When conducting contact tracing as a COVID-19 containment measure, public health officials should include people with whom the infected person had contact before that person had symptoms, according to a study published yesterday in Morbidity and Mortality Weekly Report.
The study adds to mounting evidence of this type of spread and underscores the difficulty of identifying and isolating infected people. In recognition of findings from this and other studies, the US Centers for Disease Control and Prevention recently revised its public health recommendations for managing the pandemic coronavirus disease to acknowledge that even people who appear healthy can spread the disease.
Transmission 1 to 3 days before symptoms
The researchers reviewed clinical and epidemiologic data of all 243 coronavirus cases in Singapore from Jan 23 to Mar 26 to detect presymptomatic spread. They identified seven COVID-19 clusters of two to five patients each that likely involved presymptomatic transmission; 10 of 157 locally acquired cases (6.4%) were part of the clusters.
The investigators determined presymptomatic transmission using dates of exposure and of symptom onset in cases in which there was no evidence that the secondary patient had been exposed to anyone else diagnosed as having COVID-19.
The investigators were able to determine the date of exposure to the virus in four clusters, in which transmission occurred 1 to 3 days before symptoms emerged. Exact timing of transmission could not be identified in the other three clusters because the infected people lived together and thus had continual exposure.
Clusters traced to travel, social gatherings
The study clusters involved patients 26 to 63 years old. In the first cluster, two tourists from Wuhan, China, traveled to Singapore and visited a church, where they likely spread it to four other people who attended services that day.
The second cluster involved a woman who went to a dinner party, where she had contact with someone with confirmed COVID-19. She later went to a singing class, where she likely spread it to another woman.
The third cluster involved a woman who likely passed the infection to her husband, and the fourth occurred in a man who had traveled to the Philippines, had contact with a patient with pneumonia who later died, and then presumably passed it to his wife upon his return.
In the fifth cluster, a man who traveled to Japan, where he was likely infected, spread the infection to his housemate upon his return. In the sixth cluster, a woman who was exposed to the virus at a singing class went to church, where she likely infected two people sitting one row behind her.
The seventh cluster involved a man who had traveled to Indonesia, where he was likely infected with SARS-CoV-2, the virus that causes COVID-19, and spread it to a woman he met upon his return.
Support for physical distancing
“The possibility of presymptomatic transmission of SARS-CoV-2 increases the challenges of COVID-19 containment measures, which are predicated on early detection and isolation of symptomatic persons,” the authors wrote. “Public health officials conducting contact tracing should strongly consider including a period before symptom onset to account for the possibility of presymptomatic transmission.”
The findings support previous reports of presymptomatic spread in Japan; King County, Washington; and China. The Chinese study suggested this mode of transmission in 12.6% of cases outside of Hubei province.
The authors of the new study noted that the magnitude of this type of spread depends on the extent and duration of transmissibility in presymptomatic patients, which has not been determined, and they note that presymptomatic transmission has been seen in other respiratory viruses like influenza.
“These findings also suggest that to control the pandemic it might not be enough for only persons with symptoms to limit their contact with others because persons without symptoms might transmit infection,” they said. “Finally, these findings underscore the importance of social distancing in the public health response to the COVID-19 pandemic, including the avoidance of congregate settings.”
They added that, although community transmission was limited in Singapore during the period studied and strong surveillance systems had been implemented, unknown sources could have initiated the clusters. They said their results could also have been affected by recall bias regarding the dates of symptom onset and interviewer bias that could have led to underdetection of asymptomatic disease.