Sequential Epidemiological Analyses of Real-World Data: A Tool for Prospective Drug Safety Surveillance from the...

Area of Science:

• Pharmacovigilance and Drug Safety
• Epidemiology
• Health Informatics

Background:

• Large administrative healthcare databases offer near real-time drug safety surveillance.
• This contrasts with traditional reporting-based pharmacovigilance methods.
• The study focuses on rofecoxib and cardiovascular outcomes.

Purpose of the Study:

• To develop and test a prospective drug safety monitoring system.
• To conduct sequential safety monitoring of rofecoxib and cardiovascular risks.
• To evaluate the system's ability to detect safety signals early.

Main Methods:

• Utilized Danish population-based health registers (2000-2004).
• Employed sequential analysis with case-time-control and cohort study designs.
• Assessed rofecoxib use and myocardial infarction (MI) and ischemic stroke risks.

Main Results:

• An association between rofecoxib and MI risk was detected early using both case-time-control (OR 1.42) and cohort (RR 1.22) designs.
• The case-time-control design identified the risk in study period 2.
• The cohort design identified the risk in study period 7.

Conclusions:

• The prospective monitoring setup could have detected MI risk 3.5 years before rofecoxib's market withdrawal.
• This demonstrates the potential of database-driven pharmacovigilance.
• Further research is necessary to validate this monitoring approach.