The rate of false positive results from point-of-care COVID-19 testing units sent to nursing homes has not been higher than anticipated, a top federal official in charge of the program asserted Friday.
“The last thing you said is wrong, that it was in excess of expectation,” Department of Health and Human Services (HHS) assistant health secretary Adm. Brett Giroir said on a call with reporters. “That is not shown at all. We know with every test, including molecular tests and PCR tests, that there are false positives.”
Giroir was responding to a question about reports of false positives from antigen testing units manufactured by Becton Dickinson and Quidel. HHS sent the machines to most nursing facilities in the country, with Giroir on Friday confirming the multi-month project had been completed in mid-September with a total of 13,985 units shipped.
The Wall Street Journal earlier this month reported that BD was investigating anecdotal instances of false positives from nursing home operators, while an industry survey revealed that up to 25% of facilities that had received a BD or Quidel unit experienced at least one positive result that was later rejected by a more sensitive polymerase chain reaction (PCR) test.
The Journal piece identified a facility that moved residents, who were later found to be negative for the novel coronavirus, into a dedicated COVID-19 unit based on false positives.
The antigen method was always known to be less sensitive than its PCR counterpart, but given the longer turnaround times for the latter tests, HHS has positioned frequent antigen screenings as the most effective and practical way to meet increased nursing home workforce testing mandates by the Centers for Medicare & Medicaid Services (CMS).
The overall prevalence of COVID-19 in a given population has a direct effect on the potential for false positives, Giroir said: In general, higher the actual rate of infection, the likelier each individual positive result is accurate.
“If there’s an outbreak in an institution, having a positive is probably pretty good, and you don’t need to repeat it,” Giroir said. “It’s always fine to repeat it. When you don’t have an outbreak and you get a positive, you probably need to repeat it.”
The effect applies to tests for all diseases, he added.
“I don’t want you to get the idea that you’re more likely to get a false positive when the humidity is high, or an environmental factor, or a different kind of person,” he said. “The true positives and false positives are a complete function of the prevalence within that population, and the test itself.”
The industry survey, conducted jointly by the American Health Care Association and LeadingAge, also acknowledged that even the 25% rate was not necessarily surprising, and found that most buildings that reported any false positives only had one to three instances of the effect.
“While any potential false positive results are concerning, this appears to be consistent with what you would expect to see with widespread testing of many people in communities with relative low rates of COVID-19,” the groups observed.
HHS separately has begun sending rapid tests, which do not require the use of a machine, from manufacturer Abbott to nursing homes in designated coronavirus hotspots. About 7,600 facilities had received 974,000 of the Abbott units as of last week, Giroir said Friday.