OpenEvidence — Evidence Search & Retrieval

OpenEvidence — Evidence Search & Retrieval

Overview

Primary workflow for OpenEvidence clinical evidence integration. Covers the core use case: searching clinical literature with evidence-level filters, retrieving structured citations with journal and year metadata, checking drug interactions against patient context, and looking up specialty guidelines from major bodies (ACC/​AHA, ESC, NICE). Responses include confidence scores and evidence grading to support clinical decision making. All queries support specialty filtering to narrow results to relevant domains.

Instructions

Step 1: Search Clinical Evidence

const result = await client.query({
  question: 'What is the recommended treatment for acute migraine in adults?',
  context: 'emergency_department',
  evidence_level: 'high',
  specialty: 'neurology',
  max_citations: 10,
});

console.log('Answer:', result.answer);
console.log(`Confidence: ${result.confidence} | Evidence grade: ${result.grade}`);
result.citations.forEach(c =>
  console.log(`  [${c.journal}] ${c.title} (${c.year}) — Level ${c.evidence_level}`)
);

Step 2: Filter by Specialty and Date

const recent = await client.search({
  keywords: 'GLP-1 receptor agonist cardiovascular outcomes',
  specialty: 'cardiology',
  year_min: 2024,
  evidence_level: 'meta-analysis',
  limit: 20,
});
console.log(`Found ${recent.total} results`);
recent.results.forEach(r => console.log(`  ${r.title} (${r.journal}, ${r.year})`));

Step 3: Check Drug Interactions

const interactions = await client.interactions.check({
  medications: ['metformin', 'lisinopril', 'atorvastatin'],
  patient_context: { age: 65, conditions: ['diabetes', 'hypertension'] },
});

interactions.forEach(i =>
  console.log(`${i.drug1} + ${i.drug2}: ${i.severity} — ${i.description}`)
);
if (interactions.some(i => i.severity === 'major')) {
  console.warn('WARNING: Major interaction detected — review before prescribing');
}

Step 4: Guideline Lookup

const guidelines = await client.guidelines.search({
  condition: 'hypertension',
  source: ['ACC/​AHA', 'ESC', 'NICE'],
  year_min: 2023,
});
guidelines.forEach(g =>
  console.log(`${g.source}: ${g.title} (${g.year}) — ${g.recommendation_class}`)
);

Error Handling

Output

A successful run returns evidence-backed answers with citations, drug interaction severity assessments, and guideline recommendations. Each response includes a confidence score and evidence grade for clinical decision support.

Resources

• OpenEvidence Platform
• OpenEvidence API Documentation

Next Steps

Continue with openevidence-core-workflow-b for patient case analysis and reporting.