Interpreting antibody tests
ELISA testing alone cannot be used to diagnose HIV, even if the test suggests a high probability that antibody to HIV-1 is present. In the United States, such ELISA results are not reported as "positive" unless confirmed by a Western Blot.
The ELISA antibody tests were developed to provide a high level of confidence that donated blood was NOT infected with HIV. It is therefore not possible to conclude that blood rejected for transfusion because of a positive ELISA antibody test is in fact infected with HIV. Sometimes, retesting the donor in several months will produce a negative ELISA antibody test. This is why a confirmatory Western Blot is always used before reporting a "positive" HIV test result.
Rare false positive results due to factors unrelated to HIV exposure are found more often with the ELISA test than with the Western Blot. False positives may be associated with medical conditions such as recent acute illnesses and allergies. A rash of false positive tests in the fall of 1991 was initially blamed on the influenza vaccines used during that flu season, but further investigation traced the cross-reactivity to several relatively non-specific test kits.[18] A false positive result does not indicate a condition of significant risk to health. When the ELISA test is combined with Western Blot, the rate of false positives is extremely low, and diagnostic accuracy is very high (see below).
HIV antibody tests are highly sensitive, meaning they react preferentially with HIV antibodies, but not all positive or inconclusive HIV ELISA tests mean the person is infected by HIV. Risk history, and clinical judgement should be included in the assessment, and a confirmation test (Western blot) should be administered. An individual with an inconclusive test should be re-tested at a later date.
This emphasises the point I was making in regard to the type of kit they used being a possible factor.