Write a three-page paper in which you take a definite pro or con position on the following statement: "State governments should require that the names of individuals who test positive for HIV should be submitted to a case registry maintained by the state's department of public health."

---

This question is based on a policy issued by the Centers for Disease Control in 1999 (or so) urging states to pass laws that HIV disease should be a reportable disease where physicians not only must report instances of HIV among their patients but also that they must submit the names of the patients. This led to intense opposition from AIDS activist groups like ACT-UP. We have provided a link to the ACT-UP web site that has lots of material arguing against the CDC policy recommendation.

You will notice that much of the material related to this issue is five years old. The issue is not on the front burner for ACT-UP but they do keep this page active. The issue also is a live one as our textbook indicates. Not all states have complied with this recommendation with something like 26 (or 35 in another report) going along. Our book shows us that Pennsyvlania is not one of them nor are several other large urban states with outspoken activist groups among the names reporting states.

There used to be a CDC page that gave the argument for names based reporting but I have not been able to find it. A better librarian than me might be able to unearth it since I suspect it still is on the CDC site someplace. You can get a good idea about the arguments for names based registries if you do keyword searches on MEDLINE. I entered AIDS and names based registries and got about 7 hits. AIDS and registries produced more. More experimentation with key words might produce some clearer explanations than I could find, but even these searches give an idea of what the arguments are in favor of the names-based registries.

First, Maryland experimented with a names based system in which they created a complex ID number system that removes actual names from the records and retains the key only in some central location. This means that the contents of the records would be hard to leak and there is high probability that confidentiality could be maintained at the state level.

Second, it is important to be able to track the incidence of HIV disease. This is more important now that there are effective early stage treatments. We used to be able to track AIDS when opportunistic infections showed up. Now they can lie dormant in those affected for many years and so there is no obvious way to identify patients or track the incidence of the disease.

This is especially important for some of the populations we have talked about, like heterosexual teen age women in rural areas. Since they are not included in the usual definition of risk groups they are not targets of systematic epidemiological study and so we have very little information about whether the incidence of the disease is growing in this population or how the dynamics of transfer works. I found one paper that used names based registry to report new information about this population and this seems like important and useful information to have.

Third, there is a distinction between names-based registries and non-name based registries, and there is disagreement about their relative effectiveness. The older argument for name-based registry is that it produces more accurate and higher quality information than non-names based registries. This is the case because when practitioners fill out forms reporting cases they make mistakes or leave out critical information. This makes it hard to know whether reports are accurate and it also makes it hard to do the epidemiological research we need to understand the spread of the epidemic. Names based registries improve the quality of information because we can cross-reference cases using the patients' names and create more accurate information. I saw one study that said names-based registries were over 98% accurate even with low quality information in records where non-names based registries were about 75% accurate. This was taken as an argument for non-names registries since I gather older studies found much less accuracy in the non-names registries.

On the no side of this issue, there are a number of arguments related to the importance of confidentiality and also patient control of information. One simple concern is that information may not be secure even with coding systems since the form still must be filled out in the local health setting and confidentiality may be compromised.

Whether or not this is a realistic concern, the question arises whether people will seek HIV testing if they are aware that results will be turned in to the state with their names attached. Many citizens would see this as a variety of state surveillance and they would not necessarily trust that their names would not be turned over to other government authorities. For example, if someone had a warrant out for their arrest on a felony could courts subpoena names on the AIDS registry to find a fugitive? Similarly, if the Department of Homeland Security judged an individual a terrorist threat, could they access these records? Since the Department of Homeland Security seems to have complete power to access anything, would foreign nationals residing in the U.S. who might be judged illegal aliens be willing to be tested?

One of the overriding concerns where HIV testing is concerned has always been that the potential for identification and stigma might chase people away from getting tested. Since HIV+ status is one of the most stigmatizing identities you can have people would most likely avoid getting tested, would not receive medical care, and would continue to be risky to their intimate partners if they were frightened from being tested. These fears seem especially justified when we have a government that appears to be homophobic and that also pursues an aggressive policy to imprison illegal drug users.

One of the arguments for non-names based registries is that a 75% accuracy rate is good enough for research use. You could also argue that unwillingness to be tested would introduce another bias into results so that even with 100% accuracy names-based registries would not accurately represent the infected population.