JAIDS Journal of Acquired Immune Deficiency Syndromes:
Epidemiology and Social Science
Undiagnosed HIV Prevalence Among Adults and Adolescents in the United States at the End of 2006
Campsmith, Michael L DDS, MPH; Rhodes, Philip H PhD; Hall, H Irene PhD; Green, Timothy A PhD
From the Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention, Atlanta, GA.
Received for publication May 12, 2009; accepted August 31, 2009.
The previous work by the authors “Estimated prevalence of undiagnosed HIV infection in the United States at the end of 2006” was presented at 16th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2009; Montreal, Canada. Abstract W-187.
The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
Correspondence to: Michael L. Campsmith, DDS, MPH, MS E-47, Centers for Disease Control and Prevention, 1600 Clifton Rd NE, Atlanta, GA 30333 (e-mail: email@example.com).
Objectives: To describe adults/adolescents (age 13 years and older) living with undiagnosed HIV infection in the United States at the end of 2006.
Methods: HIV prevalence and percentage undiagnosed were estimated from cumulative HIV incidence using an extended back-calculation model (using both HIV and AIDS data, the time of first diagnosis with HIV, and disease severity at diagnosis) and estimated cumulative deaths.
Results: An estimated 1,106,400 adults/adolescents (95% confidence interval = 1,056,400-1,156,400) were living with HIV in the United States at the end of 2006; overall, 21.0% (232,700; 95% confidence interval = 221,200-244,200) were undiagnosed. Whites had the lowest percentage undiagnosed (18.8%) compared with Hispanics/Latinos (21.6%), blacks/African Americans (22.2%), American Indians/Alaska Natives (25.8%), and Asians/Pacific Islanders (29.5%; all P < 0.001). Persons with a behavioral risk of injection drug use (IDU) had the lowest percentage undiagnosed (female IDU: 13.7% and male IDU: 14.5%); men exposed through heterosexual contact had the highest (26.7%) followed by men who have sex with men (23.5%).
Conclusions: Differences in undiagnosed HIV were evident across demographic and behavior groups. Effective testing programs and early access to treatment and prevention services are necessary to reduce undiagnosed HIV infections and HIV prevalence.
Current, accurate, and timely public health surveillance data on HIV prevalence are needed to guide decisions on planning for disease prevention activities, program evaluation, and resource allocation at the local, state, and national levels.1-3 However, the overall prevalence of persons living with HIV cannot be directly observed, as a percentage of persons infected with HIV has not yet been tested, diagnosed, and reported to local disease surveillance programs. Having a better understanding of the characteristics of the undiagnosed population (eg, by race/ethnicity, sex, age, and risk) is important for focusing HIV testing and prevention initiatives and to measure progress in decreasing the size of the undiagnosed population.
Research has indicated that persons who are infected but not diagnosed disproportionately contribute to the annual number of new HIV infections.4 This may be due to (1) increased infectivity due to greater levels of circulating HIV early after infection5,6 and/or (2) higher level of behavioral risks that lead to HIV transmission (unprotected sex, concurrent sex partners, and sharing of injection equipment).7,8 Thus, increasing the number of HIV-infected persons who are diagnosed and linked with effective care and prevention programs have the potential to significantly reduce new HIV infections over time. In addition, diagnosed persons can benefit from clinical treatments to prevent immune system damage and opportunistic infections.
Earlier methods to estimate HIV prevalence relied on back-calculation using the number of AIDS diagnoses and patterns of AIDS incubation to estimate the number of previous HIV infections that needed to occur over time to result in the number of new AIDS diagnoses observed.9 The use of effective therapeutic agents has modified those parameters, making HIV prevalence estimates through back-calculations exclusively from AIDS diagnosis data less accurate. Also, the 1993 change in the AIDS case definition10 to include diagnosis based on immunological criteria (CD4+ T-lymphocyte count <200/μL or <14% of total lymphocytes) makes it more difficult to interpret the data on new AIDS cases.
