In the United States, racial differences in the prevalence and incidence of HIV infection are a defining characteristic of the AIDS epidemic. Recent data from the Centers for Disease Control and Prevention reveal that whereas Blacks account for 14% of the population within states reporting HIV infection, in 2009, Blacks accounted for 44% of all diagnosed HIV infections in the United States, a rate almost 8 times higher than that observed in Whites.1 Centers for Disease Control and Prevention data further indicate that, in 2009, Black men were more than 6 times more likely than White men to be diagnosed with new HIV infection, nearly 2.5 times that of Latino men and more than twice that of Black women. Among women, the estimated rate of new HIV infections for Blacks was 15 times the rate for Whites and more than 3 times that of Latinas. Importantly, the rate of HIV infection among Blacks has nearly doubled since the previous decade when 25% of diagnosed HIV/AIDS cases were observed among Blacks.
Racial disparities in incident infections are seen among all risk groups.2 Among heterosexuals, in 2009, approximately 60% of new HIV infections were observed among Blacks and about 18% among Latinos.1 Among men who have sex with men (MSM), in 2009, 38% of new HIV infections occurred among Blacks and 21% were observed among Latinos.1 Among injection drug users (IDUs), in 2009, 48% of diagnoses of new HIV infection were among Blacks and 21% were observed among Latinos.1
Most of the understanding of racial disparities is derived from studies of MSM and heterosexuals, the 2 largest risk groups in the United States. Although these populations dominate present discourse around HIV infection, injection-related transmissions remain an important mode of HIV transmission. In 2009, 9% of all HIV/AIDS cases were attributable to injection drug use.1 Importantly, there is a significant regional variation in the prevalence and incidence of HIV infection among IDUs with the highest concentrations occurring in the northeast. For example, in Philadelphia, 30.7% (8966) of all infections have occurred among heterosexual IDUs and 3.7% (1087) among MSM/IDUs.3 Although infections caused by IDUs have decreased over the past 10 years at higher rates than sexual transmissions, racial disparities persist. Among US IDUs living with HIV, 48% were Black, 23% were White, and 26% Latino/a. Among IDUs living with an AIDS diagnosis, 52% were Black, 19% were White, and 26% were Latino/a.4 In Philadelphia, 63% (3049) of the 4842 IDUs living with HIV/AIDS are Black, 16% (764) are White, and 21% (1029) Latino/a.
There is now an accumulation of research findings examining the factors associated with racial disparities that suggests that Blacks' (and Latinos') higher rates of infection relative to Whites may not be because of higher levels of risk behavior among them but instead to "low" or "average" levels of risk behavior within networks that have a higher prevalence of HIV infection.5–13 Hallfors et al14 found that among young, predominantly heterosexual adults, Whites had elevated risk for sexually transmitted infection and HIV infection when engaging in high-risk drug and sex behaviors but Blacks' prevalence was high regardless of the level of engagement in risky behaviors. Thus, behavioral patterns alone did not predict sexually transmitted infection and HIV risk.
Similar observations have been found for MSM. In a recent meta-analysis published in The Lancet, it was found that Black MSM engaged in fewer HIV risk behaviors than MSM of other races. Black MSM reported less unprotected anal intercourse with main male partners, fewer male sex partners, more condom use, less substance use, and less substance use during sex.10 Moreover, Black MSM were more likely than other MSM to report preventive behaviors against HIV infection, including using condoms, being tested regularly, and using post- or preexposure prophylaxis.10 Despite engaging in less risky behaviors and taking more preventive action, Black MSM continue to be disproportionately infected with HIV.
Studies of IDUs have made similar observations of risk, prevalence, and race. Although Black injectors have higher HIV prevalence, rates of needle sharing are consistently lower.15–18 Other studies have reported that racial differences in HIV infections among IDUs have persisted despite increased availability of prevention services and overall reductions in HIV prevalence.16 Several studies that have identified racial/ethnic differences in HIV among IDUs have focused on social and/or structural factors that are likely fueling racial disparities. Kottiri et al18 interviewed 662 IDUs in New York City about their sex and drug risk behaviors and risk behaviors of their risk network members. They found higher prevalence and lower risk behaviors among Blacks and that sex and drug risk networks were predominantly homogeneous with respect to race.
