Immunity Passports and the Perils of Conferring Coronavirus Status

A woman is stopped by TSA.
Immunity passports, which would verify whether someone has been infected with COVID-19, could help open up travel, but their deployment raises concerns.Photograph by Marc Asnin / Redux

In January, a Swedish entrepreneur named Joakim Hultin co-founded Sidehide, a new digital app intended to streamline hotel reservations. Weeks later, some of the first confirmed cases of COVID-19 were reported in Europe. Almost instantly, Hultin told me, “demand stopped.” Before the pandemic, Sidehide was working with a London-based company called Onfido, which uses artificial intelligence and facial recognition to verify identities. Hultin learned that Onfido had created a way for users to upload a serology test to a private server and use facial biometric data to unlock the data and display the results. He asked the company to build that capacity into his app. In June, Sidehide will launch again, in a few Miami hotels, having added what is being called an “immunity passport.” Patrons who have been tested for COVID-19 antibodies and are shown to have them will have that information embedded in a QR code, to be scanned by hotel staff upon arrival. The Miami launch is a test, a proof of concept, and, if it works, Hultin is hopeful that it will help revive the travel industry. (Officials from Delta Air Lines and Heathrow Airport have expressed interest in immunity passports.) The idea, Hultin said, is to let the hotel staff know “the guest is safe.”

In April, Onfido raised a hundred million dollars in venture capital. In a press release announcing the deal, Onfido’s attorneys noted that the pandemic presented the company with the opportunity “to work on a new set of use cases,” which range from “virtual voting and passport/visa applications through to secure ways of carrying out contact tracing to track the spread of a virus without compromising user privacy.” Recently, Onfido submitted a proposal to members of the U.K. Parliament’s science and technology committee, as they consider ways to lift stay-at-home orders, including the development of immunity passports. The M.P.s are following the lead of the Chilean government, which is issuing documents that allow anyone who has recovered from COVID-19 to go back to work and move around freely during quarantine. (Onfido has not worked on that program.) After pushback from members of the Chilean medical community—and from infectious-disease specialists around the world, who say that it’s too early to know if people who have had the virus are immune to reinfection or incapable of spreading the disease—the government started calling the documents “release certificates” rather than immunity passports. Either way, they confer the same benefits. Onfido’s C.E.O., Husayn Kassai, told me, “As a society, is it acceptable for people who have had the virus and recovered to signal that or not? It’s a good question, but not one for us to answer.”

Other companies are pitching similar products. COVI-Pass is an immunity passport that uses VCode—a proprietary cryptographic image, similar to a QR code—that can be scanned at a distance of a hundred metres. It will reveal if a person has tested positive for antibodies, flash green if a person has tested negative for the virus, and red if tests show that they have the virus or don’t have antibodies, or if their test result has expired. (A yellow light shows when it’s time to be retested.) The app can “geo-fence” a point of entry, prohibiting visitors with a red light from entering a building or stadium or school. “I believe that all governments will move toward a global health passport, which will be as common as carrying a driver’s license or a passport, because this is not going to be the last pandemic we have,” Adam Palmer, the C.O.O. of COVI-Pass, said. A recent press release from the company, which is also based in the U.K., states that the passes were in the process of being shipped “to both the private sector and governments in over fifteen countries,” including Italy, France, India, and the United States. COVI-Pass would not say, specifically, which companies and governments in these countries had purchased the product. But Louis-James Davis, the C.E.O. of VST Enterprises and the creator of VCode, did mention that the company’s V-Health Passport, which also documents users’ COVID-19 status, would be rolled out at sports venues in the U.K. this summer. The passport will have to be scanned in order to buy a ticket online, and then again at the stadium, before approaching the gate.

The deployment of immunity passports raises a number of logistical and ethical concerns. In the United States, testing could become a barrier to getting a pass: diagnostic testing has been limited, and serology tests, which measure COVID-19 antibodies, are still being developed. Even if testing were universally available, people without antibodies will be shut out from the activities of everyday life, creating a divided landscape of antibody haves and have-nots. “The reality is that access to society with [this kind of] certification will likely reflect existing systemic biases, corruption, and discrimination in the system,” Alexandra Phelan, an assistant professor in infectious diseases and law at Georgetown University, told me. And, if testing positive for COVID-19 antibodies becomes a prerequisite for being able to work, there may be a perverse incentive for people to put themselves in the path of the virus. “This would not be like bringing your kids to chicken-pox parties,” Phelan said. “Individuals who have economic pressures to get back to work may already be in socioeconomically vulnerable positions, and we know that these are the people who are disproportionately more likely to get more severely ill.”