This article describes results from an extended back-calculation method11 that overcomes some of the shortcomings of the original (ie, AIDS data only) back-calculation approaches. Using HIV surveillance data from states that had integrated name-based HIV and AIDS surveillance systems, and AIDS data from states both with and without name-based HIV reporting, we estimated the number of diagnosed and undiagnosed prevalent cases of HIV infection among adults and adolescents (age 13 years and older) in the United States at the end of 2006.
Since 1982, all 50 states and the District of Columbia have reported AIDS cases to the Centers for Disease Control and Prevention (CDC) using a standardized case report form. In 1994, CDC implemented data management for national reporting of HIV integrated with AIDS case reporting, and over time, areas have adopted name-based HIV reporting and submitted these data for inclusion in the national surveillance database. We estimated HIV prevalence in the United States at the end of 2006 using information from the national HIV/AIDS reporting system on persons ≥13 years of age diagnosed with HIV before the end of 2006 and reported to CDC by the end of June 2007. AIDS data were reported by all states and the District of Columbia for the entire reporting period. Forty states provided data on both HIV and AIDS diagnoses, whereas 10 states (California, Delaware, Hawaii, Illinois, Maryland, Massachusetts, Montana, Oregon, Rhode Island, and Vermont) and the District of Columbia provided only AIDS data. Model parameters included year of HIV diagnosis, year of AIDS diagnosis, state of residence at diagnosis, sex, race/ethnicity, HIV transmission category, and age at first diagnosis. The data were adjusted for reporting delay, detection and elimination of duplicate reports, and misclassification of the first diagnosis date.12,13 Cases reported without HIV risk factor information were redistributed among transmission categories based on the transmission category of cases diagnosed 3-10 years earlier and initially reported without risk factor information but later reclassified based on information obtained through follow-up investigations.13
We used an extended back-calculation approach11 based on the number of HIV diagnoses by calendar year and disease severity (ie, whether the individual received an AIDS diagnosis in the same calendar year as the HIV diagnosis) to estimate the total number of infections (known diagnosed cases plus estimated undiagnosed cases) and then subtracted the estimated number of deaths (obtained from national HIV/AIDS surveillance data) to arrive at estimated prevalence. The expected distribution of the observed HIV diagnoses was specified using basic variables of interest: (1) AIDS hazard (the AIDS hazard in a designated year is the probability that an individual is diagnosed with AIDS in that year given that he/she was AIDS free at the beginning of the year) by time since infection in untreated HIV-infected individuals, (2) HIV testing hazard by year in HIV-infected individuals before AIDS diagnosis, and (3) number of HIV infections by year. The model estimates HIV infections by estimating the expected distribution of the time of infection for the observed HIV diagnoses in combination with estimating the probabilities that infections occurring in each of these prior years would remain undiagnosed at the present time. The model is not dependent on effects of treatment in delaying time from HIV diagnoses to AIDS, as we are interested in disease history only up to the point of initial HIV diagnosis.
To obtain reasonable and stable estimates, assumptions were made about the actual values or structure of the variables described above. The AIDS diagnosis hazards were specified, not estimated, using information from prior studies.14 The model specified time periods within which, respectively, HIV testing hazards (probability that an HIV-infected person was tested in a designated year given that he/she had not tested positive in a previous year) and the number of HIV infections were assumed to be approximately constant. The HIV testing hazard was restricted to be dependent on calendar time, not time since infection. Also, the HIV testing parameters in the extended back-calculation model do not refer to the overall rate of HIV testing in the general population (most of whom are HIV negative) but to the rate of HIV testing among persons who are HIV positive.
