Meta-analysis and the science of research synthesis
Open accessThe DOI matches its Crossref record; 101 of the 101 references it cites were found in a registry or an index; 9 of the 69 sources its claims cite were checked for support.
Abstract
Meta-analysis is the quantitative, scientific synthesis of research results. Since the term and modern approaches to research synthesis were first introduced in the 1970s, meta-analysis has had a revolutionary effect in many scientific fields, helping to establish evidence-based practice and to resolve seemingly contradictory research outcomes. At the same time, its implementation has engendered criticism and controversy, in some cases general and others specific to particular disciplines. Here we take the opportunity provided by the recent fortieth anniversary of meta-analysis to reflect on the accomplishments, limitations, recent advances and directions for future developments in the field of research synthesis.
Overview
CoChat Summary
TL;DR
This review reflects on meta-analysis as quantitative research synthesis since the 1970s, its contributions to evidence-based practice, and its methodological development across medicine, social sciences, and ecology/evolution/conservation. It argues that meta-analysis has resolved contradictory evidence and enabled broad generalizations, but faces a “midlife crisis” from poor-quality applications, publication/research bias, and misuse of the term.
Key findings
- Meta-analysis was first introduced in the 1970s; the term was coined by Gene Glass in 1976, and the paper marks its 40th anniversary. It has helped establish evidence-based practice, resolve seemingly contradictory results, and spread across many scientific fields.
- In medicine, a meta-analysis of 20 clinical studies found maternal obesity associated with increased neural tube defect risk: a 1.7-fold rise for moderately obese women and over a 3-fold risk for severely obese women. In social science, a meta-analysis of 35 studies found no significant differences between multisystemic therapy and conventional social services for serious juvenile offenders.
- In ecology, evolution and conservation, meta-analysis was adopted in the early 1990s. Parmesan and Yohe used data on over 1,700 species to show significant range shifts and advancement of spring events; Benayas and colleagues found ecological restoration increased biodiversity and ecosystem services, though values remained lower in restored versus intact reference ecosystems.
- Methodological advances include fixed- and random-effects models, meta-regression, heterogeneity quantification (τ and I²), publication-bias tools, network meta-analysis, phylogenetic meta-analysis, and software such as MetaWin and OpenMEE. PRISMA includes a 27-item checklist for systematic review reporting.
Methods
- Design: narrative review/commentary on meta-analysis and research synthesis, not a new meta-analysis. The authors state they highlight principles, limitations, utility, and future developments.
- Data or sample: draws on published literature and illustrative examples across medicine, social sciences, and ecology/evolution/conservation; includes Box 1 on meta-analytic models and Table 1 on EEC milestones.
- Main analysis: conceptual and methodological synthesis; no primary data collection, systematic search protocol, or quantitative pooling is reported in the paper.
Limitations
- The paper states that meta-analysis and systematic reviews can highlight evidence deficiencies but cannot overcome them; publication bias and research bias may be suspected and sometimes estimated but cannot truly be corrected; selective or incomplete reporting can constrain synthesis.
- The paper states that too many published meta-analyses are of low quality, the term “meta-analysis” is frequently misapplied, and vote-counting persists despite being discredited; it also notes concerns about academic inequality and the “research parasites” controversy.
- The text makes evident that, as a narrative review, it does not provide a systematic search strategy, formal inclusion criteria, or a quantitative synthesis; its examples are illustrative rather than exhaustive.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
4 Valid · 4 Not valid · 37 Not checked · 33 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Similarly, a synthesis may be constrained by either selective or incomplete reporting 66,67 .”
