At the height of the pre-Delta variant COVID-19 pandemic, Stephen Rawlings, MD, PhD, was doing HIV and SARS-CoV-2 research when a pair of men, a couple, came to him. Both of them had had COVID-19. Both wanted to donate convalescent plasma or otherwise contribute to stopping the virus. Rawlings had to tell them no.

For Rawlings and his principal investigator (PI), it was a bizarre moment.

“Here the PI was a gay man, I was a gay man, and there were these two men – and all of us couldn’t donate,” Rawlings said. “But we were still, in our own way, forging better science.”

In this case, the would-be volunteers were able to contribute antibodies, not whole blood – that is, blood that hasn’t yet been divided into its component parts, such as platelets and plasma. That’s because since the early HIV epidemic, gay, bisexual, and other men who have sex with men (MSM) could not donate whole blood. At first, it was ever, even if they’d had sex with one man in 1978 and never again. In 2018, the U.S. Food and Drug Administration (FDA) allowed gay men to give blood if they’d been abstinent for 12 months. And then, in 2020, considering the COVID-19 pandemic’s impact on the blood supply, the FDA changed the policy again to reduce the abstinence interval to 3 months.

Many, including officials at the White House, have recently argued that the current blood donation policy is out of date and unscientific. But what might a more science-based blood donation policy look like?

What we call the U.S. blood donation policy is actually an amalgam of criteria that people must meet before they are allowed to donate. It includes everything from medical and prescription histories to a slew of questions on recent behavior. These include questions about having had sexually transmitted diseases, nonprescription injection drug use, and recent tattoos, among others. In addition, men also must not have had sex with another man in the last 3 months, and women must not have had sex with a man who’s had sex with another man in the last 3 months. Once a person gets through all those hurdles, staff at blood centers like Impact Life, which provides blood to several dozen hospitals in the U.S., will inspect a person’s arms for needle tracks.

Finding none, the visitor “lays down and bleeds,” said Louis Katz, MD, acting chief medical director for Impact Life. Blood vials then go down the hall, where they undergo blood typing and testing for everything from cytomegalovirus to West Nile virus to Zika to hepatitis B and C.

“In heterosexuals, we tolerate a high level of risk and it’s almost like zero tolerance with gay men,” he said of the original lifetime deferral policy. “If you were a heterosexual man who had condomless sex with a sex worker, you might get deferred for 6 months. But it was a lifetime deferral for gay men. It just doesn’t make any sense if you’re trying to protect people.”

So what protects the maximum number of people most efficiently, with the least risk and least amount of discrimination? Here’s what science says.

According to CDC data, receptive condomless anal sex carries the highest risk of HIV transmission, followed by being the insertive partner in condomless anal sex. Then receptive partners in vaginal sex have the next highest risk. The lowest risk is for two women having non-insertive sex. And while gay men are most likely to have receptive anal sex, they aren’t the only ones who do. A 2020 modeling study in the American Journal of Reproductive Immunology suggested that 41% of new HIV diagnoses in American women were the result of condomless anal sex. There’s also some data that suggest that, for unknown reasons, cisgender women’s rectal lining may be more vulnerable to HIV than men’s.

And gay men aren’t the only people who get HIV. For instance, nearly 1 in 4 new HIV transmissions in the U.S. occurred in heterosexual adults in 2019. People who inject drugs made up 7% of new diagnoses, according to the CDC. Right now, the FDA isn’t studying deferral policies for heterosexuals at higher risk of HIV, though, said Custer.

Blood centers do defer people from donating blood if they have a history in the last 3 months of sex work or injecting drugs, however.

A more science-based approach may be to ask everyone about their sexual behavior. But that’s not what the Assessing Donor Variability And New Concepts in Eligibility (ADVANCE) study is doing. Designed and funded by the FDA, the study is considering other options for how to determine when gay and bisexual men can give blood. The study asks gay and bisexual men questions about their sexual behavior, such as how many partners men have had recently and if they use condoms or HIV prevention medications, known as pre-exposure prophylaxis (PrEP), said Custer. The study won’t be testing those questions among heterosexuals.

“These are the questions that might one day be on a future blood donor history questionnaire,” said Custer. “A contemporary, science-based approach to policy is exactly what we’re trying to do.”

Back when the ban on gay men donating blood came out, there were no tests for HIV. A ban was a blunt instrument. But it has long since been replaced by accurate HIV tests, said Katz.

“By the late 1990s, we had nucleic acid tests – PCR, basically – that could detect HIV within 7 to 10 days,” he said.

And that should mean, Rawlings said, that a gay man in a mutually monogamous relationship for the last 10 years where neither partner has HIV ought to be able to donate the same as anyone else. “I’m fairly confident that monogamous gay men can be shown to be safe” donating blood, said Katz.

Rawlings is himself one such man.

“If I’d had a new sex partner and it was unprotected and in the last 7 days, I should not give blood,” he said. “Current policy does not account for that.”

There might be one glitch in that recommendation, said Custer, of the Vitalant Research Institute: HIV transmissions while someone is using HIV prevention meds inconsistently, though rare, could result in just enough medication in the blood to keep the level of virus in the blood very low. It’s unclear whether current tests would catch that, he said.

That’s where the ADVANCE study comes in. In addition to asking questions about men’s sexual behavior and testing blood for HIV, researchers will also test blood for concentrations of one of the most common PrEP drugs, tenofovir. After 2 weeks, participants return to find out the results of the HIV test and fill out another, longer questionnaire on personal behavior. By correlating concentrations of the drug to HIV testing, they hope to figure out whether that theoretical risk is a real risk. Custer said they hope to have study results early next year.

It’s also possible, he said, that people who take PrEP are at lower risk of donating blood containing HIV because the medication is 99% effective in preventing HIV. Right now it’s a paradox.

“I don’t know if it will be enough,” he said of the data they’ll be gathering. “Through a mix of testing biomarkers for infection and PrEP use, as well as self-reported behaviors, we’re going to come as close as we can to having the evidence.”

But maybe it doesn’t need to be enough. Since the FDA reduced the MSM deferral period to 3 months in April 2020, the Transfusion-Transmissible Infection Monitoring System (TTIMS) has been following whether there’s been an uptick in donations containing HIV. So far, researchers haven’t analyzed that data, but they expect to have results by the end of 2022.

It’s also unclear whether the ADVANCE study will result in changes to the FDA’s deferral policies for gay men. If it does, the U.S. will join the U.K., France, and Greece, among others, to donate based on behavior, not identity. That could make 4.2 million people eligible. Based on the percentage of eligible people who actually donate, that could be an additional 615,300 pints of blood.

Whatever happens, Katz said it’s clear the FDA will need new data to make its decision.

According to him, a better science-based policy might defer non-monogamous gay men from giving blood for 7 to 10 days after their last sexual experience, so that HIV tests could pick up a new infection, and then base donations on actual individual behavior instead of a blanket deferral based on identity.

“There’s no reason we can’t do better,” he said. “Bottom line, we do not have a good sense yet for the best behavioral approaches, and that is what the research is asking.”