Dexamethasone, a potent steroid medication, is in the news as a possible treatment for people with Covid-19.
A June 2020 press release for the UK-based RECOVERY clinical trial announced that dexamethasone reduced the risk of death in hospitalized Covid-19 patients who needed mechanical ventilation or additional oxygen. The following month, results of the trial were published in the New England Journal of Medicine.
Public attention increased even more in early October when President Donald Trump was given dexamethasone after he was admitted to Walter Reed Medical Center for Covid-19. He was given the steroid presumably because he had a severe case of the disease.
The Food and Drug Administration has not approved dexamethasone to treat patients with Covid-19 in the hospital or at home. In addition, Covid-19 treatment guidelines from the National Institutes for Health specifically recommend against using dexamethasone use in patients who do not need oxygen, as the RECOVERY trial did not show it benefitted these patients and potentially showed evidence of harm.
Despite that, some clinicians might prescribe dexamethasone “off-label” to non-hospitalized Covid-19 patients in the same way that clinicians have prescribed other unproven Covid-19 treatments, such as hydroxychloroquine and azithromycin. To explore this possibility, we examined national data on prescriptions for oral dexamethasone that were dispensed from pharmacies across the U.S. between October 2017 and September 2020. These data, compiled by IQVIA, are considered the gold-standard of prescription dispensing data in the U.S.
Before April 2020, the number of dispensed prescriptions for oral dexamethasone averaged just under 260,000 per month. That number dropped to approximately 180,000 during April 2020 — likely because fewer people sought medical care due to fears of contracting Covid-19 — and rose to 237,000 during June 2020.
In July, the number of prescriptions shot up to 335,000, then tapered off a bit during August and September. During these three months, the total number of dispensed dexamethasone prescriptions was almost 175,000 higher (22.7%) compared with the number of prescriptions dispensed during the same three-month period in 2019.
The prescribers accounting for the greatest increases in dispensed dexamethasone prescriptions between these two time periods were internal medicine physicians (up 51,900 prescriptions, or 46.8%), nurse practitioners (up 38,300 prescriptions, or 28.3%), and family medicine physicians (up 35,000 prescriptions, or 39.8%).
The increase in prescriptions between the two time periods occurred only in adults aged 20 years and older, for whom dispensed dexamethasone prescriptions increased 33.0%. Among those aged 0 to 19 years, prescriptions decreased by 37.3%.
To see if the rise in dexamethasone prescription dispensing during July through September 2020 was driven by greater demand for oral steroids in general, we also examined dispensed prescriptions for other oral steroids such as prednisone, which is often prescribed for conditions like asthma attacks or poison ivy. As with dexamethasone, the monthly number of dispensed prescriptions for other steroids fell sharply between March and April 2020, from 6.2 million to 4.0 million, again likely because fewer people sought medical care. But unlike oral dexamethasone, the number of dispensed prescriptions for other steroids rose to just 4.5-4.6 million during July through September, a level far below the pre-pandemic baseline.
The rise in dispensed dexamethasone prescriptions beginning in the summer of 2020 coincided with the publication of results of the RECOVERY trial, occurred only in adults, was driven by prescribers who usually work in clinics and other non-hospital settings, and did not reflect increased demand for oral steroids generally.
Collectively, these findings suggest that clinicians are increasingly prescribing dexamethasone off-label to non-hospitalized patients with Covid-19.
It is difficult to predict if increased prescribing of dexamethasone will continue, although we suspect that its administration to President Trump during his hospitalization raised public awareness of the drug’s potential as a Covid-19 treatment. This could lead patients with Covid-19 to request this medication, increasing pressure for their physicians to oblige.
We raise this issue because there is no evidence that dexamethasone benefits Covid-19 patients who aren’t severely ill and because it has known side effects, such as depression, mood swings, and psychosis. Until there is evidence of effectiveness and safety, we believe it is premature for clinicians to prescribe dexamethasone for Covid-19 patients who are not severely ill.
We also raise this issue to highlight our broader concerns about off-label prescribing of unproven treatments for Covid-19. In March 2020, there was a large increase in prescriptions for hydroxychloroquine to treat Covid-19. This occurred despite the lack of evidence that this medication was effective against the disease and despite knowledge that hydroxychloroquine increases risk of dangerous irregular heart rhythms in other patient populations. Subsequent clinical trials demonstrated that hydroxychloroquine does not benefit patients with Covid-19.
Off-label use of prescription drugs can also be costly to insurers and patients. Both dexamethasone and hydroxychloroquine are inexpensive generic drugs, so widespread off-label use would not entail high costs. However, many potential Covid-19 drugs that might be prescribed off-label are high-priced, branded drugs.
The future will bring many potentially promising but unproven drugs for Covid-19. We understand why clinicians may be tempted to prescribe these drugs off-label. It is difficult not to intervene when facing a disease as dangerous as Covid-19. Yet this difficulty does not justify prescribing unproven drugs that could easily do more harm than good.
Kao-Ping Chua is a primary care pediatrician and assistant professor in the department of pediatrics of the University of Michigan Medical School and the department of health management and policy at the University of Michigan School of Public Health. Adam S. Cifu is a general internist and professor of medicine at the University of Chicago. Rena M. Conti is an associate professor in the department of markets, public policy, and law at Boston University Questrom School of Business and the associate research director of biopharma and public policy for the Boston University Institute for Health System Innovation and Policy. Access to the prescription data was provided through the IQVIA Institute’s Human Data Science Research Collaborative, the purpose of which is to promote research into the effects of the Covid-19 pandemic on the U.S. health care system. IQVIA was not involved in the analysis in any way.