National HIV Reporting Approaches, But Privacy Remains Paramount

TIME TO RETHINK: The old discussions about reporting HIV infection must be reviewed in light of a changing epidemic, says CDC's John Ward. As the AIDS epidemic enters a new phase of changing demographics and combination drug therapy, support is mounting for a national system of HIV case reporting. Even some civil rights activists who previously opposed HIV reporting now are admitting the need for it to help contain the elusive disease. However, the long-debated question of whether HIV-positive people should be identified by name remains to be resolved, and the issue may yet pose ethical concerns for some researchers. "We're at a turning point of this epidemic," notes John Ward, chief of HIV/AIDS surveillance in the Atlanta-based Centers for Disease Control and Prevention (CDC). "After 16 years, for the first time we're beginning to change, in a profound way, how we monitor it. Because of that, all the old discussions about the values and risks of reporting HIV infection have to be rephrased." Ward did just that in coauthoring an article published October 16 (L. Gostin et al., New England Journal of Medicine, 337:1162-7, 1997) that called for a national system of HIV case reporting. The article, which Ward says was CDC's first official call for a national system, was based on a consultation on HIV surveillance conducted by CDC and the Council of State and Territorial Epidemiologists. NEJM also published an editorial by deputy editor Robert Steinbrook in its September 11 issue (337:779-80, 1997) that urged adoption of a national HIV reporting standard. Both articles noted recent improvements in treating HIV with a combination regimen of protease inhibitors and reverse-transcriptase inhibitors. The paper by Ward and his coauthors asserted, "Unless we revise our surveillance system, health authorities will not have reliable information about the prevalence, incidence, and future directions of HIV infection, the kinds of behavior that currently increase the risk of HIV transmission, or the heightened impact on specific subpopulations, such as racial and ethnic minorities and women." CDC's Ward and NEJM called only for HIV case reporting, not for HIV name-reporting. No federal statute requires national reporting of any disease, but state laws nevertheless require reporting the details of many disorders to CDC. For example, all states have laws or regulations on reporting tuberculosis and various sexually transmitted diseases, including AIDS. The names of people with such diseases are collected confidentially at the state level. The names always are deleted and the remaining data are encrypted for aggregation at the federal level. Only 26 states currently require HIV surveillance for adults and adolescents. Two more states, Florida and New Mexico, plan to begin surveillance in late 1997 or early 1998, while Texas, Oregon, and Connecticut mandate surveillance only for pediatric cases. Two other states, Texas and Maryland, gather HIV data without collecting names, instead assigning each case a mathematically derived unique identifer (UI) number. But neither California nor New York, the two states that have by far the most cases of AIDS, require HIV reporting of any kind. Although CDC is not a regulatory body, its call for a national HIV reporting system will carry clout with the states. For example, CDC funding for state AIDS prevention and treatment programs is contingent upon the states' having AIDS surveillance regulations. Ward now expects that CDC will build a consensus with the states on a national system and announce a policy next year. The article by Ward and his coauthors notes that the 26 states currently requiring HIV surveillance accounted for only 24 percent of AIDS cases reported to CDC through 1996. Thus, the primary source of national data remains that of full-blown AIDS cases-people infected, on average, a decade ago. However, such data are of diminishing use in targeting education, prevention, and treatment programs to the newer population subgroups with HIV, who may have different behavioral patterns from those of the gay men, injected drug users, and commercial sex workers who were the predominant victims of the epidemic in earlier years. Nevertheless, there is worry in states such as California and New York that HIV name reporting might cause those with as-yet-asymptomatic HIV infection to suffer the same stigma and discrimination that long have accompanied the full-blown disease. Confidentiality is therefore essential. Because CDC is not requesting names for its HIV registry, its position on national reporting is supported by the Washington, D.C.