Skip to content

Introducing the Study Assessment Laboratory

We're excited to showcase the Study Assessment Laboratory, a powerful new plugin in BMLibrarian that automatically evaluates research quality, study design, and trustworthiness of biomedical evidence.

What Does It Do?

The Study Assessment Laboratory takes a publication's abstract and performs a comprehensive AI-powered analysis, providing researchers with an instant quality assessment. Simply enter a PubMed ID, and the system extracts and evaluates the study across multiple dimensions.

Study Assessment Laboratory

Key Features

Study Classification

The plugin automatically identifies and extracts critical study metadata:

  • Study Type: Categorizes the research (e.g., Meta-analysis, RCT, Cohort study)
  • Study Design: Describes the methodology used
  • Evidence Level: Assigns an evidence hierarchy rating (Level 1-5)
  • Sample Size: Extracts participant numbers when available
  • Follow-up Period: Identifies longitudinal study duration

Quality Metrics

Each assessment includes quantitative scores:

  • Quality Score: A 0-10 rating based on methodological rigor
  • Confidence: The AI's confidence level in its assessment
  • Explanatory Notes: Context for why certain scores were assigned

Design Characteristics

Quick visual indicators show study design features such as retrospective vs. prospective design, controlled vs. uncontrolled comparisons, and other relevant characteristics.

Balanced Evaluation

The plugin provides both Strengths and Limitations of each study:

Example Strengths:

  • Comprehensive literature search across multiple databases
  • Inclusion of randomized controlled trials
  • Pooled analysis providing quantitative estimates

Example Limitations:

  • Limited number of studies in the analysis
  • Heterogeneity in methodology
  • Potential publication bias not addressed

Current Status

At this stage, the Study Assessment Laboratory analyzes data extracted from publication abstracts. This provides rapid assessments while we complete development of full-text analysis capabilities, which are nearing completion.

Full-text analysis will enable:

  • Deeper methodology evaluation
  • Assessment of results and discussion sections
  • Detection of potential conflicts of interest
  • More comprehensive bias assessment

Try It Yourself

The Study Assessment Laboratory is available in the BMLibrarian desktop application. Load any PubMed ID to get an instant quality assessment of the research.

Stay tuned for updates on full-text analysis support!