McDavid, Kathleen PhD, MPH; Li, Jianmin DPE; Lee, Lisa M. PhD
Since the beginning of the AIDS epidemic in the United States in 1984, an estimated 170,000 women have received a diagnosis of AIDS, and 81,900 of them have died; by the end of 2003, almost 89,000 women were living with AIDS.1 In 1984, adult and adolescent women accounted for 6% of AIDS cases2; by 1999, this proportion had increased to 23%, and by 2003, it had increased to 27%.1,3 Of HIV cases with and without AIDS for diagnosis year 2003, 30% of those reported to the Centers for Disease Control and Prevention (CDC) were in women.1
National surveillance reports to date have used AIDS data more prominently than HIV data. This is mainly because by 1985, AIDS was a reportable condition in all states and consistent reporting methods had been established, enabling examination of trends over time for all surveillance areas.4 Uniform case reporting for HIV (without AIDS) began in 1994 and became a reportable condition over the next decade; all states required some form of HIV case reporting by 2003. Since the introduction of highly active antiretroviral therapy in 1996, the number of AIDS cases has declined and AIDS case counts alone no longer provide sufficient information about what is occurring early in the disease closer to the time of HIV infection. HIV diagnoses provide a better picture of this earlier phase; therefore, we report HIV diagnoses.
We examined recent trends in the number and rate of estimated HIV diagnoses for adolescent and adult women residing in 29 US surveillance areas in 1999, in 32 US surveillance areas in 2000, and in 33 US surveillance areas from 2001 through 2003. Diagnoses are reported by age group, race and/or ethnicity, transmission category, and region.
A case of HIV infection was defined as a confirmed HIV diagnosis with or without a concurrent diagnosis of AIDS. All cases met the CDC surveillance case definition for HIV infection,4,5 were reported through June 2005, and have been adjusted for reporting delay and presented as estimated case counts using a maximum likelihood statistical procedure as previously described for AIDS data.6 Routinely, cases reported to the CDC are classified into hierarchic transmission categories, such as male-to-male sex, injection drug use, and heterosexual contact with a person diagnosed with or at high risk for HIV infection, on the basis of the most likely mode of transmission. For women in 2004, 36% of cases were reported without a risk factor. Using historical patterns of classification of cases originally reported without a risk factor, the proportion of cases originally reported without a risk factor that are subsequently classified into a known transmission category is calculated. These proportions are then used to distribute current cases reported without a risk factor into transmission categories (stratifying by sex, race, and region). This method also uses the procedure described for AIDS data.6
Adult and adolescent women at least 13 years old at diagnosis were included in the analysis within grouped age categories as follows: 13 to 19 years of age, 20 to 29 years of age, 30 to 39 years of age, 40 to 49 years of age, and 50 years of age and older. Race and/or ethnicity categories were non-Hispanic white (white), non-Hispanic black (black), Hispanic, Asian/Pacific Islander (API), and American Indian/Alaska Native (AI/AN) and were classified by a combination of the patients and health care providers. We examined differences in percentage of cases and demographic variables by geographic region of the United States. Figure 1 displays the 33 surveillance areas included in this study as well as the regions to which each belongs. The Mid Atlantic region comprises New Jersey and New York; the East North Central region comprises Indiana, Michigan, Ohio, and Wisconsin; the West North Central region comprises Iowa, Minnesota, Missouri, Nebraska, North Dakota, South Dakota, and Kansas; the South Atlantic region comprises Florida, North Carolina, South Carolina, Virginia, and West Virginia; the East South Central region comprises Alabama, Mississippi, and Tennessee; the West South Central region comprises Arkansas, Louisiana, Oklahoma, and Texas; the Mountain region comprises Arizona, Colorado, Idaho, New Mexico, Nevada, Utah, and Wyoming; and the Pacific region comprises Alaska.
