Rapid COVID-19 self-tests could restore America to normalcy quickly, so why is the FDA blocking them?
April 26, 2020
It’s Sunday afternoon and I just took a COVID-19 antibody test, at home, and obtained results in less than 15 minutes. I’m not symptomatic but having worked in South Korea last year and in Milan this February I had a lingering curiosity whether I was exposed.
While I cannot divulge how I obtained the test, because, according to current FDA regulations, such procurement is illegal, I can advocate for expanding public access to the technology. The test kit comes from China through a US company, as do many diagnostic tests and components. In fact, the FDA has granted several US companies an EUA, or “emergency use authorization”, allowing them to distribute the tests to clinicians only. All of these rapid diagnostic tests (RDT’s) have been validated clinically to some extent per FDA’s current guidelines for SARS CoV-2 tests.
The test is similar to a home pregnancy kit in design, and no more complex or painful than a blood glucose finger prick.
So if the test is so easy to take, and already authorized for use by health care professionals, why is the FDA withholding it from the public? After all, home-based COVID-19 RDT’s could be the answer to a fast re-opening of the economy, schools, sports and other facets of society.
The reason is likely two-fold. First, there are capitalistic motives in the diagnostics industry that are in favor of more profitable clinic-based assays rather than a jump to rapid home-based tests. In the 25 years I have worked in the biotech industry I cannot recall a single diagnostic technology, in the US at least, that went directly to the consumer without first being sold to hospitals and clinics where margins are significantly greater. (Case in point — just as I am writing this editorial, Quest Diagnostics has announced they are offering a “home” antibody test for COVID-19 at about $120 — their test however requires having blood drawn at one of their labs. The self test I took costs one-third that amount and is self-administered. )
The other reason, and by no means mutually exclusive, is simply whether rapid antibody tests for COVID-19 are accurate enough. This is a valid concern. In scientific terms this is described as the specificity and sensitivity of the test or assay. In a perfect world every test would be highly sensitive — capable of detecting 100% of the positive samples in a population — generating no false negative results where someone who has the disease is not picked up. And every test would be highly specific — correctly identifying 100% of the negative individuals; never telling someone they are positive when they in-fact are not.
But no diagnostic testing is ever perfectly accurate. The scientific community accepts a certain number of false positives and false negatives for any qualitative assay as being part of the risk-benefit calculation.
The published numbers on the test I took are 87% for sensitivity to the COVID-19 IgM — the first antibodies that form within days of infection. In other words, out of 100 positive samples in the clinical validation the test picked up 87 of them; 13 samples were falsely diagnosed as negative for the antibody. The test also detected nearly 100% of the COVID-19 IgG samples — the antibody that forms 10–18 days after infection, an indicator of possible immunity. This test did not generate any false positives — having 100% specificity. The studies are admittedly small, and researchers are working to better assess the validity of these assays (Johns Hopkins has assembled an informative page on all currently available serology tests here.)
Despite only being validated in a small sample size, however, the test is good enough in my opinion to put in the hands of the public during a time when more information is better than perfect information. And using them as a “screening” test, not as a confirmation, would not only help generate larger data sets but reduce the burden on the healthcare system. Calling your doctor with this result in hand would be much more efficient, and cost saving than for her to write you a test requisition and bring you into the clinic.
The risk to be considered is what is the ramification of a wrong result. In the 1990’s, a false positive HIV test result, for instance, raised the concern of suicide by distraught individuals who felt this was a shameful death sentence. While there is no such stigma for a positive COVID-19 result, some worry that positive, or false positive, results may lead people to believe they are immune to reinfection and to take less precautions. The duration of immunity after recovering from an initial bout of COVID-19 is not well known currently.
Of higher concern, generally, with diagnostic tests, are false-negative results — individuals who have the virus or disease that the test does not detect as positive. Similarly, with HIV, the concern that individuals could be erroneously diagnosed as free of the virus, and then unknowingly spread it to sexual partners, or through blood donations, kept home-based tests off the market for decades. The Orasure at-home HIV test commonly used now isn’t perfect — it will yield a false negative about 10% of the time.
False results, however, are of concern even if the test is run in the clinic — and is a risk with every diagnostic test ever made and that likely ever will be made. This is where the benefits of obtaining some amount of true results, despite some false ones, must be considered. Take security screening at the airport for instance — metal detectors generate false positives (poor specificity) all the time but this is required to ensure the one possible person with a weapon is found. For a rapidly spreading pathogen like SARS COV-2 public health officials would likely agree that the risk of a small percent of false negatives (poor sensitivity) far outweighs the risk of a much larger number of what I term ignorance-based false negatives — infected individuals who, in the absence of evidence to the contrary, believe they are negative because they “feel ok.”
All at-home RDT’s come with an important caveat — that the results may not be accurate and that you must follow-up with your doctor to confirm a positive result, or re-test at some point in the future if you’ve received a negative result. In all scenarios individuals with symptoms are asked to seek medical advice and obtain a confirmatory result. A COVID-19 self-test would warrant require similar disclaimer. (The Quest diagnostic COVID-19 antibody test run in the clinic comes with these statements: There is not enough evidence at this time to suggest that people who have IgG antibodies are protected against future COVID-19 infection. Positive or negative antibody tests do not rule out the possibility of COVID-19 infection. Results also do not provide any information on whether you can spread the virus to others.)
Aside from being informative to the individual and household, rapid home-based COVID-19 testing would provide America with significant epidemiological information allowing a return to normalcy sooner rather than later. Currently, the only people being tested for COVID-19 are those who are symptomatic or perhaps through contact tracing effort (or those prominent or connected enough to get access to testing.) Sadly, even this isn’t entirely true — many individuals are still unable to get tested before succumbing to the disease. But because COVID-19 spreads so rapidly, and the shortage of clinical testing throughput, there is no way to perform proactive testing on mildly symptomatic or asymptomatic people. Current calculations estimate it would take nearly 4 years to screen all Americans using current clinical testing methods. In the few examples (New York and Santa Clara, CA) where studies have been performed to measure antibody levels in the general population, the results have shown much higher than calculated infection rates — providing invaluable new information by which to forecast for healthcare and economic planning.
Furthermore are the issues of cost and scaleability — testing at clinics and hospitals is expensive and serial, testing at home can be done in parallel — you don’t have to wait for the person in front of you to be tested and you don’t have to pay a lab technician and physician. The test I took at home costs just several dollars to manufacture and could be distributed by mail to households by county health officials, with follow-up, to gather and aggregate the results.
Rapid antibody testing for COVID-19 is by no means a panacea. It is also important to note that antibody tests are just one of the diagnostic tools available — and not the primary tool to diagnose active infection. Direct detection/measurement of the viral RNA through PCR, or antigen testing — both performed on nasal swabs, would remain the gold standard for determining active infection.
In either case, the more information available to individuals, health care providers and public health officials, the better. But until the FDA releases its chokehold on true COVID-19 self-tests, and allows authorized manufacturers to sell directly to consumers, America will likely have a long and slow path towards returning to normalcy. Allowing individuals to self-inform, and public health officials to aggregate this data, can only benefit our predicament. Healthcare workers, seniors, teachers and others at high risk or high risk of exposure could self-test and determine whether they have already been exposed. Those who show positive results for antibodies could make an informed decision on whether to seek follow-up if symptomatic, or return to work or other activities if asymptomatic. And those who test negative, as I did, would continue to take precautions to avoid infection.