Editor’s note: Several passages have been removed from this story because the source of those passages, Mickyel Bradford, has admitted to fabricating them. The passages include descriptions of a lunch in Bradford’s town and a ball that Bradford claimed he attended with a man identified as Seth. Bradford now confirms that neither of those events occurred as described. Additionally, Bradford admits the two men never discussed getting tested for HIV. All passages concerning the two men have been removed.

ATLANTA — In the early 2000s, a young researcher at the Centers for Disease Control named Greg Millett set out to solve an epidemiological mystery. Nobody could explain why black Americans, particularly black gay men, had such high rates of HIV infection compared to the rest of the population. How were they putting themselves in so much danger? What were they doing differently from everyone else?

Millett began with a survey of the published research, but that only seemed to raise more questions. Study after study seemed to arrive at the same conclusion: Black gay men take fewer risks in the bedroom than white gay men. They are just as, if not more, consistent about condom use and STD testing. They have fewer sex partners. They are less likely to abuse injection drugs. Despite all this, black men—gay, straight, or bisexual—are 6 times more likely than white men to contract HIV in a given year.

[emailtext title="Have you or someone you know been affected by HIV or AIDS?" linktext="Email Us"]Let us know![/emailtext]

In 2006, Millett wrote up these findings in the American Journal of Public Health, alongside a careful review of alternate theories for what was going on. Some thought that genetic risk factors or circumcision or the prison systems played a role, but Millett found that the most promising theories involved healthcare disparities. For instance, black men are more likely to have other STDs like gonorrhea or syphilis, which increases one’s risk of contracting HIV. The sheer prevalence of HIV in the community means that despite getting tested just as frequently, black men are still more likely to have an undiagnosed infection, a circumstance that makes them more likely to pass on the disease. Lack of healthcare access raises another issue, in that those who do have HIV are less likely to be on the antiretroviral therapies that could make them less infectious to others.

But those explanations didn’t seem enough to account for the vast chasm between black and white HIV rates. So Millett began to consider another theory, a stunningly simple one, that could tie all the facts together.

Millett, who is himself black and gay, was well aware that black men tend to have sex with other black men, almost exclusively. What if that behavior was putting them in danger? It’s simple mathematics: the closer-knit a community is, the more any risk factors for infection become amplified. He began to wonder: What if the most dangerous thing a gay black man could do was what came naturally to him: to date the people he was attracted to, the people he was friends with, the people who surround him?

* * *

Greg Millett grew up in Brooklyn, the son of a United Nations staffer and a microbiologist. As a child he dreamed of becoming a civil rights lawyer, of putting the ghosts of racism and inequality on trial. To steel himself for those future battles, he landed a spot at conservative Dartmouth, which in the late ’80s was rending itself apart over sexism and race. These were the fires that would forge his identity, Millett thought, where he would learn to assert himself as a proud young gay black man.

Then his friends back home started getting sick.

As the AIDS crisis reached its crescendo in the late ’80s, Millett began to realize that the injustice was right here, in his calendar filling up with funerals and in the indifference of the federal government. By the age of 25, he had watched 20 of his friends and acquaintances die of AIDS. So Millett set aside his law school plans and began to volunteer at the Gay Men’s Health Crisis, a non-profit dedicated to fighting AIDS. He organized clinics to teach men of color about the risks of HIV and how they could have safer sex.

Even back then, he says, it was clear to him that black communities were disproportionately affected by the virus. He couldn’t prove it yet, but he went to graduate school to learn about how he could help. In the late ’90s he joined the CDC to organize studies of gay and bisexual men. By then, new drug cocktails were holding the disease at bay. The number of people who died of AIDS fell for the first time in 1996. TIME named HIV researcher David Ho its Person of the Year, calling him “the man who could beat AIDS.”

But the crisis had not abated. In 2001, a landmark study from the CDC found alarmingly high numbers of young gay men testing positive for HIV. Infection rates were particularly bad for black gay men: Nearly one third of black gay men between 23-29 were found to be HIV-positive.

Inside the CDC, an order came from high up: Find out what was going on with black gay men. At the time, Millett recalls, there were few black scientists in the department, and certainly nobody who was black and gay and out. Millett was assigned to untangle the different theories about why this community—his community—was getting hit hardest by HIV.

