Skip to main content
Glama
Sign In
Cochrane Meta-Analysis MCP Server

Cochrane Meta-Analysis MCP Server

by matheus-rech
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Local

Cochrane Meta-Analysis MCP Server

An MCP (Model Context Protocol) server that provides AI-assisted meta-analysis workflows following Cochrane methodological standards.

Features

  • RevMan Import: Parse RevMan 5 (.rm5 XML) and Cochrane CSV exports

  • Data Validation: Comprehensive validation against Cochrane standards

  • Meta-Analysis: R-based statistical analysis using metafor/meta packages

  • Forest Plots: Publication-ready visualizations

  • Publication Bias: Funnel plots, Egger's test, trim-and-fill

  • Reporting: Automated Cochrane-style HTML/PDF reports

Related MCP server: ResearchStat AI

Installation

npm install
npm run build

Prerequisites

  1. Node.js 18+

  2. R 4.0+ with packages:

    • metafor

    • meta

    • ggplot2

    • jsonlite

Install R packages:

install.packages(c("metafor", "meta", "ggplot2", "jsonlite"))

Configuration

Add to Claude Desktop config (~/.config/claude/claude_desktop_config.json):

{
  "mcpServers": {
    "cochrane-meta": {
      "command": "node",
      "args": ["/Users/YOUR_USERNAME/Documents/cochrane-meta-mcp/dist/index.js"]
    }
  }
}

Available Tools

1. import_revman_data

Import and parse RevMan 5 files or Cochrane CSV exports.

{
  "file_path": "/path/to/review.rm5",
  "format": "rm5"
}

2. validate_cochrane_data

Validate study data against Cochrane standards.

{
  "data": { ... },
  "validation_level": "comprehensive"
}

3. perform_meta_analysis

Execute meta-analysis following Cochrane Handbook Chapter 10.

{
  "data": { ... },
  "effect_measure": "OR",
  "model": "random",
  "heterogeneity_test": true
}

4. generate_forest_plot

Create publication-ready forest plot.

{
  "analysis_results": { ... },
  "plot_style": "classic",
  "output_path": "/path/to/forest_plot.png"
}

5. assess_publication_bias

Perform publication bias assessment.

{
  "analysis_results": { ... },
  "methods": ["funnel_plot", "egger_test", "trim_fill"],
  "output_path": "/path/to/funnel_plot.png"
}

6. generate_cochrane_report

Generate comprehensive Cochrane-style report.

{
  "analysis_results": { ... },
  "format": "html",
  "output_path": "/path/to/report.html"
}

Example Usage with Claude

User: I have a Cochrane CSV file with 10 RCTs comparing intervention A vs control.
      Can you help me run a meta-analysis?

Claude: I'll help you analyze this. First, let me import the data using the import_revman_data tool...

[Uses import_revman_data]

Claude: Great! I've imported 10 studies. Let me validate the data...

[Uses validate_cochrane_data]

Claude: Validation passed with 2 warnings about small sample sizes.
        Now I'll perform the meta-analysis using a random-effects model
        with odds ratios as the effect measure...

[Uses perform_meta_analysis]

Claude: The pooled OR is 0.75 (95% CI: 0.62-0.91), p=0.003, with moderate
        heterogeneity (I²=45%). Let me generate a forest plot...

[Uses generate_forest_plot]

Claude: Forest plot created. Now checking for publication bias...

[Uses assess_publication_bias]

Claude: Egger's test p=0.42, no evidence of publication bias.
        Generating final report...

