U.S. COVID trials under-enrolled women, Black and Asian people, Fred Hutch study says

Several groups, including women and Black and Asian communities, were underrepresented in many U.S. COVID-19 clinical trials, according to a new study from the Fred Hutchinson Cancer Center.

The study was published in JAMA Internal Medicine Monday and is, to the researchers’ knowledge, the first of its kind in the country to examine the demographics of COVID vaccine and treatment trials early in the pandemic. According to the paper, researchers analyzed 122 trials, involving about 175,000 people, that were officially registered or published between October 2019 and February 2022.

“The issue of equitable access to clinical research has become a very topical issue that is on the front of people’s minds,” said Joe Unger, a Fred Hutch biostatistician and health services researcher who acted as the study’s senior author. “Clinical research is key for developing new treatments for all people, so it’s really critical, not just in terms of access, but for the research that’s associated with it. The way you interpret trial results really hinges, to a large degree, on how representative those trial cohorts are.”

According to the study, women made up about 49% of COVID prevention trial participants and about 45% of treatment trial participants, although they represented about 52% of Americans with COVID.

“It wasn’t so much that females were underrepresented, perhaps, but rather that males might have been overrepresented,” Unger said. “Men, especially older men, were disproportionately impacted early on in the pandemic, and that may have contributed to it.”

Women are also historically underrepresented in clinical trials, partially because people who are pregnant are routinely excluded from trials and women of “reproductive age” generally have more concerns about the safety of treatments for themselves and their potential future children, the study says.

In addition, Black participants made up about 7% of vaccine trial participants (when Black communities represent about 14% of the U.S. population), though they were well-represented in treatment trials. Asian participants were underrepresented in prevention trials compared with the general U.S. population, but not compared with those with COVID in the U.S.

Researchers found Hispanic/Latino participants were overrepresented in COVID trials, making up about 34% of trial participants despite representing about 18% of Americans with COVID.

Part of the reason could be that more than a third of the sites Fred Hutch researchers analyzed happened to be in California, Florida and Texas — where Hispanic/Latino communities make up the largest nonwhite ethnic group in the state — researchers said. Hispanic/Latino COVID patients were also hit disproportionately hard by symptoms and hospitalized at higher rates than other groups, which could have contributed to higher participation numbers in treatment trials, Unger said.

Another concerning theme researchers found included trials’ failure to report certain demographics, despite federal requirements, said Hong Xiao, a Fred Hutch public health researcher and lead author of the study.

Sex, race and ethnicity were reported in about 89%, 78%, and 71% of U.S.-based trials, respectively, suggesting there was “a meaningful gap for many studies in this important aspect of trial reporting,” according to a Fred Hutch statement.

Industry-sponsored trials, including those run by pharmaceutical companies, were generally worse culprits of under-enrolling racially diverse participants and failing to report demographics, compared with federally sponsored trials like those run by the National Institutes of Health, Xiao said.

“That really is a huge issue we’ve seen in cancer research too,” Unger said. “I think the key reason is [industry-sponsored trials] want to accelerate new drug development. … And I think a lot of it is about access to where the trials are being conducted.”

Federally sponsored trials, meanwhile, often have a wide network of community sites that are more accessible to a larger pool of participants, Unger said.

Unger said the study wasn’t necessarily meant to poke holes in the country’s wide-ranging COVID research. Instead, he said, the analysis offers a “30,000-foot level perspective.”

“You have to look at the individual trials to see whether or not they were representative in their own right,” he said. “Ours is just to look at aggregate patterns to address whether or not there are system-level issues that need to be examined.”

In this case, he said, the results show a need for industry-sponsored trials to put more effort into community outreach.

“To me, that’s a systems-level issue,” he said. “And I do think there’s a clear policy ramification.”

The U.S. Food and Drug Administration has been attempting to tackle the issue, and published a new set of guidelines for clinical trials in April. The guidance aimed to urge trial sponsors to develop a specific approach in enrolling adequate numbers of participants from underrepresented racial and ethnic populations, a move Unger called an important first step.

While it’s unclear if there are any repercussions for those who don’t make stronger efforts to diversify clinical trials, Unger said some incentivizing approaches are being discussed — like patent extensions or tax breaks for well-represented studies.

“First and foremost, it’s about the patients,” Unger said. “And patients of all backgrounds having equal access to the new treatments available in trials. That can’t be overstated.”