EvidAsk

EvidAsk is your AI-powered research companion—ask any evidence-based question and receive accurate, cited answers drawn from the scientific literature. Think of it as having a research librarian available 24/7.

---

Overview

EvidAsk combines advanced natural language processing with comprehensive literature access to answer your research questions with evidence-based responses and proper citations.

---

What EvidAsk Can Answer

Research Questions

Methodology Questions

• "When should I use random vs fixed effects model?"
• "How do I calculate sample size for a meta-analysis?"
• "What's the difference between RoB 2.0 and ROBINS-I?"
• "How do I handle missing outcome data?"

Quick Facts

• Drug dosages and indications
• Disease statistics and prevalence
• Diagnostic criteria
• Treatment guidelines

---

How EvidAsk Works

The Answer Process

Behind the Scenes: When you ask a question, EvidAsk searches across millions of indexed papers, analyzes relevant content, synthesizes the evidence, and generates a response with proper citations.

Step-by-Step:

1. Query Understanding: AI interprets your question intent
2. Literature Search: Relevant papers identified
3. Evidence Extraction: Key findings pulled from papers
4. Synthesis: Information combined into coherent answer
5. Citation Generation: Sources linked for verification

Response Components

Each EvidAsk response includes:

---

Using EvidAsk

Asking Questions

Best Practices for Questions:

Example Conversations

Clinical Question:

You: What is the NNT for statins to prevent one cardiovascular event?

EvidAsk: Based on meta-analyses of statin therapy for primary prevention:

• The number needed to treat (NNT) ranges from 50-100 over 5 years 
  to prevent one major cardiovascular event in low-risk patients.
  
• For high-risk patients (established CVD), the NNT is much lower, 
  approximately 10-20 over 5 years.

• The CTT Collaboration meta-analysis (Lancet 2010) found that for 
  every 1 mmol/L reduction in LDL cholesterol, there was a 22% 
  relative reduction in major cardiovascular events.

Sources:
[1] CTT Collaboration. Lancet. 2010;376(9753):1670-81
[2] Chou R, et al. Ann Intern Med. 2016;164(5):360-71

Would you like more details on any specific population or outcome?

Methodology Question:

You: When should I use GRADE vs ROBINS-I?

EvidAsk: These tools serve different purposes in systematic reviews:

**ROBINS-I (Risk of Bias in Non-randomized Studies of Interventions)**
• Use for: Assessing individual study quality
• Applies to: Non-randomized studies (cohorts, case-control)
• Domains: Confounding, selection, classification, deviations, 
  missing data, measurement, reporting

**GRADE (Grading of Recommendations Assessment, Development, Evaluation)**
• Use for: Rating certainty of evidence across studies
• Applies to: Bodies of evidence for specific outcomes
• Considers: Study design, RoB, inconsistency, indirectness, 
  imprecision, publication bias

**In Practice:**
Use ROBINS-I during data extraction to assess each study's RoB, 
then use GRADE during synthesis to rate overall evidence certainty.

Sources:
[1] Sterne JA, et al. BMJ. 2016;355:i4919 (ROBINS-I)
[2] Guyatt GH, et al. BMJ. 2008;336:924-6 (GRADE)

Follow-Up Questions

Continue conversations for deeper understanding:

• "Can you elaborate on the high-risk findings?"
• "What were the limitations of these studies?"
• "Are there any contradicting studies?"
• "Can you provide more recent evidence?"

---

Advanced Features

Document Context

Upload documents for contextual Q&A:

Integration with Reviews

EvidAsk understands your active reviews:

• Reference included/excluded studies
• Answer questions about your evidence base
• Help with interpretation and synthesis
• Assist with manuscript writing

Citation Management

Easily use EvidAsk responses in your work:

• Copy citations in multiple formats
• Export to reference managers
• Link directly to source papers
• Track which answers you've used

---

Quality & Limitations

What EvidAsk Excels At

✅ Evidence-based medical/scientific questions ✅ Methodology and statistics guidance ✅ Quick fact-finding with citations ✅ Explaining concepts and relationships ✅ Comparing treatments or approaches

Limitations to Understand

Important: EvidAsk is a research support tool, not a replacement for expert judgment. Always verify important findings and consult appropriate professionals for clinical decisions.

---

Best Practices

Getting Great Answers

1. Be Specific: Detailed questions get detailed answers
2. Use Medical/Scientific Terms: Proper terminology improves accuracy
3. Ask Follow-Ups: Dig deeper on important topics
4. Verify Citations: Check sources for critical information

Efficient Usage

• Start with EvidAsk for initial exploration
• Use for quick methodology guidance
• Verify answers with primary sources
• Track useful responses for future reference