To date, no studies able to examine racial disparities have been conducted involving behavioral assessments and HIV testing on both IDUs and members of their risk networks (individuals with whom they have sex and use drugs). Data from the Philadelphia site of HIV Prevention Trials Network (HPTN) 037 provided the unique opportunity to examine relationships between risk behaviors, participation in drug treatment and needle exchange, HIV prevalence, and race among both uninfected IDUs and their drug and sexual partners. HPTN 037 tested a peer education intervention promoting HIV risk reduction among IDUs and members of their drug and sex networks.19 We examine some of the key factors that might help explain higher prevalence among Black injectors, including injection and sexual practices, use of preventive services such as syringe exchange and drug treatment, and composition of injectors' risk networks.
IDUs were recruited from community settings in Philadelphia, Pa, between December 2002 and November 2006 in an effort to identify HIV uninfected, active IDUs for participation in HPTN 037. Specific locations for recruitment of active IDUs were identified by ethnographic methods conducted in neighborhoods with a high prevalence of HIV/AIDS and drug use. Ethnographers conducted observations and initiated contact with community members (IDU and non-IDU) and local gatekeepers in an effort to secure community support and access to locations where IDUs could be recruited and interviewed. Twenty-five distinct "risk pockets" (places where drug use, drug sales, and sex exchange occur within close proximity) were identified by ethnographers.20 Mobile assessment units were then parked adjacent to these identified risk pockets. The use of these mobile units allowed for the convenient completion of confidential informed consent process followed by brief, interviewer-administered prescreening for behavioral eligibility. Of the 3079 drug users who completed prescreening interviews, 84% (n = 2582/3079) reported injection drug use in the 6 months before assessment and 39% of this group (1002/2582) met eligibility criteria based upon self-reports (18 years of age or older, uninfected with HIV, injected drugs at least 12 times in the previous 3 months, not enrolled in methadone maintenance treatment for at least 3 months, willing to recruit drug and sex partners, and without plans to relocate). Of the 3079 individuals who initially presented at the mobile assessment unit, 2.1% (66/3079) self-identified as HIV positive.
Those uninfected IDUs who met behavioral eligibility criteria were invited to participate in the protocol-screening process, which involved the completion of comprehensive assessments of drug use, HIV drug and sex risk behaviors, and HIV testing via enzyme-linked immunosorbent assay with Western blot confirmation. Information concerning the composition and characteristics of risk networks was gathered through the administration of the Social Network Inventory.21 Participants were prompted to identify social network members with whom they share private and personal conversations, borrow money, socialize, ask advice, and/or discuss personal health issues. For each network member listed, information was collected for age, gender, race/ethnicity, HIV status, frequency of contact, and risk relationship type (drug risk, sex risk).
These assessments were completed at the project's community-based office, and to be eligible, index participants had to be HIV negative and able to recruit at least one of their drug using or sexual partners. If found to be negative for HIV, willing index IDUs were given guidance on how to recruit members of their risk networks. The next and final step before enrollment was for the potential index participant to successfully refer at least one of his/her network members to the study site to complete informed consent and assessments of drug and sex risk behaviors and HIV status. It is important to note that network members were able to be enrolled regardless of their HIV status. In all, 232 HIV-negative IDUs (index subjects) and 465 of their sex and/or drug-using partners (network members) satisfied eligibility criteria and were formally enrolled in the study. Details of the study procedures and intervention are reported elsewhere.19
Given the secondary nature of the descriptive analyses presented here, we focus on comparing the prevalence of HIV infection, risk behaviors, and participation in HIV prevention services by race/ethnicity groups (Whites, Blacks, and Latinos). The analyses reported here use only data from the prescreening and baseline assessment of HPTN 037 at the Philadelphia site.
Prescreen Sample Characteristics
Of the 3079 drug-using individuals who completed prescreening assessments, 52% (N = 1604) were Black, 35% (N = 1083) White, and 12% (N = 409) Latino. The mean age was 39 years, ranging from 17 to 75 years. Sixty-four percent (N = 1984) of the sample was male, and approximately 65% (N = 2014) reported having a high school education or equivalent.