What the sources say
- Not checked yetcites [66] Stewart, L. A . et al. Preferred reporting items for a systematic review and meta-analysis of individual participant data: the PRISMA-IPD statement. J. Am. Med. Assoc. 313, 1657–1665 (2015) source
- Partly supportedcites [67] Saldanha, I. J . et al. Evaluating Data Abstraction Assistant, a novel software application for data abstraction during systematic reviews: protocol for a randomized controlled trial. Syst. Rev. 5, 196 (2016) sourceExperimental — accuracy not yet measured
The Background (page unavailable) reports that omissions and errors during data abstraction changed summary meta-analytic results, supporting the general point that a synthesis can be affected by missing information. It describes information present in articles but omitted by abstractors, not selective or incomplete reporting by the studies themselves; that specific mechanism is unsupported. (LLM, conf=0.78)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“When the number of studies is an order of magnitude greater than that, the inability of narrative processes to adequately summarize the outcomes increases exponentially 2,3 .”
What the sources say
- Not checked yetcites [2] Roberts, P. D ., Stewart, G. B. & Pullin, A. S. Are review articles a reliable source of evidence to support conservation and environmental management? A comparison with medicine. Biol. Conserv. 132, 409–423 (2006) source
- Partly supportedcites [3] Bastian, H ., Glasziou, P . & Chalmers, I. Seventy-five trials and eleven systematic reviews a day: how will we ever keep up? PLoS Med. 7, e1000326 (2010) sourceExperimental — accuracy not yet measured
Our copy supports the qualitative half of the claim: the 'How Close Are We to Archie Cochrane's Goal?' section cites Goldschmidt and Mulrow for 'how great the potential is for error in reviews of health literature that were not conducted systematically' and states that 'the staple of medical literature synthesis remains the non-systematic narrative review', while the same section (Figure 4 discussion) and 'Where to Now?' document the growth from ~14 trials a day in 1979 to 75 trials and 11 systematic reviews a day, with 'the task keeps increasing in size and complexity' ('Where Are We Now?'). What the source does NOT support is the quantitative assertion attributed to it: nowhere does our copy say the inability of narrative processes to summarise outcomes 'increases exponentially', nor tie it to study counts an order of magnitude beyond a dozen — the only dozen-scale figure in the paper is the median number of trials per systematic review (six to 16, Cochrane reviews averaging over 12), which is offered as a measure of reviewing workload and coverage, not of narrative-review failure. The exponential scaling is therefore the citing authors' characterisation, not this paper's. (LLM, conf=0.72)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“However, as the number of publications and research results has mushroomed across scientific fields, the difficulties encountered in carrying out narrative reviews to identify and summarize evidence in a transparent manner has become increasingly apparent 1 .”
What the sources say
- Not checked yetcites [1] Jennions, M. D ., Lortie, C. J. & Koricheva, J. in The Handbook of Meta-analysis in Ecology and Evolution (eds Koricheva, J . et al.) Ch. 23, 364–380 (Princeton Univ. Press, 2013) source
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, in many cases of the resolution of scientific problems, when the results of an individual study are combined with those of other studies, the resulting synthesis provides a more general and complete picture of the questions being addressed.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A statement the paper makes itself — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Sometimes, the striking results of initial studies are not confirmed by those of subsequent studies or by syntheses of a body of research.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
15 more claims
Create a free account to see all 78 claims and to verify the 60 references still open.
- It may state two separate claims as one. 32
- Probably not a substantial claim. 48
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The trade-off with our method is that it does not allow for determining the magnitude of the effect of the practices33–36 but only the proportion of impacts33,37 as positive, negative or neutral.”
From Farming practices to enhance biodiversity across biomes: a systematic review (2024) · cites this paper as [36]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Besides, publication biases that have been documented in several scientific fields90,91 can also affect ML-based science.92,93”
From Leakage and the reproducibility crisis in machine-learning-based science (2023) · cites this paper as [91]
Experimental — accuracy not yet measuredIn the citing paper
“Moreover, since a meta-analysis can be conducted only when sufficient and appropriate quantitative data from original studies are available, the number of published meta-analyses about a certain topic is an indicator of the overall understanding of the subject, allowing us to reveal research gaps (25, 26).”
