Correspondence
More on Office-Based Testing for HIV
N Engl J Med 1993; 328:1717-1718June 10, 1993
- Article
To the Editor:
Recently, Gellert and colleagues expressed strong concern about office-based test systems for HIV, believing that such tests will compromise effective programs of AIDS surveillance and prevention1. We question their judgment. As the HIV epidemic continues to rage out of control, it is not clear that current prevention strategies are effective. It is imperative that new approaches be tried -- ones that also strive to protect uninfected persons from infection, thereby interrupting the transmission cycle.
Gellert et al. elected to focus on the predictive value of a positive HIV test (office-based), which they estimated to be 50 percent, given a sensitivity of 99.9 percent, a specificity of 99.6 percent, and an HIV prevalence of 0.4 percent in the tested population. By focusing on the harmful effects of misclassification among those who test HIV-positive, Gellert and colleagues overlooked the importance of office-based or home testing for personal decision making by those who want to avoid infection. Using the example of Gellert et al., an uninfected person could have lowered his or her risk of selecting an HIV-infected partner from 0.4 percent without testing to a low of 0.00004 percent among those who test negative -- a reduction of 99.9 percent. The reduction would be even greater with confirmatory testing.
As the epidemic increasingly affects heterosexuals, many search desperately for ways to reduce their risk of HIV infection. In addressing the heterosexual spread of HIV, Hearst and Hulley stated that the single most important recommendation for their patients is to avoid choosing a sexual partner who may be at risk of carrying HIV2. Because there are no discernible signs or symptoms for an average of 10 years, testing for HIV infection in blood or saliva3 is the only practical way to detect HIV carriers. Although centers performing both confidential and anonymous testing have been established for this purpose, many heterosexual couples may be reluctant to come to them, out of concern that they might be labeled as homosexuals, drug addicts, or prostitutes. When there are no incentives other than a vague concern about the risk of HIV, the discomfort of taking a blood test or discussing intimate sexual practices may be too much to endure. For such persons, office-based or home HIV-antibody testing would be a welcome alternative as a means of greatly reducing their risk of HIV infection.
3 ReferencesRalph R. Frerichs, D.V.M., Dr.P.H.
University of California, Los Angeles School of Public Health, Los Angeles, CA 90024Eugene Seymour, M.D., M.P.H.
Saliva Diagnostic Systems, Vancouver, WA 986821
Gellert GH ,Moore DF ,Weismuller PC ,Greenwood R ,Maxwell RM . Office-based test systems for HIV antibody. N Engl J Med 1993;328:211-211
Full Text | Web of Science | Medline2
Hearst N ,Hulley SB . Preventing the heterosexual spread of AIDS: are we giving our patients the best advice? JAMA 1988;259:2428-2432
CrossRef | Web of Science | Medline3
Frerichs RR ,Htoon MT ,Eskes N ,Lwin S . Comparison of saliva and serum for HIV surveillance in developing countries. Lancet 1992;340:1496-1499
CrossRef | Web of Science | Medline
To the Editor:
The letter by Gellert et al. raises “serious concern about clinical or operational feasibility” with respect to the use of visually interpreted antibody assays to detect HIV infection. Unfortunately, the authors have done their cause a disservice by discussing two unrelated assays in a misleading way. Their letter reveals a fundamental misconception of the basis on which the Food and Drug Administration has granted approval to Fluorognost HIV-1 IFA, the only indirect immunofluorescence assay for HIV antibodies licensed to date.
Fluorognost has been licensed primarily as an additional, more specific test for validation testing in reference centers. As such, it can be used to provide conclusive information on the HIV-antibody status of serum or plasma specimens, whereas the only approved use of the “microfiltration enzyme immunoassay” mentioned is in the primary testing (or screening) of previously uncharacterized specimens. The potential of Fluorognost to resolve “indeterminate” Western blot results has been demonstrated1. This fact alone is sufficient to invalidate any direct comparison between these two assays: they belong to entirely different, well-defined categories. An additional licensed use (to screen uncharacterized specimens in physicians' offices, clinics, emergency rooms, and other settings in which enzyme immunoassays are impractical or unavailable) was granted to Fluorognost only because it has been proved to be as sensitive as, but more specific than, standard enzyme immunoassays, on the basis of data from a multicenter trial and a clinical study of 1300 subjects at high risk of acquiring HIV infection2. Only 1 false positive result was seen in 10,082 blood donors. The notion that “8 positive results [will result] per 1000 tests, 4 of which are false positive results” is therefore incorrect with respect to Fluorognost.
Furthermore, by stating that “in both assays . . . the interpretation of the assay depends purely on visual skills,” the authors imply that visual evaluation is inferior. In doing so, they fail to take into account that tools for the determination of HIV serologic status have always relied exclusively on such evaluation.
