use this approach as a substitute for a meta-analysis of controlled studies when such studies are available; uncontrolled pooling cannot establish causal efficacy. Also avoid when the available series are so heterogeneous in patient selection and outcome definitions that statistical pooling would be misleading — narrative synthesis may then be more appropriate.
Strengths & limitations
- Maximises statistical power from small, scattered case series that individually lack sufficient sample size.
- Appropriate and often the only feasible synthesis design for rare diseases and novel procedures.
- Systematic search and quality appraisal add transparency and reproducibility beyond an informal narrative review.
- Random-effects pooling honestly quantifies heterogeneity rather than assuming all series estimate the same parameter.
- Provides a baseline estimate of outcome rates that can be used to plan future controlled studies.
- Absence of a control group means treatment effects cannot be causally attributed; pooled estimates reflect outcomes, not comparative efficacy.
- Case series are vulnerable to selection bias, publication bias, and reporting bias, all of which propagate into the pooled estimate.
- High between-series heterogeneity (I² often exceeds 75% in practice) can make the pooled estimate difficult to interpret.
- Quality of the synthesis is bounded by the quality of constituent series; poor reporting in primary studies limits what can be extracted.
Frequently asked
How is a meta-analytic case series different from a standard meta-analysis?
A standard meta-analysis pools effect estimates (risk ratios, odds ratios, mean differences) from comparative studies — usually RCTs — that each have a treated and a control group. A meta-analytic case series pools single-arm outcome proportions or rates from uncontrolled series, so the result is an absolute estimate of outcome frequency rather than a comparative effect size. The statistical machinery is similar, but the interpretation differs fundamentally because no causal comparison is possible.
Which quality appraisal tool should I use?
The IHE Quality Appraisal Checklist for Case Series (Institute of Health Economics, 2012) and the criteria proposed by Murad et al. (BMJ Evidence-Based Medicine, 2018) are the most widely adopted tools. Both evaluate domains such as case selection, consecutive recruitment, completeness of follow-up, and validity of outcome measurement. Choose based on discipline norms, report items for every included series, and use quality ratings in sensitivity analyses.
What transformation should I use when pooling proportions?
The Freeman-Tukey double arcsine transformation is the most common choice because it stabilises the variance of proportions near 0 or 1 — a frequent occurrence in rare-event case series. The logit transformation is an alternative when proportions are not extreme. The raw proportion can be pooled directly only when rates are moderate and sample sizes are large. The choice should be pre-specified and a sensitivity analysis across transformations is advisable.
Can I pool individual patient data (IPD) from case series?
Yes, and IPD pooling is preferred when feasible because it allows more granular subgroup analyses, better handling of missing data, and harmonisation of outcome definitions across studies. In practice, IPD are rarely shared for case series; aggregate data meta-analysis is the default. When requesting IPD, contact corresponding authors directly and document the data-sharing outcome in the PRISMA flow diagram.
What GRADE level does a meta-analytic case series produce?
Evidence from case series starts at a low or very low GRADE certainty rating because of the high risk of bias inherent in uncontrolled observational designs. Pooling multiple series does not by itself upgrade certainty; it improves precision but does not eliminate confounding or selection bias. The certainty can be downgraded further for inconsistency (high I²), indirectness, or imprecision, and upgrading factors (large effect, dose-response) rarely apply.
Sources
- Lovato, L. C., Hill, K., Hertert, S., Hunninghake, D. B., & Probstfield, J. L. (2002). Recruitment for controlled clinical trials: literature summary and annotated bibliography. Controlled Clinical Trials, 18(4), 328–352. [For meta-analytic approaches to non-randomised series see:] Murad, M. H., Sultan, S., Haffar, S., & Bazerbachi, F. (2018). Methodological quality and synthesis of case series and case reports. BMJ Evidence-Based Medicine, 23(2), 60–63. link ↗
- Murad, M. H., Sultan, S., Haffar, S., & Bazerbachi, F. (2018). Methodological quality and synthesis of case series and case reports. BMJ Evidence-Based Medicine, 23(2), 60–63. DOI: 10.1136/bmjebm-2017-110853 ↗
How to cite this page
ScholarGate. (2026, June 3). Meta-analytic Pooling of Case Series Studies. ScholarGate. https://scholargate.app/en/epidemiology/meta-analytic-case-series
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Case seriesEpidemiology↔ compare