Evidence Synthesis for Decision Making 2: A Generalized Linear Modeling Framework for Pairwise and Network Meta-analysis of Randomized Controlled Trials
Open accessThe DOI matches its Crossref record; 96 of the 96 references it cites were found in a registry or an index; none of the 13 sources its claims cite is checked for support yet.
Abstract
We set out a generalized linear model framework for the synthesis of data from randomized controlled trials. A common model is described, taking the form of a linear regression for both fixed and random effects synthesis, which can be implemented with normal, binomial, Poisson, and multinomial data. The familiar logistic model for meta-analysis with binomial data is a generalized linear model with a logit link function, which is appropriate for probability outcomes. The same linear regression framework can be applied to continuous outcomes, rate models, competing risks, or ordered category outcomes by using other link functions, such as identity, log, complementary log-log, and probit link functions. The common core model for the linear predictor can be applied to pairwise meta-analysis, indirect comparisons, synthesis of multiarm trials, and mixed treatment comparisons, also known as network meta-analysis, without distinction. We take a Bayesian approach to estimation and provide WinBUGS program code for a Bayesian analysis using Markov chain Monte Carlo simulation. An advantage of this approach is that it is straightforward to extend to shared parameter models where different randomized controlled trials report outcomes in different formats but from a common underlying model. Use of the generalized linear model framework allows us to present a unified account of how models can be compared using the deviance information criterion and how goodness of fit can be assessed using the residual deviance. The approach is illustrated through a range of worked examples for commonly encountered evidence formats.
Overview
CoChat Summary
## TL;DR
This paper presents a unified Bayesian generalized linear model framework for evidence synthesis of aggregate randomized trial data, covering pairwise meta-analysis, multiarm trials, indirect comparisons, and network meta-analysis. It shows how different outcome types can be handled with appropriate likelihoods and link functions, and how model fit can be assessed with residual deviance and DIC.
## Key findings
- The framework covers normal, binomial, Poisson, and multinomial likelihoods with identity, logit, log, complementary log-log, and probit links, all within a common core linear predictor.
- In the worked example of 22 trials of beta-blockers after myocardial infarction, the fixed-effect model gave a posterior median pooled log odds ratio of −0.26 with 95% CrI (−0.36, −0.16), and posterior median mortality probabilities of 0.10 in control and 0.08 in treatment.
- For that example, the random-effects model fit better by residual deviance, with D̄res = 41.9 versus 46.8 for the fixed-effect model, but the paper says the DIC suggested little to choose between them.
- The paper states that the same model structure can synthesize pairwise meta-analysis, indirect comparisons, multiarm trials, and network meta-analysis “without distinction.”
## Methods
- The paper develops a Bayesian MCMC approach to evidence synthesis, implemented in WinBUGS 1.4.3, with annotated code provided in a web appendix.
- It presents a generalized linear model (GLM) formulation for aggregate trial data, using a common linear predictor with fixed-effect and random-effects versions.
- An illustrative meta-analysis of 22 beta-blocker trials is used for the binomial/logit example, and model fit/comparison is done using residual deviance, leverage, and DIC.
## Limitations
- The paper states that model choice can depend strongly on sparse data and prior distributions, and that deciding between fixed and random effects may be highly sensitive to these choices.
- It notes that in network meta-analysis, consistency across evidence sources must be checked, because inconsistency can arise from trial-level effect modifiers.