On studying the product insert for Fluorognost, Gellert et al. would have found detailed summaries of the extensive clinical studies on which the FDA has based its license.
2 ReferencesHermann A.M. Mucke, M.Sc., D.Sc.
Alexander Haushofer, M.D.
Waldheim Pharmazeutika, Ges.m.b.H., A-1091 Vienna, Austria1
Sullivan MT ,Mucke H ,Kadey SD ,Fang CT ,Williams AE . Evaluation of an indirect immunofluorescence assay for confirmation of human immunodeficiency virus type 1 antibody in U.S. blood donor sera. J Clin Microbiol 1992;30:2509-2510
Web of Science | Medline2
Instructions for laboratory training and qualification: Fluorognost HIV-1 IFA. Product insert vers. 2.02. Vienna, Austria: Waldheim Pharmazeutika, May 1992.
Author/Editor Response
The authors reply:
To the Editor: We do not agree with the perspective offered by Drs. Frerichs and Seymour on the importance of office-based or home testing for personal decision making in HIV prevention. Because of the latency period between the acquisition of the virus and the appearance of antibody to HIV, this approach to disease control, in which people would conduct tests at home before engaging in sexual relations, offers little real security to the public. People who use the test but persist in high-risk behavior can be incorrectly self-identified as uninfected when in fact they are HIV-positive. Furthermore, given the reality that many people who are not otherwise at elevated risk will engage in heterosexual sex with multiple partners in a given period, this approach is not likely to be practical from the perspective of individual compliance.
The idea of a quick test as the mainstay of disease control offers only the illusion of security. What these authors have overlooked is that successful HIV control must integrate behavior-based strategies. We also differ with the view that avoiding the selection of a partner who is at risk of carrying HIV should be the core of HIV prevention; rather, avoidance of high-risk sexual practices, such as unprotected sexual intercourse, should be central to disease control until a vaccine is developed. Any alternative is just too unreliable and risks precisely the backlash against testing that the authors cite.
Our concern is not with the concept of office-based test systems in itself, but rather that unleashing this type of technology unaccompanied by behavior-based intervention will undermine efforts at prevention. Counseling about high-risk sexual practices, voluntary notification of partners, psychosocial support, and referral for follow-up must be integral to any widespread testing strategy. Physicians can and should test their patients for HIV antibody, but they should employ these vital elements and use a laboratory with the capacity for confirmatory testing. Office-based testing requires additional support and the training of office personnel to conduct these activities. Finally, anonymous-test-site data from Orange County, California, do not support the contention that heterosexuals will not use a comprehensive testing service for fear of social stigmatization; 23 percent of 11,982 tests in 1991 and 28 percent of 14,640 tests in 1992 were provided to people whose only acknowledged risk factor was heterosexual contact (heterosexual injection-drug users are not included in these figures).
We agree with Drs. Mucke and Haushofer that the Fluorognost HIV-1 IFA should be used as a confirmatory assay. However, the product insert clearly states in the “intended use” section that it can also be used in physicians' offices. The term “primarily” with respect to use does not appear in the insert as it does in their letter. This agency has communicated to representatives of Waldheim Pharmazeutica the need to delete the statement about nonlaboratory use from the product insert. The additional licensed use in offices or other nonlaboratory settings is inappropriate, and the points made above and in our original letter about compromising prevention activities apply.
The ELISA used by most conventional laboratories is not, in fact, read visually, but is read by instrumentation, and although the results of Western blot assay are interpreted visually, it does not rely on color discrimination. We do not mean to imply that visual evaluation is inferior, but that such evaluation requires trained laboratory technologists, whom many office practices do not employ. In fact, Waldheim “strongly” advises in a 13-page supplement that technicians become qualified and be evaluated by the manufacturer1. Furthermore, these personnel should be evaluated to ensure that they can read the test and discriminate color correctly, as recommended by the College of American Pathologists2.
2 ReferencesGeorge Gellert, M.D., M.P.H., M.P.A.
Douglas F. Moore, Ph.D.
Rick Greenwood, Ph.D., M.P.H.
Roberta M. Maxwell, Ph.D.
Penny C. Weismuller, R.N., Dr.P.H.
Orange County Health Care Agency, Santa Ana, CA 92701- Citing Articles (3)
Citing Articles
1
Kathryn A. Phillips, Susan Fernyak. (2000) The cost-effectiveness of expanded HIV counselling and testing in primary care settings: a first look. AIDS 14:14, 2159-2169
CrossRef2
R.R Frerichs. (1994) Personal screening for HIV in developing countries. The Lancet 343:8903, 960-962
CrossRef3
Ralph R Frerichs, Eugene Seymour. (1993) HIV testing and blood recipients. The Lancet 342:8886-8887, 1549
CrossRef