From Agricultural diversification promotes multiple ecosystem services without compromising yield (2020) · cites this paper as [25]
In the citing paper
“Meta-analysis optimises statistical power for detecting treatment effects by combining data from multiple similar studies [22, 23].”
From Does the growing of Bt maize change abundance or ecological function of non-target animals compared to the growing of non-GM maize? A systematic review (2022) · cites this paper as [23]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Box 1: Meta-analytic models and heterogeneity
“𝜀 is the within-study sampling error (solid black lines), the deviation from the study-specific mean.”
§ Box 1: Meta-analytic models and heterogeneity
“the solid black lines are deviations from the overall mean (panels a,b) or slope (panels c,d)”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Box 1: Meta-analytic models and heterogeneity
“The assumption of the fixed-effect meta-regression is that a moderator(s) can explain all between-study heterogeneity”
§ Box 1: Meta-analytic models and heterogeneity
“the blue distributions the between-study variance (panels a-d).”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Box 1: Meta-analytic models and heterogeneity
“Two basic models of meta-analysis are the fixed-effect model (Fig 1a) and the random-effects model (Fig 1b) 6 .”
§ Box 1: Meta-analytic models and heterogeneity
“Figure 1. A variety of charts and plots common in meta-analysis. a. PRISMA diagram 20 , b. a forest plot showing overall means based on a fixed-effect model and random-effects model”
Citations
Cites
Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & for the PRISMA Group (2009) · BMJ · doi:10.1136/bmj.b2535 · Cited by 83,823
Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & The PRISMA Group (2009) · PLoS Medicine · doi:10.1371/journal.pmed.1000097 · Cited by 64,042 · cited 1×
“Procedural protocols for systematic reviews are well established especially in the medical and social sciences (Preferred Reporting Items for Systematic Reviews, or PRISMA; www.prisma-statement.org) 20 .”
Bias in meta-analysis detected by a simple, graphical test
Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997) · BMJ · doi:10.1136/bmj.315.7109.629 · Cited by 56,722
Preferred reporting items for systematic reviews and meta-analyses: the PRISMA Statement.
(2009) · PubMed Central · Cited by 44,859
Meta-analysis in clinical trials
DerSimonian, R., & Laird, N. (1986) · Controlled Clinical Trials · doi:10.1016/0197-2456(86)90046-2 · Cited by 39,552
Quantifying heterogeneity in a meta‐analysis
Higgins, J. P. T., & Thompson, S. G. (2002) · Statistics in Medicine · doi:10.1002/sim.1186 · Cited by 36,763
PRISMA-P Group, Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P., & Stewart, L. A. (2015) · Systematic Reviews · doi:10.1186/2046-4053-4-1 · Cited by 26,223
Conducting Meta-Analyses in R with the metafor Package
Viechtbauer, W. (2010) · Journal of Statistical Software · doi:10.18637/jss.v036.i03 · Cited by 20,464 · cited 1×
“…the use of machine learning to screen studies 82 for inclusion in systematic reviews and meta-analyses for large reviews, increasingly sophisticated software and models for complex meta-regression 53,83 , and integration of meta-analysis with decision support in medicine and other domains 84 .”
Wang, G., Ma, R., Qiao, G., Wada, K., Aizawa, Y., & Satoh, T. (2015) · Integrative Medicine International · doi:10.1159/000368408 · Cited by 18,657
Duval, S., & Tweedie, R. (2000) · Biometrics · doi:10.1111/j.0006-341x.2000.00455.x · Cited by 14,099
Research works printed but not linked: 3
- Hedges, L. & Olkin, I. Statistical Methods for Meta-analysis (Academic Press, 1985)
- Higgins, J. P. T . & Green, S. (eds) Cochrane Handbook for Systematic Reviews of Interventions: Version 5.1.0 (Wiley, 2011).This large collaborative work provides definitive guidance for the production of systematic reviews in medicine and is of broad interest for methods development outside the medical field.