Data presented for 1999 are from 29 states; for 2000, they are from 32 states, and for 2001 through 2004, they are from 33 states. These data reflect states that had confidential name-based reporting of confirmed HIV infection for long enough to allow for stabilization of data collection and adjustment of the data to monitor trends.1 HIV diagnosis rates by region exclude the Pacific region because it had data for only 1 state (Tables 1 and 2 of estimated counts include Alaska, however). For all regions, we estimated the number of diagnoses for age group, race and/or ethnicity, transmission category, and year of diagnosis. Using population data from the bridged-race postcensal population estimates from the National Center for Health Statistics, we calculated annual estimated HIV diagnosis rates per 100,000 women.7,8 We calculated 95% confidence intervals (CIs) for the estimated number and rate, taking into account the uncertainties in the estimates adjusted for reporting delays.
Estimated annual percent change (EAPC) was calculated for race and/or ethnicity, age group, and region by fitting a regression line to the natural logarithm of the rates using calendar year as the regressor. The standard error was generated from the fit of the regression.9 Adjusted EAPC was also calculated for race and/or ethnicity, age group, and transmission category, adjusting for each of the other 2 factors, but there was no difference between adjusted and unadjusted estimates.
From 1999 through 2004, women accounted for 60,323 (28.8%) of 208,952 HIV cases reported to the CDC. HIV-infected women younger than the age of 30 years made up 29.3% (n = 17,665) of the women with a new HIV diagnosis in 1999 through 2004. Most (68.6%) women in our group were black, followed by white (16.6%). After reclassifying cases with no initial reported risk factor information, most (77.0%) HIV cases in women have heterosexual contact with a person diagnosed with or at high risk for HIV infection as the transmission category, followed by 21.4% with injection drug use as the transmission category (see Table 1).
From 1999 through 2004, in the 33 surveillance areas (29 surveillance areas for 1999 and 32 for 2000) more cases were reported for black women (n = 41,328) than 2 times the number of cases for white (n = 10,010) and Hispanic women (n = 7905) combined. Black women accounted for a low of 29.3% (Mountain region) and a high of 79.0% (East South Central region) of estimated HIV cases in women in the respective regions with at least 10 cases (see Table 2).
Regional differences in risk factors and demographic characteristics were apparent. For example, high-risk heterosexual contact as the transmission category during the surveillance period varied from a high of 84.4% in the East South Central region to 67.3% in the Mid Atlantic region (see Table 2). In the East South Central region, 13- to 19-year-olds made up 7.3% of the women with HIV, whereas in the Mid Atlantic region, the same age group constituted less than 3%. From 1999 through 2004, the proportion of 20- to 29-year-old women with a new HIV diagnosis in the West North Central region was almost twice that of women with a new HIV diagnosis in the Mid Atlantic region (31.8% vs. 17.3%).
Almost two thirds (60.5%) of all HIV cases in women reported to the CDC from 1999 through 2004 were from the South Atlantic, East South Central, and West South Central regions (see Table 2), yet 46.6% of women in the 33 surveillance areas that report to CDC live in these 3 regions.
The annual estimated rate of HIV diagnosis for all women was stable in 1999 and 2000 but decreased from 16.1 (95% CI: 16.0 to 16.3) in 2001 to 13.2 (95% CI: 12.7 to 13.7) in 2004, mainly because of the addition of New York. The annual estimated rate of HIV diagnoses for black and Hispanic women decreased significantly from 82.7 and 25.3 in 2001, when New York was added, to 67.0 and 16.3 in 2004, respectively (Fig. 2), but the EAPC was not significant for either group. For all other races and/or ethnicities, rates remained relatively stable during the 4-year period. In 2004, the rate for black women was 21 times higher than that for white women, 9 times higher than that for AI/AN women, and 4 times higher than that for Hispanic women.
Rates varied considerably by age group. From 2001 to 2004, HIV diagnosis rates decreased significantly for all age groups except those aged 50 years and older, who experienced no significant change (EAPC was not significant for any age group; Fig. 3). Rates also varied by region of residence at the time of diagnosis, with the highest rate occurring in the Mid Atlantic region, where the EAPC was not significant. Rates decreased in the Mid Atlantic and South Atlantic regions from 40.5 (95% CI: 39.7 to 41.2) and 23.7 (95% CI: 23.3 to 24.0), respectively, in 2001 to 23.2 (95% CI: 21.4 to 25.0) and 20.8 (95% CI: 19.8 to 21.8) in 2004. Rates increased slightly in the West North Central region from 3.4 (95% CI: 3.2 to 3.6) in 2001 to 4.7 (95% CI: 3.7 to 5.6) in 2004 and in the Mountain region from 3.6 (95% CI: 3.4 to 3.9) in 2001 to 5.3 (95% CI: 4.2 to 6.4) in 2004 (Fig. 4).
