AI Pharmacovigilance Signal Assessment Playbook

A pharma company's post-market safety team has received a signal from its pharmacovigilance database: a disproportionality analysis shows a reporting odds ratio of 4.2 for a serious cardiac adverse event in patients taking its approved drug. The signal was not present in pre-approval trials. The company has 15 business days to assess the signal under ICH E2E guidelines.

When to use this playbook

• Use this playbook when the decision looks like the situation above: A pharma company's post-market safety team has received a signal from its pharmacovigilance database: a disproportionality analysis shows a reporting odds ratio of 4.2 for a serious cardiac adverse event in patients taking its approved drug.
• It is a fit when you have source files in hand and need a structured, reviewable analysis — not a generic chat answer about "Pharmacovigilance Signal Assessment".
• Do not use it as a substitute for licensed, legal, clinical, or authorized official judgment in the domain.

What you'll need

• Pharmacovigilance database disproportionality analysis (ROR 4.2 for cardiac adverse events)
• Pre-approval clinical trial safety data
• Literature review of the cardiac adverse event and drug class
• Patient case narratives for the top 20 reported cases
• Company pharmacovigilance standard operating procedures

Attachments: Documents (Documents)

You are a pharmacovigilance medical officer assessing a post-market safety signal for a marketed drug. I am attaching:

Work only from the attached source files. If a conclusion is not supported, say so.

Produce:
1. Assess the ROR of 4.2: is this a strong or weak signal, what is the confidence interval, and how does it compare to background rates for this cardiac event?
2. Analyze the 20 case narratives: is there a consistent clinical pattern (timing, dose, patient demographics, confounders) suggesting a causal relationship?
3. Review the literature: is there a mechanistic basis for the drug to cause this cardiac event based on its pharmacology?
4. Determine the regulatory obligation: does this signal require an expedited report (15-day), a periodic safety update report (PSUR) update, or a label change?
5. Tell me the signal assessment conclusion (signal, no signal, inconclusive), the proposed action, and the FDA communication strategy.

Call out where independent models are likely to disagree, and list follow-up documents a reviewer should request.

What to expect

• Signal strength assessment with confidence interval
• Case narrative pattern analysis
• Mechanistic plausibility review
• Regulatory reporting obligation determination
• Signal conclusion, proposed action, and FDA communication strategy

Review before you act

• Validate this output against source files before relying on it: Assess the ROR of 4.2: is this a strong or weak signal, what is the confidence interval, and how does it compare to background rates for this cardiac event?.
• Validate this output against source files before relying on it: Analyze the 20 case narratives: is there a consistent clinical pattern (timing, dose, patient demographics, confounders) suggesting a causal relationship?.
• Validate this output against source files before relying on it: Review the literature: is there a mechanistic basis for the drug to cause this cardiac event based on its pharmacology?.
• Validate this output against source files before relying on it: Determine the regulatory obligation: does this signal require an expedited report (15-day), a periodic safety update report (PSUR) update, or a label change?.
• Confirm every cited figure, date, counterparty, or requirement against the attached originals — models compress and can drop a qualifier.
• Treat disagreement between models as a review item, especially on classification, materiality, and recommended next action.
• Do not authorize an operational, clinical, legal, credit, or enforcement action solely because the models agree.