Among those completing prescreens, 87.4% (N = 2690) reported previous testing for HIV, and of these, 2.5% (N = 66) reported having been told that they had a positive result. The rates differed significantly by race (χ2 = 10.2, P = 0.017). The highest rate of infection was reported among Blacks (3.3%), followed by Latinos (2.5%) and Whites (1.3%).
Black participants engaged in fewer drug-related risk behaviors than did White and Latino participants (Table 1). Black participants were less likely than White and Latino participants to report injection drug use (71.2% vs. 98.8% and 94.1%, respectively; χ2 = 390.24, P < 0.001). For those who reported use of injection drugs, Black participants were also less likely than White and Latino participants to report sharing needles (21.5% vs. 46.0% and 34.8%, respectively; χ2 = 145.9, P < 0.001).
Similar rates of sex activity were reported by each racial/ethnic group (χ2 = 5.665, P > 0.1). Approximately 82% of the sample reported being sexually active in the past 3 months (Table 1). Reports of multiple partnerships were also similar between race/ethnic groups, with approximately 54% of each race group reporting having more than one partner. However, rates of other sex risk behaviors, such as condom use, varied by group. Whites were more likely than Blacks (83.1% vs. 77.6%) to use condoms inconsistently (χ2 = 10.7, P < 0.05). Blacks and Latinos were similar in the degree to which they used condoms consistently. Black men (2.8%) were less likely than White (6.0%) and Latino (8.2%) men to report same sex partners (χ2 = 34.3, P < 0.001).
Black participants were less likely than White and Latino participants to report receipt of drug treatment (13.7% vs. 25.2% and 24.7%, respectively; χ2 = 65.9, P < 0.001). They were also less likely to report use of needle exchange services (24.4% vs. 41.5% and 29.3%, respectively; χ2 = 87.5, P < 0.001).
An examination of the racial homophily between network indices and their network members revealed that Blacks have the most racially homophilous networks with 78% of Black indices having only Black network members, 60.5% of White indices had only White network members, and 46.2% of Latino indices had only Latino network members. These racial differences in the degree of homophily were significant (χ2 = 21.4, P < 0.01).
Among the 465 network members who were screened and enrolled in the study, an overall HIV prevalence rate of 9% was found. As shown in Figure 1, significantly higher HIV prevalence (P < 0.05) was found in network members of Black (11.4%) and Latino (10.4%) injectors than in the networks of White injectors (5.3%).
Research has demonstrated that racial disparities in HIV cannot be explained by individual behavior patterns alone. Previous research has shown that Blacks' and Latinos' higher rates of infection as compared with Whites are not because of higher levels of risk behavior among them but instead to low or average levels of risk behavior but within risk networks that have a higher baseline rate of HIV infection. Our study provides additional evidence supporting this notion. Furthermore, our unique data show that HIV-negative injectors who are Black and Latino are more likely than White injectors to have HIV in their risk networks.
Consistent with other studies, our data reveal that Black and Latino injectors engage in less risk behaviors than White injectors but bear a disproportionate burden of HIV disease. Our results show that neither injection risk behaviors nor sex behaviors alone explain disparities. However, when we examine syringe exchange use and substance abuse treatment, we saw that Blacks were significantly less likely to access both services. Syringe exchange programs and substance abuse treatment are the 2 effective HIV prevention services for IDUs; so, it might be concluded that lower use of these services among Black injectors partially explains the disparities. However, the reduced use of syringe exchange programs and drug treatment among Blacks has not translated into greater injection risk. Data from this study and others show that the prevalence of needle sharing among Blacks is low, and thus, risk reduction messages and practices have penetrated Black injector networks. The results from this study indicate that continuing racial/ethnic disparities in HIV are related to the significantly higher prevalence of HIV in the networks of Black, and to a lesser degree Latino, HIV-negative IDUs.