- Evident from the text: many assumptions are required for specific models, such as constant hazards for Poisson/log-link and cloglog models, and proportional competing risks for the multinomial competing-risks model; the paper presents these as model assumptions rather than validated facts.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 0 Not valid · 11 Not checked · 19 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The posterior median of the pooled log odds ratio of beta-blockers compared to control in the FE model is −0.26 with 95% credible interval (CrI; −0.36, −0.16)”
§ Evidence Synthesis for Decision Making 2 > Development of The Core Models: Binomial Data with Logit Link > Worked Example: Binomial Likelihood, Logit Link (Appendix: Example 1)
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 result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In comparison of the fit of the FE and RE models (Table 3), the posterior mean of the residual deviance indicates that although the RE models is a better fit to the data, with D¯res= 41.9 against 46.8 for the FE model”
§ Evidence Synthesis for Decision Making 2 > Development of The Core Models: Binomial Data with Logit Link > Worked Example: Binomial Likelihood, Logit Link (Appendix: Example 1)
hedgedstates an association- 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.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The DIC suggests that there is little to choose between the two models—differences of less than 3 or 5 are not considered important”
§ Evidence Synthesis for Decision Making 2 > Development of The Core Models: Binomial Data with Logit Link > Worked Example: Binomial Likelihood, Logit Link (Appendix: Example 1)
hedged- 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.
In the paper
“Further, the p D, and therefore the DIC, calculated in the way that we suggest, is not precisely the same as that calculated in WinBUGS, except in the case of a normal likelihood.”
§ Evidence Synthesis for Decision Making 2 > Development of The Core Models: Binomial Data with Logit Link > Worked Example: Binomial Likelihood, Logit Link (Appendix: Example 1)
What the sources say
- Not checked yetcites [34]
plain- 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.
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
“An advantage of this approach is that it is straightforward to extend to shared parameter models where different randomized controlled trials report outcomes in different formats but from a common underlying model.”
§ Evidence Synthesis for Decision Making 2 > Abstract We set out a generalized linear model framework for the synthesis of data from randomized controlled trials. A common model is described, taking the form of a linear regression for both fixed and random effects synthesis, which can be implemented with normal, binomial, Poisson, and multinomial data. The familiar logistic model for meta-analysis with binomial data is a generalized linear model with a logit link function, which is appropriate for probability outcomes. The same linear regression framework can be applied to continuous outcomes, rate models, competing risks, or ordered category outcomes by using other link functions, such as identity, log, complementary log-log, and probit link functions. The common core model for the linear predictor can be applied to pairwise meta-analysis, indirect comparisons, synthesis of multiarm trials, and mixed treatment comparisons, also known as network meta-analysis, without distinction. We take a Bayesian approach to estimation and provide WinBUGS program code for a Bayesian analysis using Markov chain Monte Carlo simulation. An advantage of this approach is that it is straightforward to extend to shared parameter models where different randomized controlled trials report outcomes in different formats but from a common underlying model. Use of the generalized linear model framework allows us to present a unified account of how models can be compared using the deviance information criterion and how goodness of fit can be assessed using the residual deviance. The approach is illustrated through a range of worked examples for commonly encountered evidence formats. Meta-analysis of randomized controlled trials is widely used in the medical research literature,1–3 and methodology for pairwise meta-analysis is well developed.4–9 Recently, meta-analysis methods have been extended to indirect and mixed treatment comparisons, also known as network meta-analysis (NMA), which combine data from randomized comparisons, A v. B, A v. C, B v. D, etc., to deliver an internally consistent set of estimates while respecting the randomization in the evidence.10 NMA is particularly useful in decision-making contexts.11–21 We present a single unified approach to evidence synthesis of aggregate data from randomized controlled trials, specifically, but not exclusively, for use in probabilistic decision making.22 To cover the variety of outcomes reported in trials and the range of data transformations required to achieve linearity, we adopt the framework of generalized linear modeling (GLM).23 This provides for normal, binomial, Poisson, and multinomial likelihoods, with identity, logit, log, complementary log-log, and probit link functions, and common core models for the linear predictor in both fixed effects (FE) and random effects (RE) settings. Our common core models can synthesize data from pairwise meta-analysis, multiarm trials, indirect comparisons, and NMA without distinction. Indeed, pairwise meta-analysis and indirect comparisons are special cases of NMA. We take a Bayesian Markov Chain Monte Carlo (MCMC) approach using the freely available software WinBUGS 1.4.3.24 We include an extensive web appendix with fully annotated WinBUGS code for all models to run a series of worked examples. This code is also available at www.nicedsu.org.uk.25
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.
