WASHINGTON (NYTIMES) - A law firm in Scottsdale, Arizona, tested employees who hoped, with the prick of a finger, to learn if they might be immune.

In Laredo, Texas, community leaders secured 20,000 of the new tests to gauge how many residents had been infected. In Chicago, a hospital screened firefighters to help determine whether they could safely stay on the job.

In recent weeks, the United States has seen the first rollout of blood tests for coronavirus antibodies, widely heralded as crucial tools to assess the reach of the pandemic in the United States, restart the economy and reintegrate society.

But for all their promise, the tests - intended to signal whether people may have built immunity to the virus - are already raising alarms.

Officials fear the effort may prove as problematic as the earlier launch of diagnostic tests that failed to monitor which Americans, and how many, had been infected or developed the disease the virus causes.

Criticised for a tragically slow and rigid oversight of those tests months ago, the federal government is now faulted by public health officials and scientists for greenlighting the antibody tests too quickly and without adequate scrutiny.

The Food and Drug Administration has allowed about 90 companies, many based in China, to sell tests that have not gotten government vetting, saying the pandemic warrants an urgent response. But the agency has since warned that some of those businesses are making false claims about their products; health officials, like their counterparts overseas, have found others deeply flawed.

Tests of "frankly dubious quality" have flooded the US market, said Scott Becker, executive director of the Association of Public Health Laboratories. Many of them, akin to home pregnancy tests, are easy to take and promise rapid results.

And the federal guidance that does exist is so confusing that health care providers are administering certain tests unaware that they may not be authorised to do so. Some are misusing antibody test results to diagnose the disease, not realising that they can miss the early stages of infection.

"People don't understand how dangerous this test is," said Michael Osterholm, an infectious disease expert at the University of Minnesota.

"We sacrificed quality for speed, and in the end, when it's people's lives that are hanging in the balance, safety has to take precedence over speed."

Even as government agencies, companies and academic researchers scramble to validate existing tests and create better ones, there are doubts they can deliver as promised.

Most tests now available mistakenly flag at least some people as having antibodies when they do not, which could foster a dangerously false belief that those people have immunity.

And even if the tests do improve, their availability could be hampered by the same manufacturing shortages that have prevented the Covid-19 diagnostic tests from scaling up adequately.

As President Donald Trump presses to reopen the country and several states are considering lifting lockdowns in the next few weeks, widespread screening is considered critical.

On Friday (April 17), Trump cheered the FDA's emergency approval of some antibody tests, saying they would support efforts to get Americans back to work "by showing us who might have developed the wonderful, beautiful immunity".

Epidemiologists are testing for antibodies in hot spots to better measure the extent of the outbreaks, and government officials intend to use those results to help decide when and how to return residents to daily life. But many scientists and political leaders say the country is nowhere close to deploying enough diagnostic and antibody tests at the speed and volume required.

"The more testing, the more open the economy," Governor Andrew Cuomo of New York said on Wednesday. He has pushed for the production of antibody tests as a central cog in plans to ease stay-at-home restrictions, saying that New York would eventually screen 100,000 people a day.

Recent testing around the country demonstrates the challenges of using the new products. At the Chicago hospital, for example, the city's Public Health Department intervened, warning that it should not use antibody tests to determine whether emergency workers were actively infected.

Soon after it helped screen the Rose Law Group, the firm in Arizona, a lab stopped providing rapid tests to other clients, fearing they might not comply with federal guidelines, and switched to more sophisticated lab-based tests.

In Laredo, officials discovered the tests they received were woefully inadequate. The local health department found them to have a reliability of about 20 per cent, far from the 93% to 97 per cent the company had claimed. A police investigation led to a federal seizure of the tests.

FREE-FOR-ALL

More than 90 companies have jumped into the market since the FDA eased its rules and allowed antibody tests to be sold without formal federal review or approval.

