domains / families / research / protocol-deviation-classifier
protocol-deviation-classifier
separate
Classify clinical trial protocol deviations as major or minor based on ICH E6/GCP guidelines. Three-impact-dimension assessment (safety, data integrity, scientific validity), confidence scoring, and regulatory compliance reporting with recommended actions.
When to use
- Use this skill when the task needs Determine whether an incident in a clinical trial is a "major deviation.
- Use this skill for data analysis tasks that require explicit assumptions, bounded scope, and a reproducible output format.
- Use this skill when you need a documented fallback path for missing inputs, execution errors, or partial evidence.
When not to use
- Do not use for classifying adverse events (use a MedDRA coding tool).
- Do not use as the final authority on deviation severity — classification must be confirmed by clinical QA personnel.
- Do not use for non-clinical-trial compliance issues (e.g., manufacturing, lab QC).
- Do not use when the deviation description is too vague to assess impact on safety, data integrity, or scientific validity — request clarification first.
Skill page
Source