[Uses generate_cochrane_report]

Claude: Complete! Your meta-analysis report is ready at report.html

CSV Format

For CSV imports, use this format:

study_id,authors,year,title,intervention,comparison,outcome,events_treatment,n_treatment,events_control,n_control
Study1,Smith 2020,2020,RCT of intervention,Drug A,Placebo,Mortality,10,100,20,100
Study2,Jones 2021,2021,Another RCT,Drug A,Placebo,Mortality,15,150,30,150

For continuous outcomes:

study_id,authors,year,title,intervention,comparison,outcome,mean_treatment,sd_treatment,n_treatment,mean_control,sd_control,n_control

Development

# Watch mode
npm run dev

# Build
npm run build

# Test (coming soon)
npm test

Architecture

  • TypeScript MCP Server: Handles tool requests from Claude

  • R Bridge: Executes statistical analyses via Rscript

  • Validation Layer: Zod schemas for data validation

  • Tools: Modular tool implementations for each MCP capability

Integration with Existing Tools

This MCP server integrates with your existing meta-analysis infrastructure:

  • Uses your R meta-analysis scripts (~/meta_analysis_workflow.R)

  • Compatible with medical research multi-agent system

  • Can leverage AI citation processors for literature extraction

Cochrane Compliance

Follows:

  • Cochrane Handbook for Systematic Reviews (Chapter 10)

  • PRISMA reporting guidelines

  • Cochrane risk of bias (RoB 2) recommendations

  • GRADE framework for evidence certainty

License

MIT

Version

0.1.0

Author

Matheus Rech

Available Tools

6 tools
assess_publication_biasB

Perform publication bias assessment using funnel plots, Egger's test, and trim-and-fill method.

Inputs: analysis_results (required), methods, output_path.

generate_cochrane_reportB

Generate comprehensive meta-analysis report in Cochrane format (HTML/PDF).

Inputs: output_path (required), analysis_results (required), format, include_code.

generate_forest_plotB

Create publication-ready forest plot following Cochrane standards.

Inputs: output_path (required), analysis_results (required), plot_style, confidence_level.

import_revman_dataA

Import and parse meta-analysis data from CSV, Excel (XLSX), or JSON formats. Supports both structured Cochrane datasets and flat record arrays. Extracts study metadata, outcomes, and analysis parameters following Cochrane standards.

Inputs: format (required), file_path (required).

perform_meta_analysisB

Execute meta-analysis using R metafor/meta packages. Follows Cochrane Handbook Chapter 10 guidelines.

Inputs: data (required), effect_measure (required), model, heterogeneity_test.

validate_cochrane_dataA

Validate study data against Cochrane standards. Checks PICO criteria, sample sizes, effect sizes, and data quality.

Inputs: data (required), validation_level.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.7/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool targets a distinct stage in the meta-analysis workflow: import, validate, analyze, visualize, bias assessment, and report generation. No overlap or ambiguity between tool purposes.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun snake_case pattern (import_, validate_, perform_, generate_, assess_, generate_). The pattern is uniform and predictable.

    Tool Count5/5

    Six tools cover the essential steps of a Cochrane meta-analysis pipeline without being excessive. Each tool serves a clear and necessary role, making the set well-scoped.

    Completeness4/5

    The toolset covers the core workflow from data import to report generation. Minor gaps exist (e.g., sensitivity analysis, subgroup analysis, data export), but the primary domain is well represented.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers

    • Cochrane MCP

      aliildan
      A
      license
      Not graded
      quality
      D
      maintenance
      Turns the Cochrane Library into a tool for AI agents to search and retrieve clinical evidence, including systematic reviews, trials, and plain-language summaries.
      25 npm
      1
      MIT

    • ResearchStat AI

      flupke91
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI-native statistical analysis and reproducible research workflows through MCP, including natural language planning, protocol-based analysis, Python/R cross-validation, and publication-ready figure generation.
      MIT

    • PROSPERO MCP Server

      yussufembabi
      F
      license
      Not graded
      quality
      C
      maintenance
      Enables AI assistants to search registered systematic review protocols, look up individual CRD records, explore the MeSH vocabulary, check for possibly duplicate/overlapping reviews, and export results in RIS format.
      -

    • reviewfast-mcp

      taehojo
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables an AI assistant to run systematic review screening steps such as creating projects, importing records, scoring and ranking them with a zero-shot LLM classifier, presenting records for human decisions, applying a statistical stopping criterion, and exporting results. Human reviewers retain responsibility for inclusion decisions, with probabilities hidden unless requested.
      Apache 2.0