ATLANTAHuman immunodeficiency virus (HIV) infection is
one of the leading causes of morbidity and mortality in the United
States. Testing for HIV, in conjunction with counseling and other
preventive services, can reduce the risk for HIV infection and
appropriately link infected persons to treatment. To characterize HIV
testing by region, state, and sex, the Centers for Disease Control
(CDC) analyzed data from the 1996 Behavioral Risk Factor Surveillance
System (BRFSS). This report summarizes the results of that analysis,
which indicate a high degree of variability in HIV testing throughout
the United States.
The BRFSS is a state-specific, random-digit-dialed telephone survey
of the US population 18 years of age and older. In 1996, all 50
states and the District of Columbia participated in the BRFSS.
The 1996 survey included 14 questions about HIV/acquired
immunodeficiency syndrome (AIDS)related knowledge and attitudes
and HIV-antibody testing history. The questions were restricted to
persons under age 65 years, except in California, where the questions
were asked of persons under age 45 years.
In 1996, 97,006 persons responded to these questions (state-specific
range, 899 to 3,653). Data were weighted by demographic
characteristics and by selection probabilities. Confidence intervals
were calculated using SUDAAN software to account for the complex
A mean of 42% of persons surveyedrange, 26% (South Dakota) to
60% (Washington, DC)answered yes to the question Have you
ever had your blood tested for HIV? Persons who answered
yes were asked What was the main reason you had
your last blood test for HIV? Responses were divided into two categories:
Those who chose to be tested for personal or health reasons (ie,
voluntarily tested). These people gave responses such as just
to find out if infected, for routine check-up,
doctor referral, sex partner referral,
because of pregnancy, or other.
Those who were tested for other reasons (eg, military induction,
insurance, and employment).
A mean of 22% of personsrange, 10% (South Dakota) to 45%
(Washington, DC)reported obtaining HIV-antibody tests for
The rate of AIDS cases in 1996 was compared with HIV testing
percentages in 1996. In general, in states where the AIDS rate was
high, HIV testing also tended to be high (see Figure).
For example, Washington, DC, had the highest AIDS rate and the
highest testing percentage; Florida ranked third in both categories.
In comparison, rates of overall testing and voluntary testing were
lower in the Midwest, where the rate of AIDS is low.
A mean of 44% of men reported having ever been tested for
HIVrange, 28% (South Dakota) to 62% (Washington, DC). A mean of
40% of women reported having ever been tested for HIVrange, 23%
(North Dakota) to 57% (Washington, DC). In 45 states and Washington,
DC, a greater percentage of men reported ever being tested for HIV
than women. The states with the greatest difference by sex of ever
being tested for HIV were North Dakota (11%), Hawaii (10%), and New
York (9%). The states with the smallest differences were Alaska,
Delaware (both 0.5%), and Texas (0.6%).
A mean of 20% of men reported that their most recent HIV test was
voluntaryrange, 8% (South Dakota) to 46% (Washington, DC). A
mean of 25% of women reported that their most recent HIV test was
voluntaryrange, 12% (North Dakota) to 45% (Washington, DC). In
49 states, a greater percentage of women reported being voluntarily
tested than men. The sex-specific difference in reports of being
voluntarily tested ranged from 0.1% in New York and Indiana to 13% in California.
Editorial Note From the CDC
The findings in this report document a high degree of state-specific
variability in self-reported HIV-antibody tests. Previous reports
suggest that this variability probably represents state-specific
differences in such factors as prevalence of HIV infection and the
activities of HIV prevention and education programs.
The success of a health-promotion program depends on the level of
participation of clients. Although HIV testing and counseling do not
affect behavior change similarly across all population groups, in
general, persons who voluntarily receive HIV testing are more likely
to undergo counseling and to modify their behaviors than are those
who receive testing for other reasons. As a result, tracking overall
testing rates and voluntary testing rates can help target
The findings in this report are subject to at least two limitations.
First, because BRFSS excluded persons without telephones, some
persons at high risk for HIV infection probably were excluded.
Second, because the BRFSS relies on self-reported data, some bias is expected.
HIV testing can help reach at-risk persons with counseling and other
prevention services, and link infected persons with needed health
care services. General population surveys, such as BRFSS, provide
data to assess the use of HIV testing services across geographical
areas. However, not all persons need to be tested. The CDC recommends
HIV counseling and testing services for persons with specific risk
factors for HIV infection and in specific screening settings (eg,
tissue donation and pregnancy). HIV prevention programs should be
structured to increase the proportion of at-risk persons who receive HIV-testing