- Cooper, H. M ., Hedges, L. V . & Valentine, J. C. (eds) The Handbook of Research Synthesis and Meta-analysis 2nd edn (Russell Sage Foundation, 2009).This book is an important compilation that builds on the ground-breaking first edition to set the standard for best practice in meta-analysis, primarily in the social sciences but with applications to medicine and other fields.
Cited by
The PRISMA 2020 statement: An updated guideline for reporting systematic reviews
Page, 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., et al. (2021) · BMJ · doi:10.1136/bmj.n71 · Cited by 108,283
The PRISMA 2020 statement: an updated guideline for reporting systematic reviews
Page, 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., et al. (2021) · Systematic Reviews · doi:10.1186/s13643-021-01626-4 · Cited by 13,874
The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews
Page, 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., et al. (2021) · International Journal of Surgery · doi:10.1016/j.ijsu.2021.105906 · Cited by 11,748
The PRISMA 2020 statement: An updated guideline for reporting systematic reviews
Page, 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., et al. (2021) · PLoS Medicine · doi:10.1371/journal.pmed.1003583 · Cited by 5,148
“Systematic reviews therefore generate various types of knowledge for different users of reviews (such as patients, healthcare providers, researchers, and policy makers).[1,2]To ensure a systematic review is valuable to users, authors should prepare a transparent, complete, and accurate account of why the review was do…”
BackgroundThe PRISMA 2020 statement: an updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J., Bossuyt, P., Boutron, I., Hoffmann, T., mulrow, C., Shamseer, L., Tetzlaff, J., Akl, E., Brennan, S. E., et al. (2020) · doi:10.31222/osf.io/v7gm2 · Cited by 5,177
Metadata
- Authors
- Gurevitch, J., Koricheva, J., Nakagawa, S., & Stewart, G.
- Affiliations
- Stony Brook University (US)
- Royal Holloway University of London (GB)
- Garvan Institute of Medical Research (AU)
- Newcastle University (GB)
- Corresponding author
- Gurevitch, Jessica
- Koricheva, Julia
- Nakagawa, Shinichi
- Stewart, Gavin
- Venue
- Nature
- Published
- 2018-03
- Volume
- 555
- Issue
- 7695
- Pages
- 175-182
- ISSN
- 0028-0836 (ISSN-L)
- 1476-4687
- OpenAlex
- W2791681873
- Semantic Scholar
- ac5ad1932e31c3ea141366e8e073d6d95869c4ef
- Cited by
- 1,709
- References
- 101
- Citations by year
- 2026: 177 · 2025: 230 · 2024: 231 · 2023: 199 · 2022: 234 · 2021: 235 · 2020: 204 · 2019: 130 · 2018: 64
- FWCI
- 31.20
- Type
- journal article
- Language
- en
- Open access
- green
- hdl.handle.net
- Licence
- cc-by-nc-nd
- www.springer.com/tdm (tdm)
- Locations
- Nature (publishedVersion)
- PubMed (publishedVersion)
- UNSWorks (UNSW Sydney) (submittedVersion, cc-by-nc-nd)
- RePEc: Research Papers in Economics (submittedVersion)
- Newcastle University ePrints (Newcastle Univesity)
- Funders
- National Science Foundation (1262402)
- Laura and John Arnold Foundation
- Topics
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Social Sciences › Library and Information Sciences › scientometrics and bibliometrics research
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Psychology Research and Bibliometrics
- Keywords
- meta-analysis, evidence synthesis, history of meta-analysis, evolutionary biology, conservation biology, artificial intelligence, evidence-based practice
- MeSH terms
- Animals, Biological Science Disciplines, Ecology, Biological Evolution, Research, Meta-Analysis as Topic, History, 20th Century, History, 21st Century
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-04, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-04 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Meta-analysis and the science of research synthesis. https://app.cochat.ai/papers/10.1038/nature25753
- The paper
- Gurevitch, J., Koricheva, J., Nakagawa, S., & Stewart, G. (2018). Meta-analysis and the science of research synthesis. Nature. https://doi.org/10.1038/nature25753
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.