Rob Davidson, an emergency-room physician in western Michigan and the executive director of the Committee to Protect Medicare, an advocacy group, worries that immunity passports based on tests with high rates of false positives will be especially pernicious. “If we have tests with a five-per-cent false-positive rate, and you are forced to take a test that is not very good, and you test positive, you will have a false sense of security,” he told me. “The government and the employers probably know the shortcomings of the test, but a lot of the workers probably don’t.” In those cases, immunity passports may increase the risk of infection to both their holders and to the community at large. (Both COVI-Pass and Sidehide say that they are partnering with local, well-regarded testing facilities.)

Privacy, too, may be at risk, even with immunity passports that do nothing more than display a user’s test results. “We are in a less-than-ideal situation with the coronavirus,” Onfido’s Kassai told me. “This is one of the ways people can signal they are safe without divulging personal information.” But immunity status is personal information. If a person’s immunity status were linked to an insurance policy, or attached to an employee’s human-resources file, it could cause rates to rise, or promotions to be denied. Already, officials at the U.S. Defense Department have proposed banning anyone who has had COVID-19 from enlisting in the military. And a recent story in the Times inventories a long list of people who have been shunned as a consequence of others knowing their disease status: “The veterinarian who refused to treat a recovered woman’s dog. The laundromat worker who jumped at seeing an elected official whose illness had been reported on the local news. The gardener who would not trim the hedges outside a recovered man’s home. The neighbor who dropped off soup, and said not to bother returning the Tupperware it was in. And the sick teenager whose solace during his long illness was the thought of fishing with friends, only to have them ghost him when he recovered.”

There remains another obstacle to using COVID-19 antibodies as a delimiting metric: science. The biology of this novel coronavirus is so new that no one knows whether COVID-19 antibodies actually confer immunity—and, if they do, for how long. In April, the World Health Organization issued a scientific brief that concluded that “there is not enough evidence about the effectiveness of antibody-mediated immunity to guarantee the accuracy of an ‘immunity passport’ or ‘risk-free certificate.’ ” Even the most precise serological test, from the Swiss pharmaceutical company Roche, which is reported to be a hundred-per-cent accurate at detecting the coronavirus antibodies, can only determine the presence of COVID-19 antibodies—it can’t say what they mean. “We hope antibodies will be protective, but we don't know for sure yet,” Tom Frieden, who was the director of the Centers for Disease Control and Prevention during the Obama Administration, told me. “It now appears likely that at least some infections will provide at least some degree of immunity. Which infections, how much immunity, and for how long are crucially important but still unanswered questions.”

Frieden added that, if these questions could be resolved—and if a vaccine were still absent—he could imagine a role for immunity passports in certain circumstances. “For example, you might want to make sure that only people who—if there is documented immunity—work in nursing homes to the greatest extent possible with non-immune residents, because that’s a population of enormous concern if immunity is confirmed,” he said. “I think it’s too soon to say that these will be useful, but it’s not too soon to begin thinking about how they might be used if antibodies are indeed protective.”

A group of scientists, working under the auspices of DARPA, the research arm of the Defense Department, is pursuing a different path. The scientists are developing a method of testing for certain host signatures of gene expression caused by the virus—what is known as epigenetics—to detect the presence of COVID-19 well before there are symptoms. Eric Van Gieson, a program manager at the Biological Technologies Office, who is overseeing this effort, told me that the goal is to create a test that is as easy to use and as prevalent as a strep test or a test for flu that you might take at the doctor’s office. “We want to identify as many people as early as possible,” Van Gieson said. “We also want to make sure that we’re not depending on viral shedding for diagnosis.” This is crucial, because, once someone is shedding the virus, they will be spreading it unless they are isolated. This test will be especially useful once there are effective treatments, but even in their absence it has the capacity to slow the spread of the virus by preventing asymptomatic people from becoming vectors of disease. And, because epigenetics also exploits the body’s innate immune memory to reveal when a person has been exposed to a pathogen, this test could prevent cluster outbreaks when paired with robust contact tracing. If COVID-19 antibodies do not turn out to be protective, or if the protection they offer is limited, or if societies reject immunity-passport schemes, frequent epigenetic tests may be a way to bypass large-scale lockdowns.

Van Gieson would not put a time line on the production of an epigenetic test, because there is still research to be reviewed and regulatory procedures to negotiate. He did say that scientists involved in the effort, some of whom have been sick with COVID-19, are working nonstop. In the meantime, governments and businesses, under increasing financial and social pressures to end lockdowns, may find it politically and economically troublesome to wait for the science. But, Van Gieson told me, “We know there’s a cure out there and we know there’s a diagnostic tool out there. We know there’s a vaccine out there. So let the science do the work, right? I mean, at the end of the day, the truth is going to come down to the science of it, not what people say or how people react—the panic, all that stuff. You just have to do thorough, rigorous scientific work and only believe your data. Don’t believe anything else.”


A Guide to the Coronavirus