What Do HIV Tests Test For?

Dear Christine,

I have a close friend in the AIDS industry who says that HIV tests are 99% accurate in detecting HIV infection. When I brought up information that questions the accuracy of the tests, she insisted rather vehemently that these ideas had been proven wrong. She also told me that Dr David Ho created an improved HIV test and that with this, Ho proved Duesberg and other AIDS critics wrong. She could not actually say though what HIV test actually finds the virus itself.

How would you respond to what my friend says? What exactly do the HIV tests that are on the market test for?

Linda C

Dear Linda,

According to test kit literature, none of the HIV tests on the market are intended for use in diagnosing actual HIV infection. If you read far enough into the fine print, you will find that they all make a point of mentioning this fact.

HIV antibody tests like the ELISA and Western Blot are widely believed to detect the presence of HIV specific antibodies, and these antibodies are then assumed to indicate the presence of HIV infection. But the tests can only register a person’s antibody response to a series of proteins thought to be specific to HIV. In fact, antibodies produced in response to a variety of non-HIV viruses, bacteria and other antigens such as foreign chemicals or sperm are noted in the medical literature as able to cause positive readings on HIV antibody tests. Further, the proteins used in HIV tests are not at all unique to HIV.
A careful reading of the ELISA test kit literature reveals that “prior pregnancy, blood transfusion or other exposures” may cause “non-specific reactions” which result in positive results. More condemning to the notion these tests are able to detect specific HIV antibodies is the statement in the test that “there is no recognized standard for establishing the presence or absence of HIV antibody in human blood.”

Moreover, no HIV antibody test has ever been validated for accuracy in determining HIV infection by establishing what percentage of people with positive antibody results have actual HIV that can be found in their fresh, uncultured plasma. This is the only way to effectively establish the accuracy of an antibody test in identifying those with current active infection. Sadly, AIDS researchers have never established how many people labeled HIV positive actually have the virus via the direct finding of HIV in their blood. Because of this, the ELISA test can state only that “a person who has antibodies…is presumed to be infected with the virus,” and that its claims for accuracy are based on assumptions and estimates rather than on studies that verified how many people who react positively actually have HIV.

Regarding the idea of improved tests, I believe your friend is mistaken. Even the most recently approved HIV tests use the same non-specific proteins and have no percentage of accuracy that’s been established by the direct finding of HIV in people who test positive.

Interestingly, the newest antibody test to hit the market characterizes HIV as “the virus correlated with AIDS.” The second most recent test calls HIV “the virus thought to cause AIDS.” If all were as scientifically certain about HIV as we are led to believe, wouldn’t these tests describe HIV as the virus proved to cause AIDS? Or at least the virus shown to cause AIDS?

The "viral load" tests your friend refers to was popularized by Dr David Ho and is widely believed to be able to detect actual HIV (virus). But her disagreement with those that challenge this notion is misguided as the test manufacturer places a carefully worded disclaimer in their literature that states "this test is not intended to be used as a screening test for HIV or as a diagnostic test to confirm the presence of HIV." So much for new and improved.

Your friend may not realize that Ho's claims of finding massive quantities of HIV rely on a complex mathematical formula that assigns a “viral value” to bits of RNA or DNA assumed to belong to HIV and multiplies these out to arrive at a number now known as an HIV viral load. Ho’s work essentially avoids addressing the obvious: If HIV is present in people who test positive in the large quantities he claims, it should be readily found in their fresh, uncultured plasma without all the mathematical gymnastics. Instead Ho asks us to believe that the detection of fragments of genetic material associated with HIV is more accurate in determining the presence and quantity of HIV than the detecting and quantifying of the actual virus.

To me, this is makes as much sense as an expert detective determining whether there’s an intruder in your house not by checking for or finding the actual culprit, but by finding feet print associated with a criminal. Would you be convinced by this indirect evidence that your house was safe or dangerous? Or like me, would you want the expert to just go from room to room to see if an intruder were actually there? While this is not a perfect example, it’s similar to what Ho’s viral load theory says AIDS doctors should do: Look for indirect evidence correlated with HIV, not look for whole infectious virus, and from this indirect evidence, say whole infectious virus has been found.

Even if Ho's viral load theory could be taken as evidence HIV is present in people who test antibody positive, presence alone is not enough to prove HIV causes AIDS or that Duesberg is wrong. To date, scientists are still without an explanation as to how or why HIV causes AIDS, giving that little sentence in the new tests (“HIV, the virus correlated with AIDS”) an undeniable significance.

Thanks for writing,


HIV Antibody Test Certificate of Accuracy

Related Articles


to Questioning The Test FAQ's - Why Don't People Who Test False Positive Sue?

to T Cells and Viral Load