HHS Warns Against Using Antigen Results for Nursing Home Cohorting, But Sharply Criticizes State Ban

A top U.S. coronavirus testing czar had harsh words for a Nevada directive banning the use of antigen point-of-care COVID-19 tests in skilled nursing facilities during a press call on Friday, calling it “an uninformed and unlawful unilateral prohibition.”

But Department of Health and Human Services (HHS) assistant secretary for health Adm. Brett Giroir also noted that the results from the devices in question, the BD Veritor and the Quidel Sofia, should not be used to cohort new admissions without a confirmatory polymerase chain reaction (PCR) test.

“Antigen testing should not be used in a normal person coming in a nursing home to screen them and cohort them in a COVID ward,” he said on the call. “That absolutely should be repeated by a PCR test. If they were positive, you should isolate them pending the PCR test or confirmatory test. You should not make a clinical decision on an otherwise well individual, not in an outbreak, just based on the single test.”


The call came after the Nevada Department of Health and Human Services issued a technical bulletin on October 2 requiring SNFs “to immediately discontinue the use of all COVID-19 point of care (POC) antigen tests until the accuracy of the tests can be better evaluated.”

The directive called into question the sensitivity, or the tests’ capacity to identify people with COVID-19, and specificity, or the capacity to identify people without COVID-19, of the two test types, describing them as “based on extremely limited data.”

Nevada based its directive on a survey of the 12 facilities performing antigen testing, with 3,725 total antigen tests performed and 60 positive test results. Of the 12 facilities, eight collected specimens for confirmatory PCR testing; 39 of the 60 positive tests had samples collected with results available at the time of the state’s survey.


Out of those 39, 16 were true positives as confirmed by PCR testing, and 23 were false positives as confirmed by PCR testing.

The rate of false positives for antigen COVID-19 tests to true ones will be affected by the prevalence of the disease in the population tested, Giroir said on the HHS call when asked about the false positives in Nevada.

According to Giroir, the 16 out of 39 — or 40% true positive rate — is “an outstanding rate, and shows the test is performing very well within the circumstances” of low prevalence, in the sense that it keeps infectious people out of nursing homes. The assistant secretary cited a September Rockefeller Foundation report that examined testing and found that in areas of low prevalence, 1,000 tests could produce 30 false positives for every true one, even for a test with 97% specificity rate.

Nevada has seen at least 11 new COVID-19 deaths and 471 new cases as of October 8, according to The New York Times’s COVID-19 tracker. Over the past seven days, it has seen 3,464 cases, or 112 cases per 100,000 residents, according to the same tracker.

In an October 8 letter responding to the Nevada directive, Giroir called on the Silver State to “cease the improper unilateral prohibition” on using the antigen tests in SNFs.

“While we absolutely welcome the opportunity to discuss any concerns with these two tests, or any other tests, and to report them appropriately to the FDA for further evaluation, the Department of Health and Human Services will, as our letter clearly stipulates, take swift and appropriate steps to protect Nevada seniors, if the state of Nevada does not immediately reverse its unwise, uninformed and unlawful unilateral prohibition,” Giroir said on the call.

He also noted that the move violates HHS’s coverage of these tests under the Public Readiness and Emergency Preparedness Act (PREP Act), which was expanded specifically to override state bans on antigen test use. At the time, Nevada was not one of the states specifically noted.

HHS announced in July that it would send the point-of-care testing kits to all nursing homes in the U.S., and at the time, Giroir emphasized the then-recommendation to test staffers to ensure they did not bring COVID-19 into nursing homes unawares. The Centers for Medicare & Medicaid Services (CMS) would later implement stricter mandates for testing of staffers and residents, along with fines for non-compliance, on August 25.

That initial announcement also included the acknowledgement that the antigen tests were less sensitive than the PCR tests.

The shipment of the BD and Quidel devices to all eligible nursing homes was completed in last month, Giroir noted on the call, with 13,850 facilities receiving a total of 13,985 instruments and more than 4.9 million tests by September 14.

There were 65 nursing homes in Nevada that received more than 25,000 point-of-care antigen tests from BD and Quidel, Giroir said; that does not include shipments to nursing homes of the rapid Abbott Labs BinaxNOW tests as part of a parallel program announced in September.

Questions about the accuracy of the antigen tests for screening asymptomatic people arose over the summer, though the initial focus was on false negatives. Stories of false positives began to emerge soon after, with the Wall Street Journal reporting that BD was investigating claims of false positives in the nursing home setting. Two major industry groups, the American Health Care Association and LeadingAge, found in a survey that almost a quarter of nursing homes had reported at least one false positive COVID-19 result.

At the time of those reports, Giroir also pointed out the issue of population spread; in general, he said in September, the higher the actual rate of infection in a population, 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 at the time. “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.”

In addition, the PREP Act guidance for the antigen tests includes a range of updates from the FDA and the Centers for Disease Control and Prevention (CDC) on the question of using them to test asymptomatic individuals for COVID-19. While more sensitive tests were preferred, the feasibility and turnaround times for such tests were also factors for providers to consider, according to the FDA’s recommendations to health care providers on screening for COVID-19.

LeadingAge, for its part, recommended on September 17 that its members use the point-of-care tests while following guidance from the CDC on when to conduct confirmation testing of positive antigen tests. The association, which represents non-profit senior living and care providers, also recommended confirmation testing with PCR tests in situations the CDC guidance did not address, such as asymptomatic residents or staff testing positive with an antigen test.

“If asymptomatic residents test positive with an POC antigen test, they should be placed into TBC [to be confirmed] and single room if possible but not moved to a COVID-19 (+) or cohorted with a known positive resident until confirmation testing with PCR is obtained,” the association noted.

In the October 9 call, Giroir emphasized that even though a false positive might take a staff member out for a couple days while waiting on a PCR test, “this is a minuscule price to pay identifying and isolating those who are infected, infectious, and who can devastate our elderly in nursing homes.”

“Bottom line, the recommendations in the Nevada letter are unjustified and not scientifically valid,” he said on the call. “They must cease their prohibition immediately.”

Companies featured in this article:

, , , , , ,