Comparators in Pharmacovigilance: A Quasi-Quantification Bias Analysis.

Area of Science:

• Pharmacovigilance and Drug Safety
• Biostatistics and Epidemiological Methods
• Regulatory Science

Background:

• Disproportionality analyses are key in pharmacovigilance for detecting adverse drug events.
• The choice of comparator group significantly influences the reliability of these analyses.
• Optimal comparator selection for bias reduction remains an open question.

Purpose of the Study:

• To evaluate how different comparator strategies affect bias in disproportionality analyses.
• To assess the impact of restricted versus unrestricted comparators on signal detection.
• To investigate the influence of time-based restrictions on bias directionality.

Main Methods:

• Utilized the US Food and Drug Administration Adverse Event Reporting System (FAERS).
• Examined two drug-event combinations: rivaroxaban/hepatotoxicity and canagliflozin/acute kidney injury.
• Compared three disproportionality estimates (ROR, PRR, IC) using unrestricted and restricted comparators across defined time periods.

Main Results:

• No consistent bias directionality was observed across comparators for either drug-event pair.
• Time-based restrictions, particularly around external events, significantly altered results.
• Restricted comparators showed varied performance, sometimes worsening and sometimes improving signal detection.

Conclusions:

• No single comparator consistently reduces bias in disproportionality analyses.
• External events and associated time restrictions substantially impact analytical outcomes.
• Further research is needed to establish robust comparator methodologies in pharmacovigilance.