We assumed that the diagnosis counts have a Poisson distribution with expectations equal to a linear function of the 3 sets of model parameters. We used an expectation-maximization algorithm15 to estimate the unknown parameters in the extended back-calculation model. After specifying some initial starting values for the unknown parameters, the algorithm alternates between an expectation step (which calculates an “expanded” version of the observed dataset that is consistent with both the specified model structures and current “working” parameter values) and a maximization step (which reestimates the parameter values using the observed and the expanded data). In this case, the expanded dataset consists of the number of diagnoses by time of infection, severity of diagnosis, and time of diagnosis. Each case has 3 possible outcomes for each time interval: (1) HIV and AIDS diagnosis within the same year, (2) HIV diagnosis without an AIDS diagnosis in the same year, and (3) no diagnosis (ie, the case remains undetected). Model fit was measured by comparing the observed and expected diagnoses by time period and severity, using a log-likelihood ratio comparison suitable for Poisson random variables.
The population denominators used for rate calculations were based on official postcensus estimates for 2006 from the US Census Bureau16 and bridged-race estimates for 2006 obtained from the National Center for Health Statistics.17 The data analyses were generated using SAS software, Version 9.1 of the SAS System for Windows, SAS Institute, Inc18 and APL*PLUS III, Manugistics, Inc.19
Figure 1 displays the changes in estimated HIV prevalence among adults and adolescents (age 13 years and older) living with HIV infection in the United States. At the end of 2006, there were an estimated 1,106,400 [95% confidence interval (CI) = 1,056,400-1,156,400] persons living with HIV infection in the United States (Table 1). Males made up three-quarters (74.8%) of the estimated persons living with HIV (prevalent cases). By race/ethnicity, blacks/African Americans made up 46.1% of estimated prevalent cases, whites 34.6%, Hispanics/Latinos 17.5%, Asians/Pacific Islanders 1.4%, and American Indians/Alaska Natives 0.4%. By HIV transmission category, men who have sex with men (MSM) made up the largest percentage of persons estimated to be living with HIV (48.1%).
Overall, 21.0% (232,700; 95% CI = 221,200-244,200) of estimated prevalent HIV cases were undiagnosed. The percentage of estimated cases of HIV infection that was undiagnosed varied by demographic and risk factors (Table 1). The difference in the percentage of undiagnosed HIV by sex was small (21.7% for males vs. 19.1% for females). Greater differences were observed by race/ethnicity, age, and transmission category. Whites had a significantly lower percentage of undiagnosed HIV (18.8%) compared with Hispanics/Latinos (21.6%), blacks/African Americans (22.2%), American Indians/Alaska Natives (25.8%), and Asians/Pacific Islanders (29.5%). The youngest age group (13-24 years) had the greatest estimated percentage of undiagnosed HIV (47.8%), and the percentage undiagnosed significantly decreased with age up to age 55 years (Table 1).
By HIV transmission category, cases associated with injection drug use (IDU) had the lowest percentage of undiagnosed HIV, male IDU: 14.5%; female IDU: 13.7%; and male-to-male sexual contact and IDU: 12.1% (Table 1). The highest percentage of undiagnosed HIV (26.7%) was among men with a transmission category of high-risk heterosexual contact (HRHC), defined as reporting heterosexual contact specifically with a person known to have, or to be at high risk for, HIV infection (eg, an injection drug user). The second highest percentage of undiagnosed HIV was among MSM (23.5%). We observed differences in the percent of undiagnosed HIV when risk was stratified by sex and race. In ascending order, the percentage of undiagnosed HIV among male HRHC by race was black/African American male HRHC (25.7%), Hispanic/Latino male HRHC (27.9%), white male HRHC (28.5%), Asian/Pacific Islander male HRHC (33.3%), and American Indian/Alaska Native male HRHC (38.8%; all comparisons to black/African American male HRHC, P < 0.003). In ascending order, the percentage of undiagnosed HIV among MSM by race was white MSM (19.8%), American Indian/Alaska Native MSM (25.0%), Hispanic/Latino MSM (26.3%), black/African American MSM (27.2%), and Asian/Pacific Islander MSM (28.6%; all comparisons to white MSM, P < 0.001). A similar pattern was seen for female HRHC, with white female HRHC having a lower percentage of undiagnosed HIV (18.0%) than Hispanic/Latino female HRHC (20.3%), black/African American female HRHC (22.0%), American Indian/Alaska Native female HRHC (25.0%), and Asian/Pacific Islander female HRHC (30.4%; all comparisons to white female HRHC, P < 0.001).
