Analysis suggests approach may be particularly useful when testing resources are limited.
We’ve all heard the mantra for the age of COVID-19: test (smarter, but not harder) and trace. However, at least to date, addressing the first part has been easier said than done.
Quality problems and supply shortages have plagued (pardon the expression) RT-PCR testing initiatives in even the wealthiest nations since the start of the pandemic, with problems even more pronounced in low- and middle-income countries.
Thankfully, the authors of a new analysis published online on June 23rd by the Journal of the American Medical Association (JAMA) offer a potential solution: pool testing.
“From our simulations, given the test characteristics and proper grouping, pool testing may be substantially better than individual testing, both in terms of relative cost-savings and efficiency, without increasing the probability of false-negative results,” coauthor Alhaji Cherif, PhD, a research mathematician at the Renal Research Institute, told Contagion®.
According to Dr. Cherif and his colleagues, pool testing uses pooled samples from multiple patients. If the results from the pool test are negative for COVID-19, the disease caused by the new coronavirus, SARS-CoV-2, all patients in the pooled sample are considered negative for the virus. However, if the results of the pool are positive, each patient sample is tested separately.
To assess the sensitivity of a pool testing approach for COVID-19—under the theory that a sensitivity of less <100% would yield false negatives for the entire pool—Cherif and his colleagues constructed a model based on a 2-stage pool testing protocol in a population characterized by imperfect testing, taking into account disease prevalence, test sensitivity, and test specificity. They assumed that the probability of a true-positive result pool test equals sensitivity, the probability of a false-positive result equals 1 - specificity , test sensitivity and specificity would be unaffected by the number of patients in a pool, and that all tests are identically distributed.
In their report in JAMA, the team presented findings generated by their model for a typical number of RT-PCR tests for 94 patients, a specificity of 100%, a COVID-19 prevalence from 0.001% to 40%, and sensitivities from 60% to 100%. In general, they found that a pool testing strategy was an improvement over individual testing when prevalence is <30%.
For a realistic scenario, such as a sensitivity of 70% and prevalence of 1%, the optimal strategy required 13 patients per pool, they said. With this optimal pool size, only 16% as many tests would be required by subgroup tests than by individual tests, they found.
Overall, risk for false negatives was virtually nonexistent, and cost savings were substantial, they added.
“Given the uncertainty associated with mild and asymptotic infections, and global testing shortages, pool testing would allow expansion of current testing capacities, thereby increasing detection in the community in a cost-efficient manner,” Cherif said. “The strategy can be very useful in natural group settings, for example, in first responders, shift workers, classrooms, hospital departments, local clusters, households, conventions and events, long-term care facilities, to name but a few.”
According to Cherif, several countries in Europe—including Germany—are considering using pool testing strategies as they reopen. Considering the spike in infections seen in many states across the US that have reopened, it may prove an important strategy here as well.