Do HIV Tests Detect HIV?

March 5, 2012

(Español) In 1993, Australian researchers published a report that questioned whether Western Blot HIV tests can prove that an HIV infection exists.  Although their report and references raised significant questions about the reliability of such tests, it’s technical language and 161 scientific references make it difficult to understand. 

Since then, scientists like  Henry Bauer Ph.D, medical doctors like Etienne de Harven MD, and writers like Liam Scheff have posted more readable reports and links about the topic.  But because it is difficult to explain the pharmaceutical shell-game of HIV test marketing, it is still hard for many readers to navigate through the shell-game that these researchers have painstakingly untangled.

This report does not attempt to improve the excellent reporting from these or other writers.  Instead, it corroborates previous analyses in plainly-written language so that readers can easily understand the problems associated with HIV testing.


To date, FDA-approved diagnostic tests for medical use related to HIV neither claim to, nor even attempt to, confirm the actual presence of HIV in a patient’s sample.  In fact, none of the HIV-related diagnostic tests or devices currently listed on the FDA’s Website claim to confirm the actual presence of HIV in any sample with any degree of stated accuracy.

The standard lingo “HIV test,” “testing HIV positive,” or a “confirmatory test for HIV,” is misnomer because none of these tests confirm the actual presence of the virus in a patient’s sample.  As such, all diagnoses of HIV infection in patients are presumptive, based on their behavioral and environmental issues, risk factors, immune status, and clinical condition (actual health) in combination with other indirect diagnostic evidence for either the past or current presence of HIV in a blood, urine, or saliva sample.

While FDA-approved “HIV tests” are intended for use in diagnosing the presence or absence of either antibodies to HIV in a sample, or of molecular fragments of HIV in a sample, none of these tests detect an actual virus.  Nevertheless, based on guidelines and recommendations put forth by the CDC in 1987, virtually all diagnoses of HIV infection since that time have been, and still are, based solely on combinations of HIV antibody tests.

Antibodies are proteins that the body’s immune system produces in response to the presence of germs, such as bacteria and viruses, and do not represent the germ itself.  In fact, these antibodies assist in the elimination of the germ by binding to it, thereby interfering with its ability to replicate, and marking it for digestion by other cells in the immune system.

Germ-specific antibodies produced in this way remain at detectable levels in the body for several months to several years, even after the complete elimination of the infectious agent from the body.  By analogy, footprints from a mountain lion in a forest do not necessarily mean there is a danger.  It could mean the lion has gone away or has been captured.  Similarly, antibodies to a germ do not mean the germ is present.  Indeed, in the vast majority of cases (such as polio, small pox, measles, hepatitis, chicken pox), it means the germ has been neutralized and eliminated from the body.

The two primary classes of so-called “HIV tests” for detecting antibodies to HIV are known as the ELISA (or EIA) and the Western Blot (WB). The detection of antibodies to HIV in a sample is intended to aid physicians in their presumptive diagnosis of infection with HIV.  However, neither test confirms, or is intended to confirm, the actual presence or absence of virus in a sample.

According to current CDC guidelines and recommendations, when a blood, urine, or saliva sample from a patient is presented for testing, it is initially screened for the presence or absence of antibodies to HIV using the ELISA test. If positive, it is retested and, if positive again, the probability that antibodies to HIV are present is considered to be high.  However, because “non-specific results [false positives] are found commonly when screening tests are used in low risk populations,” and “(s)ince the psychosocial and medical implications of a positive antibody test may be devastating,” samples repeatedly reactive on screening assays such as the ELISA should be further tested using an additional, more specific test for HIV antibodies, such as WB. (Epitope/Organon Plasma)

The FDA-approved WB kits are intended for use “as an additional, more specific test for HIV-1 antibodies,” in persons already found to be repeatedly reactive on screening tests such as ELISA.  However, paradoxically, the manufactures of such products inform us that a sample testing positive on both the EIA screening test and the WB can only be “presumed to be positive for antibody to HIV-1.” (emphasis added. OraSure, Epitope/Organon, and Bio-Rad)

The reason that manufacturers of these “more specific tests” no not estimate the probability that such a sample has antibodies to HIV, is because there is currently “no recognized standard for establishing the presence or absence of [HIV] antibody in human blood” (Abbott ELISA).  The manufactures of HIV-antibody tests also emphasize: “A person who has antibodies to [HIV] is presumed to be infected with the virus.” (emphasis added. Epitope/Organon, Bio-Rad, Abbott ELISA.) The reason one can only presume this is that according to the scientific literature, “there is no ‘gold standard’ laboratory test that defines the true infection status” (Sheppard (1991).  Recognizing that a sample testing positive on both ELISA and WB is only presumed to be positive for antibodies, and further presumed to therefore be positive for the virus, the manufacturers of WB tests agree that the best one can conclude from such a result is that it “… may indicate infection with the HIV-1 virus” (emphasis added. OraSure, Epitope/Organon, Cambridge Summary).

Consequently, persons testing “positive” on both ELISA and WB are only presumed to be positive for antibodies to HIV and presumed to be infected with the virus, which means they may (or may not) be infected with HIV.  Despite these facts, the CDC recommends – and it has become the standard practice of “the medical community” to use such a combination of test results to tell patients that they have, with certainty, a confirmed infection with virus itself (HIV).  This is also the case for asymptomatic patients (those without symptoms) or risks consistent with the possibility of infection.

This is particularly disturbing when one considers that with the release of the first WB approved for medical use, the FDA announced: “The significance of antibodies in an asymptomatic individual is not known.” (emphasis added. Suzan Cruzan. FDA News Release. April 30, 1987; P87-11.)  The manufacturers of WB tests currently on the market likewise emphasize: “The clinical implications of antibodies to HIV-1 in an asymptomatic person are not known.” (emphasis added. Cambridge, Bio-Rad.)

This is why test manufacturers warn: “Do not use this kit as the sole basis of diagnosis of HIV-1 infection;” (Bio-Rad, Epitope/Organon, OraSure, Cambridge) and emphasize for all persons testing positive on both ELISA and WB: “Clinical correlation is indicated … to decide whether a diagnosis of HIV infection is accurate.” (Bio-Rad, Epitope/Organon).

These facts suggest that the currently accepted practice of telling people that they are infected with a deadly virus that has no cure, based solely on the results of antibody tests (i.e. not actual confirmation of the virus) falls well outside of the FDA-approved use of these products and is scientifically unsound.   In part, this disparity between the positions taken by the FDA and CDC can be explained by the fact that the FDA functions under a legal mandate to base product approval on reproducible scientific observations demonstrating statistically significant results.  The CDC, however, has no such legal restrictions and can therefore base recommendations solely on theoretical considerations, or simply “expert opinion” – experts who are typically paid to represent the interests of test manufacturers.

On the surface, the fact that the CDC would recommend the use of antibody tests for the purpose of telling persons that they are, with certainty, infected with a deadly and contagious virus – when the manufacturers of these tests warn that the tests cannot be used to establish (with any degree of stated certainty) that this is the case – may seem so incredulous that it cannot be true.  This is particularly the case for persons who were young adults in the late 1980s and 1990s who recall reading in reputable organs of the popular press that the tests being used to diagnose infection with HIV are “greater than 99.9 percent accurate.”

Next Page

Pages: 1 2

Tags: , , , , , ,

You must be logged in to comment

Log in