Rates per 100,000 population were calculated for adults/adolescents living with undiagnosed HIV infection in the United States at the end of 2006. Table 2 displays the estimated rates and 95% CI of undiagnosed HIV infection by race/ethnicity and sex for persons aged 13 years and older. Black/African American males had the highest rate of undiagnosed HIV infection (556.5 per 100,000). The next highest rates were among black/African American females (225.7 per 100,000) and Hispanic/Latino males (201.6 per 100,000). Overall, whites represented the greatest percentage of the adult/adolescent population of the United States at the end of 2006, 69.0% overall, but a lower percentage of estimated living HIV cases (34.6% overall; 39.6% of males and 19.7% of females). The estimated rate of undiagnosed HIV among whites was 42.2 per 100,000. In contrast, blacks/African Americans made up 12.0% of the adult/adolescent population but 46.1% of the estimated persons living with HIV (40.2% of males and 63.8% of females). The estimated rate of undiagnosed HIV among blacks/African Americans (380.3 per 100,000) was 9 times the rate for whites. Similarly, Hispanics/Latinos made up 13.4% of the adult/adolescent population but 17.5% of estimated persons living with HIV (18.4% of males and 15.0% of females). The estimated rate of undiagnosed HIV among Hispanics/Latinos (126.4 per 100,000) was 3 times the rate for whites.
The number of persons in the United States living with HIV infection continues to increase each year. A major factor contributing to this increase is reduced mortality due to the use of highly active antiretroviral therapy among persons diagnosed with HIV.20-23 From 1995 to 1998, the estimated number of deaths among persons with AIDS declined 63%, from 51,670 to 18,82324; from 2002 through 2005, the estimated number of deaths averaged 17,189 per year.25 Additionally, the estimated number of annual HIV infections has remained relatively stable over the past decade,11 and these new infections contribute to the number of persons living with HIV.
The burden of HIV infection, both prevalence and percentage undiagnosed, is disproportionate across population groups. Racial/ethnic minorities made up less than one-third (31.0%) of the adult/adolescent population in the United States at the end of 2006 but accounted for nearly two-thirds (65.4%) of persons estimated to be living with HIV. Blacks/African Americans accounted for slightly less than half (46.1%) of all adults/adolescents living with HIV despite comprising only 12.0% of the population. Each racial/ethnic minority group had a significantly greater percentage of undiagnosed HIV infection compared with whites. Blacks/African Americans had the highest rates of undiagnosed infection, with black/African American men showing the highest rate overall. These findings demonstrate the differential impact of HIV on racial/ethnic populations. Because race/ethnicity is not itself a risk factor for HIV infection, differences in HIV disease burden across population groups are likely due to differences in other factors, including perception of risk and risk behaviors26-28 and relative lack of access to-and utilization of-health care resources, particularly HIV testing and treatment.29-31
We also observed differences in HIV prevalence and percent undiagnosed by behavioral risk factor. Sexual contact is the main behavioral risk for both men and women diagnosed with HIV infection.32 Our analysis found that men with a behavioral risk factor of male to male sex comprise nearly half (48.1%) of the estimated adults/adolescents living with HIV at the end of 2006. Although not precisely known, the percentage of MSM in the general population is estimated to be much lower. Data from CDC's National Survey of Family Growth indicate that among males aged 15-44 years, 3.7% ever had anal sex with another male and the percentage of men who had a male sexual partner in the past 12 months was 2.9%.33 MSM also had a significantly greater percentage of undiagnosed HIV infection (23.5%) compared with the overall percentage undiagnosed (21.0%). Again, there were differences in the percentage of undiagnosed HIV infection among MSM by race, with minority MSM having significantly higher percentages undiagnosed compared with white MSM. A similar pattern was seen in a study among MSM in 5 US cities.27 That study also found 48% of MSM diagnosed with HIV were unaware of their infection, twice the undiagnosed percentage of MSM from our national estimate. This finding is likely due to differences in the analysis populations. Persons whose HIV behavioral risk factor was HRHC made up over one-quarter (27.6%) of estimated prevalent HIV cases. Two-thirds (66.0%) of the estimated living HIV cases attributed to HRHC were among women, with two-thirds of those being black/African American women.
