When the healthy immune system is exposed to HIV antigens (antigens are bits of the protein structure of the virus), it will produce proteins called antibodies that stick to the viral particles and help clear them from the body. That HIV ultimately evades this system is why the disease is so serious. HIV tests use artificially produced HIV antigens that can capture antibodies if they are present in a person's blood. If a person has never been exposed to HIV, then no antibodies will stick to the antigen in the test and it will read negative. But if the antibody test is positive, it means that the person has had an immune reaction to HIV in the past and is probably infected. These tests are called immunoassays.
The usefulness of an HIV antibody test is judged by its sensitivity and specificity. Sensitivity is the ability of a test to detect HIV even when present in very small amounts. If a test is not sensitive enough, some positive samples will slip though the screen. But for a test to be sensitive enough to detect 99.9 percent of positive samples, it may sometimes read positive when no HIV is present. This is called a false positive result. The other measure of a test's reliability is its specificity. This tells how well the test discriminates between detecting HIV and other somewhat similar antibodies. There is usually a tradeoff between sensitivity and specificity, with highly sensitive tests being more likely to be fooled into giving a false positive answer by similar but non-HIV antibodies.
The trade-off between sensitivity and specificity is why a positive result on an initial HIV test must always be confirmed with a second, more specific, HIV test before telling someone they are positive. Highly sensitive tests are ideal for jobs such as screening the blood supply or performing anonymous surveillance of HIV prevalence in a population. In particular, blood screening is a job where it is better to be safe than sorry; a few false positives are not of concern. But a positive HIV diagnosis is of huge concern to the person who gets one, so the system is careful to be sure the diagnosis is correct.
Rapid testing brings a new challenge to the established system of confirming a diagnosis before telling someone his or her results. If a simple finger prick device can give a result within 15 minutes (similar to a home pregnancy test), how does a test provider deal with a positive result? One way is to tell individuals that they have had an inconclusive result then draw more blood to send to a lab for the conventional test and confirmation process. So while a single rapid test may be fine for alleviating the anxiety of the uninfected, it may not be the best solution for someone who really has HIV.
In parts of the world where conventional laboratory-based tests are often not available, a number of rapid tests have come into common use over the past decade. The World Health Organization (WHO) has developed a set of protocols for using combinations of rapid tests to deliver reliable, confirmed results, depending on the purpose and context of the test.
The result is a system capable of providing test results with the same confidence as laboratory-based testing but at a far lower cost. In addition, the availability of same-day results means that many more people learn their HIV status and remain available for counseling about their health. Studies have shown that people who learn their diagnosis are more likely to begin practicing risk reduction behaviors than those who fail to return for their results.
The WHO protocol recommends confirmation with multiple different rapid tests depending on the objective of the test and the background prevalence of HIV in the region. A single test may be sufficient for screening blood or performing surveillance studies in high prevalence regions. For individuals with symptoms of HIV disease living in a high prevalence area, a single test may also suffice. In areas with a lower background prevalence of infection, the proportion of false positives from too-sensitive tests demands additional confirmatory testing.
In settings where multiple confirmatory tests are expected, the first test performed should have very high sensitivity to insure that all true positives are captured. Since these first pass tests will tend to report a higher number of false positives, the second, confirmatory test should be highly specific for HIV. In multiple test systems, each test should use a different antigen to avoid overlapping specificity. One drawback to rapid assays is that a new sample may need to be obtained if an individual requires a confirmatory test. WHO recommends collecting serum, plasma or dried blood spots if multiple test strategies are used, to avoid having to collect multiple samples.
Back to the GMHC Treatment Issues February 2002 contents page.