Estimated rates of HIV diagnosis for women living in urban areas were generally the same as for all women, with a few percentage point differences in groups with the highest rates. For example, in 2004, the rate for black women was approximately 10 percentage points higher for urban women than for women overall (77.3 [95% CI: 73.5 to 81.1] and 67.0 [95% CI: 64.1 to 70.0], respectively), and the rate for 30- to 39-year-old women in urban areas was approximately 6 percentage points higher than the rate for women overall of the same age group (30.5 [95% CI: 28.3 to 32.8] and 24.5 [95% CI: 22.9 to 26.1], respectively).
Although the rate of HIV diagnosis in adult and adolescent women decreased significantly from 2001 to 2004, rates of HIV diagnoses remained disproportionately high for Hispanic women and especially for black women. In 2004, rates for Hispanic women and black women were 5 and 21 times, respectively, the rate for white women. Among cases reported from the surveillance areas included here, heterosexual contact with a person diagnosed with or at high-risk for HIV infection is the most common way by which women acquire HIV infection and rates of HIV diagnosis are highest for women aged 20 to 49 years.
Although the HIV diagnosis rate for black women decreased during the study period, the disparities remain enormous. Black women are disproportionately affected by HIV; this is true for every subgroup analyzed here. Blacks represented almost 70% of women with a new HIV diagnosis from 1999 through 2004, yet they represented approximately 13% of the population in the 33 surveillance areas in 2004.10 The huge disparity (21 times higher for blacks) in rates of HIV diagnosis between black women and white women is not matched in magnitude by any other major disease that the authors could find.11-13
Women with HIV who were younger than 30 years of age accounted for 29.3% of the women with a new HIV diagnosis in 1999 through 2004, whereas women 50 years of age and older accounted for 12.4% (17% in New York). Health care providers may not pay enough attention to women at the older end of the age spectrum, who are a growing segment of the US population and may be at risk for HIV. For example, in Dallas County, Texas, primary care physicians rarely or never (40%) asked patients older than 50 years of age about HIV risk factors.14 Furthermore, Speer and colleagues15 reported that more than 60% of black and white women with AIDS who were at least 44 years old did not remember receiving prevention information before receiving their diagnoses. Zablotsky and Kennedy16 discuss the behavioral aspects of HIV that have been overlooked in older women. They cite, for example, the need to tailor prevention messages and the need for greater physician awareness and assessment of sexual risk factors for older women, and they call for a more thorough approach to this age group.
High rates of HIV diagnoses for women in the southern regions have been reported elsewhere4,17,18 and are cause for concern. The Mid Atlantic region has a higher reported rate (23 vs. 21 per 100,000 in 2004), however, and the data are not sufficiently complete to focus entirely on regional differences. The notably steep decline of the HIV diagnosis rate in the Mid Atlantic region is driven by New York, which began HIV reporting in 2001.
Reasons for the higher rate of HIV diagnoses in the southern regions than in other regions may be attributable, at least partially, to increased transmission through heterosexual contact with men19-23 on the "down low" (heterosexually identified men who have sex with men without their female partners knowing it).24 The extent to which this is true cannot be fully assessed, because approximately 36% of all HIV (not AIDS) cases in women in 2004 were reported without a risk factor,1 and women may be unaware of their male partners' behavioral risk factors for HIV infection. The Healthy People 2010 for HIV/AIDS initiative has morbidity-based objectives related to new cases acquired heterosexually. Given the current trends reported here, these objectives cannot be achieved.25
Currently, the CDC applies an algorithm that uses cases initially reported without a risk factor to adjust for the proportion without risk factor information after completion of epidemiologic follow-up. It is unclear whether this adjustment is robust enough to reflect the distribution of risk factors accurately in the face of a growing proportion of cases reported without a risk factor and a smaller proportion of cases initially reported with a risk factor. Analyses from an assessment of HIV risk factors in 3 states indicate that the current redistribution method underestimates the proportion of cases that are heterosexually transmitted to men and women.26 CDC and surveillance area staff are developing procedures and educational materials for collecting risk factor information in an attempt to redouble efforts to ascertain these data.
