The Centers for Disease Control and Prevention (CDC) estimate that between 1 039 000 and 1 185 000 people in the USA are infected with HIV, and approximately 75% are aware of their infection and 25% are not . Understanding the extent to which these groups transmit new HIV infections through sexual behavior can help guide public health strategies and allocation of resources to fight the HIV/AIDS epidemic in the USA.
A recent meta-analytic review of USA findings demonstrated that the prevalence of unprotected anal or vaginal intercourse (UAV) with at-risk (i.e., HIV-negative or unknown serostatus) partners was substantially lower in HIV-positive persons aware they were infected relative to HIV-positive persons unaware of their status . Building on those findings, we estimate the proportion of sexual transmission of HIV attributable to HIV-positive aware and unaware persons in the USA.
The formula presented here reflects the relative contribution of these two awareness groups in sexually transmitting new HIV infections. The parameters used to calculate the estimate include the number of people aware and unaware they are infected with HIV, sexual behavior differences between these two groups (i.e., relative reduction in the prevalence of UAV with at-risk partners in the aware group and differences between groups in the average number of at-risk UAV sex partners), as well as consideration of viral load (VL) which affects transmission risk. The formula includes parameters based on empirical data and parameters based on assumptions when empirical data were not available. The formula is given below, followed by a description of each component.
* Ta is the proportion of sexual transmission from the HIV-positive aware group; the proportion from the unaware group (Tu) is 1−Ta.
* Na is the number of people aware they are HIV positive, and Nu is the number who are unaware they are HIV positive. For illustrative purposes, we use a population of 100 HIV-positive persons, 75 of whom are aware they are infected, and 25 of whom are not, corresponding to the estimated 75%/25% distribution in the USA .
* U is the proportion of HIV-positive unaware persons who engage in UAV with at-risk partners.
* b is a parameter applied to U and reflects the magnitude of reduction in the prevalence of UAV with at-risk partners among HIV-positive aware persons relative to the prevalence among HIV-positive unaware persons. The b parameter is based on a meta-analysis of findings from between-group comparisons (cross-sectional data comparing independent groups of aware and unaware persons) and within-subject comparisons (comparing sexual behavior before and after learning one's HIV-positive status) . The combined findings demonstrated a 57% relative reduction (range 52–59%) in the prevalence of UAV with at-risk partners in the aware (versus unaware) group. Importantly, the findings were consistent across studies conducted before and after highly active antiretroviral therapy became available and across studies in which the unaware group's prevalence of UAV varied from 29% to 63%. Thus, the reduction applies independently of the prevalence of UAV in that group. In our formula, factor b equals 0.43 (reflecting the 57% reduction).
* SPa is the aware group's average number of at-risk UAV sex partners, and SPu is the unaware group's average number of at-risk UAV sex partners. In a comprehensive search of the literature, we did not find data that compared the two awareness groups on this specific behavioral dimension. Nevertheless, it is important to include these components in the formula. Accordingly, we calculated the formula using three different assumptions regarding the ratio of SPu to SPa: (i) 1 to 1, (ii) 1.5 to 1, and (iii) 2 to 1.
* An important biologic factor affecting transmission is the VL of the infected person. Many HIV-positive aware persons are in medical care for their HIV  and some have undetectable or low VL which reduces risk of infecting sex partners [4–7]. Thus, it is important to include a parameter to reflect that some persons aware of their infection may have minimal risk of infecting a partner because of low or undetectable VL. As described below, Rvl is the proportion of HIV-positive aware persons who potentially transmit to at-risk partners and is calculated as Rvl = 1 − (x * y). x is the proportion of people aware they are HIV positive in the USA who are in medical care for their HIV disease. This parameter is estimated at 0.60 based on a recently published study on the percentage of recently diagnosed HIV-positive persons who had at least one medical visit for HIV in the past 6 months . y is the proportion of HIV-positive aware persons in care whose VL is less than 500 copies/ml. Studies of heterosexual transmission found that no HIV-positive persons who had a VL less than 1500 copies/ml  or less than 1094 copies/ml  infected a sex partner. We used a cut-point of less than 500 copies/ml to be more conservative. Among those in care, the proportion with a VL less than 500 is estimated at 0.545 based on cross-sectional findings from HIV-positive patients sampled at six HIV clinics . Rvl is estimated at 0.67. A VL adjustment was not applied to the unaware group under the assumption that unaware persons are not in treatment and none are on antiretroviral therapy. Indeed, some may be in the primary infection stage characterized by particularly high levels of virus which promotes transmission of infection [9,10].
