Speaker
Description
The serial interval of an infectious disease – the length of time between symptom onset of an infector and infectee – is an important quantity in epidemiology, but its estimation requires knowledge of individuals' contacts and exposures, typically obtained through resource-intensive contact tracing efforts or household studies. Under partially sampled data, purported transmission pairs (infector-infectee) may be separated by one or more unsampled cases, biasing estimates of the serial interval.
We present a method to estimate the serial interval distribution that explicitly takes incomplete sampling into account and can use either contact data or virus sequence data for inference of transmission pairs. This genomic approach allows us to estimate the serial interval in a broad range of settings beyond households e.g. in workplace, school or healthcare clusters, and thereby compare how speed of transmission varies in different settings and over time. We present an application to clusters of COVID-19. Through its applicability to outbreaks with low sampling rates and large population sizes, our method is well suited for diseases with widespread community transmission tracked by routine genomic surveillance.
| Expedited Notification | No thanks, I do not require Expedited Notification |
|---|
