Does this skill support binary outcomes like Odds Ratios?

Yes, it provides implementation patterns for Odds Ratios (OR), Risk Ratios (RR), and Risk Differences (RD) using the escalc and metabin functions.

Is Bayesian meta-analysis supported?

Yes, it includes templates for fitting Bayesian random-effects meta-analysis models using the brms package with custom priors.

Can I perform subgroup analysis or meta-regression?

Yes, the skill covers subgroup analysis for categorical moderators and meta-regression for continuous or mixed moderators using the rma function.

How does the skill handle publication bias?

It includes methods for creating contour-enhanced funnel plots, performing Egger's regression and Begg's rank correlation tests, and applying the trim-and-fill method.

Which R packages are utilized for these meta-analysis methods?

The skill primarily utilizes the 'metafor' and 'meta' packages for frequentist analysis, and 'brms' for Bayesian meta-analysis.

Meta-Analysis Methods in R

Name: Meta-Analysis Methods in R
Author: choxos

bychoxos

0•

Data Science & ML

Implements comprehensive meta-analysis workflows in R, including effect size calculation, heterogeneity assessment, and publication bias detection.

This skill provides a robust framework for conducting pairwise meta-analyses using leading R packages like metafor and meta. It covers the complete evidence synthesis pipeline, from calculating effect sizes for continuous, binary, and survival data to fitting complex fixed and random effects models. Designed for researchers and data scientists, it streamlines advanced statistical tasks such as meta-regression, sensitivity analysis, and Bayesian modeling, while offering high-quality visualization patterns for forest and funnel plots.

Key Features

01Fixed and random effects modeling with multiple tau estimators

02Advanced heterogeneity diagnostics including I-squared and prediction intervals

03Comprehensive effect size calculation (SMD, OR, RR, HR, ZCOR)

04Publication bias detection using Egger's regression and trim-and-fill methods

050 GitHub stars

06Customizable forest, funnel, and Baujat plots for evidence visualization

Use Cases

01Performing meta-regressions to identify study-level moderators of heterogeneity

02Synthesizing clinical trial data to determine overall treatment efficacy

03Evaluating potential publication bias and small-study effects in research literature

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add choxos/tidyrmodelling meta-analysis

For use in Claude.ai and ChatGPT

Download Skill