October 12, 2007
In the United States, Hispanics are affected disproportionately by human immunodeficiency virus (HIV) infection and acquired immunodeficiency syndrome (AIDS). Although Hispanics accounted for 14.4% of the U.S. population in 2005,1 they accounted for 18.9% of persons who received an AIDS diagnosis.2 The rate of HIV diagnosis among Hispanics also remains disproportionately high; in 2005, the annual rate of HIV diagnosis for Hispanics was three times that for non-Hispanic whites. To better characterize HIV infection and AIDS among Hispanics in the United States, CDC analyzed selected characteristics of Hispanics in whom HIV infection was diagnosed during 2001-2005 and those living with AIDS in 2005. The results indicated that the mode of HIV infection for Hispanics varied by place of birth, suggesting that all HIV-prevention measures might not be equally effective among Hispanics and that HIV educational activities should address cultural and behavioral differences among Hispanic subgroups.
This analysis includes cases of HIV/AIDS diagnosed among Hispanic adults and adolescents aged >13 years during 2001-2005 in 33 states and cases of Hispanics living with HIV or AIDS in 50 states and the District of Columbia in 2005. Included are HIV cases reported to CDC from the 33 states* that have conducted name-based HIV reporting since at least 2001. Confidential name-based HIV and AIDS reporting has achieved high levels of accuracy and reliability (CDC, unpublished data, 2005). HIV/AIDS cases include those with 1) a diagnosis of HIV infection that have not progressed to AIDS, 2) a diagnosis of HIV infection followed by a diagnosis of AIDS, 3) and concurrent diagnoses of AIDS and HIV infection (i.e., in the same month).
Cases were classified according to the following transmission categories: 1) male-to-male sexual contact (i.e., among men who have sex with men [MSM]); 2) injection-drug use (IDU); 3) MSM with IDU; 4) high-risk heterosexual contact (i.e., with a person of the opposite sex known to be HIV infected or at high risk for HIV/AIDS [e.g., MSM or injection-drug user]); and 5) other modes of infection (e.g., receipt of transfusion of blood, blood components, or tissue transplant) and unknown risk factors. Cases reported with unknown risk factors were reclassified into transmission categories (e.g., MSM, IDU, MSM and IDU, high-risk heterosexual contact, and other) in accordance with methods described previously.3 Potential duplicate cases were identified based on unique identifiers and selected demographic characteristics and were eliminated on both state and national levels.
For 2005, annual HIV/AIDS diagnosis rates per 100,000 population were calculated for Hispanics, non-Hispanic whites, and non-Hispanic blacks. Data were adjusted for reporting delays.,3 The number of Hispanics living with HIV or AIDS at the end of 2005 was calculated based on reported cases adjusted for delays in reporting and deaths; this calculation does not account for undiagnosed cases.
During 2001-2005, a total of 184,167 adults and adolescents had HIV/AIDS diagnosed in the 33 states and reported to CDC. Of these, 33,398 (18%) were Hispanics; 93,017 (51%) were non-Hispanic blacks; 54,029 (29%) were non-Hispanic whites; 1% were Asian/Pacific Islanders; and <1% were American Indian/Alaska Natives. The mode of HIV infection for 61% of Hispanic males was male-to-male sexual contact, 17% of infections occurred through high-risk heterosexual contact, and 17% occurred through IDU. Among Hispanic females with HIV/AIDS diagnoses, 76% were exposed through high-risk heterosexual contact, and 23% were exposed through IDU (Table 1).
In 2005, the overall annual rate of HIV/AIDS diagnosis among Hispanic males was 56.2 per 100,000 population and among Hispanic females was 15.8 per 100,000 population. For Hispanic males, the highest rate of HIV diagnosis (86.3 per 100,000) occurred among those aged 30-39 years; for Hispanic females, the highest rate (25.0 per 100,000) occurred among those aged 40-49 years. The overall rates for non-Hispanic white and non-Hispanic black males in 2005 were 18.2 and 124.8, respectively, and the rates for non-Hispanic white and non-Hispanic black females were 3.0 and 60.2, respectively.
The mode of HIV infection among Hispanics varied by place of birth (Table 2). Infection through male-to-male sexual contact was more common among Hispanics born in South America (65%), Cuba (62%), and Mexico (54%) than among Hispanics born in the United States (46%). A greater proportion of Hispanics born in the Dominican Republic (47%) and Central America (45%) were infected through high-risk heterosexual contact, compared with Hispanics born in the United States (28%). Hispanics born in Puerto Rico had a greater proportion of HIV infections attributed to IDU (33%) than those born in the United States (22%).
In 2005, in the 33 states, the rate of living with HIV infection among Hispanics was estimated at 173.0 per 100,000 population (Table 3). Estimated HIV prevalence among Hispanics ranged from 34.3 per 100,000 population in Wyoming to 443.0 in New York. In the 50 states and DC, the rate of living with AIDS among Hispanics was estimated at 244.2 per 100,000 population. Estimated AIDS prevalence ranged from 28.7 per 100,000 population in Montana to 1,165.8 per 100,000 population in DC.
Reported by: L Espinoza, DDS, KL Dominguez, MD, RA Romaguera, DMD, X Hu, LA Valleroy, PhD, HI Hall, PhD, Div of HIV/AIDS Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD and TB Prevention, CDC.
