By David Brown
Washington Post Staff Writer
Friday, April 21, 2006
Although male prisoners have a relatively high rate of HIV infection, very few of them acquire the virus while behind bars, according to a federal study that is the largest and longest one to look at the issue.
About 90 percent of HIV-positive men in Georgia's prison system -- the nation's fifth largest -- were infected before they arrived, the study found. Over a 17-year period, 88 men became infected in prison by the virus that causes AIDS, chiefly through same-sex intercourse. Georgia prisons currently house about 45,000 men.
The study, published yesterday by the Centers for Disease Control and Prevention, refutes the widespread impression that U.S. prisons are hotbeds of the AIDS epidemic, and that incarceration contributes directly to the high rate of HIV among black men, who make up the majority of male prisoners in many states.
The study revealed a surprising and unexplored aspect of HIV infection in prisoners. Of the men who became infected behind bars and acknowledged having gay sex there, half reported that their partners were prison staff members, not other inmates. Whether guards were the source infection, or became infected themselves, in any of those liaisons is unknown.
Nearly three-quarters of inmates who reported having had sex with other men described it as consensual. Nearly one-third of those said they used condoms or an improvised protective method such as rubber gloves or plastic wrap.
"The popular assumption is that prison is a very good place to contract HIV infection," said Richard Tewksbury, a professor of justice administration at the University of Louisville, who has studied HIV-prevention strategies inside and outside prison.
"Both inmates and society as a whole have long held the belief that transmission is common among prison inmates. The interesting thing about this study is that it directly contradicts that," he said.
Terry Butler, a CDC spokeswoman, said: "Media coverage of this issue over the past several years has been characterized by misperceptions that HIV transmission in prisons is widespread." Most news stories provide little evidence for that idea, she said.
The findings, published in CDC's Mortality and Morbidity Weekly Report, raise numerous policy questions that prison officials in Georgia, and possibly other states, may need to address.
An obvious one is whether to make condoms available. They are contraband in Georgia prisons, as in nearly all U.S. correctional institutions. Another is whether prison staff members should be tested for HIV, as incoming prisoners are. Guards are now tested only for tuberculosis.
"We can now begin the policy discussion about ways to prevent [new infections] in the prison population," said Brian Owens, assistant commissioner of the Georgia Department of Corrections. He would not discuss what, if any, changes officials are considering.
Patrick Sullivan, a CDC epidemiologist who helped lead the study, said prison is an opportunity to get AIDS-prevention messages to a hard-to-reach, high-risk population -- something that is especially important given that 97 percent of inmates nationwide eventually are released.
In 1988, Georgia began testing convicts on arrival at prison. As of February 2005, HIV had been diagnosed in 88 men who arrived uninfected. Two-thirds were black and one-third were white. As of last fall, Georgia prisons held 856 HIV-positive men.
To try to determine how the in-house infections occurred, CDC and Georgia state epidemiologists compared each newly infected prisoner with an uninfected one who had served a similar length of time.
Not all 88 were available: Eleven had been released, two had died and some declined to cooperate. In the end, 68 infected men and a similar number of uninfected ones were interviewed, by computer and in person.
The infected men were 10 times more likely than the uninfected to have had same-sex intercourse in prison, and 13 times more likely to have received a tattoo there.
Among the 68 infected men, 20 reported no sex in prison, 61 reported no use of injected drugs and 28 reported receiving no tattoos. Six of the 68 reported none of those activities. Although tattooing is considered a risk factor, there are no confirmed cases linking it to HIV infection, Sullivan said.
Forty-five of the infected men reported having same-sex liaisons in prison, and 22 of those said they had had sex with "male prison staff." (That was also true of four of 68 uninfected men.) Of the 48 who reported sex of any type, 15 infected men reported liaisons with female staff members, as did six of the 68 comparison inmates.
Sex between inmates and staff members, even if consensual, is a felony in Georgia.
The District, which has been criticized for its lack of HIV data citywide, does not know the incidence of infection in the D.C. jail. However, from October to January, 986 inmates agreed to take an HIV test when they arrived, and 4.6 percent tested positive.
According to a Maryland Division of Correction spokesman, 1,104 of about 25,000 men and women in prison were infected with HIV as of September, and 420 of them had AIDS. The system has no statistics on how many became infected while incarcerated. The infection rate in Virginia prisons was not available yesterday.
A Justice Department study of allegations of sexual activity in prison reported to state officials in 2004 found that more of them involved staff-inmate liaisons than nonconsensual acts between inmates.
Staff writer Susan Levine contributed to this report.