13 more claims
Create a free account to see all 30 claims and to verify the 13 references still open.
- It may state two separate claims as one. 8
- Probably not a substantial claim. 6
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
“This approach follows the hierarchical Bayesian framework for network meta-analysis described by Lu and Ades (2004) and Dias et al. (2013), which allows continuous outcomes, including standardized mean differences, to be modeled at the arm level using a normal likelihood [37, 38].”
From The Effect of Different Psychological Interventions on Athletes' Competitive Anxiety: A Systematic Review and Bayesian Network Meta-Analysis. (2026) · cites this paper as [38]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“suggesting the network better estimates treatment effects than pairwise analyses alone with no evidence of inconsistency.9”
From Psychological and pharmacological interventions for social anxiety disorder in adults: a systematic review and network meta-analysis (2014) · cites this paper as [9]
Experimental — accuracy not yet measuredIn the citing paper
“The posterior mean of the residual deviance was 165·3 in the standard network meta-analysis model compared with 176·3 in the independent-effect model that compares favourably with the number of treatment groups (n=148), suggesting the network better estimates treatment effects than pairwise analyses alone with no evidence of inconsistency.9”
From Psychological and pharmacological interventions for social anxiety disorder in adults: a systematic review and network meta-analysis (2014) · cites this paper as [9]
Citations
Cites
Statistical power analysis for the behavioral sciences
(1990) · Computers, Environment and Urban Systems · doi:10.1016/0198-9715(90)90050-4 · Cited by 65,269
Measuring inconsistency in meta-analyses
Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003) · BMJ · doi:10.1136/bmj.327.7414.557 · Cited by 63,139
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
Cochrane Handbook for Systematic Reviews of Interventions: Cochrane Book Series
(2008) · doi:10.1002/9780470712184 · Cited by 25,048 · cited 2×
“However, dividing estimates through by the sample standard deviation introduces additional heterogeneity and distortion(49) and produces results that some find less interpretable.(8) A procedure that would produce more interpretable results would be to divide all estimates from a given test instrument by the standard…”
BackgroundMethodGRADE: an emerging consensus on rating quality of evidence and strength of recommendations
Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R., Falck-Ytter, Y., Alonso-Coello, P., & Schünemann, H. J. (2008) · BMJ · doi:10.1136/bmj.39489.470347.ad · Cited by 23,305
McCullagh, P., & Nelder, J. A. (1989) · doi:10.1007/978-1-4899-3242-6 · Cited by 16,836
Statistical Methods for Meta-Analysis
Wasserman, S., Hedges, L. V., & Olkin, I. (1988) · Journal of Educational Statistics · doi:10.2307/1164953 · Cited by 10,326
Schulz (1995) · doi:10.1001/jama.1995.03520290060030 · Cited by 5,634
WinBUGS - A Bayesian modelling framework: Concepts, structure, and extensibility
Lunn, D. J., Thomas, A., Best, N., & Spiegelhalter, D. (2000) · Statistics and Computing · doi:10.1023/a:1008929526011 · Cited by 5,368
Salanti, G., Ades, A. E., & Ioannidis, J. P. A. (2011) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2010.03.016 · Cited by 3,766
Self-citation
Research works printed but not linked: 7
- Zelen M, 1971, Biometrika, 58, 129
- Welton NJ, 2005, Appl Stat, 54, 385
- 10.1002/(sici)1097-0258(19961230)15:24 · doi:10.1002/(sici)1097-0258(19961230)15:24
- National Institute for Health and Clinical Excellence, 2008, Guide to the Methods of Technology Appraisal
