Clinical Evidence Synthesis Tool

Synthesize clinical evidence from multiple biomedical sources -- including PubMed, the Cochrane Library, clinical practice guidelines, and systematic review databases -- into structured, quality-graded evidence summaries. This skill enables AI agents to perform rapid evidence appraisal, generate GRADE-framework evidence tables, assess risk of bias across studies, and produce actionable clinical evidence briefs that support guideline development, clinical question answering, and evidence-based practice at the point of care.

Quick Install

npx skills add Open-Medica/open-medical-skills --skill evidence-synthesis-ai

What It Does

• Multi-source evidence aggregation pulling relevant studies from PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), WHO ICTRP, and published clinical practice guidelines (AHA, ACC, IDSA, NICE, WHO) for a given clinical question
• Study quality appraisal applying validated tools for risk-of-bias assessment: Cochrane RoB 2 for randomized trials, ROBINS-I for non-randomized studies, AMSTAR 2 for systematic reviews, and Newcastle-Ottawa Scale for observational studies
• GRADE evidence profiling generating Grading of Recommendations Assessment, Development and Evaluation (GRADE) evidence tables that rate certainty of evidence as high, moderate, low, or very low across outcomes, with transparent justification for up- or down-grading
• Forest plot data extraction identifying and structuring key quantitative data (effect sizes, confidence intervals, heterogeneity statistics, sample sizes) from systematic reviews and meta-analyses for rapid comparison
• Evidence brief generation producing concise, structured clinical evidence summaries formatted as PICO-based answers (Patient/Population, Intervention, Comparison, Outcome) with strength-of-recommendation ratings and key caveats

Clinical Use Cases