Theoretical Framework and Key Considerations for Time-to-Onset Analysis in Spontaneous Reporting Systems

Area of Science:

• Pharmacovigilance and Drug Safety
• Biostatistics and Epidemiology

Background:

• Spontaneous reporting databases are crucial for drug and vaccine safety monitoring.
• Traditional methods focus on reporting frequency, underutilizing temporal information of adverse events.
• Time to onset (TTO) provides valuable data on the timing of adverse event manifestation.

Purpose of the Study:

• To review the definition, calculation, and limitations of TTO analyses in spontaneous reporting databases.
• To provide an overview of statistical signal detection methods that incorporate TTO information.
• To discuss the complementary role of TTO analyses alongside frequency-based methods.

Main Methods:

• Review of existing literature on TTO analysis in pharmacovigilance.
• Discussion of nonparametric distribution-comparison tests (e.g., Kolmogorov-Smirnov, Anderson-Darling) for TTO data.
• Exploration of survival analysis and Weibull modeling for TTO data, noting limitations.

Main Results:

• TTO analysis captures temporal patterns of adverse events not evident in frequency-based methods.
• Nonparametric tests are suitable for spontaneous reporting data due to unknown population size and exposure.
• Combined use of disproportionality and TTO analyses improves sensitivity and interpretability of safety signals.
• Survival analysis and Weibull models have limitations in spontaneous reporting systems due to data characteristics.

Conclusions:

• TTO analysis is a valuable exploratory tool for characterizing adverse event onset patterns.
• TTO helps generate hypotheses and inform future epidemiological and safety studies.
• TTO analyses should not be used for estimating population risks or inferring causality due to inherent data limitations.