During those years, the CDC was heavily invested in programs to educate the public, the same kind of work that Millett had undertaken while he was living in New York at the height of the crisis. It was common-sense stuff: hammering home the importance of regular testing, of using condoms, of knowing your status and your partner’s.

Few of the programs at the time, though, were tailored to minorities, and some thought that cultural barriers were to blame for the high rate of infections among black gay men. The Tuskegee syphilis experiment had taught many black Americans to be wary of the medical establishment. Some still believed HIV was a government conspiracy, or that only white men could contract the disease.

“We’ve failed in developing prevention messages for young black men who have sex with men,” one black activist complained to The Washington Post in 2001.

Yet, when Millett went back to review the previous research, it seemed that HIV knowledge had indeed percolated into the community. Black men were already practicing safe sex and getting tested. Millett and his colleagues found a trail of evidence extending back 20 years, in paper after paper, the same results and the same question: How were black men, for all their careful behavior, getting infected at such high rates?

Only 12.2 percent of the American population is black, but the black community makes up 44.4 percent of people living with HIV in America.

“One of the things that’s really incredible about scientists is that we publish a lot of things, but then the community as a whole seems to have this collective amnesia of the things that have been done previously,” Millett says. “People keep asking a lot of the same questions, saying, ‘Well, we’re not really sure if this has ever been investigated before.’ By and large, you find that it has been investigated before—but people haven’t really taken the time to just put it all together into one place.”

In this case, putting together all the pieces meant that Millett and his colleagues were challenging a pillar of the CDC’s HIV prevention strategy. Just how effective was it to pour more dollars into trying to change the way that people had sex, especially when it seemed that by and large, the hardest hit community—the black community—had already learned that lesson?

* * *

Other researchers were beginning to wonder the same thing.

In 1999, an ambitious experiment called EXPLORE had been launched to determine if intense one-on-one counseling could help gay men stay HIV-negative. Up until that point, no HIV education campaign had ever been shown to reduce HIV infections among gay men. Some could demonstrate increases in safe habits like condom use, but none had gone on to measurably affect rate of new infections.

EXPLORE, which ran for four years, was the Cadillac of these so-called behavioral interventions. Counselors used techniques from cognitive behavioral therapy, a psychological treatment that had been used effectively to treat people with addiction, eating disorders and depression. Over the course of ten one-hour sessions, participants learned about condom negotiation and the risks of unprotected sex. They learned how to communicate with their partners. They talked frankly about their fears and their motivation to stay negative.

But EXPLORE, too, turned out to be a disappointment. Although the men reported having less unprotected sex, they still caught HIV at roughly the same rate as those who didn’t get counseling. (The small difference was so slight as to be statistically nonexistent.) Four years out, the differences in condom use also vanished.

“It was just knocking on the door of efficacy,” says Patrick Sullivan, a professor of epidemiology at Emory University. In Sullivan’s estimation, the gay community has already been so saturated with messages about HIV that additional campaigns have a hard time making a difference. So much changed during the AIDS crisis, he says. Gay men started to use condoms and began to have fewer sexual partners. “When we talk about behavioral interventions now, in the gay community, we’re talking about incremental changes that would be on top of a huge shift in behaviors that occurred early on in the epidemic.”

In the mid-2000s, when Sullivan was running the CDC’s HIV surveillance branch, public health scientists were just beginning to converge on this realization. They started to question the assumption that risky behavior was the chief enabler of the ongoing AIDS epidemic and that addressing risky behavior was the chief solution. These ideas may have been valid at one point, but they seemed less relevant now.

Sullivan recalls the stir caused by Millett’s widely cited reports on gay black men. The most obvious explanation for the black community’s high HIV rates—promiscuity and sexual risk-taking—had been shot down by the research. And the most obvious solution—a blitz of safe-sex messaging at Larry Kramer volumes—was failing to make a difference among gay men at large. Researchers began to look for for new ways to combat the epidemic.

* * *

In 2009, Millett left his research job for the White House. The Office of National AIDS Policy needed advisers and a friend put his name in. Millett had no background in policy, but he knew that someone had to advocate for black gay men, and for gay men in general, who for years have struggled to get their fair share of HIV care and prevention dollars. The President’s National HIV/AIDS Strategy, which was released a year later, had Millett’s fingerprints all over it.

