Runnin’ Numbers

Posted in: News 2005-Older

Breakin’ it Down: The Complicated Made Simple

For a decade now the CDC has been saying the same thing about HIV infections: We log 40,000 new ones a year. Everyone accepts the figure and moves on. But the reality is it’s an incredibly broad estimate, and we can say with little confidence whether it’s wildly inflated or deflated.

At the National HIV Prevention Conference in Atlanta earlier this month, the CDC announced plans to change that troubling fact. Before offering shocking new estimates on the virus’ spread–over a million infected, roughly half of them Black; half of all Black gay and bisexual men likely positive–CDC number cruncher Dr. Jeff Greenberg spelled out the agency’s plan for finally drawing a detailed, accurate portrait of HIV in America. By this time next year, he pledged, public health will have far more insight into the virus’ movements.

The new knowledge could both drastically reshape how we view the epidemic and overhaul how the federal government hands out prevention and care dollars.

From the beginning, public health’s ability to track the AIDS epidemic has been hamstrung by its broader social and political implications. It was, after all, originally known as “Gay Related Immune Disorder–at a time when this month’s Gay Pride celebrations were still considered radical.

Spend a few days below the Mason Dixon line today, and it’s clear that infection still carries massive stigma. AIDS and gay activists alike have worried–and still worry–that collecting too much information about those getting tested will drive away the people who most desperately need one. An untidy compromise public health made with this social reality was to collect data only on diagnosed AIDS cases, rather than on those who were infected but not actually in immediate danger of illness.

In late 1997, after new drugs drastically prolonged the time between testing positive and developing AIDS, CDC officials first began publicly musing about the need to track HIV infections instead–and to do so by using the name-based reporting system employed to keep up with other communicable diseases. The idea ignited a hot debate in the AIDS world

Slowly, however, most states gave in to the inevitable. While the CDC never made name-based HIV reporting mandatory, the agency made clear its intention to build a national surveillance system anchored on names reporting. It didn’t take local officials long to realize Washington would ultimately use that system to divvy up resources, and that states not meeting its standards risked getting short-changed. Today, 32 states monitor new infections by collecting names; the remaining states use some combination of coded identifiers and names.

With that glacial revolution now in irreversible motion, the CDC has launched a multi-pronged plan to more closely examine the much-cited 40,000 figure. As a result, you’ll be hearing a lot of new numbers in the coming years, some of which are bound to be controversial. With our minds on that proverbial demon lurking in the details, we give you a brief snapshot of how the CDC’s gonna do its math.

What they want answered: How’d you get it?
How they’ll answer it: The National Behavioral Surveillance System. This system, already up and running, generated the most shocking stat to come out of Atlanta: 46 percent of Black men who have sex with men may already be infected, and two thirds of those folks don’t know they’ve got it. The idea is to take a closer look at those who are most statistically at-risk, asking what they are doing and how they are interacting with services that public health has set up to help them stay well.

The system will track people who fit three behavioral categories: Men who have sex with men, injection drug users and what the prevention wonks have started calling “high risk heterosexuals.” Yes, the scintillating CDC jargon is distracting, but try to focus–this is the important part. In alternating 12-month cycles, researchers will focus in on one group and study their behavior, so that they’ll get a fresh batch of information on each group every three years. They will draw their subjects from the same cities each time, in order to create clean data on trends over time.

Men who have sex with men drew the short straw and went first. CDC has already completed research on them for the 2004-2005 cycle, in which researchers interviewed over 14,000 men at 17 cities in 13 states and territories. The information released in Atlanta came from five of those cities, and the full report is expected by year’s end.

What they want answered: How many are infected each year and who are they?
How they’ll answer it: The HIV Incidence Surveillance System. This is the agency’s tool for getting past its annual 40,000-newly-infected estimate. CDC has chosen 34 sites around the country to focus on each year. In those places, researchers will drill down on every newly-diagnosed infection to get, among other things, the person’s demographics and clinical info on that person’s virus. Importantly, they will also run newly-developed tests to determine how recently the person has been infected. This last bit will be key to deciphering where in society the virus is speeding up and slowing down, thereby allowing CDC to target prevention resources in a way that has not been previously possible.

Half of the 34 sites have already launched the new system. The other half are expected to begin next month, so that by December CDC will have six months of complete data and, by this time next year, we’ll hear something more precise than 40,000.

In Atlanta, Greenberg also cited the CDC’s ongoing HIV Case Surveillance network as working in tandem with this new model. CDC will continue to draw its overall estimate of the national HIV case load from the 32 states that track new HIV infections by recording the names of those who test positive.

What they want answered: What’s the total number of infected?
How they’ll answer it: HIV Prevalence Estimates. Again, nothing new. CDC takes data from the 32 states using name-based HIV reporting systems, runs it through two statistical modeling processes to factor in those who may be positive but have never been tested, and comes up with its total number.

In Atlanta, Greenberg updated the current estimate, saying that CDC believes between 1,039,000 and 1,185,000 Americans were living with HIV as of 2003. He also estimated that anywhere from 24 percent to 27 percent of those folks didn’t know they were infected. It is the first time CDC has put the positive tally above one million, meaning more people are living with HIV today than ever before.

Greenberg also noted the long-standing AIDS Case Surveillance system will continue to be a tool for building this sort of big-picture data. This is the original surveillance system, which tallies the number of newly diagnosed AIDS cases and deaths reported by all states each year. As of 2003, around 400,000 of the one million positive folks Greenberg cited had advanced to an AIDS diagnosis.

What they want answered: If you’re positive, how are you doing in treatment?
How they’ll answer it: The Morbidity Monitoring Project. CDC not only wants to know how people who are in treatment are faring, it’s also curious about what’s happening to those who are not in treatment—why aren’t they there and are they doing something harmful to themselves or others while they’re on the outside of the care bubble.

So researchers have selected 26 sites in 21 states from which to pull medical records of those in care and ask some questions. They’ll get demographics, lab results and history of antiretroviral use, among other things. They will then do interviews with both people who are in care at those sites and those in the area who are not. In the interviews, they’ll add questions about access to health services and HIV-risk behaviors.

The project is to begin this summer.

Leave a Reply

Your email address will not be published. Required fields are marked *