CDC guidelines for HIV surveillance support call for name-based reporting

Follow guidelines or risk losing CDC funding

A year after releasing its proposed HIV surveillance guidelines and two years after first calling for HIV surveillance, the Centers for Disease Control and Prevention in Atlanta has issued its final recommendations, which include support for name-based reporting.

States and territories will have to follow the CDC’s guidelines, which took effect Jan. 1, 2000, or risk losing CDC funding for monitoring HIV/ AIDS. The guidelines require surveillance systems to use reporting methods that provide case reporting that is at least 85% complete and has 66% reported within six months of diagnosis. Also, there should be fewer than 5% duplicate case reports and fewer than 5% incorrectly matched case reports.

The final version of the guidelines is nearly identical to the proposed version. (See AIDS Alert, February 1999, p. 13.)

HIV surveillance will permit the CDC to track the epidemic with as current a snapshot as possible, says Helene D. Gayle, MD, MPH, director of the CDC’s National Center for HIV, STD, and TB Prevention. Gayle is one of the authors of the CDC guidelines.

For example, when the CDC compares HIV reporting data from 25 states to AIDS data from all states, it finds that HIV data show a greater proportion of African-Americans and women and a trend of HIV spreading through heterosexual transmission. The AIDS data give an older picture of the infection rate because there often is such a long lag time between when a person’s HIV infection becomes known and when he or she is diagnosed with AIDS.

Tracking HIV infections is a laudable goal, but the CDC’s emphasis on names reporting will defeat its purpose, charges Daniel Zingale, outgoing director of AIDS Action in Washington, DC.

"There are still hundreds of thousands of Americans living with HIV who don’t know it, and they’re at the highest risk of passing on the virus," Zingale says. "So there’s still an urgent need to bring those people into voluntary HIV testing, and the implementation of names reporting is still one barrier to bringing people in for testing."

The guidelines, published in the Dec. 10, 1999, issue of the Morbidity and Mortality Weekly Report, refer to studies that suggest name-based HIV reporting has not served as a major deterrent to testing.

A study conducted by researchers at the University of California at San Francisco found that concern about name-based reporting of HIV infections to the government was a factor for avoiding HIV testing in 28% of men who had sex with men, 18% of injecting-drug users, and 13% of heterosexuals.1 By comparison, nearly half of the respondents said they avoided testing out of a fear of finding out they had HIV infection or a belief they were not infected.

Zingale says while the primary deterrent to testing is that people are afraid to find out their HIV status, names reporting is still high on the list of disincentives, and it’s counterproductive to do anything that could stand in the way of bringing in people for testing.

Plus, states could avoid the stigma that name-based reporting has and go with a unique identifier system, Zingale adds. "There’s mounting evidence that unique identifier systems are more reliable, and I assume that’s because of people’s lack of confidence in a name-based system," he says.

The CDC report disagrees. "On the basis of studies of coded identifier systems conducted in at least eight states, published evaluations of name-based and code-based surveillance systems, and CDC’s assessment of the quality and reproducibility of the available data, CDC has concluded that confidential name-based HIV/ AIDS surveillance systems are most likely to meet the necessary performance standards, as well as to serve the public health purposes for which surveillance data are required," it states.

For instance, the CDC conducted a three-year evaluation of reporting systems in Maryland and Texas, which used social security number-based, non-named unique identifiers. The evaluation found that about 30% to 40% of the reports had incomplete codes and only 25% to 50% of the reports were complete. Plus, neither state could assess the number of duplicate case reports.

When the CDC conducted additional research in California and New Jersey, the agency concluded that neither state found a non-name system that performed as well as a name-based method.

Another recent study questions whether the government should even spend money on a name-based HIV surveillance program because its public benefit is apparently so limited. The study, published in the Nov. 16, 1999, issue of Annals of Internal Medicine, was conducted by a team of AIDS researchers from the University of California at San Francisco and Berkeley who say name-based reporting programs for HIV infection are not producing any specific public health benefits related to controlling the epidemic.2

Data showed no effect either way

"We found both the positive and the negative ways that name-based surveillance programs supposedly affect public health interventions — specifically partner notification and access to health care — to be exaggerated," says Dennis Osmond, PhD, lead author of the study and an associate professor in the UCSF department of epidemiology and biostatistics and the UCSF Center for AIDS Prevention Studies.

"Our data showed no effect one way or the other," he adds. "The findings do open up discussion as to whether public health funds spent on these interventions are being used in the most effective way."

Gayle says the CDC merely wants states to collect accurate and complete HIV data, and the agency doesn’t dictate how they can do this.

"It’s been a difficult issue, and people should decide what is the best policy given the epidemic in their state," she says. "There’s enough of a concern about name-based reporting, and it’s important these days to have people know their HIV status, so we definitely wouldn’t want that to be overshadowed by surveillance needs."

This is why the CDC will work closely with states to assess whether their reporting practices have any impact on HIV testing behavior, Gayle adds.

However, if states do use a name-based system, and so far 34 states have passed laws to do so, then they should educate the public that the names are not sent to the CDC or stored and used for any purpose other than to make sure that one person with HIV isn’t counted twice.

That’s how all surveillance is done’

"The only reason the name is used is because that’s how all surveillance is done so far in this country. It’s the way you are coded on your medical record, and it’s the simplest way to enter somebody into a database," Gayle says.

There was one case in Florida where a state employee illegally disclosed the names of HIV-infected people, and that person was prosecuted. To prevent those types of occurrences, the CDC also recommends that states follow these security measures:

• Protect electronic HIV/AIDS surveillance data with computer encryption during data transfer.

• Place HIV/AIDS surveillance records in a physically secure area that is protected by coded passwords and computer encryption.

• Restrict access to the surveillance registry to a minimum number of authorized staff who have been trained in confidentiality procedures and are aware of penalties for unauthorized disclosures.

• Require security and confidentiality protections and penalties for unauthorized disclosures by all public health programs that receive HIV/ AIDS information.

• Researchers using surveillance data should be approved by appropriate institutional review boards and must sign confidentiality statements.

• Surveillance data used for epidemiological analyses must not include names or other identifying information.

• State and local health departments should promptly investigate and report any security breaches in HIV/AIDS surveillance data.


1. Fleming PL, Ward JW, Janssen RS, et al. Guidelines for national human immunodeficiency virus case surveillance, including monitoring for human immunodeficiency virus infection and acquired immunodeficiency syndrome. MMWR 1999; 48(RR-13):1-37.

2. Osmond DH, Bindman AB K, Vranizan JS, et al. Name-based surveillance and public health interventions for persons with HIV infection. Ann Intern Med 1999; 131.