Standard HIV antibody tests are currently done in central laboratories using batch-oriented enzyme immunoassays. A test result is reported as negative if the enzyme immunoassay result is negative. The result is reported as positive if the enzyme immunoassay results are repeatedly reactive and if the results of a more specific, supplemental confirmatory test, such as the Western blot or the immunofluorescence assay, is also positive. This testing can theoretically be done within 24 hours, but physicians and patients often wait more than a week for results because of test batching or a dearth of qualified personnel [8].

Enzyme immunoassay routinely takes 3 to 4 hours, but rapid tests for HIV, such as the one studied by Irwin and colleagues [7], can typically be done in 10 to 15 minutes. The sensitivity and specificity of rapid tests range between 0.99 and 1.00 [7, 9-15]. Thus, a rapid test for HIV could be substituted for the traditional enzyme immunoassay in the standard testing algorithm. Persons could then be notified of their preliminary results at the time of testing. Because a rapid test has a negative predictive value equivalent to that of traditional enzyme immunoassays, persons with negative test results would not need to return for confirmed results. Persons with reactive test results could be notified of their potentially positive HIV status, counseled accordingly, and asked to return for confirmatory tests and further post-test counseling.

When HIV testing was first implemented, the U.S. Public Health Service (USPHS) guidelines recognized the value of reporting preliminary positive results [16]:

If additional more specific test results are not readily available, persons in high-risk groups with strong repeatedly reactive ELISA [enzyme-linked immunosorbent assay] results can be counseled, before any additional test results are received regarding their probable infection status, their need for medical follow-up, and ways to reduce further transmission.

As the availability of supplemental tests increased and the results of these tests could be reported on the second visit, subsequent USPHS guidelines [17] recommended that

no positive test results be given to clients/patients until a screening test has been repeatedly reactive on the same specimen and a supplemental, more specific test such as the Western blot has been used to validate these results.

New developments in rapid testing technology allow patients to receive enzyme immunoassay results during the initial visit. Existing USPHS guidelines may need to be reevaluated in light of these new developments.

Although providers of HIV tests are not accustomed to providing counseling on the basis of preliminary results, many medical precedents exist for reporting the results of tests used to screen for other potentially fatal diseases, such as mammography for breast cancer before biopsy and Papanicolaou smears for cervical cancer before colposcopy. In addition, reporting the preliminary results of HIV assays may significantly increase the number of persons who are aware of their HIV serostatus [8], because follow-up for HIV test results is poor in many settings. National statistics in 1994 showed that 40% of clients at publicly funded HIV testing sites failed to return for test results and post-test counseling [18].

In hospital settings, standard testing for HIV routinely takes 24 hours or longer. By the time a physician discovers that a patient is HIV positive, that patient has often been discharged. In some cases, the patient never learns his or her HIV status. Irwin and colleagues [7] describe the performance characteristics of the Genie HIV-1 and HIV-2 assay (Genetic Systems, Seattle, Washington), an investigational rapid HIV test. In their study, 16 of 45 infected patients never returned for their HIV test results; reporting preliminary results of rapid tests would have allowed more persons to learn their HIV status and to receive more specific medical treatment of HIV-related complaints. An opportunity was also missed to inform and counsel noninfected patients, more than one third of whom never returned to learn the results of their tests. Rapid testing for HIV also has the potential to improve clinical outcomes in prenatal settings, in which women at risk for HIV infection often present late in pregnancy without having been tested. Reactive rapid test results could indicate the need for urgent confirmatory testing, providing timely information that could assist in decisions about preventive antiretroviral therapy and obstetric procedures to minimize risk for transmission [4]. Rapid testing of the source patient could also assist in decisions about preventive antiretroviral therapy for hospital staff members percutaneously exposed to infected blood [19].

Rapid HIV testing could improve the delivery of medical care and HIV-prevention services, but several barriers block implementation of this technology. Providers of HIV testing and counseling may be reluctant to notify and counsel patients on the basis of a screening test result because of the possibility of false-positive results. Many providers are concerned that persons with false-positive results may be placed at unnecessary risk for psychological stress or suicide. However, the literature in this area is reassuring: Notification of positive or negative HIV status is associated with decreased suicidal ideation or decreased hopelessness [20]. Furthermore, studies evaluating suicide rates among persons notified of positive HIV test results did not show that these persons had an increased risk for suicide [21]. In addition, a study of HIV prevention counseling that used a rapid HIV antibody test at a sexually transmitted disease clinic and at an anonymous testing site found high rates of counselor acceptance of and patient satisfaction with counseling done on the basis of preliminary results of rapid tests [22].

If preliminary positive results are given on the basis of a single rapid HIV test, the positive predictive value of the rapid test (which depends on the prevalence of HIV in the population being tested) must be taken into account during counseling. If additional rapid tests become available for clinical use in the future, the positive predictive value of the screening test result could be improved by the use of two different rapid tests in combination [23]. Unfortunately, this algorithm cannot currently be used because only one rapid HIV test, the Single Use Diagnostic System HIV-1 Test, is approved by the Food and Drug Administration (FDA) and available for clinical use.

The Genie HIV-1 and HIV-2 assay performed well in the study by Irwin and colleagues [7] and may be an excellent second test to use in combination with the Single Use Diagnostic System HIV-1 Test. However, the manufacturer is not seeking FDA approval for clinical use in the United States, and manufacturing of the product has been moved to France. This story is quite common. Since 1987, at least 14 rapid HIV tests have been developed [9-15, 24]. Only 2 of these tests, the Single Use Diagnostic System HIV-1 Test [9] and Recombigen-LA-HIV-1 [24], have received FDA approval for use in the United States. Recombigen-LA-HIV-1 was the first rapid test to be approved by the FDA, but it did not gain wide acceptance because it was not as accurate as traditional enzyme immunoassays in some settings [15]. Therefore, this test is not currently available in the United States. Although many rapid tests for HIV have been developed by manufacturers in the United States, none of these manufacturers is currently pursuing FDA approval. Most companies have moved their manufacturing sites to countries in which rapid tests are used for clinical diagnosis and for blood screening before transfusion. The reasons for producing and marketing rapid HIV tests outside the United States are complex and include the perception of a small market in the United States, the cost and time associated with approval, and difficulties in securing patents for various test components. Manufacturers of rapid HIV tests should be encouraged to seek approval of their products in the United States so that 2 rapid HIV tests could be used in combination to allow for more precise counseling of persons with positive screening test results.

Irwin and colleagues [7] show the potential benefits of rapid HIV testing in hospital settings. Given these benefits, the broader implementation of this technology combined with appropriate counseling should be considered for persons presenting to emergency departments who have symptoms of infections that might be related to HIV or who have behaviors or conditions suggesting a high risk for HIV infection. Rapid tests for HIV can be accurate and cost-effective [8], and their use may improve clinical outcomes and facilitate HIV-prevention activities in various clinical settings. If the question is how best to provide HIV testing and counseling services, rapid testing for HIV may be an important part of the answer.

Dr. Kassler: Division of HIV/AIDS Prevention, Mailstop E-45, Centers for Disease Control and Prevention, Atlanta, GA 30333.