These results confirm a previous report of disproportionate rates of HIV diagnosis among Hispanics, who have the second highest rate among all racial/ethnic groups in the United States.4 During 2001-2004, HIV-diagnosis rates among Hispanics declined by 4.7% and 13.0% among Hispanic males and females, respectively.4 These decreases among Hispanics might have resulted from decreased incidence of HIV infection (e.g., in response to prevention measures) or a decrease in HIV testing among Hispanics. However, this report indicates that Hispanics are not a homogenous group, and risk factors differ for Hispanic subpopulations.
Nearly half of U.S. Hispanics in whom HIV infection was diagnosed were not born in the United States. Hispanics born in Mexico and elsewhere often migrate to the United States to work as laborers and in service occupations. Migration might contribute to an increase in HIV risk behaviors, perhaps because change in residence can be followed by homelessness, loneliness, isolation, separation from usual sex partners, and financial instability. These factors can be associated with new sex partners, illegal drug use, and inadequate access to health information and health-care services.5
During 2001-2005, the primary mode of HIV infection among Hispanic males was male-to-male sexual contact. A recent study of HIV risk behaviors among MSM reported that Hispanic and non-Hispanic black MSM were more likely than non-Hispanic white MSM to report inconsistent condom use during anal sex.6 However, male-to-male sexual contact is not the most common transmission category for Hispanics for certain places of birth. High-risk heterosexual contact was more common among Hispanics born in Central America and the Dominican Republic than Hispanics born in South America, Cuba, Mexico, Puerto Rico, and the United States. In addition, HIV knowledge and perceptions of risk differ among U.S. Hispanic subgroups. Immigrants born in Cuba, Mexico, and Puerto Rico who were injection-drug users reported less AIDS knowledge than U.S.-born injection-drug users.7
The finding that a greater proportion of Puerto Rico-born Hispanics residing in the 33 states are infected through IDU is consistent with previous reports8 and might be the result of both greater prevalence of IDU and increased levels of high-risk behaviors related to IDU (e.g., frequency of injecting and sharing syringes) compared with other Hispanics.9 U.S. Hispanic subgroups of varied national origin or ancestry differ in IDU-related behaviors. Puerto Rico-born injection-drug users are more likely to share syringes, cotton, or rinse water and to inject more frequently than Puerto Ricans born in the United States.10
The findings in this report are subject to at least four limitations. First, although AIDS is a reportable condition in all 50 states, name-based HIV data were available from only 33 states. These states represented an estimated 63% of all AIDS cases and 56% of AIDS cases among Hispanics in the United States during 2001-2005. The exclusion2 of data from some states with high AIDS prevalence and a large Hispanic population (e.g., California) results in an underrepresentation of cases among Hispanics. Second, the assumptions by which the approximately 32% of cases that had no known risk factors were redistributed among transmission categories might no longer be valid; these assumptions are being reevaluated. Third, misclassification of Hispanics as members of other races/ethnicities or inability to include undocumented migrant workers might have resulted in underestimations of the number of Hispanics overall and in Hispanic subgroups. Finally, birthplace information was missing for approximately 24% of Hispanics in this analysis. Depending on the distribution of birthplaces for persons with missing information, transmission-category prevalences for certain subgroups might have been larger or smaller.
The disproportionate rate of HIV infection among Hispanics might reflect the failure of HIV-prevention programs to reach Hispanics at high risk for acquiring or transmitting HIV infection. More specifically, the difference in HIV transmission categories among Hispanics by place of birth might represent differences in acculturation, linguistic ability, socioeconomic status, and stigma associated with homosexuality or male-to-male sex. CDC recently established an internal committee to develop a National Plan of Action to reduce the number of new HIV infections among Hispanics and to increase access to culturally appropriate prevention, care, and treatment services. The plan is aimed at enhancing research, policy, and community involvement to increase capacity to deliver appropriate HIV-prevention services to Hispanics. CDC will expand its partnerships with other federal agencies, state and local health departments, academic institutions, and community-based organizations to identify specific steps to implement the National Plan of Action. Because the Hispanic population in the United States is expected to nearly triple between 2000 and 2050,§ additional attention to the impact of HIV on this population is warranted.
* Alabama, Alaska, Arizona, Arkansas, Colorado, Florida, Idaho, Indiana, Iowa, Kansas, Louisiana, Michigan, Minnesota, Mississippi, Missouri, Nebraska, Nevada, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, West Virginia, Wisconsin, and Wyoming.
Reporting delays (i.e., time between diagnosis and report) can differ by geographic location, age, sex, transmission category, and race/ethnicity. Adjustments for reporting time were calculated for HIV and AIDS cases using a maximum likelihood statistical procedure that accounts for differences in reporting time for the preceding characteristics while assuming the reporting delay has remained constant over time. Adjustments also were made based on the redistribution of cases across transmission categories by sex, race/ethnicity, and geographic region for cases diagnosed 3-10 years earlier and initially classified as reported with unknown risk factors but later reclassified.
§ Information available at www.census.gov/ipc/www/usinterimproj/natprojtab01a.pdf.
Click image to enlarge.
Click image to enlarge.
Click image to enlarge.