Probability sampling to slow spread of SARS-CoV-2

By Dr. Ananya Mandal, MDOct 14 2020

The COVID-19 pandemic has infected over 38 million people around the world and is showing no signs of going away any time soon. Researchers worldwide are frantically working on developing an effective vaccine to prevent infection with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) that causes COVID-19 disease.

At present, there are no effective drugs for the treatment of COVID-19 nor vaccine to block SARS-CoV-2 infection. Several non-pharmacological measures are being practiced to break the chain of human to human transmission of the infection.

Researchers from the University of Washington worked on one of the non-pharmacological measures to break the chain of transmission and analyzed its efficacy in stopping the pandemic.

Their study titled, “An Innovative Non-Pharmaceutical Intervention to Mitigate SARS-CoV02 Spread: Probability Sampling to Identify and Isolate Asymptomatic Cases,” was released at the pre-publication and pre-peer review site medRxiv*.

Transmission from asymptomatic cases

Studies have shown that many cases of COVID-19 are transmitted from individuals who do not exhibit symptoms – asymptomatic cases. One way to stop transmission of the infection has been identification, isolation, and treatment of the symptomatic cases. To prevent viral spread, there have been complete lockdowns of nations, including businesses and schools, for many months now. Despite this, there has been a steady rise in the number of infected individuals, many of whom are killed by the infection.

NPIs

Non-pharmaceutical interventions (NPIs) are implemented to prevent the spread of the infection before an effective vaccine or drug is found. These NPIs, include social distancing and a ban on gatherings, mask use, and hand hygiene. With the closure of businesses and schools over months, the well-being of individuals and populations is challenged.

What can be done?

The authors wrote that the main questions asked by them include, “can less-disruptive non-pharmaceutical interventions substantially mitigate the COVID-19 epidemic and can they mitigate the epidemic as much as or more than disruptive interventions like school and work closures?”

They suggest that isolating the whole population in lockdowns can severely impact the economy, and it could be “financially, physically, and socially,” much less harmful if only those who are infected are isolated only for the period in which they are infectious. Both symptomatic and asymptomatic infections thus need to be identified and isolated to prevent the spread of the infection.

Figure 4a shows highest daily prevalence of symptomatic infections. Figure 4b shows cumulative number of symptomatic infections at day 250. All STQ scenarios use 50,000 tests implemented once every seven days. Random seed used: 17392. See Table S4 in the paper for numerical presentation of these results.

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

Asymptomatic carriers

Studies have shown that 79 percent of the transmission of CVODI-19 is caused by asymptomatic cases, and the number of asymptomatic cases could be 3 to 40 times the symptomatic cases. This makes the presence of undetected asymptomatic carriers a huge public health problem. Testing is mainly done for symptomatic cases, they write.

How to detect asymptomatic cases and STQ

One of the best ways to detect asymptomatic cases is universal and ongoing testing and thorough contact tracing.

The authors devise a method called “sampling-testing-quarantine (STQ).” STQ involves testing and quarantine of the possible contacts. In addition, a probability sampling of the general population is done regardless of who has symptoms. For this study, the authors used simple random sampling, cluster sampling, and pooled sampling methods to test and detect asymptomatic cases.

To see if their STQ method was working, they used an agent-based model (ABM) to see if the transmission could be reduced.

Sampling methods

Three methods used were described as follows:

• Simple random sampling – randomly, people were chosen from the population with equal probability.
• Cluster sampling – First, people were divided into community clusters in which they live, or children were separated based on schools they attended. Then from the community or school cluster, the same proportion of people were randomly selected.
• Pooled sampling – people were divided into groups of two and groups of five. A sample is tested from everyone in the group by the pooling method. If it returns a positive for SARS-CoV-2, the whole group is isolated.

Results

Results showed that STQ could substantially slow and reduce the spread of COVID-19. This was even in the absence of school, and business shutdowns, they wrote. Some of the main results the team found were:

• If no interventions were followed, a peak number of 77,000 was reached in around 136 days. If symptomatic cases were isolated, the peak was reduced to 50,000 cases
• If the household contacts of symptomatic individuals were isolated, the peak came down to 26,000.
• STQ helped in identifying asymptomatic cases
• Isolating only the asymptomatic cases reduced the peak to 66,000.
• Isolating asymptomatic cases, as well as their family contacts, reduced peaks to 53,000.
• If all symptomatic and asymptomatic cases and their household members could be isolated, the peak came down to 18,000 cases at peak prevalence. This was 24 percent of the pandemic that was allowed to progress without any intervention.
• Further cluster sampling could be one of the optimum methods of sampling for STQ, they noted.
• The team found, “With just 10,000 tests used every seven days for the STQ procedure, the peak symptomatic prevalence decreases by two thirds, from approximately 77,000 to 25,000 cases, and is delayed by 50 days.”

Conclusions and implications

These results showed that STQ procedure and random selection from the general population for testing could help identify enough asymptomatic but infectious cases and successfully reduce and decrease the pandemic. Thus even if schools and businesses are open, the cases could be brought down by identification and isolation of all cases of COVID-19 irrespective of symptoms.

This study was funded by the University of Washington, Department of Sociology.

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.