1 figure, 1 table omitted
In 2003, an estimated 1 million persons in the United States were living with human immunodeficiency virus (HIV) infection.1
Approximately 25% were unaware of their infection1
; however, that percentage might have been greater among persons at high risk for HIV infection, including racial/ethnic minority populations.2,3
To increase the proportion of persons aware of their HIV serostatus,
CDC launched the Advancing HIV Prevention initiative in 2003.4 One strategy of the initiative is to implement new models for diagnosing HIV infections outside medical settings.
During 2004-2006, CDC funded a demonstration project to provide rapid HIV testing and referral to medical care, targeted to racial/ethnic minority populations and others at high risk in outreach and other community settings. This report summarizes the results of that project,
which indicated that, of 23,900 clients who received a rapid HIV test,
39% were non-Hispanic blacks, 31% were Hispanics, 17% reported male-male sex, and 6% were injection-drug users. A total of 267 (1%) persons had confirmed HIV-positive test results; of these, 195 (74%) were either non-Hispanic blacks or Hispanics. The project results demonstrate that rapid HIV testing in outreach and other community settings can identify large numbers of persons in racial/ethnic minority populations and others at high risk who are unaware they are infected with HIV.
Rapid HIV testing was conducted by eight community-based organizations (CBOs) in seven U.S. cities: Boston, Massachusetts; Chicago, Illinois;
Detroit, Michigan; Kansas City, Missouri; Los Angeles, California;
San Francisco, California; and Washington, D.C. (DC). CBOs identified testing venues where persons at high risk congregated, resided, or sought medical care (e.g., parks, shelters, hotels, clubs, health fairs, syringe-exchange sites, and community clinics). Trained CBO staff members offered counseling and rapid HIV testing to clients either in mobile testing units or inside venues. Persons eligible for testing were those capable of providing written, informed consent who met age of consent criteria for HIV testing in the state in which the CBO was operating; persons not meeting these criteria and persons with a previous diagnosis of HIV infection were excluded. CBO staff members collected information from persons tested regarding their demographic characteristics, risk behaviors, and HIV testing history.
HIV testing was performed with rapid tests (Oraquick® Rapid HIV-1
Antibody Test or OraQuick® Advance™ Rapid HIV-1/2 Antibody Test [OraSure Technologies, Bethlehem, Pennsylvania]) on either oral fluid or whole-blood specimens, and results were provided to clients 20-40 minutes after specimens were collected. For persons with reactive (i.e., preliminary positive) rapid test results, testing staff members collected either oral fluid or whole-blood specimens for confirmatory Western blot testing and scheduled a follow-up appointment to give the client the confirmatory test results. HIV-positive persons who returned for confirmatory test results were referred to clinics affiliated with participating CBOs or to other local health-care providers for medical care.
Of 24,172 persons who agreed to be tested, 44 persons did not meet age of consent criteria, and 84 persons reported a previous diagnosis of HIV infection. Data on the total number of persons offered testing were not collected. Of the 24,044 persons who met eligibility criteria for participation and agreed to be tested, 144 were excluded from the analysis because they either did not receive their rapid HIV test results or had missing test-result information. A total of 23,900
persons were included in the analysis: 5,536 from Los Angeles; 5,162
from Boston; 4,586 from DC; 2,985 from Kansas City; 1,931 from San Francisco; 1,868 from Detroit; and 1,832 from Chicago. Among participants,
39% were non-Hispanic blacks, 31% were Hispanics, and 21% were non-Hispanic whites. Sixty-three percent of participants were male, 50% reported not having any public or private health insurance, 40% reported not visiting a health-care provider during the preceding year, and 9%
reported being homeless.
Sixty-six percent of participants reported having multiple sex partners, 17% reported male-male sex, and 6% reported injection-drug use during the preceding year. A total of 7,034 (30%) participants had never been tested for HIV; among the 16,543 (70%) who had been tested, 6,982 (43%) had not been tested during the preceding year.
Of 14,096 persons who had seen a health-care provider during the preceding year, 6,257 (44%) had received an HIV test during that period, and 3,299 (24%) had never been tested for HIV, including 19 persons who were confirmed to have HIV infection.
A total of 331 persons (1%) had a preliminary positive rapid HIV test result; of these, 286 (86%) received a confirmatory test.