The creation of social networks occurs not by chance but according to predictable patterns of behavior, with racial preferences being of particular importance. In particular, risk behaviors occur as a constellation of behaviors where the presence of one risk behavior generally indicates the presence of others as well.22 There is not a great deal of research concerning the formation of IDU risk networks, but normative social network formation demonstrates that individuals are more likely to include people in their network if they are of the same race,23 social circumstances,24 and life stage.25 These social networks tend to have consistent membership over time, with a solid core membership that often lasts over a decade.24 Ties that are cross-gender or cross-race are more likely to be dropped than are ties among demographically similar friends.26
Racial homogeneity within networks has implications for sex partnering and disease transmission. Mixing patterns between individuals in different sex activity classes (ie, different rates of partner change) vary according to race/ethnicity.27 In the United States, individuals tend to select sex partners within their own race/ethnic group, but this has adverse consequences for Blacks given higher background prevalence of infectious diseases within Black communities. Laumann and Youm28 demonstrated this with a nationally representative sample, showing that same race partnering results in lower risk "peripheral" Blacks partnering with higher risk "core" Blacks; this disassortative mixing facilitates spread of infection through the Black population.
The results of this study indicate the need for more research and interventions that target individuals and their networks. Without including sex and drug partners in HIV research and interventions, preventing HIV among all races will be greatly stalled. Furthermore, although the data show that there is a high degree of assortative racial mixing within networks, there is still a fair degree of mixing: 20%–40% among Blacks and Whites and over 50% among Latinos. This suggests the possibility of infection spread between races. Latinos, in particular, may operate as a bridging group. More research is needed on Latino injectors and their network and behavioral norms.
Observers of the racial disparities in prevalence and incidence of HIV infections and AIDS diagnoses in the United States have concluded that these disparities exist because prevention messages, supplies, and/or interventions do not effectively reach those at greatest risk of infection. In essence, such interpretations suggest that Blacks and Latinos continue to practice more risk behaviors than Whites. Results reported here suggest that this is not the case but instead that Black and Latino IDUs who are HIV negative have a higher prevalence of HIV in their social and risk networks and thus a higher probability of encountering the virus even with relatively low rates of risk behaviors. Interventions with greater impact among Blacks will be required to reduce transmission. This conceptualization has implications for both research and prevention. For Black and Latino injectors, minimal engagement in risky behaviors brings with it a higher likelihood of infection. Risk reduction messages, supplies, and interventions targeted toward HIV-negative IDUs remain important prevention activities and linkage to care efforts for HIV-positive injectors.29
1. Centers for Disease Control and Prevention. HIV/AIDS Fact Sheet: Estimates of New HIV Infections in the United States, 2006-2009. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2011.
2. Kraut-Becher J, Eisenberg M, Voytek C, et al.. Examining racial disparities in HIV: lessons from sexually transmitted infections research. J Acquir Immune Defic Syndr. 2008;47:S20–S27.
3. Aids Activities Coordinating Office. 2012 Surveillance Report. Philadelphia, PA: Philadelphia Department of Public Health; 2012.
4. Centers for Disease Control and Prevention. Updated Slide Set: HIV Surveillance in Injection Drug Users Through 2010. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2012.
5. De P, Cox J, Boivin JF, et al.. The importance of social networks in their association to drug equipment sharing among injection drug users: a review. Addiction. 2007;102:1730–1739.
6. Jolly AM, Muth SQ, Wylie JL, et al.. Sexual networks and sexually transmitted infections: a tale of two cities. J Urban Health. 2001;78:433–445.
7. Klovdahl AS. Social networks and the spread of infectious diseases: the AIDS example. Soc Sci Med. 1985;21:1203–1216.
8. Leigh Brown A, Lycett S, Weinert L, et al.. Transmission network parameters estimated from HIV sequences for a nationwide epidemic. J Infect Dis. 2011;204:1463–1469.
9. May R, Anderson R. Transmission dynamics of HIV infection. Nature. 1987;326:137–142.
10. Millett GA, Peterson JL, Flores SA, et al.. Comparisons of disparities and risks of HIV infection in black and other men who have sex with men in Canada, UK, and USA: a meta-analysis. Lancet. 2012;380:341–348.
