Identifiers are coming, but names or numbers?
CDC discourages unique identifiers
The imperative to adapt a nationwide HIV reporting surveillance system has been accepted by most states now that there are more than twice as many HIV cases reported in the United States as AIDS cases. Some states, however, are leaning toward alternatives to name-based reporting, and that raises concern at the Centers for Disease Control and Prevention.
The most recent and dramatic arguments for mandatory HIV reporting came through new HIV surveillance figures the CDC released in May, showing that 52,690 cases of HIV were reported between January 1994 and June 1997, compared to 20,215 AIDS cases among the 25 states that have provided HIV data during that period.1
"The article shows just how handicapped a state would be in dealing with the problem of HIV/AIDS if it didn't have these kind of data, how it would really miss the target if prevention is being driven simply by AIDS case surveillance alone," says Kevin DeCock, MD, director of the CDC's Division of HIV/AIDS Prevention, Surveillance, and Epidemiology.
Compared with AIDS data, which describes the epidemic after it has matured, the HIV data in these states show where the epidemic is flourishing at the moment. Not surprisingly, more women are reported with HIV (28%) than with AIDS (17%). African-Americans make up 57% of HIV cases, compared to 45% in AIDS. And young people age 13 to 24 make up 14% of HIV cases, compared to 3% of AIDS cases.
"This illustrates what a huge picture we are seeing," DeCock says. "And as delaying progression to AIDS becomes more widespread, we should expect this trend to continue."
Although 29 states have adopted HIV surveillance, they account for fewer than one-third of the AIDS cases in the country. Until New York joined the fray, the only high-incidence states with HIV name reporting were Florida and New Jersey.
CDC guidance delayed by controversy
To help states adopt HIV surveillance systems, the CDC plans to publish a report this summer providing technical guidance on what should be included in a comprehensive HIV name reporting system. The document has been delayed by revisions, an indication of how controversial the issue is. The delay in CDC guidance has only added to the misunderstanding and lack of coherent support for name reporting.
"Many people feel the objective is having the names as opposed to doing the surveillance," says Helene Gayle, MD, MPH, director of the agency's Center for HIV, STD, and TB Prevention. "Lack of understanding of how surveillance is done has really made this one of the most difficult communication efforts we have dealt with in a long time."
The CDC has successfully moved the debate from whether HIV reporting is needed to what type of HIV reporting system should be implemented. Several states, including Hawaii and Washington, are considering joining Maryland in using unique identifiers rather than names, which would offer more protection should there be a breach in confidentiality. At the same time, Texas, which experimented with a unique identifier system, is going back to a name-based system because it found unique identifiers unworkable, DeCock says.
After studying both systems, the CDC is convinced that any unique identifier system falls short of the basic requirements of a comprehensive surveillance system - completeness, timeliness, and absence of duplication. Maryland's system, for example, cannot determine which cases may be duplicates. HIV reporting also was incomplete there, particularly in listing a person's risk factors. Also, Maryland's data cannot be linked to other national surveillance databases, such as the tuberculosis registry, DeCock says.
Moreover, Maryland uses the date of birth and the last four digits of a person's Social Security number - information that when combined can possibly be linked to a name, says a CDC consultant.
Taken together, these limitations pose enough potential problems that the CDC has not supported unique identifiers as a viable alternative. Indeed, because the agency has developed no expertise in the use of unique identifiers and doesn't plan to, states that adopt them may have trouble getting technical assistance from the CDC.
"I guess we will have to fund every state no matter what they do, but if some states do elect to do unique identifiers and then ask us for technical assistance, I don't know what we will do because we have no expertise there," DeCock says. "It's just not how we do surveillance at the national level."
Appealing to confidentiality fears
Sensitive to fears that names in a government database are at risk of being used for unintended purposes, the CDC has proposed adding another layer of security by offering to keep the key to a state's encryption code. By chance, if high-ranking government officials or politicians asked for HIV reporting data, state officials could deny them, saying the CDC has the key.
"We had hoped that would address many community concerns about confidentiality or data being accessed for non-public health purposes by legislators, but it isn't as straightforward as that," DeCock says.
The public distrust over name-based reporting may simply reflect a gap in communication over how public health conducts disease surveillance in general, says Helen Fox-Fields, a spokeswo man for the Association of State and Territorial AIDS Directors. Many people, for example, don't realize that names are included for most disease reporting, she notes.
"Public health assumed that this wasn't a dialogue we needed to include the community in," she says.
Reference
1. Centers for Disease Control and Prevention. Diagnosis and reporting of HIV and AIDS - 25 states, January 1994-June 1997. MMWR 1998; 47:309-314.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content