NEW YORK — A federal effort to arm nursing homes with rapid coronavirus tests is stumbling on concern the tests return false positives, putting at least one state at odds with federal officials over the value of the tests.
Nevada this month ordered nursing homes to stop using the point-of-care tests after they found more than 20 instances where positive findings were overturned by more precise assays. That was more than half of the positive samples re-tested. Other states are now questioning their accuracy as well.
Federal officials, meanwhile, said states don’t have the authority to bar the tests, and the U.S. Department of Health and Human Services said it would take “appropriate action” if Nevada didn’t cease the ban. Brett Giroir, an assistant health secretary leading the administration’s testing efforts, said Nevada was putting nursing home residents at risk.
Nevada’s directive “can only be based on a lack of knowledge or bias, and will endanger the lives of our most vulnerable,” Giroir wrote to Nevada officials Friday.
The Trump administration expanded testing in long-term care settings this summer, shipping thousands of point-of-care testing systems to nursing homes. The goal was to allow facilities to quickly detect new COVID-19 cases among residents and staff.
The rapid antigen tests works in 15 minutes, rather than the day-or-longer turnaround time for more common polymerase chain reaction tests. They also don’t require a lab to process and are cheaper. But a well-established limitation is that they can return false negatives, missing people who are infected.
In recent weeks, a new concern has emerged around false positives, which could prompt nursing homes to isolate residents or even dangerously move them to COVID-19 wards when they’re not infected.
“Then they go and get COVID,” said Jeff Engel, senior adviser at the Council of State and Territorial Epidemiologists, or CSTE. “We think there’s a lot more work that needs to be done on accuracy” of point-of-care tests.
CSTE has heard about false positives in almost every state, though Nevada is the only state known to have recommended halting the tests, Engel said.
Manufacturers of the screenings, government officials and an independent expert, say that false positives occur with all tests and that the concerns are based on a small sample.
Of 3,725 antigen tests performed at Nevada nursing homes, 39 positive antigen test samples were also sent for PCR testing. According to the Nevada directive this month, 16 of those were true positives and 23 were false positives.
Mark McClellan, director of the Duke-Margolis Center for Health Policy at Duke University and a former top official at the U.S. Food and Drug Administration and the Centers for Medicare & Medicaid Services, has been working with the Rockefeller Foundation to develop COVID-19 testing protocols.
McClellan pointed out the Nevada antigen tests had detected many true positives, which could have sparked outbreaks and might not otherwise have been detected in time. “Averting even one nursing home outbreak is a huge economic and health benefit,” he said.
So far nursing homes have had the most experience with COVID-19 antigen tests. The government, though, is pushing to expand rapid testing more broadly, including in schools.
When issues with false positives emerged in Vermont this summer, developer Quidel Corp said the company investigated but found no issues with its tests. The problem likely lay not with Quidel’s product but with the PCR tests used to verify the antigen results, Chief Executive Officer Douglas Bryant said.
Company officials at Becton Dickinson and Co., meanwhile, said the Nevada numbers represent a 0.6% rate of false positives, consistent with the company’s stated rate of up to 2% false positives with its test.
The state’s sample represents just 12 of 11,000 nursing homes which have the BD tests, and data the company is now tracking also doesn’t show anything unexpected, said Dave Hickey, president of integrated diagnostic solutions. The company plans to meet with Nevada officials early next week.
Washington state health officials are also aware of false positives at long-term care facilities, said Scott Lindquist, state epidemiologist for communicable diseases. However, the state hasn’t required require faulty results to be reported, so it’s hard to gauge the scope of the problem, he said.
It’s now scrambling to create a protocol for verifying positive antigen tests with a PCR test and, if that’s negative, to label the initial result as a false positive. The state recommends isolating long-term care employees or residents following a positive antigen test, while awaiting confirmation from a PCR test.
Louisiana is taking a “cautious approach” and is finalizing more specific guidance on how to use the rapid tests in nursing homes that it plans to release it soon, said Mindy Faciane, a public information officer at the Louisiana Department of Health.
“We are aware of the existing concerns regarding rapid point-of-care antigen tests, are aware of the challenges other states are experiencing, and are certainly aware of how high the stakes are,” given the importance of protecting vulnerable nursing home residents, she said.
The American Health Care Association and LeadingAge, two organizations representing nursing homes, recommend their members continue using antigen tests. A majority of facilities reported less than five results that didn’t match further tests to verify them, according to a September survey of members.
Federal guidance, meanwhile, says the tests have “lower sensitivity, but similar specificity” as PCR diagnostics that take longer to perform. That means that while they may not detect as many positive cases, they are about as accurate in correctly telling when people are COVID-negative.
The CDC says that nursing homes using antigen testing should consider confirming a positive result for an asymptomatic individual with a PCR test within 48 hours, especially if the county has low virus rates.
The difference in accuracy between what test makers report to regulators and the experience of nursing home may stem from the conditions that the tests are deployed in, said Eric Blank, the chief program officer of the Association of Public Health Laboratories.
Manufacturers trying to prove to the Food and Drug Administration that their diagnostics work are “essentially doing it under perfect conditions and selecting perfect data,” he said.
Conditions can be messier in practice, with tests performed by people without formal training. “If you’re in a nursing home what do you do with a positive individual test?” Blank said. “It’s a lot of attendant issues with the uncertainty of tests results.”
©2020 Bloomberg News
Visit Bloomberg News at www.bloomberg.com
Distributed by Tribune Content Agency, LLC.