11. Perisse ARS, Langenberg P, Hungerford L, et al.. The use of supplementary techniques to increase recall of sex partners in a network-based research study in Rio de Janeiro, Brazil. Sex Transm Dis. 2008;35:674–678.
12. Sharpe TT, Harrison K, Dean H. Summary of CDC consultation to address social determinants of health for prevention of disparities in HIV/AIDS, viral hepatitis, sexually transmitted diseases, and tuberculosis. Public Health Rep. 2010;125(suppl 4):11–15.
13. Ward H. Prevention strategies for sexually transmitted infections: importance of sexual network structure and epidemic phase. Sex Transm Infect.2007;83(suppl 1):i43–i49.
14. Hallfors DD, Iritani BJ, Miller WC, et al.. Sexual and drug behavior patterns and HIV and STD racial disparities: the need for new directions. Am J Public Health. 2007;97:125–132.
15. Amesty S, Rivera AV, Fuller CM. Overview of HIV among injection drug users in New York City: critical next steps to eliminate racial/ethnic disparities. Subst Use Misuse. 2001;46:285–294.
16. DesJarlais DC, Arasteh K, Hagan H, et al.. Persistence and change in disparities in HIV infection among injection drug users in New York City after large-scale syringe exchange programs. Am J Public Health. 2009;99:S445–S451.
17. Friedman SR, Chapman TF, Perlis TE, et al.. Similarities and differences by race/ethnicity in changes of HIC seroprevalence and related behaviors among drug injectors in New York City, 1991–1996. J Acquir Immune Defic Syndr. 1999;22:83–91.
18. Kottiri BJ, Friedman SR, Neaigus A, et al.. Risk networks and racial/ethnic differences in the prevalence of HIV infection among injection drug users. J Acquir Immune Defic Syndr. 2002;30:95–104.
19. Latkin CA, Donnell D, Metzger D, et al.. The efficacy of a network intervention to reduce HIV risk behaviors among drug users and risk partners in Chiang Mai, Thailand and Philadelphia, USA. Soc Sci Med. 2009;68:740–748.
20. Perkins WE, Metzger D. AIDS in Philadelphia: emerging from the shadow of crack. In: Bowser B, Quimby E, Singer M, eds. When Communities Assess Their AIDS Epidemics. Lanham, MD: Lexington Books; 2007:88–101.
21. Latkin CA, Mandell W, Vlahov D. The relationship between risk networks' patterns of crack cocaine and alcohol consumption and HIV-related sexual behaviors among adult injection drug users: a prospective study. Drug Alcohol Depend. 1996;42:175–181.
22. Friedman S, Neaigus A, Jose B, et al.. Sociometric risk networks and risk for HIV infection. Am J Public Health. 1997;87:1289–1296.
23. Latkin C, Mandell W, Vlahov D, et al.. People and places: behavioral settings and personal network characteristics as correlates of needle sharing. J Acquir Immune Defic Syndr Hum Retrovirol. 1996;13:273–280.
24. Suh T, Mandell W, Latkin C, et al.. Social network characteristics and injecting HIV-risk behaviors among street injection drug users. Drug Alcohol Depend. 1997;47:137–143.
25. Latkin C, Mandell W, Vlahov D, et al.. Personal network characteristics as antecedents to needle-sharing and shoting gallery attendance. Soc Networks. 1995;17:219–228.
26. Neaigus A, Friedman S, Jose B, et al.. High-risk personal networks and syringe sharing as risk factors for HIV infection among new drug injectors. J Acquir Immune Defic Syndr Hum Retrovirol. 1996;11:499–509.
27. Turner KM, Garnett GP, Ghani AC, et al.. Investigating ethnic inequalities in the incidence of sexually transmitted infections: mathematical modeling study. Sex Transm Infect. 2004;80:379–385.
28. Laumann EO, Youm MA. Racial/ethnic differences in the prevalence of sexually transmitted diseases in the United States: a network explanation. Sex Transm Dis. 1999;26:250–261.
29. DesJarlais DC, Cooper HLF, Bramson H, et al.. Racial and ethnic disparities and implications for the prevention of HIV among persons who inject drugs. Curr Opin HIV AIDS. 2012;7:354–361.