Interestingly, our analysis found that persons exposed to HIV through IDU had significantly lower percentages of undiagnosed HIV infection. This may be the result of injection drug users interacting with the health care system through the use of emergency departments, needle exchanges, drug treatment facilities, or community outreach programs;34-37 in such settings, IDUs may have a greater chance of being offered an HIV test. Also, it may be that injection drug users are more likely to acknowledge their exposure risk for HIV and thus have a greater predilection to accept HIV testing when it is offered.35-38
Our analysis is subject to some limitations. HIV data from the 40 states used in the extended back-calculation model represent only a portion of persons in the United States who are diagnosed with HIV infection. Several high-morbidity areas did not contribute HIV surveillance data, including California, Illinois, Maryland, and the District of Columbia. Thus, the national data on diagnosed cases of HIV infection are incomplete. Additionally, the data presented here have been statistically adjusted to account for reporting delays for new cases and for deaths, and cases reported without risk factor information have been redistributed among other transmission categories. These adjustments are based on assumptions (eg, reporting delays have not changed over time) that may no longer be accurate.12,13
The continued increase in the prevalence of persons living with HIV infection, both diagnosed and undiagnosed, is an ongoing challenge for providing medical and social services. The financial costs of care for persons diagnosed with HIV continue to grow, with most of the care dollars provided by the federal government. The federal budget request for fiscal year 2007 included $13.2 billion for medical care for persons with HIV.39 The discounted lifetime cost for treating a person entering HIV care with a CD4 count less than 350 has been estimated to be $385,000 in 2004, with most of the cost attributed to HIV-related medications.40 That figure reflects the substantial costs associated with treating HIV infection for a projected period of 24.2 years after initiation of antiretroviral therapy. As the number of diagnosed cases of HIV infection increases, those dollar amounts will continue to grow ever larger.
Persons living with HIV infection who are not yet diagnosed are not able to benefit from early monitoring and appropriate treatment of their disease condition, which have been shown to reduce morbidity and mortality.41,42 Persons who are unaware of their positive HIV status are also more likely to engage in HIV transmission risk behaviors compared with infected persons who have been diagnosed: Studies have shown that transmission risk behavior decreases among persons newly diagnosed with HIV infection.7,43,44 Thus, recent national HIV prevention strategies have focused efforts on routinizing HIV testing and working with HIV-positive persons to initiate and maintain HIV risk reduction behaviors, with the goal of reducing new HIV infections in the United States.2,45
The epidemic of HIV infection in the United States is now in its third decade. Better treatments are allowing many infected people to live longer; however, there is still no cure for HIV disease. Despite major advances in the scientific understanding of HIV, development of a safe and effective vaccine against HIV remains elusive.46,47 Thus, prevention will continue to be the main component of HIV disease control activities. Innovative approaches to reduce transmission risk behaviors are needed to decrease the number of new HIV infections. Additionally, new and creative public health programs that include sufficient and sustained funding are necessary to increase the percentage of persons with HIV infection who are diagnosed and provided appropriate care and prevention services.
1. Centers for Disease Control and Prevention. Guidelines for national human immunodeficiency virus case surveillance, including monitoring for human immunodeficiency virus infection and acquired immunodeficiency syndrome. MMWR Recomm Rep. 1999;48(RR13):1-28.