The most common reason cited by persons with preliminary positive HIV test results for refusing confirmatory testing was that they wanted to have the testing performed elsewhere. Of the 286 persons who received a confirmatory test, 267 (93%) were confirmed to have HIV infection,
and 17 had negative confirmatory test results (i.e., false preliminary positive rapid HIV test results). The positive predictive value of a preliminary positive rapid result for a confirmed test was 94% (267
of 284). Of the 267 persons with newly diagnosed HIV infection, 200
(75%) received their confirmatory test results. The most common reason cited by participating sites for why clients with preliminary positive test results did not receive their confirmatory test results was that the clients could not be located. Of the 200 persons who received their confirmatory results, 171 (86%) accepted referrals to medical care for HIV; the reasons that 29 persons (14%) did not accept referrals to medical care are not known. Referral to care encompassed a range of actions, including escorting clients to medical care, scheduling medical appointments, or providing contact information for clients to schedule their own appointments.
D Aguirre, Bienestar Human Svcs; A Mares-DelGrasso, AIDS Healthcare Foundation, Los Angeles; C Emerson, Tenderloin Health, San Francisco,
California. J Tsang, The Night Ministry, Chicago, Illinois. J Pincus,
MD, Dotwell, Dorchester, Massachusetts. C Calhoun, Community Health Awareness Group, Detroit, Michigan. H Buckendahl, Kansas City Free Clinic, Missouri. D Dekker, PhD, Whitman Walker Clinic, Washington,
DC. K Jafa-Bhushan, MBBS, K Bowles, MPH, H Clark, MPH, B Song, MS,
PS Sullivan, PhD, JD Heffelfinger, MD, Div of HIV/AIDS Prevention;
J Cleveland, MS, National Center for HIV, Viral Hepatitis, STD, and TB Prevention; E Tai, MD, EIS Officer, CDC.
The Advancing HIV Prevention demonstration project described in this report provided rapid HIV testing to 23,900 persons, including 30% who had never been tested previously for HIV, and identified 267
newly diagnosed cases of HIV infection. Seventy percent of those tested were in racial/ethnic minority populations at greater risk for HIV infection, and 66% had multiple sex partners. These results suggest that rapid HIV testing in outreach and other community settings can effectively target substantial numbers of persons at high risk for HIV infection. Overall, approximately 1% of persons tested had newly diagnosed HIV infection. This is comparable to the 1% rate of positive test results at CDC-supported HIV counseling and testing sites, although clients differed in referral status, race/ethnicity, and risk behaviors.5
In this project, the percentage of persons who had been tested previously for HIV (70%) was lower than the percentage (73%-88%) who reported being tested previously in a 2002 survey of populations at high risk for HIV infection.6
in this project, 75% of persons with confirmed positive HIV tests received their results, a rate similar to those reported previously from six rapid HIV testing studies.7
improved strategies might increase that proportion and also the proportion of clients who receive their results and accept referral to medical care. One strategy to improve the rate of referral might be to refer persons with preliminary positive HIV test results immediately to medical care rather than waiting until results of confirmatory testing are available. This strategy would eliminate the need for clients to return to the testing site to receive confirmatory results before being referred to medical care. Another way to increase acceptance of referral might be to use a combination of rapid HIV tests rather than a Western blot test to confirm preliminary positive HIV results.
This practice would allow clients to receive a preliminary positive HIV test result and a confirmed test result rapidly and be linked to health-care and prevention services the same day.8 CDC currently is evaluating use of a confirmatory algorithm with a combination of rapid tests. However, until this strategy can be validated, preliminary positive tests should always be confirmed with Western blot tests.
In this project, 19 persons with newly diagnosed HIV infection had visited a health-care provider during the preceding year but had never been tested for HIV; these persons appear to represent missed opportunities to test medical patients routinely in populations at high risk for HIV infection. In 2006, CDC published revised recommendations for HIV testing in medical settings, including routine HIV testing for patients aged 13-64 years in all health-care settings.9
Routine testing without risk assessment can identify persons with undiagnosed HIV infection and reduce the reluctance associated with testing protocols that require assessment of risk behavior.10
The findings in this report are subject to at least three limitations.
First, the project did not track the number of persons who were offered testing; therefore, the rate of acceptance of rapid HIV testing in outreach and other community settings cannot be calculated. However,
rapid HIV testing has been preferred over conventional HIV testing.8 Second, selection of venues for HIV testing by the CBOs was not systematic; therefore, those persons tested might not be representative of all persons served by the CBOs, and other risk factors for HIV infection might exist that were not elicited.
Finally, information regarding whether the 171 persons with newly diagnosed HIV infection who accepted referral to medical care were actually linked to HIV care (e.g., made at least one follow-up medical visit) was either incomplete or unavailable for most participating CBOs.
This project demonstrated that rapid HIV testing in a range of settings can effectively target multiple populations at high risk for HIV infection. Offering rapid HIV testing in outreach and other community settings provides opportunities to identify HIV infections and to link persons with positive test results to prevention and medical care.
Rapid HIV Testing in Outreach and Other Community Settings—United States, 2004-2006. JAMA. 2008;299(3):280-282. doi:10.1001/jama.299.3.280