In Defense of the FDA

The FDA recently announced that it is lifting its ban on blood donation for previously (but not currently) gay men, ending a year-long public comment period:

The gay community's feelings on the news have been conflicted. It feels like a progressive victory but it leaves gay men in practically the same position they were a year ago: you probably aren't allowed to donate blood.

My own opinion is nuanced—I think the FDA's policy is practical and effective, but unethical. I work with this agency regularly and hopefully I can lend some insight into their thought process and why their policy seems so disconnected from reality.


The FDA has a long history of inertia on matters of gay rights. In the early days of the AIDS epidemic, most treatments had unknown efficacy. The FDA was unwilling (or at least slow-moving) to approve these treatments despite public demand, leading to an underground drug trade for AIDS patients desperate for anything that might help.

We know today that most of these drugs were actually not very effective in treating HIV or AIDS and the FDA would have been right to deny their approval, but in historical context the inaction was perceived as (and probably due in part to) a combination of indifference to the gay community's plight and moral stigma attributed to those suffering with the disease.

All of this led to a surreal occupation of FDA headquarters in 1988, where AIDS rights activists laid siege to the FDA itself. It's a very long story, with entire documentaries devoted to it, but the short of it is that it took some dramatic action on the part of the community to even get a seat at FDA's table when AIDS treatments were being discussed.

We have their attention, today, but the relationship has always been a tenuous one. While most of this essay may defend or explain the FDA's policy, don't mistake it for support. The Agency wants us to be quiet and go away and we have to work to keep their attention.


The FDA has a responsibility to protect the blood supply. One way in which they do this is by identifying high risk groups and excluding them from donating blood. Far and away, the highest incidence group for HIV infection is gay men:

HIV Incidence by Population

The CDC also attributes gay sex for over half of the HIV-infected population. The unfortunate reality is that gay men are the most burdened population when it comes to HIV. We're also a minority: only 3.4% of the population identifies as LGBT (though due to stigma we can inflate that estimate).

Setting aside fairness, ethics or legality, if the FDA barred gay men and black people from donating blood, it would eliminate the vast majority of HIV infected and at risk people from the donation supply while only disenfranchising a small fraction of the US adult population. It's the obvious, practical choice, and it's a choice that can scale easily to a national level with minimal effort and expense.


We have tests for HIV and people can lie about their sexual encounters in order to donate blood, so aren't there more effective ways to protect the blood supply? Let's examine that.

HIV testing is effective but imperfect as a prophylactic against infecting the blood supply. Over 9 million people donate blood every year. With the prevalence of undiagnosed HIV in the population (assuming very few people would knowingly donate infected blood), upwards of 4,000 donors could be HIV positive. If an HIV test is 99.9% accurate, that would still lead to over 4 HIV-positive samples infecting the blood supply every year.

Barring gay men from donating blood effectively adds another layer of protection to the blood supply. Of those 4,000 undiagnosed positive donors, 63% of them would be disqualified and now only 1 or 2 infection events might occur in a year (and those are further mitigated by other means).

As for people lying, the effectiveness of the donor screening questionnaire has been studied to death. If there's one thing to know about the FDA, it's that they fucking love validation studies. Let's stop having this argument: it could always work better, but it fundamentally works. Even if a significant percentage of people are so incentivized to give blood that they lie on the questionnaire—it still presents an incredibly effective screen for the most at-risk donors.


That's about the limit to which my defense of the FDA's antiquated policy extends. The policy doesn't exclude black people from donating blood because we agree that it is unethical to discriminate against someone purely based on race, no matter how statistically sound the argument. At the time of the blood ban, discrimination based on sexual orientation was ethical; it was still hotly debated whether sexuality was a lifestyle choice.

The debate rages on but the writing is on the wall: the United States is quickly accepting that sexual orientation is not a choice or, to the extent it is, it should be a protected one. Once we're there, it will be clear that the FDA's policy banning gay men is unethical despite being practical and effective.

If we want to replace the ban, we have to be working today on validating effective and practical proxies. Can we instead ask about regular condom use, having multiple partners, or frequency and environment of sexual encounters? Will people be more or less honest about these issues than they are about sexual orientation?

I believe there's a solution, but we haven't found it yet. Yes, we can complain about how the FDA should be proactive in making this happen for us. But as an LGBT community, if we really care, we need to be funding and executing these studies ourselves. We need to collect and present the evidence that there are practical alternatives to discriminating against us.

Just like in 1988, nobody else will do it for us.