Pooled testing offers improved strategy for identifying COVID-19
A new approach to pooled novel coronavirus (COVID-19) testing can be a highly effective tool for curbing the SARS-CoV-2 pandemic, even if infections are widespread in a community, according to new research by the Harvard T.H. Chan School of Public Health and the Broad Institute of the Massachusetts Institute of Technology and Harvard University, published in the journal Science Translational Medicine.
By identifying infected individuals so that they can be treated or isolated, SARS-CoV-2 testing is a powerful tool for curbing the COVID-19 pandemic and safely reopening schools and businesses. But limited and sometimes costly testing throughout the pandemic has hampered diagnosing individuals and has hamstrung public health efforts to curtail the virus's spread, according to the study.
Pooled testing, in which multiple individual samples are processed at once, could be a powerful tool to increase testing efficiency. If a pooled test comes back negative, all samples in that pool are considered negative, thus eliminating the need for further testing. If a pooled sample is positive, the individual samples within that testing group need to be tested again separately to identify which specific samples are positive.
Although pooled testing has been implemented during the COVID-19 pandemic, its usefulness is curtailed when the pathogen is widespread in a community. Under those circumstances, most pooled samples could be positive and require additional testing to identify the positive individuals in each pool. This confirmatory testing eliminates any efficiencies gained by pooled testing, the researchers said.
To identify ways to make pooled testing more useful during widespread outbreaks, the researchers developed a model for how quantities of viral RNA, which are used to identify SARS-CoV-2 infection, vary across infected people in the population during an outbreak. This gave the researchers a very detailed picture of how test sensitivity is affected by pool size and SARS-CoV-2 prevalence.
The researchers then used the model to identify optimal pooled testing strategies under different scenarios. Using the model, testing efforts could be tailored to the available resources in a community to maximize the number of infections identified using as few tests as possible. Even in labs with substantial resource constraints, the researchers created simple pooled testing schemes that could identify as many as 20 times more infected individuals per day compared with individual testing.
Simple pooled testing schemes could be implemented with minimal changes to current testing infrastructures in clinical and public health laboratories, the researchers said.
"Our work helps quantify pooled testing's tradeoffs between losses in sensitivity from sample dilution and gains in efficiency," said Brian Cleary, PhD, corresponding author of the study, in a statement. "We show how to identify simple strategies that require no expertise to implement and that result in the greatest number of infections identified on a fixed budget."