Epidemiology and Social: Concise Communication
Risk factors driving the emergence of a generalized heterosexual HIV epidemic in Washington, District of Columbia networks at risk
Magnus, Manyaa; Kuo, Irenea; Shelley, Katharinea; Rawls, Anthonya; Peterson, Jamesa; Montanez, Luza; West-Ojo, Tiffanyb; Hader, Shannonb; Hamilton, Florac; Greenberg, Alan Ea
aThe George Washington University School of Public Health and Health Services, USA
bDistrict of Columbia, Department of Health, HIV/AIDS Administration, USA
cFamily and Medical Counseling Service Inc., Washington, District of Columbia, USA.
Received 19 November, 2008
Revised 21 February, 2009
Accepted 5 March, 2009
Correspondence to Manya Magnus, PhD, MPH, Department of Epidemiology and Biostatistics, The George Washington University School of Public Health and Health Services, Ross Hall Suite 120b, 2300 Eye Street NW, Washington, DC 20037, USA. Tel: +1 202 994 3024; fax: +1 202 994 0082; e-mail: email@example.com
Objectives: Washington, District of Columbia has the highest HIV/AIDS rate in the United States, with heterosexual transmission a leading mode of acquisition and African–American women disproportionately affected. The purpose of this study was to examine risk factors driving the emergence of the local epidemic using National HIV Behavioral Surveillance data from the District of Columbia.
Design: The design of the study is cross-sectional.
Methods: Individuals at high risk for HIV based on connection to areas with elevated AIDS and poverty were collected from December 2006 to October 2007. Analyses characterized participants from a respondent-driven, nonclinic-based sample; factors associated with preliminary HIV positivity were assessed with logistic regression.
Results: Of 750 participants, 61.4% were more than 30 years of age, 92.3% African–American, and 60.0% with an annual household income of less than $10 000; 5.2% (95% confidence interval, 2.9–7.2%) screened HIV positive; women were more likely to screen positive than men (6.3 versus 3.9%). Of those, 47.4% (95% confidence interval, 30.9–78.7%) did not know their status prior to the study. Last vaginal sex was unprotected for 71.2% of respondents; 44.9% reported concurrent sex partners, and 45.9% suspected concurrency in their partners. Correlates of screening HIV positive were identified.
Conclusion: This study suggests that a generalized heterosexual HIV epidemic among African–Americans in communities at risk may be emerging in the nation's capital alongside concentrated epidemics among men who have sex with men and injecting drug users. Innovation of prevention strategies is necessary in order to slow the epidemic in District of Columbia.
The District of Columbia is experiencing a ‘Modern HIV Epidemic’  – modern in its scope and heterogeneity. With an HIV/AIDS case rate nearly 10 times the United States rate, its AIDS case rate is higher than that of Baltimore, Philadelphia, New York City, Detroit, and Chicago; one in 20 to one in 50 of approximately 600 000 residents of District of Columbia is living with HIV/AIDS [1,2]. Black individuals are at disproportionate risk, heterosexual transmission is the leading cause of newly reported HIV infections (∼37%), and the number of women with AIDS has increased by 76%  over the past 6 years, with newly reported HIV cases elevated among women (47 per 100 000 residents ). These data suggest that alongside concentrated epidemics among men who have sex with men (MSM) and injecting drug users (IDU), there may be a generalized HIV epidemic emerging among heterosexuals in the capital of the United States.
Understanding the local epidemic may enable the development of effective prevention efforts . Several studies [4–16] have shown that multiple sex partners, paid sex, coinfection with sexually transmitted infections, early sexual debut, lack of circumcision, age mixing, sex partner concurrency, and herpes simplex virus-2 are associated with increased heterosexual transmission. Centers for Disease Control and Prevention (CDC) data indicating more incident infections among Black individuals in the United States than previously thought  underscore the need to identify population-specific prevention interventions [5,6]. This study describes risk factors among heterosexuals in the District of Columbia in order to inform prevention strategies.
