scholarly journals Individual participant data meta-analysis of continuous outcomes: A comparison of approaches for specifying and estimating one-stage models

2018 ◽  
Vol 37 (29) ◽  
pp. 4404-4420 ◽  
Author(s):  
Amardeep Legha ◽  
Richard D. Riley ◽  
Joie Ensor ◽  
Kym I.E. Snell ◽  
Tim P. Morris ◽  
...  
Keyword(s):  
Meta Analysis ◽  
Individual Participant Data ◽  
Stage Models ◽  
One Stage ◽  
Continuous Outcomes ◽  
Individual Participant

Estimating interactions in individual participant data meta-analysis. A comparison of methods in practice.

2021 ◽  
Author(s):  
Ruth Walker ◽  
Lesley Stewart ◽  
Mark Simmonds
Keyword(s):  
Random Effects ◽  
Meta Analysis ◽  
Individual Preference ◽  
Detailed Examination ◽  
International Studies ◽  
Individual Participant Data ◽  
Stage Models ◽  
One Stage ◽  
Interaction Terms ◽  
Individual Participant

Abstract Medical interventions may be more effective in some types of individuals than others and identifying characteristics that modify the effectiveness of an intervention is a cornerstone of precision or stratified medicine. The opportunity for detailed examination of treatment-covariate interactions can be an important driver for undertaking an individual participant data (IPD) meta-analysis, rather than a meta-analysis using aggregate data. A number of recent modelling approaches are available. We apply these methods to the Perinatal Antiplatelet Review of International Studies (PARIS) Collaboration IPD dataset and compare estimates between them. We discuss the practical implications of applying these methods, which may be of interest to aid meta-analysists in the use of these, often complex models. Models compared included the two-stage meta-analysis of interaction terms and one-stage models which fit multiple random effects and separate within and between trial information. Models were fitted for nine covariates and five binary outcomes and results compared. Interaction terms produced by the methods were generally consistent. We show that where data are sparse and there is low heterogeneity in the covariate distributions across trials, the meta-analysis of interactions may produce unstable estimates and have issues with convergence. In this IPD dataset, varying assumptions by using multiple random effects in one-stage models or using only within trial information made little difference to the estimates of treatment-covariate interaction. Method choice will depend on datasets characteristics and individual preference.

scholarly journals Deriving percentage study weights in multi-parameter meta-analysis models: with application to meta-regression, network meta-analysis and one-stage individual participant data models

2017 ◽  
Vol 27 (10) ◽  
pp. 2885-2905 ◽  
Author(s):  
Richard D Riley ◽  
Joie Ensor ◽  
Dan Jackson ◽  
Danielle L Burke
Keyword(s):  
Meta Analysis ◽  
Information Matrix ◽  
Indirect Evidence ◽  
Parameter Estimates ◽  
Individual Participant Data ◽  
One Stage ◽  
Individual Participant ◽  
Meta Regression ◽  
Multiple Treatments ◽  
Analysis Models

Many meta-analysis models contain multiple parameters, for example due to multiple outcomes, multiple treatments or multiple regression coefficients. In particular, meta-regression models may contain multiple study-level covariates, and one-stage individual participant data meta-analysis models may contain multiple patient-level covariates and interactions. Here, we propose how to derive percentage study weights for such situations, in order to reveal the (otherwise hidden) contribution of each study toward the parameter estimates of interest. We assume that studies are independent, and utilise a decomposition of Fisher’s information matrix to decompose the total variance matrix of parameter estimates into study-specific contributions, from which percentage weights are derived. This approach generalises how percentage weights are calculated in a traditional, single parameter meta-analysis model. Application is made to one- and two-stage individual participant data meta-analyses, meta-regression and network (multivariate) meta-analysis of multiple treatments. These reveal percentage study weights toward clinically important estimates, such as summary treatment effects and treatment-covariate interactions, and are especially useful when some studies are potential outliers or at high risk of bias. We also derive percentage study weights toward methodologically interesting measures, such as the magnitude of ecological bias (difference between within-study and across-study associations) and the amount of inconsistency (difference between direct and indirect evidence in a network meta-analysis).

scholarly journals One‐stage individual participant data meta‐analysis models for continuous and binary outcomes: Comparison of treatment coding options and estimation methods

2020 ◽  
Vol 39 (19) ◽  
pp. 2536-2555
Author(s):  
Richard D. Riley ◽  
Amardeep Legha ◽  
Dan Jackson ◽  
Tim P. Morris ◽  
Joie Ensor ◽  
...  
Keyword(s):  
Meta Analysis ◽  
Estimation Methods ◽  
Binary Outcomes ◽  
Individual Participant Data ◽  
One Stage ◽  
Individual Participant ◽  
Analysis Models

scholarly journals One-stage individual participant data meta-analysis models: estimation of treatment-covariate interactions must avoid ecological bias by separating out within-trial and across-trial information

2016 ◽  
Vol 36 (5) ◽  
pp. 772-789 ◽  
Author(s):  
Hairui Hua ◽  
Danielle L. Burke ◽  
Michael J. Crowther ◽  
Joie Ensor ◽  
Catrin Tudur Smith ◽  
...  
Keyword(s):  
Meta Analysis ◽  
Individual Participant Data ◽  
Ecological Bias ◽  
One Stage ◽  
Individual Participant ◽  
Analysis Models

scholarly journals Multilevel Meta-Analysis of Individual Participant Data of Single-Case Experimental Designs: One-Stage versus Two-Stage Methods

2020 ◽  
pp. 1-20
Author(s):  
Lies Declercq ◽  
Laleh Jamshidi ◽  
Belén Fernández Castilla ◽  
Mariola Moeyaert ◽  
S. Natasha Beretvas ◽  
...  
Keyword(s):  
Single Case ◽  
Meta Analysis ◽  
Experimental Designs ◽  
Individual Participant Data ◽  
Stage Versus ◽  
Two Stage ◽  
One Stage ◽  
Individual Participant

scholarly journals Individual Participant Data Meta-Analysis for a Binary Outcome: One-Stage or Two-Stage?

