Replies

I would be seriously interested in the immunological evidence behind the "most" in that sentence. Because I have a hunch it may have been pulled out of somebody's tailbone, and that actually they flat do not know how many are in the one category, and how many are in the other.

Well, since most patients are progressors and those patients are the ones whose CD8 functions are impaired, and the small balance of non-progressors has a certain percentage whose CD8 functions are more like normal, it's not an unreasonable sentence.

You have to include everything in the selecting of non-progressors. Which does not begin from "patients". When someone develops AIDs, you know they have it. (They have HIV, because that is part of the definition of AIDs). Then some group of contacts know they are at risk and are tested. Some get tested on their own initiative because they know they are in risk groups, or because they are particularly nervous or careful or can afford it or some institution makes them. Nothing like everyone exposed gets tested, and those that are involve detection failures, and a distribution of time lags between exposure and detection, etc.

Then for another monkey wrench, take the case of a mutated strain that infected 7 in Australia through a blood screening failure. None of the 7 has developed AIDs. Why do we even know it happened? Because of the manner of infection - transfusion - and screening procedures caught the HIV positives despite a complete absence of symptoms. Now, can that happen with sexual tranmission in some subpopulation in New Orleans? Sure. Would be ever know about it? We'd probably eventually see a portion of the HIV positives, because of a random testing "hit" within that subpopulation etc. Would we see all of it? No. Would such cases show up in a general survey of the incidence of HIV positives? A random one yes, a patient one no.

You have to be very careful with definitional and observational bias issues when making statements about epidemic categories...