Results and discussion
Table 1 shows the estimated proportion of sexually transmitted HIV attributable to HIV-positive aware (Ta) and unaware (Tu) persons. Depending on the assumption regarding the ratio of the number of at-risk UAV partners, the proportion of sexually transmitted HIV from the unaware group was estimated to range from 0.54 (assuming no difference in average number of at-risk UAV partners between groups) to 0.70 (assuming twice as many at-risk UAV partners in the unaware group). Using the lower bounds, we conservatively estimate that just over half of new sexually transmitted HIV infections in the USA stem from the 25% of the infected persons in the USA who are unaware of their seropositive status. The precision of the estimate will increase when data become available on numbers of at-risk UAV sex partners for each awareness group.
The other behavioral parameter in the estimation was the relative reduction in the prevalence of UAV with at-risk partners among HIV-positive aware (versus unaware) persons. This parameter was based on a prior meta-analysis of independent findings from cross-sectional as well as cohort studies. The cross-sectional studies included persons who had been aware of their HIV infection for several years. Thus, this parameter was not based only on persons who had known of their seropositive status for a short time. Although sexual behavior may fluctuate in the years following an HIV-positive diagnosis, evidence shows that the prevalence of UAV remains at a considerably lower level than that observed during the unaware period [2,11].
Our formula used a VL cut-point of less than 500 copies/ml to represent that portion of the HIV-positive aware group who had a low likelihood of infecting a sex partner. However, having a low or undetectable VL does not completely eliminate the possibility of transmitting infection. Thus, the Rvl parameter, which essentially reduces the number of aware persons who are transmitters based on their low VL, may underestimate the number of transmitters who are aware of their infection. This may be offset, however, by the number of unaware persons who are in the primary infection stage with attendant high VL which substantially increases risk of infecting a partner [9,10].
Viral load is not the only biologic factor affecting transmission risk. The presence of a sexually transmitted disease (STD) such as syphilis or gonorrhea may increase the likelihood of transmitting or contracting HIV . Several studies show that the prevalence of STD is higher among people who are unaware than aware of their HIV infection [13–16]. Thus, the contribution to transmission from the unaware group may be even larger than the estimates given here.
Approximately 80% of HIV diagnoses in the USA are among people who become infected through sexual exposure . Thus, of the approximately 40 000 new HIV infections each year in the US , 32 000 are sexual transmissions. Our results indicate that the HIV-positive unaware group contributes disproportionately to these new infections relative to their percentage of the HIV-positive population in the USA. Of the approximately 1 000 000 people living with HIV in the USA, about 750 000 are aware and 250 000 unaware of their seropositive status. Applying the conservative estimate from our analysis (i.e., 0.54 contribution from the unaware group) to the 32 000 cases, about 17 280 sexual transmissions may be from those who are unaware of their infection for a transmission rate of 6.9% (17 280/250 000), and 14 720 may be from those who are aware of their infection for a transmission rate of only 2.0% (14 720/750 000). The transmission rate is 3.5 times higher in the unaware group.
Our results indicate a need to expand HIV counseling and testing to increase the number of HIV-positive persons who are aware of their infection. If all people unaware of their infection could learn of their serostatus, and the prevalence of UAV with at-risk partners declined by 57% (based on the empirical data from the meta-analysis), then new sexual HIV infections could, theoretically, be reduced from 17 280 to 7430 among the HIV-positive unaware group. This reduction would represent a 31% reduction in the overall number of new sexual infections per year (from 32 000 to 22 150).
