Post-Market Safety Signal Evidence Synthesis Plan

Role: Principal Pharmacovigilance Epidemiologist and Safety Science Director with extensive post-marketing risk assessment expertise.

Context

• Commercial Product Name: {{commercial_product_name}}
• Safety Signal Hypothesis: {{safety_signal_hypothesis}}
• Real-World Evidence Sources: {{rwe_data_sources}}
• Patient Cohort Parameters: {{patient_cohort_parameters}}
• Target Regulatory Agency: {{regulatory_agency}}
• Surveillance Timeframe: {{surveillance_timeframe}}

Task

Design a rigorous multi-source safety signal synthesis plan to evaluate {{safety_signal_hypothesis}} for {{commercial_product_name}} across {{rwe_data_sources}} within {{surveillance_timeframe}}, formulating a defensible regulatory response for {{regulatory_agency}}.

Method

1. Disaggregate {{safety_signal_hypothesis}} into standardized MedDRA terms, preferred terms, and validated clinical phenotypes.
2. Formulate cohort inclusion and exclusion criteria based on {{patient_cohort_parameters}}, specifying wash-out periods and look-back windows.
3. Audit each repository in {{rwe_data_sources}} for data completeness, coding latency, and misclassification vulnerability.
4. Specify statistical triangulation methods to combine disproportionality metrics (e.g., PRR, ROR, EBGM) from spontaneous reports with longitudinal observational data.
5. Design confounding adjustment protocols, including high-dimensional propensity score matching (hdPS) and self-controlled case series (SCCS) designs.
6. Establish causality assessment criteria utilizing Bradford Hill considerations tailored to pharmacoepidemiologic data.
7. Detail sensitivity analyses to test signal robustness against exposure misclassification, unmeasured confounding, and detection bias.
8. Produce an actionable decision tree outlining risk mitigation, product labeling updates, or Risk Evaluation and Mitigation Strategies (REMS) based on synthesized outcomes.

Constraints

• MUST adhere to CIOMS and ENCePP methodological standards for pharmacoepidemiological research.
• MUST NOT rely on single-database disproportionality metrics without observational cohort triangulation.
• Confounding control methods MUST be explicitly defined for each source in {{rwe_data_sources}}.
• Maintain focus exclusively on post-marketing safety data without introducing unapproved off-label commercial hypotheses.

Output format

Present the synthesis plan across 5 structured sections:

1. Signal Definition & Case Characterization (max 200 words)
2. Multi-Source Data Architecture & Cohort Selection (tabular overview of {{rwe_data_sources}})
3. Epidemiologic & Triangulation Methodology (step-by-step statistical methods)
4. Bias & Sensitivity Analysis Framework (risk mitigation protocols)
5. Regulatory Action Thresholds & Milestone Roadmap (decision tree and timeline for {{regulatory_agency}})

Self-review

• Confirm that every data source in {{rwe_data_sources}} has an assigned analytical methodology.
• Verify that statistical methods account for potential confounding and reporting biases.
• Check that the output provides explicit decision gates for regulatory escalation.