The key message was that money had to be shifted toward the people at highest risk—gay men; black men and women; Latinos and Latinas. Minorities are the face of the epidemic these days, their tightly clustered social networks helping HIV to flourish. Invoking Millett’s own body of work, the report warned readers not to blame the victim: “Many individuals in [minority] groups may not engage in greater risk behaviors than others, but they still can be more likely to become infected with HIV. Research has shown that the higher risk for these groups is associated with the sheer number of HIV-positive persons in the communities where they live.”

The AIDS Strategy also acknowledged how weak the evidence was for behavioral interventions. Use them if they can be done cheaply, it advised, but don’t rely on them exclusively. Instead, the report said, focus on what could be done: getting people diagnosed, and helping those who test positive stay healthy and on medication.

These also happen to be areas where Millett showed that gay black men have fallen behind. They are less likely to know if they are positive, and more likely not to be in treatment.

There is so much HIV among black gay men that they should be tested much more often, Millett says—every couple of months, instead of once or twice a year. And the confluence of poverty, stigma and lack of healthcare access has always made it difficult for black gay men to obtain and stay in treatment.

How much these factors contribute to the disparities between the black and white HIV rates is still unknown. But it has become clear in the past couple of years that diagnosis and treatment make a big difference.

There had long been evidence showing a correlation between the number of copies of virus in a person’s bloodstream—his or her viral load—and how contagious that person is. But this relationship wasn’t nailed down until 2011, when the results of a five-year randomized experiment showed that being on HIV medication, and starting it early, can help prevent someone from transmitting the virus. The researchers saw a 96 percent decrease in HIV transmission, a result so dramatic that they immediately halted the study and advised all of the participants to start drug therapy if they hadn’t already.

Science magazine called it the “Breakthrough of the Year,” and it has birthed a new catchphrase in the prevention community. “Treatment as prevention” describes an approach that puts attention on those who are HIV positive—identifying them and giving them medication so they are less contagious. The phrase also refers to using drugs—“treatment”—preemptively in healthy people to prevent HIV infection. This was the recommendation of another set of breakthrough studies that showed a once-a-day dose of anti-retrovirals, if taken regularly, could reduce the risk of HIV infection by 90 percent. Together, these drug-based strategies jigsaw happily with the White House’s mandate to deliver more funding to those with greatest need. Instead of sprinkling prevention dollars on everyone indiscriminately, the latest programs target groups with the highest rates of HIV, black gay men especially.

* * *

Even with this solution in hand, the question that Millett had posed back in 2006 remains unanswered. If it isn’t behavior that is driving high HIV rates in the black gay community, what is it?

Since then, a lot of research had been accumulating around the idea that social networks—who you’re friends with, where you live and the kinds of people you date—shape the way that men are exposed to risk.

Patrick Sullivan is one of the researchers still at work on this mystery. In the course of his 20-year career, Sullivan has become one of the nation’s experts on the whys and wherefores of HIV transmission among gay men. Since 2010, the soft-spoken, sandy-haired professor has run a study recruiting single gay men in Atlanta to investigate their sex lives. (It is, by coincidence, the same study that Bradford had participated in.)

The small size of the black gay community, and the sheer concentration of the virus within it, makes everyone more vulnerable, no question about it. Dating within the black community means a higher chance of encountering someone who is HIV positive — or worse, someone who is positive but doesn’t know it. But the theory remains stubbornly unproven. It’s a touchy subject. If scientists showed with experimental clarity that black men were putting themselves at risk by dating each other, what advice could they give? That black men should avoid each other?

Sullivan is interested in untangling the other factors that might contribute to the black community’s high HIV rates. His hunch is that stigma and the environment where people live also play a big role in putting the community at risk. For instance, black gay men report experiencing more homophobia in their lives, which then affects how often they visit the clinic, or even how willing they are to discuss a partner’s HIV status.

Sullivan’s preliminary results in Atlanta confirmed the huge racial differences that persist. Among black gay men, 43 percent were HIV positive, compared to 13 percent of white gay men, even though the black gay men had fewer sex partners and less unprotected sex. Everyone also reported a tendency to date within the same race. Black gay men said that 80 percent of their partners in the past year were also black. The white gay men said that 73 percent of their partners were also white.

Crucially, there was also a significant difference in whether the men talked about HIV with their sexual partners. White gay men asked about HIV status about 70 percent of the time. Black gay men only asked about half of the time.