PLoS ONE ◽  
2013 ◽  
Vol 8 (4) ◽  
pp. e60650 ◽  
Author(s):  
Thomas P. A. Debray ◽  
Karel G. M. Moons ◽  
Ghada Mohammed Abdallah Abo-Zaid ◽  
Hendrik Koffijberg ◽  
Richard David Riley
Keyword(s):  
Meta Analysis ◽  
Individual Participant Data ◽  
Binary Outcome ◽  
Two Stage ◽  
One Stage ◽  
Individual Participant

scholarly journals Meta-analysis using individual participant data: one-stage and two-stage approaches, and why they may differ

2016 ◽  
Vol 36 (5) ◽  
pp. 855-875 ◽  
Author(s):  
Danielle L. Burke ◽  
Joie Ensor ◽  
Richard D. Riley
Keyword(s):  
Meta Analysis ◽  
Individual Participant Data ◽  
Two Stage ◽  
One Stage ◽  
Individual Participant

scholarly journals The use of one-stage meta-analytic method based on individual participant data for binary adverse events under the rule of three: a simulation study

PeerJ ◽  
2019 ◽  
Vol 7 ◽  
pp. e6295 ◽  
Author(s):  
Liang-Liang Cheng ◽  
Ke Ju ◽  
Rui-Lie Cai ◽  
Chang Xu
Keyword(s):  
Adverse Events ◽  
Coverage Probability ◽  
Meta Analysis ◽  
Random Effect ◽  
Individual Participant Data ◽  
Average Width ◽  
Rule Of Three ◽  
One Stage ◽  
Individual Participant ◽  
The One

Objective In evidence synthesis practice, dealing with binary rare adverse events (AEs) is a challenging problem. The pooled estimates for rare AEs through traditional inverse variance (IV), Mantel-Haenszel (MH), and Yusuf-Peto (Peto) methods are suboptimal, as the biases tend to be large. We proposed the “one-stage” approach based on multilevel variance component logistic regression (MVCL) to handle this problem. Methods We used simulations to generate trials of individual participant data (IPD) with a series of predefined parameters. We compared the performance of the MVCL “one-stage” approach and the five classical methods (fixed/random effect IV, fixed/random effect MH, and Peto) for rare binary AEs under different scenarios, which included different sample size setting rules, effect sizes, between-study heterogeneity, and numbers of studies in each meta-analysis. The percentage bias, mean square error (MSE), coverage probability, and average width of the 95% confidence intervals were used as performance indicators. Results We set 52 scenarios and each scenario was simulated 1,000 times. Under the rule of three (a sample size setting rule to ensure a 95% chance of detecting at least one AE case), the MVCL “one-stage” IPD method had the lowest percentage bias in most of the situations and the bias remained at a very low level (<10%), when compared to IV, MH, and Peto methods. In addition, the MVCL “one-stage” IPD method generally had the lowest MSE and the narrowest average width of 95% confidence intervals. However, it did not show better coverage probability over the other five methods. Conclusions The MVCL “one-stage” IPD meta-analysis is a useful method to handle binary rare events and superior compared to traditional methods under the rule of three. Further meta-analyses may take account of the “one-stage” IPD method for pooling rare event data.

A general framework for the use of logistic regression models in meta-analysis

2016 ◽  
Vol 25 (6) ◽  
pp. 2858-2877 ◽  
Author(s):  
Mark C Simmonds ◽  
Julian PT Higgins
Keyword(s):  
Logistic Regression ◽  
Random Effects ◽  
Regression Models ◽  
Meta Analysis ◽  
Test Accuracy ◽  
Individual Participant Data ◽  
Logistic Regression Models ◽  
One Stage ◽  
Individual Participant ◽  
Meta Analyses

Where individual participant data are available for every randomised trial in a meta-analysis of dichotomous event outcomes, “one-stage” random-effects logistic regression models have been proposed as a way to analyse these data. Such models can also be used even when individual participant data are not available and we have only summary contingency table data. One benefit of this one-stage regression model over conventional meta-analysis methods is that it maximises the correct binomial likelihood for the data and so does not require the common assumption that effect estimates are normally distributed. A second benefit of using this model is that it may be applied, with only minor modification, in a range of meta-analytic scenarios, including meta-regression, network meta-analyses and meta-analyses of diagnostic test accuracy. This single model can potentially replace the variety of often complex methods used in these areas. This paper considers, with a range of meta-analysis examples, how random-effects logistic regression models may be used in a number of different types of meta-analyses. This one-stage approach is compared with widely used meta-analysis methods including Bayesian network meta-analysis and the bivariate and hierarchical summary receiver operating characteristic (ROC) models for meta-analyses of diagnostic test accuracy.

Sign in / Sign up

Export Citation Format

Share Document