HIV Counseling and Testing-United States, 1993

Counseling and testing are important components of state and localHIV prevention programs [1]. Analysis of national data sourcesindicates that HIV antibody tests are obtained from a varietyof testing sites, including private physicians, hospitals, andoutpatient clinics (66.7%), and publicly funded sites (33.1%)[2]. This report uses data from CDC's 1993 Behavioral Risk FactorSurveillance System (BRFSS) to examine variations in rates ofuse of private and public HIV counseling and testing sites bystate.

In 1993, a total of 49 states and the District of Columbia participatedin the BRFSS, a state-specific, population-based, random-digit-dialedtelephone survey that collects information monthly from U.S. adultsaged 18 years or older. Thirteen questions about HIV/AIDS-relatedknowledge and attitudes, and HIV antibody testing history duringthe preceding year were asked only of respondents aged 65 yearsor younger. In 1993, a total of 84,039 persons responded to thesequestions (state-specific range: 993 to 3,367) [3]. Data for eachstate were weighted by demographic characteristics and by selectionprobability; results are representative of persons aged 18 to65 years in each state.

A median of 25.5% of persons (range: 14.4% [Iowa] to 37.5% [Alaska])answered yes to the question: "Except for donating or givingblood, have you ever had your blood tested for the AIDS virusinfection?" The number (weighted estimate) of adults whohad ever been tested for HIV was highest in California (6.3 million).

A median of 9.6% of persons (range: 4.1% [Maine and South Dakota]to 16.9% [District of Columbia]) reported obtaining HIV antibodytests primarily for diagnostic reasons. For this study, diagnosticHIV antibody tests were defined as those administered primarilyto learn infection status, rather than voluntary tests to qualifyfor insurance, military induction, immigration, marriage licenseapplication, or employment. Persons categorized as having obtaineddiagnostic HIV antibody tests were identified by one of threeresponses to the question "What was the main reason you hadyour last AIDS blood test?":

1. "To find out if infected"

2. "Because of referral by a doctor or health departmentor sex partner"

3. "For routine checkup." This response was includedin "diagnostic" reasons to avoid excluding respondentswho initiated a routine examination to determine whether theywere infected with HIV.

In 43 states and the District of Columbia, at least 50% (median:60.9%) of respondents had obtained their last diagnostic testfrom a private physician, health maintenance organization, orprivate outpatient clinic. A median of 16.2% of persons (range:5.0% [North Dakota] to 37.6% [Mississippi]) had obtained theirlast diagnostic test at a publicly funded prevention site (includinghealth departments; AIDS, sexually transmitted disease [STD],or tuberculosis clinics; and drug treatment programs).

The estimated number of persons who obtained a diagnostic testat a publicly funded site during the preceding year correlatedwith the number of tests reported to CDC's HIV Counseling andTesting System by publicly funded sites in each state [4].

A median of 60.7% of persons who had obtained their most recentdiagnostic HIV antibody test at a publicly funded site (range:30.8% [New Jersey] to 95.7% [Oklahoma]) received counseling withtheir test results. In comparison, a median of 28.2% of personswho had obtained their tests from a private site (range: 7.7%[Kentucky] to 77.3% [Oklahoma]) received counseling. In 90% ofthe reporting areas, the number of persons who received counselingwith their HIV test results was at least 1.5 times greater forpersons tested at publicly funded sites than those tested at privatesites.

Editorial Note from the CDC: The findings from the 1993BRFSS document a high degree of state-specific variability inself-reported HIV antibody tests in the United States. This variabilitymay reflect state-specific differences in such factors as theprevalence of HIV infection, and HIV testing in high-risk groups,the presence and impact of HIV prevention programs, and age distribution.The BRFSS estimates of the number of persons last tested for voluntaryor diagnostic reasons at a publicly funded clinic correlated highlywith estimates from CDC's HIV Counseling and Testing System, andthe median percentage of respondents ever tested for HIV (25%)is consistent with estimates based on CDC's National Health InterviewSurvey (22%).

Health-care visits to seek and obtain HIV tests are importantopportunities to counsel persons about the risk for HIV infectionand methods to reduce such risk.1 The data in this report indicatethat, in most states, approximately threefold more persons reportedhaving obtained their HIV test from a private provider than froma public site; however, persons who had obtained their test froma private provider were substantially less likely to have reportedreceiving counseling than those who obtained tests at a publicsite. This finding underscores the need for physicians and otherhealth-care providers in private settings to offer HIV counselingat the time patients receive their HIV test results.

Limitations

The findings in this report are subject to at least two limitations.

First, the sample size of persons who reported having had an HIVantibody test in individual states did not enable stratificationby other respondent characteristics. For example, state-specificsample sizes precluded analysis to determine whether specifichigh-risk populations that obtained HIV antibody testing alsoreceived counseling.

Second, because the BRFSS is a telephone-based system, some personsat high risk for HIV infection most likely were excluded fromthe survey.

The BRFSS is a unique source for information about HIV antibodytesting behaviors of U.S. adults--particularly patterns of HIVtesting outside of public clinics--and can be used both at thefederal and state levels to improve HIV prevention and interventionprograms. Questions about CT in the 1993 BRFSS were developedbased on input from state health departments; subsequent BRFSSsurveys may incorporate additional HIV-related behavioral questions.

References:

1. Hinman AR: Strategies to prevent HIV infection in the UnitedStates. Am J Publ Health 81:1557-1559, 1991.

2. CDC: HIV counseling and testing services from public and privateproviders-United States, 1990. MMWR 41:743,749-752, 1992.

3. CDC: 1993 BRFSS quality control report. Atlanta, US Departmentof Health and Human Services, Public Health Service, CDC, 1994.

4. CDC: HIV counseling and testing data system: National profile,1993. Atlanta, US Department of Health and Human Services, PublicHealth Service, CDC, 1994.