February 20, 2004
The analysis included persons aged >13 years with HIV; infections were categorized as either heterosexually acquired§ or nonheterosexually acquired. Heterosexually acquired HIV infections were further categorized as diagnosed with AIDS (i.e., during the same calendar month) or diagnosed without AIDS. New diagnoses of HIV infections were examined for 1999-2002. Data were adjusted for reporting delays, and HIV-exposure data were adjusted for reclassification of cases initially reported with no mode of exposure into categories according to historical patterns of reclassification.3 CDC calculated confidence intervals (CIs), taking into account adjustments for reporting delays and reclassification to exposure categories, and variance estimates were derived from monthly data submissions to CDC.4
During 1999-2002, a total of 101,877 HIV infections were diagnosed in the 29 states and reported to CDC, including 36,084 (35%) acquired through heterosexual contact (Table). Among states, the median prevalence of heterosexually acquired HIV infections was 27% (range: 13%-47%).
The proportion of females was greater among persons with heterosexually acquired HIV infections (64%; 23,205 of 36,084) than the proportion of females among persons exposed through injection-drug use, blood products, transfusions, and undetermined modes of exposure (36%; 6,661 of 18,732). Among age groups, prevalence for heterosexually acquired HIV infections was greatest (35%) among persons aged 30-39 years.
Non-Hispanic blacks accounted for 26,748 (74%) of persons with heterosexually acquired HIV infections. A total of 5,257 (15%) were non-Hispanic white; 3,498 (10%) were Hispanic; and <1% were Asian/Pacific Islander or American Indian/Alaska Native. By comparison, among persons with nonheterosexually acquired HIV infections, non-Hispanic blacks accounted for 29,607 (45%), and non-Hispanic whites accounted for 26,731 (41%). During 1999-2002, an overall increase in heterosexually acquired HIV infections from 8,925 (95% CI = 8,606-9,243) in 1999 to 9,156 (95% CI = 8,713-9,600) in 2002 was not statistically significant.
Among the 36,084 persons with heterosexually acquired HIV infections, 7,395 (20%) (Table) received a concurrent diagnosis of AIDS. Diagnosis of HIV/AIDS was more common among males (25%; 3,223 of 12,879) than among females (18%; 4,172 of 23,205).
Females accounted for 89% of heterosexually acquired HIV infections among persons aged 13-19 years (Figure 1). Females also accounted for 70% of such cases reported among non-Hispanic whites, 64% among non-Hispanic blacks, and 56% among Hispanics (Figure 2).
H. I. Hall, Ph.D., L. M. Lee, Ph.D., M. K. Glynn, D.V.M. and R. Song, Ph.D. are with the Division of HIV/AIDS Prevention, National Center for HIV, STD, and TB Prevention. L. Espinoza, D.D.S. is EIS Officer at the CDC.
During 1999-2002, approximately 64% of heterosexually acquired HIV infections reported in the United States occurred in females. The proportion of infected females was highest among persons aged 13-19 years, consistent with a previous finding.5 Survey data suggest that females in this age group engage in behaviors that place them at increased risk for acquiring HIV infections; the high proportion of infected females might be associated with sexual contact with older males, who are more likely to be infected. In addition, young females might have more opportunities for HIV testing and diagnosis (e.g., routine family planning and gynecological services) than young males.
Persons in certain racial/ethnic populations continue to have disproportionate numbers of HIV infections. Non-Hispanic black and Hispanic populations constituted 21% of the total population of the 29 states in the study, according to the 2000 U.S. Census, yet these populations accounted for 84% of heterosexually acquired HIV infections during 1999-2002. HIV infections are concentrated in populations that traditionally have had limited access to prevention services, medical care, and effective therapies. Lack of knowledge about HIV, decreased perception of risk, use of drugs or alcohol, and different interpretations of so-called "safe sex" might contribute to the risk for HIV infection among non-Hispanic blacks and Hispanics.6 In addition, because of social patterns, non-Hispanic black and Hispanic females are more likely than other females to be exposed to HIV because of a higher prevalence of infection among non-Hispanic black and Hispanic males.7
Diagnosis of HIV and AIDS in the same calendar month occurred with 20% of the heterosexually acquired HIV infections, reflecting HIV diagnosis late in the course of infection and suggesting late testing in the course of the disease. A previous study determined that 41% (43,089 of 104,780) of persons with reported HIV infections also received an AIDS diagnosis within 1 year, which might indicate treatment failure or late testing.8
The findings in this report are subject to at least three limitations. First, although AIDS is a reportable condition in the United States, during 1999-2002, name-based HIV case data were available from only 29 states, which reported an estimated 39% of all AIDS cases. Nationwide reporting of HIV diagnoses would improve data regarding the HIV-infected population. Second, cases with no identified mode of exposure were classified into exposure categories on the basis of follow-up investigation. Cases with follow-up information were assumed to constitute a representative sample of all cases initially reported with no identified exposure, and the distribution among exposure categories was assumed to be consistent during the preceding 10 years. Finally, completeness of reporting and potential duplicate reporting by states are being evaluated in accordance with CDC's performance standards for HIV/AIDS surveillance.2 Reported HIV infections are estimated to represent >85% of all HIV infections.9
CDC recommends reporting on the prevalence of HIV infection to detect patterns in HIV transmission. New testing technology that distinguishes between recent and long-term infections will allow for better characterization of HIV-transmission patterns and more rapid and targeted preventive measures.10 CDC is working in areas of high morbidity (i.e., >300 AIDS cases per year) to integrate this technology into routine HIV case surveillance.
Racial/ethnic disparities continue among persons with HIV infections. Culturally sensitive HIV-prevention messages are needed to target those populations most affected. Prevention and education programs targeting heterosexually active teens, especially females and persons in certain racial/ethnic populations, should be developed. In addition, non-Hispanic black and Hispanic populations, which historically have less access to treatment and prevention services, are affected disproportionately by HIV. Barriers to care and prevention services for these populations should be removed.
* Alabama, Arizona, Arkansas, Colorado, Florida, Idaho, Indiana, Iowa, Louisiana, Michigan, Minnesota, Mississippi, Missouri, Nebraska, Nevada, New Jersey, New Mexico, North Carolina, North Dakota, Ohio, Oklahoma, South Carolina, South Dakota, Tennessee, Utah, Virginia, West Virginia, Wisconsin, and Wyoming.
? Five additional states (Alaska, Georgia, Kansas, New York, and Texas) have implemented name-based HIV/AIDS reporting that meets CDC standards. Pennsylvania also has implemented such reporting, but only in areas outside Philadelphia.
§ An HIV infection was categorized as heterosexually acquired if a patient reported specific heterosexual contact with a person with HIV infection or with a person at increased risk for HIV infection (e.g., an injection-drug user).