2. Centers for Disease Control and Prevention. Advancing HIV prevention: new strategies for a changing epidemic-United States, 2003. Morb Mortal Wkly Rep. 2003;52:329-332.
3. Health Resources and Services Administration.The Ryan White HIV/AIDS Treatment Modernization Act of 2006. Public Law 109-45. Available at: http://hab.hrsa.gov/law/reauthor06.htm
. Accessed December 1, 2008.
4. Marks G, Crepaz N, Janssen RS. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS. 2006;20:1447-1450.
5. Hollingsworth TD, Anderson RM, Fraser C. HIV-1 transmission, by stage of infection. J Infect Dis. 2008;198:687-693.
6. Lavrey L, Baeten JM, Panteleeff DD, et al. High levels of cervical HIV-1 RNA during early HIV-1 infection. AIDS. 2006;20:2389-2390.
7. Marks G, Crepaz N, Senterfitt JW, et al. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States. Implications for HIV prevention programs. J Acquir Immune Defic Syndr. 2005;39:446-453.
8. West GR, Corneli AL, Best K, et al. Focusing HIV prevention on those most likely to transmit the virus. AIDS Educ Prev. 2007;19:275-288.
9. Karon JM, Rosenberg PS, McQuillan G, et al. Prevalence of HIV infection in the United States, 1984 to 1992. JAMA. 1996;276:126-131.
10. 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 Recomm Rep. 1993;41(RR-17):1-19.
11. Hall HI, Song R, Rhodes P, et al. Estimation of HIV incidence in the United States. JAMA. 2008;300:520-529.
12. Green TA. Using surveillance data to monitor trends in the AIDS epidemic. Stat Med. 1998;17:143-154.
13. Song R, Hall HI, Frey R. Uncertainties associated with incidence estimates of HIV/AIDS diagnoses adjusted for reporting delay and risk redistribution. Stat Med. 2005;24:453-464.
14. Aalen OO, Farewell VT, De Angelis D, et al. A Markov model for HIV disease progression including the effect of HIV diagnosis and treatment: application to AIDS prediction in England and Wales. Stat Med. 1997;16:2191-2210.
15. Becker NG. Uses of the EM algorithm in the analysis of data on HIV/AIDS and other infectious diseases. Stat Methods Med Res. 1997;6:24-37.
18. SAS (for Windows) [computer program]. Version 9.1.3. Cary, NC: SAS Institute Inc; 2002.
19. APL*Plus III [computer program]. Rockville, MD: Manugistics Inc; 1995.
20. Fleming PL, Ward JW, Karon JM, et al. Declines in AIDS incidence and deaths in the USA: a signal change in the epidemic. AIDS. 1998;12:S55-S61.
21. McNaghten AD, Hanson DL, Jones JL, et al. Effects of antiretroviral therapy and opportunistic illness primary chemoprophylaxis on survival after AIDS diagnosis. AIDS. 1999;13:1687-1695.
22. Cain LE, Cole SR, Chmiel JS, et al. Effect of highly active antiretroviral therapy on multiple AIDS-defining illnesses among male HIV seroconverters. Am J Epidemiol. 2006;163:310-315.
23. Bhaskaran K, Hamouda O, Sannes M, et al. Changes in the risk of death after HIV seroconversion compared with mortality in the general population. JAMA. 2008;300:51-59.
24. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, 2001. Vol 13 (No 2). Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2002.
25. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, 2006. Vol 18. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2008.
26. Millett G, Malebranche D, Mason B, et al. Focusing on “down low”: bisexual black men, HIV risk and heterosexual transmission. J Natl Med Assoc. 2005;97:S52-S59.
27. Centers for Disease Control and Prevention. HIV prevalence, unrecognized infection, and HIV testing among men who have sex with men-five U.S. cities, June 2004-April 2005. MMWR Morb Mortal Wkly Rep. 2005;54:597-601.
28. Centers for Disease Control and Prevention. Racial/ethnic disparities in diagnoses of HIV/AIDS-33 states, 2001-2005. MMWR Morb Mortal Wkly Rep. 2007;56:189-193.