Data were obtained through the CDC-funded National HIV Behavioral Surveillance (NHBS) District of Columbia site [7–10]. NHBS methods have been described elsewhere [7–11]. Briefly, cross-sectional data from MSM, IDU, and heterosexuals at high risk of HIV infection (HET) are collected in repeated, annual cycles. For NHBS-HET, a nonclinic-based sample considered to be at risk for HIV due to its connection with geospatial areas of high risk defined by high AIDS prevalence (reported between 2001 and 2006) and poverty levels as defined by the 2000 census  was recruited. Eligible individuals were males and females 18–50 years old who had sex with a member of the opposite sex in the past year, resided in, or were referred by a friend, sex partner, or family member who lived in the high-risk area (HRA), and had not injected drugs in the past year. Participants were recruited between December 2006 and October 2007 via respondent-driven sampling (RDS) [13–21], a chain-referral method that accesses hard-to-reach populations, and, if all assumptions are met [participants know one another as members of the population, dense networks to sustain long chains, and recruitment waves without too much insularity (homophily)], allows for generalizability to the population of networks from which it was drawn [14,15]. Nonrandomly identified ‘seeds’ who were otherwise eligible, were knowledgeable about the community, and were dynamic individuals capable of beginning the recruitment chains. Seeds were given coupons for three people from their social and/or sexual networks to join the study. In this study, RDS assumptions regarding waves, referral chain length, and homophily were met. All individuals met eligibility criteria, had a coupon, completed the survey in English or Spanish, and provided informed consent.
Participants completed an anonymous interviewer-administered survey on sexual, drug use, and health behaviors, and a rapid oral HIV screening test (OraQuick ADVANCE Rapid HIV-1/2 Antibody Test; OraSure Technologies Inc., Bethlehem, Pennsylvania, USA) . Participants received $25.00 for the interview, $10.00 for the test, and $10.00 for each eligible participant referred. Interviews were conducted at a local community health clinic. Those screening HIV positive were referred for western blot and care at the clinic that provides HIV care downstairs from the research unit. Participants could complete both the survey and the HIV test (95.6%) or the survey without the HIV test (4.4%). Not all participants returned for western blot (of those screening positive, n = 17 refused western blot), thus for this analysis only preliminary positives were analyzed; this differs from the CDC's strong recommendations to use only western blot-confirmed tests. Participants who tested negative on the western blot were coded as HIV negative. All activities were overseen by the CDC, approved by the District of Columbia Department of Health (DOH) and George Washington University (GWU) Institutional Review Boards (IRBs), and guided by a community advisory board.
Individualized weights incorporating network characteristics and referral patterns were generated and applied to adjust for interrelatedness of RDS-recruited participants. RDS allows for a final sample independent from the seeds, provided that RDS assumptions are met. If met, calculation of sampling probabilities that provide population-based estimates of variables under study is possible [19,23–33] is possible. Per CDC protocol, recruitment chains were truncated, with persons living outside of an HRA or having injected drugs in the past year being ineligible to recruit. Chi-square tests were used to compare weighted versus unweighted estimates; with no significant differences between weighted and unweighted estimates (P > 0.10), unweighted logistic regression was used to model having a reactive HIV screening result. Variables moderately significant in bivariate analysis were tested for inclusion in the models and remained if they were statistically significant or if addition or removal resulted in a change (±5%) in the estimates. Respondent Driven Sampling Analysis Tool (RDSAT), version 5.6.0 (Ithaca, New York, USA), Statistical Analysis Software (SAS) 8.1 software (SAS Institute Inc., Cary, North Carolina, USA), and Stata 9.0se (College Station, Texas, USA) were used for analysis.
Overall, 1106 nonseed participants were screened; 915 (82.7%) were eligible, 785 (86.3%) had not injected drugs in the past year, and 750 (95.5%) had complete data. Of 191 ineligible participants, 41.1% were ineligible due to age and 38.3% due to no sex in the past year.
The majority of participants was more than 30 years old, black, never married, heterosexual, at least high school graduate, and unemployed, with nearly two-thirds reporting an annual household income of less than $10 000 (Table 1). More than half of the participants had ever been to jail, prison, or detention. Women were significantly more likely than men to be bisexual, to be of high school level or higher, and earn less than $10 000 annually; men were more likely than women to lack health insurance and to have ever been in jail, prison, or juvenile detention and to have been arrested by police and booked in the past 12 months.
Overall, 5.2% [95% confidence interval (CI), 2.9–7.2%] of participants screened HIV positive. Women were more likely to screen positive than men [6.3% (95% CI, 3.3–9.7%) versus 3.9% (95% CI, 1.6–5.7%)]. Nearly half [47.4% (95% CI, 30.9–78.7%)] of those who screened HIV positive did not know their HIV status prior to study participation. Despite the fact that more than three-quarters had seen a healthcare provider in the past 12 months, only half were offered an HIV test.