In practice, a combination of targeted HIV testing campaigns and broad-based screening campaigns will be needed to increase people's knowledge of their HIV serostatus. Targeted testing may be most important for men who have sex with men (MSM). For example, recent surveillance data from five urban centers indicated that, of MSM who tested HIV positive in the surveillance, 48% were unaware of their seropositive status and many (79%) reported that had been tested on other occasions and tested negative . Ethnic minority MSM had a higher prevalence of unrecognized infection than white MSM . Thus, for MSM, particularly MSM of color, the proportion of sexual HIV transmissions from the unaware group may be higher than the estimate given here for the general USA population. Other people with HIV may not consider themselves to be at risk for infection. Accordingly, screening programs in health-care settings and other agencies may be very useful for identifying people who are infected but unaware of it. Such screening programs may also help link people to medical and social services. It will take a greatly expanded testing effort to make a large impact on HIV transmission.
1. Glynn M, Rhodes P. What is really happening with HIV trends in the United States? Modeling the national epidemic. National HIV Prevention Conference
. June 2005. Atlanta [abstract T1-B11–13].
2. Marks G, Crepaz N, Senterfitt JW, Janssen RS. 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.
3. Gardner LI, Metsch LR, Anderson-Mahoney P, Loughlin AM, del Rio C, Strathdee S, et al
. Efficacy of a brief case management intervention to link recently diagnosed HIV-infected persons to care. AIDS 2005; 19:423–431.
4. Quinn TC, Wawer MJ, Sewankambo N, Servadda D, Chuanjun L, Wabwire-Mangen F, et al
. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med 2000; 342:921–929.
5. Tovanabutra S, Robison V, Wongtrakul J, Sennum S, Suriyanon V, Kingkeow D, et al
. Male viral load and heterosexual transmission of HIV-1 subtype E in northern Thailand. J Acquir Immune Defic Syndr 2002; 29:275–283.
6. Castilla J, del Romero J, Hernando V, Marincovich B, Soledad G, Carmen R. Effectiveness of highly active antiretroviral therapy in reducing heterosexual transmission of HIV. J Acquir Immune Defic Syndr 2005; 40:96–101.
7. Pedraza MA, del Romero J, Roldan F, Garcia S, Eyerbe MC, Noriega AR, et al
. Heterosexual transmission of HIV-1 associated with high plasma viral load levels and a positive viral isolation in the infected partner. J Acquir Immune Defic Syndr Hum Retrovirol 1999; 21:120–125.
8. Richardson JL. Dataset: Partnership for Health Study. Los Angeles CA: University of Southern California; 2001.
9. Koopman JS, Jacquez JA, Welch GW, Simon CP, Foxman B, Pollack SM, et al
. The role of early HIV infection in the spread of HIV through populations. J Acquir Immune Defic Syndr Hum Retrovirol 1997; 14:249–258.
10. Wawer MJ, Gray RH, Sewankambo NK, Serwadda D, Xianbin L, Laeyendecker O, et al
. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J infect Dis 2005; 191:1403–1409.
11. Weinhardt LS. HIV diagnosis and risk behavior. In: Kalichman SC, editor. Positive Prevention: Reducing HIV Transmission among People Living with HIV/AIDS. New York: Kluwer Academic/Plenum; 2005. pp. 29–63.
12. Fleming DT, Wasserheit JN. From epidemiological synergy to public health policy and practice: the contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex Transm Infect 1999; 75:3–17.
13. Chamot E, Couglin SS, Farley TA, Rice JC. Gonorrhea incidence and HIV testing and counseling among adolescents and young adults seen at a clinic for sexually transmitted diseases. AIDS 1999; 13:971–979.
14. Otten MW, Zaidi AA, Wroten JE, Witte JJ, Peterman TA. Changes in sexually transmitted disease rates after HIV testing and posttest counselling, Miami, 1988 to 1989. Am J Public Health 1993; 83:529–533.
15. George N, Green J, Murphy S. Sexually transmitted disease rates before and after HIV testing. International J STD AIDS 1998; 9:291–293.
16. Allen S, Serufilira A, Bogaerts J, van de Perre P, Nsengumuremyi F, Lindan C, et al
. Confidential HIV testing and condom promotion in Africa. JAMA 1992; 268:3338–3343.
17. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, 2000
. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. 2005. 16
18. 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. Morb Moral Wkly Rep
. 2005. 54