Glossary of Terms
A comprehensive reference guide to systematic review methodology, statistical concepts, and EvidAI platform terminology.
Systematic Review Methodology
A-E
| Term | Definition |
|---|
| Active Learning | Machine learning approach where the AI model improves by learning from your screening decisions in real-time |
| AMSTAR-2 | A Measurement Tool to Assess Systematic Reviews (version 2) - 16-item critical appraisal tool |
| Audit Trail | Complete, time-stamped record of all actions and decisions made during a systematic review |
| Bias | Systematic error that leads to deviation from true findings |
| Boolean Operators | Logical connectors (AND, OR, NOT) used to combine search terms |
| Citation Chaining | Search technique that follows references to identify additional relevant studies |
| Cochrane Handbook | The definitive methodological guide for conducting systematic reviews in healthcare |
| Confidence Interval (CI) | A range of values that likely contains the true population parameter |
| Data Extraction | Systematic process of collecting relevant data from included studies |
| Deduplication | Removing duplicate references that appear when searching multiple databases |
| Effect Size | Quantitative measure of the magnitude of a phenomenon |
| Eligibility Criteria | Pre-specified characteristics for study inclusion/exclusion |
F-M
| Term | Definition |
|---|
| Fixed-Effect Model | Meta-analysis model assuming all studies estimate the same underlying effect |
| Forest Plot | Visual representation of meta-analysis results |
| Funnel Plot | Scatter plot used to assess publication bias |
| GRADE | Framework for rating certainty of evidence and strength of recommendations |
| Grey Literature | Publications produced outside traditional commercial channels |
| Heterogeneity | Variability in study results beyond what would be expected from chance |
| I² Statistic | Measure indicating percentage of variability due to true differences |
| Living Systematic Review | Systematic review that is continually updated with new evidence |
| Meta-Analysis | Statistical technique for combining quantitative results from multiple studies |
| MeSH Terms | Medical Subject Headings - controlled vocabulary for indexing articles |
N-Z
| Term | Definition |
|---|
| Narrative Synthesis | Non-quantitative approach to synthesizing evidence |
| Newcastle-Ottawa Scale | Quality assessment tool for non-randomized studies |
| PICO Framework | Structure for clinical questions: Population, Intervention, Comparator, Outcome |
| PRISMA | Preferred Reporting Items for Systematic Reviews and Meta-Analyses |
| PROSPERO | International prospective register of systematic reviews |
| Publication Bias | Tendency for positive results to be more likely published |
| Random-Effects Model | Meta-analysis model accounting for between-study heterogeneity |
| Rapid Review | Streamlined systematic review for timely evidence |
| Risk Ratio (RR) | Ratio of the probability of an event between groups |
| RoB 2.0 | Cochrane Risk of Bias tool version 2 for randomized trials |
| Scoping Review | Review that maps the evidence landscape on a broad topic |
| Sensitivity | Proportion of truly relevant studies correctly identified |
| Specificity | Proportion of truly irrelevant studies correctly excluded |
EvidAI Platform Terms
| Term | Definition |
|---|
| Evidence Studio | Core systematic review workflow module |
| Compose | Module for Protocol Builder, Library Manager, and Manuscript Writer |
| Discover | AI-powered research discovery module |
| Knowledge Suite | Analytics and visualization tools |
| EvidAsk | Evidence-based Q&A system with cited literature |
| Multi-Agent Consensus | 4 AI models voting independently for screening accuracy |
| Confidence Score | AI-generated indicator (0-100%) of certainty |
| Living Review Mode | Automated monitoring and updates for completed reviews |
Statistical Terms Quick Reference
| Symbol | Meaning | Interpretation |
|---|
| p < 0.05 | P-value less than 0.05 | Statistically significant at 5% level |
| I² > 50% | High I² value | Substantial heterogeneity present |
| 95% CI | 95% Confidence Interval | Range containing true effect with 95% probability |
| OR = 1 | Odds ratio of 1 | No association (null effect) |
| RR < 1 | Risk ratio less than 1 | Intervention reduces risk |
