US Prescriptions During COVID-19: Hydroxychloroquine Soars, CVD Drugs Stable

Physicians latched quickly onto hopes that antimalarial drugs could help, but weren’t swayed by ACE inhibitor/ARB theories.

US Prescriptions During COVID-19: Hydroxychloroquine Soars, CVD Drugs Stable

As the COVID-19 pandemic was first hitting the United States, pharmacy-filled prescriptions for hydroxychloroquine and chloroquine shot up, but those for ACE inhibitors and ARBs remained stable, an analysis of US prescribing patterns shows.

On the one hand, the numbers hint at how quickly both physicians and the public latched onto hopes that the antimalarial drugs might prove helpful in this disease despite insufficient evidence, the authors say. On the other, concerns that the most commonly prescribed cardiovascular medications might increase the risk of SARS-CoV-2 infection or cause more serious illness did not appear to sway physician behavior.

More of TCTMD's coverage on our COVID-19 hub.
More of TCTMD's coverage on our COVID-19 hub.

The ACE inhibitor/ARB data are reassuring, senior author Haider Warraich, MD (VA Boston Healthcare System, MA), told TCTMD. “Those initial concerns maybe led to more papers being published and conversations on social media rather than actually impacting the care patients were receiving, which I think is a good thing,” he said. “It does show that physicians for the most part did not alter their behavior based on hearing some initial rumors about the potential harm of these medications in people with COVID-19.”

The hydroxychloroquine/chloroquine numbers, by contrast, are more worrisome, he said, pointing out that an emergency use authorization by the FDA was restricted to the hospital setting, yet the spike in prescriptions was in the outpatient setting, where close monitoring for heart rhythm disturbances and other adverse reactions would have been minimal.

“If people were in fact taking these medications, this is without any kind of monitoring so it's potentially even more unsafe than the use of these in the inpatient setting, which has in fact been studied and has in some reports been shown to be harmful. The impact cannot be overstated that this is a concerning finding in general,”  Warraich said. “To date, we have no data whether prophylactic use of this medication does anything, really, or at least anything beneficial.”

Prescription Patterns, Then and Now

For the analysis, led by Muthiah Vaduganathan, MD, MPH (Brigham and Women’s Hospital, Boston), and published online May 28, 2020, as a research letter in JAMA, investigators reviewed weekly prescriptions for the 10 most commonly prescribed drugs of 2019, as well as hydroxychloroquine/chloroquine and azithromycin, between February 16 and April 25, 2020. These were compared with weekly prescribing patterns during the same weeks in 2019. Data came from deidentified prescription claims of more than 58,000 US pharmacies representing 17 million claims from 50 states, which Warraich characterized as “pretty representative” of prescribing patterns across the US.

Filled prescriptions for almost all medications increased in the week of March 15 to 21, around the time US states were initiating lockdowns. To Warraich, that suggests that many people were stocking up on medications for chronic conditions the same way they were toilet paper, given uncertainties over how long and how restrictive the stay-at-home orders would become. “To me, that's an encouraging sign, that most physicians and their patients were really going into this with their eyes open and were prepared,” he said.

Prescriptions that week for amlodipine, for example, increased by 32% over the same week in 2019, atorvastatin by 31%, lisinopril by 23%, and losartan by 32%. Prescriptions for antidepressants and drugs for ulcers, hypothyroidism, and pain were also up. For cardiometabolic drugs over the entire 10-week period, prescriptions were stable or declined only slightly compared with 2019 estimates, the authors say.

Of note, the paper documents a decrease in antibiotic medications during the study period, which Warraich suggested likely reflects the “precipitous drop” in doctor’s office visits for things like upper respiratory tract infections. Prescriptions for azithromycin increased by only 8.7% that week and dropped thereafter.

Hydroxychloroquine and Chloroquine Scripts Soar

The biggest spike that week, however, was for filled hydroxychloroquine/chloroquine prescriptions, which, for fewer than 28 tablet fills, soared by nearly 2,000% compared with 2019 (2,208 prescriptions in 2019 versus 45,858 in a single week). That sharp increase coincides with “the World Health Organization declaring a global pandemic on March 11, the United States declaring a national emergency on March 13, a single-group nonrandomized study published on March 17, and President Trump’s support of the drug on March 19,” the authors write.

Prescription fills for larger numbers of tablets during this week were increased but not as radically, and by the end of the study period, prescriptions for these agents were still up, but to a far lesser degree. For the whole 10-week period there were 483,425 “excess fills” of hydroxychloroquine/chloroquine in 2020 compared with 2019.

Early concerns were raised that physicians—who intuitively would be more apt than the general public to understand the paucity of evidence for these antimalarials for COVID-19—were helping to drive the use of these medications through off-label self-prescribing, at the same time that the public was requesting the pills.

Both of these likely contributed to the spike, Warraich hypothesized to TCTMD. He pointed to an earlier paper that tracked internet searches for drugs in the first months of the pandemic. The trends Warraich and colleagues captured in terms of filled prescriptions track with that earlier paper, he noted.

“I certainly think there were physicians who were afraid of catching this infection for themselves or for their family members and were prescribing it for either themselves or their loved ones, and then there was a lot of interest from patients and the general public about these medications,” he said. “A lot of people have warned about misinformation in the wake of this pandemic and I think this paper really quantifies just how widespread the impact of misinformation can be. . . . I've never written an outpatient prescription for hydroxychloroquine and there was never any reason to do so, and yet many [physicians] kind of crossed that bridge in the time of this pandemic. So I really hope we can look back at this time and ask: what could we have done better as a medical community?”

I really hope we can look back at this time and ask: what could we have done better as a medical community? Haider Warraich

So far there are no data proving unequivocally that hydroxychloroquine/chloroquine can help with the treatment or prevention of COVID-19, he stressed. “I think many people were anxious to want to do something in that data-free zone, but I really wish that more thought had been put into these decisions as both patients and physicians were going through this.”

The need for reliable data to support medical therapies—and widespread trust and acceptance of those data—is only going to become more critical in the coming months, Wairrach predicted.

“Right now there is a clock ticking to develop a vaccine, but there is another clock ticking that requires us to overcome the suspicions people have of medicine. If we develop a vaccine that is very effective, but a large chunk of the population doesn't take it, then what good is that vaccine? As much as we need to be thinking about how to develop the best possible therapeutics, the best possible vaccines, [and] the best data and evidence to limit the impact of this pandemic, I think there's another, parallel path, and that is, how do we restore people's trust in science, which I think you could argue is potentially even harder than anything we do in the lab.”

Sources
Disclosures
  • Vaduganathan reports being supported by the KL2/Catalyst Medical Research Investigator Training award.
  • Warraich reports no relevant conflicts of interest.

We Recommend

Comments