Since the onset of the pandemic, scientists have been searching for existing, available medications that might be repurposed to treat COVID-19.
Laboratory studies and observations in patients suggested that medications used to treat depression—especially so-called serotonergic antidepressants such as fluoxetine (Prozac) and fluvoxamine (Luvox)—might have antiviral properties and lessen the severity of disease.
But a recent large study conducted by investigators at McLean Hospital and published in Frontiers in Psychiatry counters these findings.
- A large analysis of patient data counters earlier research suggesting that medications used to treat depression may lessen the severity of COVID-19
- The findings highlight the importance of rigorous studies to test treatments that may initially appear promising
Using a patient data registry from the Mass General Brigham health care system, the McLean researchers reviewed de-identified records of 500 hospitalized patients with COVID-19 who were receiving antidepressants at the time of hospital admission, and compared them against two other patient groups, each comprising more than 500 hospitalized individuals with COVID-19 who were not receiving antidepressants.
The data analysis adjusted for patients’ demographic features (such as age, gender, and race/ethnicity), co-occurring medical conditions, and a measure estimating overall level of health.
The McLean analysis revealed that antidepressants had no effect on preventing the need for a breathing tube or on preventing death, regardless of whether the investigators assessed patients receiving all types of antidepressants or subgroups of patients receiving specific types of antidepressants.
To further examine why their study generated conflicting results from some of the previous research, the McLean investigators examined three other recent studies of hospitalized patients with COVID-19, one that reported a positive effect of antidepressants and two that reported weak or negative results.
Although no single factor explained the differences in the findings, all of the studies were vulnerable to various limitations.
Of note, in April 2022, after reviewing all available data, the U.S. Food and Drug Administration rejected an application for an Emergency Use Authorization to use the antidepressant fluvoxamine to treat COVID-19.
The agency noted that two unpublished additional trials of fluvoxamine for COVID-19 had failed to show any benefit, and both had been terminated prematurely for futility.
“Several of the antidepressant studies used the same dataset to evaluate many different medications. However, some of these studies failed to use sufficiently rigorous statistical procedures for this kind of analysis, and hence made errors in the direction of producing overly optimistic findings regarding the effectiveness of certain individual medications or types of medications,” said James I. Hudson, MD, ScD, SM.
“It is important to be cautious when evaluating such findings, especially in the early stages of drug evaluation.”
Hudson is director of the Biological Psychiatry Laboratory at McLean Hospital and the senior author of the study who supervised the statistical analyses.
Brian P. Brennan, MD, MMSc, the lead author of the study, offered a similar assessment.
“When searching for treatments for a new and threatening disease, it is common and understandable that investigators will become enthusiastic about a promising potential treatment, and that enthusiasm often spreads to the general public. But as more rigorous and methodologically sound studies are performed, it often proves that the initial enthusiasm was not justified,” Brennan said.
“For example, drugs such as hydroxychloroquine and ivermectin were initially hailed as promising breakthroughs for the treatment of COVID-19—but ultimately, high-quality studies with rigorous methodology failed to find these drugs effective. Unfortunately, our recent study, together with emerging studies from other research groups, suggests that antidepressants are likely following a similar trajectory from enthusiasm to disappointment.”
The research offers an example of how data from the Mass General Brigham health care system can be leveraged to uncover important information about health and disease, and to ultimately impact patient care.
“Use of health care data is allowed in a secure, protected environment through the Mass General Brigham-supported Research Patient Data Registry, which is governed by the Institutional Review Board. It provides a powerful resource for clinical scientists to ask questions about the care of our patients and factors that influence their outcomes,” said Shawn N. Murphy, MD, PhD, chief research information officer at Mass General Brigham.
Journalist or member of the media? We are available 24/7 for media requests.