Our report provides results from the largest population-based database on new HIV diagnoses available. It is based on the most comprehensive data available to date on HIV/AIDS among women in the United States. It includes data from 33 states (which represent 64% of the US population), including most of the states in the southeast, where HIV diagnoses among women seem to be stable or slowing, and New York. Other states that have high HIV/AIDS morbidity and racial and/or ethnic distributions different from these 33 states were not included. These data represent persons who have been tested for HIV and whose results were reported to the 33 surveillance systems included here. Therefore, these data do not include persons who have not been tested, who have been tested anonymously, or whose results were reported to other than the 33 states. Completeness of reporting for HIV infection (not AIDS) is estimated at more than 85%.27
More should be done to prevent new HIV infections from occurring disproportionately in minority populations, especially blacks. Continued efforts using individual level interventions do not seem to be sufficient in preventing new HIV infections.28,29 To avert new infections, policy makers should consider more ecologic approaches to improving public health, as called for by the Institute of Medicine30 (eg, improve socioeconomic equality, gender relationships, and power dynamics28,29). Other potential strategies could focus on women aged 13 to 19 years, because it is in the subsequent age groups that the HIV diagnosis rate increases approximately 4-fold and previous opportunities for prevention with this group may have been missed. Prevention messages aimed at these adolescent women should include awareness and education of the problem and should use strategies that are proven effective and appropriate for this age group. Recommendations for effective school health education to prevent HIV infection are available on the CDC's Web site.31
More complete information on which to base important prevention, programmatic, and care decisions for women with HIV in the United States should come from changes in the way HIV risk factors are collected, the ability to identify incident HIV infections, and the Advancing HIV Prevention (AHP) initiative activities (emphasizing working with HIV-positive persons on HIV prevention [eg, promoting prevention among HIV-infected men who have HIV-uninfected female sex partners]).32
Prevention is clearly the key to controlling the spread of HIV; great opportunities are before us. Future prevention efforts must clearly target black women, for whom HIV is the leading cause of death among those aged 25 to 34 years,33 and their partners with effective age- and culturally appropriate messages. In keeping with the AHP initiative, women should be encouraged to be tested, encourage their male partners to be tested, and learn skills to negotiate sexual practices that protect both partners when the serostatus of either is unknown.
The authors thank Lynne Stockton, VMD, MS, for editing this manuscript and Ruiguang Song, PhD, for statistical input in calculating the adjusted EAPC.
1. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, 2003, vol. 15. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2004. Available at: http://www.cdc.gov/hiv/stats/hasrlink.htm
. Accessed November 30, 2004.
2. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, December 1985. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention.
3. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, 1999. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention.
4. Centers for Disease Control and Prevention. CDC guidelines for national human immunodeficiency virus case surveillance, including monitoring for human immunodeficiency virus infection and acquired immunodeficiency syndrome. Morb Mortal Wkly Rep. 1999;48(RR-13):1-31.
5. Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR. 1992;41(RR-17):1-19.
6. Green T. Using surveillance data to monitor trends in the AIDS epidemic. Stat Med. 1998;17:143-154.
7. National Center for Health Statistics. Bridged-race vintage 2004 postcensal population estimates for July 1, 2000-July 1, 2004, by year, county, single-year of age, bridged-race, Hispanic origin, and sex (September 8, 2005). Available at: http://www.cdc.gov/nchs/about/major/dvs/popbridge/popbridge.htm
. Accessed October 11, 2005.
8. National Center for Health Statistics. Bridged-race intercensal population estimates for July 1, 1990-July 1, 1999, by year, county, 5-year age group, bridged-race, Hispanic origin, and sex (April 15, 2003). Available at: http://www.cdc.gov/nchs/about/major/dvs/popbridge/popbridge.htm
. Accessed April 15, 2004.