- Spiegelhalter D. Two brief topics on modelling with WinBUGS. IceBugs: a workshop about the development and use of the BUGS programme. Available from: http://www.math.helsinki.fi/openbugs/IceBUGS/Presentations/SpiegelhalterIceBUGS.pdf
- Rothman KJ, 2008, Modern Epidemiology
- Ades AE, 2005, Health Econ, 14, 373
Cited by
Hutton, B., Salanti, G., Caldwell, D. M., Chaimani, A., Schmid, C. H., Cameron, C., Ioannidis, J. P. A., Straus, S., Thorlund, K., Jansen, J. P., et al. (2015) · Annals of Internal Medicine · doi:10.7326/m14-2385 · Cited by 8,269
Whelton, P. K., Carey, R. M., Aronow, W. S., Casey, D. E., Collins, K. J., Dennison Himmelfarb, C., DePalma, S. M., Gidding, S., Jamerson, K. A., Jones, D. W., et al. (2018) · Hypertension · doi:10.1161/hyp.0000000000000065 · Cited by 7,391
CINeMA: an approach for assessing confidence in the results of a network meta-analysis
Nikolakopoulou, A., Higgins, J. P. T., Papakonstantinou, T., Chaimani, A., Del Giovane, C., Egger, M., & Salanti, G. (2020) · PLOS Medicine · doi:10.1371/journal.pmed.1003082 · Cited by 1,935
Ranking treatments in frequentist network meta-analysis works without resampling methods
Rücker, G., & Schwarzer, G. (2015) · BMC Medical Research Methodology · doi:10.1186/s12874-015-0060-8 · Cited by 1,732
Khera, R., Murad, M. H., Chandar, A. K., Dulai, P. S., Wang, Z., Prokop, L. J., Loomba, R., Camilleri, M., & Singh, S. (2016) · JAMA · doi:10.1001/jama.2016.7602 · Cited by 865
Metadata
- Authors
- Dias, S., Sutton, A. J., Ades, A. E., & Welton, N. J.
- Affiliations
- University of Leicester (GB)
- Corresponding author
- Dias, Sofia
- Venue
- Medical Decision Making
- Publisher
- SAGE Publications
- Published
- 2012-10-26
- Volume
- 33
- Issue
- 5
- Pages
- 607-617
- ISSN
- 0272-989X (ISSN-L)
- 1552-681X
- PMC
- PMC3704203
- OpenAlex
- W2150529573
- Semantic Scholar
- c72c5a00952b53efdb446b2f8bc705bb45ec679e
- Cited by
- 1,207
- References
- 96
- Citations by year
- 2026: 89 · 2025: 69 · 2024: 90 · 2023: 80 · 2022: 94 · 2021: 144 · 2020: 123 · 2019: 106 · 2018: 104 · 2017: 82 · 2016: 77 · 2015: 71 · 2014: 52 · 2013: 25 · 2012: 1
- FWCI
- 55.08
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by-nc
- journals.sagepub.com/page/policies/text-and-data-mining-license (tdm)
- Locations
- Medical Decision Making (publishedVersion, cc-by-nc)
- Europe PMC (PubMed Central) (publishedVersion, cc-by-nc)
- PubMed Central (publishedVersion, cc-by-nc)
- PubMed (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- Explore Bristol Research (submittedVersion, cc-by-nc)
- Topics
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Keywords
- generalized linear models, meta-analysis, randomized controlled trials, Bayesian meta-analysis, WinBUGS, link functions, network meta-analysis, indirect comparison, shared parameter models, deviance information criterion, Markov chain Monte Carlo, evidence synthesis
- MeSH terms
- Network Meta-Analysis as Topic, Decision Making, Models, Theoretical, Meta-Analysis as Topic, Likelihood Functions, Randomized Controlled Trials as Topic, Evidence-Based Medicine
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.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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: Evidence Synthesis for Decision Making 2: A Generalized Linear Modeling Framework for Pairwise and Network Meta-analysis of Randomized Controlled Trials. https://app.cochat.ai/papers/10.1177/0272989x12458724
- The paper
- Dias, S., Sutton, A. J., Ades, A. E., & Welton, N. J. (2013). Evidence Synthesis for Decision Making 2: A Generalized Linear Modeling Framework for Pairwise and Network Meta-analysis of Randomized Controlled Trials. Medical Decision Making. https://doi.org/10.1177/0272989x12458724
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.