Case Study

Twenty-two countries participated in the studies herin referenced. A total 0f 3,867 patients and/or subjects were tested and reported. Paired specimens of serum and saliva were tested. The serum specimens were tested whether or not they were known positive or negative. The subject's serum had to be tested with both ELISA and Western Blot. The saliva samples were tested by the adjusted rapid ELISA test which is contained in the HIVhometest test kits.

The following are the results of the tests reported:

Of the 3,867 patients and/or subjects which were tested and reported, 368 were positive with the HIVhometest. Of the 368 positives with the HIVhometest, 367 were positive with ELISA test on the serum and 365 were positive with confirmation tests using Western Blot. Comparing the HIVhometest with the Western Blot gave a sensativity (frequency of a positive test in individuals with the HIV disease) of 99.18%.

Of the 3,498 patients and/or subjects which were tested and reported negative with the HIVhometest, 3,493 were negative with ELISA and 3,498 were tested negative with confirmation tests using Western Blot. Comparing the HIVhometest with the Western Blot gave a specificity (frequency of negative test results in individuals without the HIV disease) of 100%.

It must be noted that there were not any tests performed at a later date to determine if any of the negative subjects were tested during a time when the disease did not produce sufficient antibodies to be detected. Although the chance of a window (inability to detect antibodies) may occur, the author has made his conclusion based upon the assumptions that none of the tests was conducted during this window period. To eliminate this possibility, it would be necessary to test for viral load using the Polymerase Chain reaction test (PCR) for RNA or DNA.

In summary: Sensativity: 99.18% Specificity: 100%

It can be concluded from this study that the HIV ELISA is overly sensative (false positive) and inaccurate (false negative); while the saliva HIVhometest is only overly sensative.(false positive)

This information proves that the ELISA is a bad screening test. Any comment regarding this information?

Great summary. Wrong conclusions. Why don't you calculate the false positive and false negative rates for all the tests you cite and then explain why they are bad screening tests. MH