9. Kleinbaum DG, Kupper LL, Muller KE. Applied Regression Analysis and other Multivariable Methods. 2nd ed. Belmont, CA: PWS-Kent; 1988:266-268.
11. Office of Disease Prevention and Health Promotion. Healthy People 2010. Washington, DC: US Department of Health and Human Services. Available at: http://www.healthypeople.gov
. Accessed November 11, 2004.
13. US Cancer Statistics Working Group. United States Cancer Statistics: 2001 Incidence. Atlanta: Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; 2002.
14. Skiest DJ, Keiser P. Human immunodeficiency virus infection in patients older than 50 years: a survey of primary care physician's beliefs, practices, and knowledge. Arch Fam Med. 1997;6:289-294.
15. Speer DC, Kennedy M, Watson M, et al. Ethnic, demographic, and social differences among middle and older adults with HIV/AIDS. AIDS Patient Care STDS. 1994;13:615-624.
16. Zablotsky D, Kennedy M. Risk factors and HIV transmission to midlife and older women: knowledge, options and the initiation of safer sexual practices. J Acquir Immune Defic Syndr. 2003;33(Suppl):S122-S130.
17. Wortley PM, Fleming PL. AIDS in women in the United States: recent trends. JAMA. 1997;278:911-916.
18. Davis SF, Rosen DH, Steinberg S, et al. Trends in HIV prevalence among childbearing women in the United States, 1989-1994. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;19:158-164.
19. Kalichman S, Roffman R, Picciano J, et al. Risk for HIV infection among bisexual men seeking HIV prevention services and risks posed to their female partners. Health Psychol. 1998;17:320-327.
20. Stokes JP, Peterson JL. Homophobia, self-esteem, and risk for HIV among African American men who have sex with men. AIDS Educ Prev. 1998;10:278-292.
21. Chu SY, Peterman TA, Doll LS, et al. AIDS in bisexual men in the United States: epidemiology and transmission to women. Am J Public Health. 1992;82:220-224.
22. Diaz T, Chu SY, Frederick F, et al. Sociodemographics and HIV risk behaviors of bisexual men with AIDS: results from a multistate interview project. AIDS. 1993;7:1227-1232.
23. McKirnan DJ, Stokes JP, Doll D, et al. Bisexually active men: social characteristics and sexual behavior. J Sex Res. 1995;32:65-76.
24. Millett G. Men on the 'down low': more questions than answers [abstract]. Presented at: 11th Conference on Retroviruses and Opportunistic Infections; 2004; San Francisco.
26. McDavid K, Gerstle J, Hammett T, et al. Results of the Expanded HIV Risk Assessment Project (EHRAP). AIDS Care. 2006;18(1):77-81.
27. Centers for Disease Control and Prevention. Diagnosis and reporting of HIV and AIDS in states with integrated HIV and AIDS surveillance-United States, January 1994-June 1997. MMWR. 1998;47:309-314.
28. Sanders-Phillips K. Factors influencing HIV/AIDS in women of color. Public Health Rep. 2002;117(Suppl):S151-S156.
29. Adimora AA, Schoenbach VJ, Martinson FEA, et al. Concurrent partnerships among rural African Americans with recently reported heterosexually transmitted HIV infection. J Acquir Immune Defic Syndr. 2003;34:423-429.
30. Committee on Assuring the Health of the Public in the 21st Century, Institute of Medicine. The Future of the Public's Health in the 21st Century. Washington, DC: National Academy Press; 2002.
31. Centers for Disease Control and Prevention. Guidelines for effective school health education to prevent the spread of AIDS. MMWR. 1988;37(S-2):1-14.
32. Centers for Disease Control and Prevention. Advancing HIV prevention: new strategies for a changing epidemic-United States, 2003. MMWR. 2004;52:329-332.
33. Anderson RN, Smith BL, Centers for Disease Control and Prevention. Deaths: leading causes for 2001. National Vital Statistics Reports 2003. Hyattsville, MD: National Center for Health Statistics. 2003. Available at: http://www.cdc.gov/nchs/data/nvsr/nvsr52/nvsr52_09.pdf
. Accessed October 11, 2005.
© 2006 Lippincott Williams & Wilkins, Inc.