Some of those companies are startups; others have established records. In a federal guidance document on March 16, the FDA required them to validate their results on their own and notify the agency that they had done so.

Their products vary. Some test only for a transient antibody that spikes while the body is in the throes of an active infection.

An antibody that peaks about four weeks after infection and marks longer-term immunity is a separate target. There are tests that look for both antibodies; others also look for a third involved in respiratory infections.

The most reliable ones involve a laboratory technique called Elisa that can indicate the amount of antibodies a person may have. Higher levels generally mean a stronger physiological response, but it is unclear what levels are needed for immunity to the new coronavirus - or how long it would last.

"We're really far from that," Osterholm said. "We're not even in the second inning of a nine-inning game at this point." Most of the tests offered are rapid tests that can be assessed in a doctor's office - or, eventually, even at home - and provide simple yes-or-no results.

Makers of the tests have aggressively marketed them to businesses and doctors, and thousands of Americans have already taken them, costing a patient roughly US$60 (S$85) to US$115.

Rapid tests are by far the easiest to administer. But they are also the most unreliable - so much so that the World Health Organisation recommends against their use.

This month, the FDA warned that some firms marketing their antibody tests in the United States were falsely claiming that they had formal federal approval or that they could diagnose Covid-19.

In an effort to speed up access, the FDA apparently did not fully consider how these tests would be administered. The agency released a guidance document saying that antibody tests could be performed at "point-of-care" settings, indicating that doctors, nurses and others could give them to patients in their offices. But agency officials also acknowledged that under federal law, if a test has not been authorized by the agency, it must be conducted in so-called high-complexity laboratories, like some large commercial facilities or public health labs. The officials declined to provide additional clarification.

"If you are getting an antibody test, and it's being conducted in your physician's office, it's a red flag," said Kelly Wroblewski, director of the infectious disease programmes for the Association of Public Health Laboratories.

No action has been taken against doctors, but as companies realised the ambiguity of the federal guidelines, some changed course, shifting to lab-based tests or pursuing formal federal approval, which would allow their products to be used at doctors' offices and even at home.

The FDA has received requests for emergency-use authorisation from 120 antibody-testing developers. So far it has granted formal approval to just four: Cellex, Ortho Clinical Diagnostics, Chembio Diagnostic Systems and the Mount Sinai Laboratory.

IN SEARCH OF A TEST THAT WORKS

While political leaders and some health officials say that antibody testing will be essential to reopening the country, it is unlikely to meet expectations anytime soon.

Less than 5 per cent of the US population may be infected, and even in hot zones like New York or New Orleans, the prevalence may not be higher than 10 to 15 per cent, according to Osterholm. In China, early screening in hard-hit Wuhan indicates that only about 3 per cent of the population has antibodies against the new coronavirus.

When the proportion of people exposed is that low, the tests' false-positive rate - signalling antibodies where there are none - can limit the tests' utility.

Even Cellex's FDA-authorized test has a false-positive rate of about 5 per cent. That is still a significant margin of error: In a community where 5% of people have had the virus, Osterholm said, there would be as many false positives as true ones.

"What are you going to do with that? Are you going to say you're not going to distance?" he said. "I don't think that would give me peace of mind at all."

Some tests on the market without federal approval are likely to have even higher false-positive rates. It is unclear how these companies have validated their tests or how they stack up against one another.

Researchers at the National Institutes of Health are validating some of the tests in cooperation with the FDA. Last week, a German group published a small analysis of nine tests and found, as might be expected, that some performed better than others. A research group backed by the Chan Zuckerberg Initiative is working to assess all available tests using the same set of samples for each, with early results expected next week.

"There's an urgent need to know which of these tests we can rely on; I think the only way we can know is head-to-head testing," said Dr Alexander Marson, one of the leaders of the project and a microbiologist at the University of California, San Francisco.

"The ideal outcome is not to pick a winner but to pick as many winners as possible so the world can have a diversified list of tests," he added.