-based AIDS Action, a lobby group representing 2,000 community-based service organizations nationwide for people living with AIDS. "But we're not convinced that having names is necessary," says AIDS Action legislative representative Julio Abreu. "Certainly, with AIDS surveillance, the information they've gotten has resulted in better targeting of AIDS prevention dollars. Hopefully, that will also be the case for HIV [reporting]," he adds. "Let's face it, many people think the crisis in AIDS is over, that it has become a chronic, manageable disease, when, in fact, infection is still occurring at a high rate." CDC estimates there are between 650,000 and 950,000 HIV- infected people in the United States. Ward estimates that 500,000 people have been diagnosed as HIV-positive in America and presumably know their status. His best guess is that about one-third of all people in America who are carrying the virus do not know it. BILL'S SPONSOR: Rep. Tom Coburn (R-Okla) introduced the proposed HIV Prevention Act, which would mandate a national reporting system. The American Medical Association and 47 state and local societies have lent their backing to the proposed HIV Prevention Act. Introduced in the House by Rep. Tom Coburn (R-Okla.), it would be the first bill ever to mandate a national system for any reportable disease. The proposed law, similar to one Coburn unsuccessfully introduced in 1996, is due to be considered by the House in February 1998. A Senate companion bill has not yet been scheduled for hearings. AIDS activists charge that the Coburn proposal would institute name-reporting of HIV-positive people to a federal registry, but Coburn's office is adamant that the same no-name system used for national reporting of other disorders would apply to HIV. This difference of opinion may become moot. Although CDC policy doesn't carry the force of law, its call for a national HIV reporting system might preempt the need for any federal legislation, allowing "holdout" states to comply at their own pace. In clinical trials of HIV drugs, the use of names is indispensable, declares Mark Jacobson, an infectious diseases internist at San Francisco General Hospital and principal investigator of the AIDS Clinical Trials Group (ACTG), an NIH-funded national network of HIV clinical research sites. "We have to know who the people are; we have to be able to contact them," he notes. He says when trial data are centralized, a random number is used for each person. At every trial site, "shadow documentation" is kept that includes the individual's name for cross-checking records. Despite antidiscrimination laws passed at various tiers of government in recent years to help protect the civil rights of HIV-positive people, activists still worry that many people avoid HIV tests because they fear that their names will be reported if they test positive. Definitive studies have not been published on this point, but in many states, clinics offer anonymous testing, which CDC supports as an option to name-reporting. NO NAMES: Anonymous HIV testing gives clients more authority and autonomy, the Whitman Walker Clinic's Troy Knighton explains. The nonprofit Whitman Walker Clinic is among several in the Washington, D.C., area that conduct anonymous HIV testing. Troy Knighton, director of HIV testing and counseling, says clients are asked to choose their own code within a prescribed formula of one letter and six numbers. "It gives them more authority and autonomy, as well as a role in the process," he explains. "Within the clinic, there really have been no problems with the codes . . . but on occasion, we've had to turn people away when they come for their results, because they can't even remember the day they came in for the test." AIDS Action 1875 Connecticut Ave. N.W., Suite 700 Washington, D.C. 2009 (202) 986-1300 Fax: (202) 986-1345 http://www.thebody.com AIDS Clinical Trials Information Service P.O. Box 6421 Rockville, Md. 20849-6421 (800) 874-2572 Fax: (301) 738-6616 Centers for Disease Control and Prevention Mail Stop E47 Atlanta, Ga. 30333 (301) 639-3311 http://www.cdc.gov Rep. Tom A. Coburn (R-Okla.) 427 Cannon House Office Building Washington, D.C. 20515 (202) 225-2701 Fax: (202) 225-3038 Council of State and Territorial Epidemiologists 2872 Woodcock Blvd., Suite 303 Atlanta, Ga. 30341 (770) 458-3811 Fax: (770) 458-8516 http://www.cste.org For HIV-positive clients, counseling includes discussing notification of current sexual partners and tracing any other contacts who might have been exposed to the virus. The client may personally assume these responsibilities or agree to the help of an agency. In the case of contact tracing, agency notification of persons at risk is done without divulging the client's identity. "I think people tend to value the anonymous testing more [than when a person's name is given], because it lends a stronger element of trust to the test process and gives clients the power to determine who they want to tell of their status," Knighton comments, "because there's still a lot of discrimination out there." For