Estimating the basic reproduction number at the beginning of an outbreak

Area of Science:

• Epidemiology
• Mathematical Biology
• Infectious Disease Modeling

Background:

• Accurate estimation of the basic reproduction number (R0) is crucial for understanding and controlling infectious disease outbreaks.
• R0 estimation methods are often developed with specific epidemiological models in mind, such as SIR, SEIR, and SEAIR.

Observation:

• This study evaluated the performance of popular R0 estimation techniques using simulated data from SIR, SEIR, and SEAIR models.
• The sensitivity of R0 estimators to model misspecification (differences between data-generating and estimation models) was examined.
• Real-world Canadian COVID-19 case data was also analyzed to assess R0 estimation in a practical setting.

Findings:

• Simulation results indicated that certain R0 estimation methods performed better than others, but no single method proved universally superior.
• The performance of estimators varied depending on whether the epidemiological model assumptions matched the data-generating process.

Implications:

• The findings suggest that practitioners should exercise caution when selecting R0 estimation methods, considering the underlying epidemiological assumptions.
• Recommendations are provided to guide the choice of R0 estimation techniques for early epidemic stages, applicable to influenza and COVID-19.