CDC Study on HIV Infections Among Gay Men Based on ‘Tiny Number’ of Cases, Sullivan Says

A recent CDC study that reported a "dramatic" increase in the number of new HIV infections among young gay men -- particularly African Americans -- is based on so few cases that the findings are more "hype" than "science," New Republic senior editor Andrew Sullivan writes in a column for that magazine. Sullivan criticizes mainstream media organizations such as the New York Times and Washington Post for engaging in "hysterical journalism" when reporting the study findings, adding that a skeptical look at the CDC's "small print ... [is] not a pretty sight." First, Sullivan says, the study is not new; rather, it is a "continuation of the same study released last February to similar headlines." For the findings released last week, researchers subjected already-collected blood samples to new tests to find out how recently HIV infections had occurred. Of 2,942 gay men studied, Sullivan says that "a grand total of 38" new HIV infections were identified -- a "tiny number" from which the CDC derived its conclusion that there was "striking evidence of recent infections." Sullivan adds that there is no data on the racial breakdown of those 38 cases, meaning that the group might include "as few as two or even one black man." Further, Sullivan notes that of the 408 gay black men studied, 122 were HIV-positive. If that number is averaged over the six cities surveyed in the study, the data on which CDC scientists based their "belief that we have 'explosive incidence rates' among black men who have sex with men is based on about 20 HIV-positive men in each city surveyed," he writes.

More Analysis, Less Hype
Looking beyond the CDC study, Sullivan concedes that it is possible that HIV infections are rising, noting that, with AIDS deaths at an "all-time" low, it is even "statistically likely" that the total number of people with HIV is growing. But "we don't know for sure." Sullivan adds, "I do not mean to impugn the CDC," which does a "difficult job honorably." But he notes that the lack of mandatory HIV reporting means data must be collected from "nonrandom samples of gay men recruited in bars and STD clinics," making any statistics about this population "dubious." While HIV infection rates should not be "dismissed," Sullivan concludes, "We need more data, more analysis, more research -- and less hysterical journalism. Please" (Sullivan, New Republic, 6/18).