29. Kellerman SE, Drake A, Lansky A, et al. Use of and exposure to HIV prevention programs and services by persons at high risk for HIV. AIDS Patient Care STDs. 2006;20:391-398.
30. Cunningham WE, Sohler NL, Tobias C, et al. Health services utilization for people with HIV infection-comparison of a population targeted for outreach with the U.S. population in care. Med Care. 2006;44:1038-1047.
31. Palacio H, Kahn JG, Richards TA, et al. Effect of race and/or ethnicity in use of antiretrovirals and prophylaxis for opportunistic infection: a review of the literature. Public Health Rep. 2002;117:233-215.
32. Centers for Disease Control and Prevention. Cases of HIV infection and AIDS in the United States, by race/ethnicity, 2000-2004. HIV/AIDS Surveil Supplemental Rep. 2006;12(1).
33. Mosher WD, Chandra A, Jones J. Sexual behavior and selected health measures: men and women 15-44 years of age, United States, 2002. Adv Data. 2005;362:1-55.
34. Hsieh YH, Rothman RE, Newman-Toker DE, et al. National estimation of rates of HIV serology testing in US emergency departments 1993-2005: baseline prior to the 2006 Centers for Disease Control and Prevention recommendations. AIDS. 2008;22:2127-2134.
35. Downing M, Knight K, Reiss TH, et al. Drug users talk about HIV testing: motivating and deterring factors. AIDS Care. 2001;13:561-577.
36. Heimer R, Grau LE, Curtin E, et al. Assessment of HIV testing of urban injection drug users: implications for the expansion of HIV testing and prevention efforts. Am J Public Health. 2007;97:110-116.
37. Centers for Disease Control and Prevention. Rapid HIV testing in outreach and other community settings-United States, 2004-2006. MMWR Morb Mortal Wkly Rep. 2007;56:1233-1237.
38. Kellerman SE, Lehman JS, Lansky A, et al. HIV testing within at-risk populations in the United States and the reasons for seeking or avoiding HIV testing. J Acquir Immune Defic Syndr. 2002;31:202-210.
39. The Henry J. Kaiser Family Foundation. U. S. Federal Funding for HIV/AIDS: The FY 2007 Budget Request
. HIV/AIDS Policy Fact Sheet; 2006. Available at: www.kff.org
, Publication #7029-03. Accessed March 10, 2009.
40. Schackman BR, Gebo KA, Walensky RP, et al. The lifetime cost of current human immunodeficiency virus care in the United States. Med Care. 2006;44:990-997.
41. Palella FJ Jr, Deloria-Knoll M, Chmiel JS, et al. Survival benefit of initiating antiretroviral therapy in HIV-infected persons in different CD4+ cell strata. Ann Intern Med. 2003;138:620-626.
42. Treatment for adult HIV infection. 2004 recommendations of the International AIDS Society-USA Panel. Special Communication. JAMA. 2004;292:1-20.
43. Centers for Disease Control and Prevention. Adoption of protective behaviors among persons with recent HIV infection and diagnosis-Alabama, New Jersey, and Tennessee, 1997-1998. MMWR Morb Mortal Wkly Rep. 2000;49:512-515.
44. Weinhardt LS, Carey MP, Johnson BT, et al. Effects of HIV counseling and testing on sexual risk behavior: a meta-analytic review of published research, 1985-1997. Am J Public Health. 1999;89:1397-1405.
45. Centers for Disease Control and Prevention. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep. 2006;55:1-17.
46. Duerr A, Wasserheit JN, Corey L. HIV vaccines: new frontiers in vaccine development. Clin Infect Dis. 2006;43:500-511.
47. Barouch DH. Challenges in the development of an HIV-1 vaccine. Nature. 2008;455:613-619.
undiagnosed HIV prevalence; risk behaviors; disease disparities
© 2010 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.