|
Abstract:
|
Estimates of the incubation period, the time between infection and disease onset, are needed to set quarantine policy during infectious disease outbreaks. Real-time evaluation of incubation periods presents statistical challenges. First, data collected early in an epidemic are subject to right truncation. Second, frailty effects could lead to earlier infection and shorter incubation periods among the first observed cases in an outbreak. Third, data from outbreaks are often coarse: the times of infection and disease onset are not observed exactly. We develop a framework to address these challenges. Simulations quantify the effects of truncation, frailty and coarse data on the bias and variance of incubation period estimates. We also consider sample size requirements for assessing incubation periods. Our methods are useful in outbreaks of new strains and emerging pathogens such as H5N1.
|
