Skip to main content

Contact Tracing Assessment of COVID-19 Transmission Dynamics in Taiwan and Risk at Different Exposure Periods Before and After Symptom Onset

Our take —

Based on contact-tracing data from 100 cases in Taiwan, the risk of transmission of SARS-CoV-2 is higher just after symptoms begin or even before symptom onset. Despite the prolonged viral shedding among those infected with SARS-CoV-2 and extended durations of clinical symptoms among COVID-19 cases, transmission to others is most likely to occur right before or after symptom onset, reinforcing the need for social distancing and protective measures even when contacts do not appear to be ill.

Study design

Case Series; Other

Study population and setting

All laboratory-confirmed cases in Taiwan between January 15 and March 18, 2020, and their contacts were monitored through an electronic tracing system to determine the transmission risk at different exposure time windows. The exposure window was defined as the time from onset of symptoms onset of the case to exposure of other individuals who were close contacts of the case (this number could be negative for contacts who were exposed prior to onset of symptoms). For asymptomatic cases, the window was based on exposure relative to date of confirmation. Close contacts were quarantined for a period of 14 days following exposure and monitored for any relevant symptoms. Those contacts with symptoms received RT-PCR testing; contacts belonging to high-risk populations were tested regardless of symptoms. The secondary clinical attack rate was defined as the ratio of symptomatic confirmed cases among the close contacts.

Summary of Main Findings

A total of 100 confirmed cases and 2761 close contacts were included. A total of 22 secondary cases, including four asymptomatic infections, were identified. The secondary attack rate was 0.7% overall (95% CI 0.4%-1.0%), and was higher among those whose exposure started within five days of symptom onset (1%, 95% CI 0.5%-1.6%]. All 22 secondary cases had initial exposure within the first 5 days of symptom onset of the index case; by contrast among those who did not develop Covid-19 symptoms, 68% had contact within that initial period. Those contacts with only pre-symptomatic exposure had an attack rate of 0.7% (95% CI 0.2%-2.4%). Attack rates were higher among household and non-household family contacts compared with health care or other settings, and were higher among those aged 40-59 years, and those 60 years and older.

Study Strengths

This study includes contacts who were exposed to cases prior to symptom onset among the cases. This allows for the study of pre-symptomatic transmission, as well as symptomatic transmission.


Though the attack rate is clearly noted as being defined by symptomatic confirmed cases, only symptomatic contacts were tested. There was also underascertainment of contacts prior to symptoms onset and this may underestimate transmissions occurring prior to symptom onset.

Value added

Although it was previously established that SARS-CoV-2 could be transmitted prior to symptom onset, the long duration of viral shedding and prolonged illness among those sickened by COVID-19, raises questions around when transmission is most likely to occur during the natural history of SARS-CoV-2. This study provides evidence that onward transmission of SARS-CoV-2 occurs most frequently before and soon after symptom onset.

This review was posted on: 16 July 2020