Maryland and Texas, the two states that use non-name UIs to tally HIV infection, Ward's paper cites "a substantial number" of incomplete and difficult-to-match records. Both states use the same UI system, consisting of the last four digits of a client's Social Security number, followed by a code for date of birth, sex, and race. "We do not yet have an off-the-shelf, secure yet easily identifiable code," Ward notes. "Is there another unique identifier that would work better? We are continuing to investigate." Sharon Melville, director of the epidemiology division in the Texas Department of Health's bureau of STD/HIV prevention, says the UI was chosen after a careful search of mathematical options, but the primary problem with it is avoiding duplication from the two main sources of data, laboratories and health care providers. "Laboratories don't routinely collect Social Security numbers. We've worked with them, but it's very difficult to get them to change their databases," she asserts. With an incomplete identifier and no name to rely on, determining whether the same patient's information has been reported twice, or not at all, is work-intensive and unremunerated for doctors' offices. "If we call a provider, we might say, 'Can you look through your records on this day, and find anyone with these last four numbers of their Social Security number who has this date of birth?'" Melville explains. From the program's initiation in April 1994 until June 1997, the state received about 25,000 reports of HIV-positive individuals from labs and doctors, but only 14,000 contained complete codes. Of those 14,000, about 3,000 could be identified as duplicates. Of the remaining 11,000, others were eliminated as full-blown AIDS cases rather than asymptomatic HIV, leaving 6,882 cases. The HIV-positive figure, however, seems too low, Melville thinks. Yet, without any way of accurately cross-checking reports from labs and doctors, she can't be sure. When the Texas AIDS database, which includes names, was matched against hospital and death registries, it proved to be 90 percent complete. Between July 1995 and June 1996, about 2,000 AIDS cases were reported, most of which should have a corresponding HIV report. "When we did that match, we found that 79 percent of the information in our AIDS database had enough information to create a UI data base [for HIV infection], but there was only about a 26 percent match," says Melville. "What the local health departments want to do-that couldn't be done even if the UI system worked 100 percent of the time- is to link those people to health services." She adds, "What we now know about HIV is that it starts replicating very early. . . .The epidemic has changed, and we need to adjust to that." In 1985, Colorado and Minnesota started the trend in instituting name reporting for HIV infection. Kristine MacDonald, assistant state epidemiologist and medical director of the acute disease epidemiology section in the disease prevention and control division of the Minnesota Department of Health, says a state law stipulates that when any disease is added to the reportable list, all carriers also must be listed. She adds that the policy has not been highly controversial, except for some concerns over partner notification. Anonymous tests are an option, should a client refuse to provide a name. "We try to treat HIV as an STD," MacDonald says. "For us in Minnesota, use of the name has worked. I think what is needed is to manage access to the names." She says that Minnesota keeps its list highly secure, few people have access to it, and the names are not used on the state's computerized reporting database. Throughout the states, name-reporting is highly confidential. There is only one documented instance, in Florida, of unauthorized release of names on a state HIV registry. Minnesota health officials have used both AIDS and HIV data to help drive community-focused programs. MacDonald gives the example of adolescent virus-carriers, who formerly were few in the state. HIV test results show their numbers are growing, and more education/prevention funds accordingly have been directed toward youth. MacDonald is on the 10-member executive committee of the Atlanta-based Council of State and Territorial Epidemiologists, representing 450 epidemiologists in every state. In June, the group unanimously passed a position statement that calls for confidential name reporting of all persons with HIV, to avoid current problems with UIs. "Our concern stems from the belief that the epidemic is still a major health problem and we still want to understand better what's going on," MacDonald explains. "At least, this [HIV reporting] would give us a better idea of the burden of the disease, of what we call the front end."

National HIV Reporting Approaches, But Privacy Remains Paramount

Interested in reading more?

Become a Member of