As part of a
serosurveillance system to monitor the prevalence of human immunodeficiency virus type 1 (HIV-1)
in the United States, the Centers for Disease Control and Prevention (CDC), in collaboration
with state and local health departments, conducted standardized anonymous unlinked
seroprevalence surveys in selected sentinel sites from 1988 through 1999. In addition, the
Department of Labor, the Department of Defense, and the American Red Cross provide CDC with
statistical data from routine HIV testing for surveillance purposes.
The objectives of the serosurveillance system are (1) to provide federal,
state, and local health officials and the general public with standardized estimates of HIV
prevalence among selected populations, (2) to describe the magnitude and changes over time of
HIV infection in these populations within regions and within selected demographic and
behavioral subgroups, (3) to recognize new or emerging patterns of HIV infection among specific
subgroups of the U.S. population, and (4) to assist in directing resources and in targeting
programs for HIV prevention and care.
From 1987 through 1999, CDC provided technical and financial assistance
to state and local health departments to conduct anonymous unlinked HIV surveys in sentinel
sites in selected metropolitan areas. The survey sites serve populations at high risk for HIV
infection, such as those at sexually transmitted disease (STD) clinics and drug treatment
centers (DTCs). Survey sites also included adolescent medicine clinics, which serve a
population at lower risk. Investigators from state and local health departments chose clinics
for participation in the surveys on the basis of client demographic and behavioral
characteristics, local public health priorities, projected sample size, availability of
voluntary counseling and testing, logistical considerations, and ability and willingness of the
clinic staff to conduct surveys in accordance with national standardized protocols.
Anonymous unlinked surveys were used because they allow unbiased
estimates of HIV infection among selected populations with increased HIV prevalence. Unlike surveys in which HIV prevalence rates are obtained through results that depend on client
testing decisions, unlinked surveys are unbiased by self-selection because anonymous specimens
from all clients are tested. Several steps were taken to ensure that these surveys were both
anonymous and ethical: (1) only residual sera from blood specimens originally collected for
routine diagnostic purposes were used in the surveys; (2) before specimens were tested for HIV,
all personal identifying information was permanently removed to ensure that neither HIV test
results nor survey information could be linked to specific individuals; (3) no interaction with
survey participants could take place solely for the purpose of the surveys; thus, the integrity
of the studies was in no way compromised and each person’s right to privacy was protected; and
(4) all clinic sites that conducted unlinked surveys offered voluntary HIV counseling and
testing allowing anyone visiting a site the opportunity to learn his or her HIV status and to
receive appropriate counseling and referral services.
CDC also monitors HIV prevalence in three populations in which HIV
screening is routinely performed. Since 1985, the American Red Cross has provided CDC with HIV
test results for blood donors. In addition, the U.S. Department of Defense has provided HIV test
results for applicants to the military service since 1985. Beginning in 1987, the U.S.
Department of Labor has provided HIV test results for entrants to the Job Corps, a federally
funded job training program for disadvantaged youth. As is true of all the unlinked surveys,
personal identifiers for participants in these screening programs are not available to CDC.
Results from routine HIV screening of Job Corps entrants, military
applicants, and blood donors provide important additional information on the evolving HIV
epidemic. Although geographically diverse, each of these groups is disproportionately composed
of persons with particular demographic and socioeconomic characteristics. Job Corps entrants
comprise young men and women who are educationally or economically disadvantaged. Military
applicants and blood donors are low-risk populations because persons with known HIV infection
are not accepted into the military and potential blood donors with known HIV infection or risk
factors for HIV infection are likely to have self-deferred.
This report complements three previous CDC prevalence reports: (1)
National HIV Prevalence Surveys, 1997 Summary; (2) National HIV Serosurveillance
Summary, Update–Results through 1993; and (3) National HIV Serosurveillance Summary,
Results through 1992. Because fewer clinics were funded by CDC to conduct unlinked surveys
from 1993 through 1997 than in earlier years, the number of clinics represented in this
report is substantially lower than the number in previous CDC reports. For example, in the
earlier CDC report of HIV trends (National HIV Serosurveillance Summary, Results through
1992), data were analyzed from 112 STD clinics in 46 cities, 78 DTCs in 35 cities, and 21
adolescent medicine clinics in 12 cities. This report presents data from 23 STD clinics in
13 cities, 22 DTCs in 14 cities, and 5 adolescent medicine clinics in 3 cities.
Included in this report are summaries of data from January 1993
through December 1997 from the unlinked prevalence surveys1 conducted in selected STD clinics,
DTCs, and adolescent medicine clinics, as well as data from routine HIV screening programs for entrants to the Job Corps,
applicants for military service, and first-time blood donors. High-risk populations include
men who have sex with men (MSM) and heterosexual patients at STD clinics and injection drug
users (IDUs) entering DTCs. Youth populations include patients at adolescent medicine
clinics and Job Corps entrants. Low-risk populations include military applicants and blood
donors (Table 1).
Go to Collection and Analysis of Data
1. The unlinked prevalence surveys were discontinued at the end of 1999. Because of the small number of participating clinics in 1998 and 1999, this report includes only data through 1997.