meta-analysis

This paper addresses the limited efficacy of conventional meta-regression methods under dual heterogeneity—both geographic and temporal. We systematically evaluate traditional approaches (random-, fixed-, and mixed-effects models) against seven novel specifications, including our proposed “location-fixed-effects + time-trend” hybrid model. Through large-scale simulation studies, we quantitatively compare methods across three key dimensions: statistical power, estimation bias, and model robustness. Results demonstrate that jointly modeling dual heterogeneity substantially outperforms unidimensional modeling: under high heterogeneity, it improves statistical power by 32% and reduces bias by 41%. Our hybrid specification achieves an optimal balance of flexibility and robustness. Furthermore, we provide a practical model-selection framework grounded in information criteria and robustness diagnostics. This work advances methodology for causal inference in settings characterized by multi-source spatiotemporal heterogeneity.

Bayesian meta-analysis remains inaccessible to researchers without programming expertise, hindering its adoption in empirical social sciences. Method: We developed and integrated the first user-oriented Bayesian meta-analysis module into the open-source statistical platform JASP. The module implements core Bayesian techniques—including estimation, hypothesis testing, model averaging, meta-regression, multilevel modeling, and publication bias adjustment—via an intuitive graphical user interface, modular workflow, and built-in visualizations (e.g., forest plots, bubble plots, marginal means plots). Contribution/Results: Officially released and empirically validated, the module substantially improves efficiency and reproducibility of rigorous cumulative evidence synthesis for non-technical users. It lowers methodological barriers, promotes standardized application of Bayesian methods in social science research, and advances open, transparent meta-analytic practice.

Traditional meta-analysis methods require programming proficiency, limiting interdisciplinary researchers’ adoption of advanced techniques. This project develops an open-source, graphical meta-analysis module for JASP—a free and open statistical software platform—integrating both standardized (e.g., fixed- and random-effects models, heterogeneity tests, publication bias assessments) and advanced meta-analytic functionalities for the first time in such software. The module features an interactive, code-free interface enabling rigorous, transparent, and fully reproducible meta-analyses from data import to reporting. Its core contribution lies in substantially lowering technical barriers, thereby promoting widespread, transparent, and reproducible meta-analytic practice aligned with open science principles. Specifically designed for psychology, education, and related fields, it delivers a user-friendly, reliable, and standards-compliant tool. Empirical validation confirms its functional completeness and result robustness, with successful application across multiple disciplinary empirical studies.

Personalized treatment for depression faces challenges in modeling treatment effect heterogeneity and integrating evidence across randomized controlled trials (RCTs). This study proposes a two-stage meta-analytic framework: in Stage I, multiple causal models—including causal forests, Bayesian additive regression trees (BART), and parametric regression—are fitted within each RCT to estimate conditional average treatment effects (CATEs); in Stage II, a hierarchical meta-analysis integrates CATE estimates across trials, jointly modeling within- and between-study heterogeneity to produce statistically valid CATE prediction intervals. The framework enhances external validity and uncertainty quantification, accommodates multi-model inputs, and supports clinical decision-making. Empirical analysis of duloxetine versus vortioxetine reveals no significant treatment-effect heterogeneity, with age emerging as a potential effect modifier; the resulting prediction intervals more comprehensively capture true uncertainty than conventional confidence intervals.

This study addresses the heterogeneity of treatment effects in clinical research, where conventional subgroup analyses lack individual-level predictive power and purely machine learning–based approaches often lack statistical guarantees. To bridge this gap, the authors propose a two-stage hybrid workflow: first, using formal statistical hypothesis testing to confirm the presence of heterogeneous treatment effects, then constructing an individualized treatment strategy evaluated via cross-fitted doubly robust estimation under a Neyman–Pearson risk constraint. This framework integrates the interpretability of statistical inference with the predictive strength of machine learning, yielding a transparent, auditable, and statistically principled approach to heterogeneity. The method demonstrates efficacy in both simulation studies and the ACTG 175 HIV trial, and is accompanied by a practical implementation checklist along with guidance for alignment with regulatory-oriented heterogeneous treatment effect (HTE) assessment protocols.

This study addresses limitations of conventional fixed- and random-effects models in meta-analysis, particularly their inadequate handling of between-study heterogeneity, hierarchical covariates, and small-sample settings. The authors propose a Bayesian hierarchical meta-analytic framework that employs analytical integration to achieve efficient and robust inference. This approach simultaneously models study-level heterogeneity and covariate effects while effectively synthesizing data across multiple studies. The method demonstrates markedly improved estimation accuracy and computational efficiency in small-sample scenarios. Extensive simulations confirm its robustness, and a real-world application comparing the safety profiles of oxcarbazepine and carbamazepine reveals significantly lower adverse event risk with the former. All code and data are publicly available.

This work addresses the widespread misuse of statistical methods—often stemming from implicit or ambiguous assumptions—which exacerbates the reproducibility crisis in scientific research, particularly in hypothesis testing and meta-analysis where formal verification mechanisms are lacking. To bridge this gap, the authors propose the first formal verification framework tailored for Python-based statistical programs. By developing a Why3-py frontend, they translate dynamically typed, runtime-polymorphic Python code into the WhyML intermediate representation and extend the StatWhy tool to support meta-analysis verification. Integrating program transformation, static analysis, and formal verification techniques, this approach enables, for the first time, automated correctness verification of statistical programs written in Python, effectively uncovering overlooked assumptions and misuses, thereby filling a critical void in the formal verification of statistical software.

This study addresses the challenge in component network meta-analysis (CNMA) that existing methods struggle to reliably rank multi-component treatments and their constituent components due to the complex set of relatively effects that can be uniquely estimated. For the first time, the authors extend treatment hierarchy construction methods to the CNMA setting, proposing an integrated frequentist–Bayesian workflow that systematically resolves misleading rankings caused by non-identifiable relative effects through rigorous assessment of estimability. The approach was successfully applied to two real-world networks—first-line treatment for depression and chronic lymphocytic leukemia—yielding accurate hierarchical rankings of both treatments and components, thereby providing a robust basis for decision-making in complex interventions.

This study addresses the underappreciated challenge of estimation and communication following multiplicity adjustment within the frequentist framework in complex clinical trials, where multiple endpoints, interim data looks, or group comparisons often introduce estimation bias and complicate interpretation, thereby undermining transparency in benefit–risk assessment. By integrating advanced methodologies such as adaptive designs and graphical approaches to multiple testing, the work illustrates through concrete examples the limitations of current strategies in conveying trial results meaningfully. The research underscores the need to critically reevaluate prevailing practices and foster interdisciplinary dialogue to enhance both the accuracy of effect estimation and the clarity of result communication, ultimately informing future methodological standards and regulatory guidance.

Traditional meta-analyses struggle to quantify the strength of evidence for the presence or absence of an effect and cannot adequately assess the sensitivity of conclusions to publication bias or small-study effects. This work proposes a Bayesian evidence auditing framework tailored to meta-analytic corpora, integrating bias-aware models with unbiased baseline specifications through Bayesian model averaging and Bayes factors. It introduces “rigor” as a composite metric that jointly evaluates the strength of evidence for an effect and robustness to bias—allowing null effects to achieve high rigor scores when supported by strong evidence. Built upon Bayesian random-effects models, the approach employs simulation and resampling strategies within the ADEMP framework, including synthetic data generation, registered-report resampling, and contour-enhanced funnel weighting. Applied to nutritional intervention studies, the method frequently attenuates conventional effect estimates, revealing that many nominally significant findings lack robust evidential support. Full reproducible resources are publicly released.