The federal government's guidance on emergency usage of the antimalarial drug hydroxychloroquine for COVID-19 patients may have actually set the medication up for failure.
That's because the guidance limits the drug's usage to those sick enough to be hospitalized. Many doctors suspect, however, that if the drug does turn out to be beneficial, it may work better early on in the course of the illness.
"That's the case for any therapeutic," Dr. George Diaz, an infectious disease specialist at Providence Regional Medical Center in Everett, Washington, said. "Generally the earlier you use it, the more effective it is."
Antiviral treatments work by reducing the amount of virus in the body. They're best given early on, as symptoms are starting and the viral load is still low. For example, Tamiflu, an antiviral used to treat the flu, works better the earlier a person takes it.
"When you give it to somebody who is already super sick, it's likely not going to make an impact because the damage is already done," said Dr. Ken Lyn-Kew, a pulmonologist in the critical care department at National Jewish Health, a hospital in Denver.
At least anecdotally, a handful of critical care physicians across the country have told NBC News they've seen no evidence the drug is helping their sickest patients in intensive care.
RELATED: Take a look that the first trials of the coronavirus vaccine in Seattle:
"We've been using it," said Dr. Hugh Cassiere, a pulmonologist and medical director of Respiratory Care Services at North Shore University Hospital in Long Island, New York, a hot spot for the pandemic in the United States. "But we really haven't seen any efficacy."
"I'm not convinced it works," said Lyn-Kew, adding, "I've not seen anybody have anything near what I would call a miraculous recovery because of hydroxychloroquine."
In Louisiana, Dr. Josh Denson, a pulmonary medicine and critical care physician at the Tulane University Medical Center in New Orleans, said that "many patients do fine and tolerate it, but I don't think it's making a difference."
I've not seen anybody have anything near what I would call a miraculous recovery because of hydroxychloroquine.
"We're using it; it's all we have," Denson added. Louisiana has also been hit hard by the coronavirus, with at least 16,284 cases and 582 deaths.
"I can't honestly say that I think that it's necessarily an effective agent at this point," said Diaz, of the Providence Regional Medical Center in Washington state. Diaz was involved in the treatment of the first patient in the U.S. diagnosed with the coronavirus.
New use for decades-old drug?
Hydroxychloroquine, and a related compound called chloroquine, is a medication that's been around for decades. It's used to treat malaria, as well as certain autoimmune diseases including lupus rheumatoid arthritis.
President Donald Trump and other political and corporate leaders have touted use of the medication after a few small studies suggested it might be effective against the coronavirus.
Last month, the Food and Drug Administration issued an emergency use authorization, allowing health care providers to use the medicine for illness, even though the drug has not been approved as a specific treatment for COVID-19.
It's unclear how many hospitals across the country have tried hydroxychloroquine. While it appears many are, some are not.
"UC San Diego Health does not support the use of medications in ways for which there is not yet scientific evidence to support safety and efficacy," the hospital system wrote in an email to NBC News, adding investigators there preparing to enter clinical trials to do their own studies of the drug.
Critical care physicians at Northwestern Medicine in Chicago aren't using it either. "It's the view of our ICU group here that the data supporting its use are insufficient for routine use for COVID-19 patients, particularly when they become critically ill," Dr. Ben Singer, assistant professor of medicine in pulmonary and critical care at Northwestern University's Feinberg School of Medicine, said.
This week, the Centers for Disease Control and Prevention pulled back on its guidance for hydroxychloroquine usage for COVID-19 on the agency's website, no longer offering recommendations for dosage. The CDC has also deleted information on those early studies of the drug.
The science so far
One of the first studies to look at hydroxychloroquine as a potential treatment for COVID-19 came out of China in early February. That study looked at how the drug behaved on a cellular level, but did not examine how it would work in patients.
But, "those reports were talking about patients who received a particular drug and then got better. We know most people with COVID-19 get better on their own," said Dr. Wesley Self, an emergency physician at Vanderbilt University Medical Center in Nashville, Tennessee, who is studying the drug.
We need real science behind this disease.
"If you took those same people, and didn't give them medicine, many of them would report, we think, similar rates of recovery," Self said.
In other words, it's possible those patients would have improved clinically whether they got hydroxychloroquine or nothing at all.
What's more, those studies were published online before undergoing what's called peer review, meaning other scientists hadn't had the chance to analyze the data and point out potential flaws. Peer review is considered essential to quality research.
"People take these tiny studies, and quote them as gospel," Lyn-Kew said. "We need real science behind this disease," adding physicians are "desperate for something to help people."
The science yet to come
Physicians across the country are investigating, in real time, to see whether the drug might indeed be beneficial to treat severe cases or even prevent coronavirus infections.
Researchers at NYU Langone in New York City and the University of Washington Medical Center in Seattle are recruiting 2,000 people who have been in contact with a COVID patient, but haven't developed symptoms themselves.
Participants will receive either hydroxychloroquine or vitamin C every day for two weeks, and undergo regular testing for COVID-19.
The study "could provide answers by summer on whether a preventive dose of the drug is safe and effective," according to a press release posted online from NYU Langone. "If so, the strategy could give health officials a much needed boost in slowing person-to-person transmission."
Download the NBC News app for full coverage of the coronavirus outbreak
New York Gov. Andrew Cuomo said Monday physicians in that state are moving forward with testing hydroxychloroquine for COVID-19 patients because "there has been anecdotal evidence that it is promising."
But until there's definitive, robust research on the matter, physicians are left with dueling anecdotes about hydroxychloroquine's effectiveness. Some anecdotes suggest a benefit; others don't.
Self, from Vanderbilt, is leading a randomized clinical trial investigating the drug.
Patients will be randomly selected to receive either hydroxychloroquine or a placebo. Neither the patients, the physicians, nor people who will assess the results at the end of the project will know which patients got the actual medication, and which patients got a dummy pill.
Ten patients have been enrolled in the past four days. But "it's just too early to know if the drug is working or not," Self said.
Potential for harm — and hope
All medications come with potential side effects. Hydroxychloroquine is no different.
The drug can lead to an irregular heartbeat, which can be deadly in some patients. Very sick people in intensive care units may be at particular risk, Self said, because they tend to be more susceptible to drug side effects in general.
"We have to understand whether the side effects of this drug are worse than any benefits," Self said.
Because it's thought the drug might work best early on, the Vanderbilt study is only enrolling patients within 10 days of developing coronavirus symptoms, such as cough, fever, extreme fatigue or shortness of breath.
Hydroxychloroquine "clearly has promise," Self said. "But we really need the clinical trial data to understand what this is doing in our patients."
"We have a duty to know that before we start prescribing to thousands and potentially millions of people."