Three-quarters reported having three or fewer sex partners in the past year, and most reported their last sex partner as being a main partner (‘someone you have sex with and who you feel committed to above anyone else’). Condom use was low, with 71.2% of the most recent vaginal sex acts unprotected. Reported concurrency (having sex with someone else during the time they were having sex with their last partner) and perceived partner concurrency were common, with half of participants reporting a concurrent sexual relationship during at least one of their sexual partnerships in the past year. Men were more likely to report concurrency than women (P < 0.05), and half of the participants believed their partner had concurrent sex outside the partnership. Use of alcohol and/or drugs with sex, older partners, having been arrested and incarcerated, and experiencing depressive symptoms and emotional and/or physical abuse were common.
Although women were more likely than men to screen HIV positive (P> 0.05), they were less likely than men to have sexual debut at less than 13 years of age, have more than one sex partner in the past 12 months, and have a casual partner (having sex with someone one does not feel committed to or does not know very well) or exchange partner (having sex with someone in exchange for money or drugs) as their last sex partner to report use of alcohol at last sex (P < 0.05); they also had fewer total sex partners in the last 12 months than men (P < 0.05).
Among women, past IDU, being incarcerated, having depressive symptoms, and using a condom at last sex were associated with screening HIV positive (Table 2). Emotional and/or physical abuse was collinear with depressive symptomatology, which functioned more stably in the models; thus abuse was excluded from the multivariable models. For men, increased age was associated with screening HIV positive.
This study offers the first estimate of HIV and risk behaviors among urban, low income, and African–Americans in the nation's capital, revealing an HIV prevalence of 5.2%, with 6.3% among women and 3.9% among men. These data corroborate those from the District of Columbia DOH surveillance system , with the prevalence among women approaching that of Tanzania (7.0%) and Uganda (7.1%) in this sample . HIV risk behaviors reported here are parallel to those found in studies of heterosexual epidemics elsewhere in the world [23,24,34] and the United States [5,8,26–34]. The lack of condom use, early sexual debut, substance abuse, depressive symptomatology, emotional and/or physical abuse, and concurrency estimates were similar to those found in other studies [25–31,35,36]. Other research suggests [3,5,25–31,35,36] that the density of sexual networks in tandem with high rates of concurrency and unprotected heterosexual sex may be causal factors in expanding the epidemic among heterosexuals in the District of Columbia. We did not find evidence of high numbers of heterosexual males engaging in MSM behavior, and women had fewer sexual risk factors than men; this lack of traditional individual-level risk factors has been reported previously [3,25–31,35,36].
As a cross-sectional study, temporality and causality cannot be inferred. Risk behaviors are based on self-report and must be interpreted cautiously. The primary outcome was oral HIV screening result. Recent evaluation of oral HIV screening in New York City found false positive rates somewhat higher than the range presented by OraSure Technologies Inc. . Applying the wider range of false positives from oral specimens found in New York City (0.11–1.11%), we estimate that no more than one additional false positive would be likely among those with unknown HIV status (possible false positive range: n = 0.02–0.17); however, these results cannot be interpreted to be the true population prevalence, but only an estimate.
All RDS assumptions were met, suggesting that this sample is generalizable to the underlying population of networks from which it was drawn [13–21,33,37,38]. Biases in the sample may have been introduced due to the exclusion of current IDU, incomplete data, or RDS itself . To the extent that RDS can, we believe the weighted estimates may be applied to the population of networks living in or connected to communities from which they were recruited. Despite its limitations, use of RDS not only provides generalizable estimates, but also allows access to hard-to-reach populations often omitted from convenience samples.
This study suggests that the emergence of an HIV epidemic among heterosexual African–American individuals in communities and networks at risk for HIV may be network-based rather than individually-based. The primary risk may simply be engaging in normative heterosexual sexual behavior within a network with a high prevalence of HIV. Risks are perpetuated through concomitant behaviors such as having unprotected sex, partner concurrency, substance use, and early sexual debut. Innovation, implementation, and evaluation of evidence-based sex-specific and race-specific prevention strategies are critical steps towards slowing the HIV epidemic in the District of Columbia.
This study was funded by District of Columbia, DOH/HIV-AIDS Administration (HAA), contract number #POHC-2006-C-0030, funded in part by grant number #PS000966-01, from the United States Department of Health and Human Services (DHHS)/CDC. All coauthors have reviewed and approved the final draft of the study, including those from the District of Columbia DOH/HAA. Under the partnership contract, the District of Columbia DOH/HAA had the right to review and approve the manuscript. The content of this publication does not necessarily reflect the views or policies of DHHS/CDC, and responsibility for the content rests solely with the authors.
For their assistance and expertise throughout the study, the authors acknowledge Dr Amanda Castel, Dr Dante Verme, Ms Michelle Folkers, Ms Dodi Hathazi, Ms Sarah Jackson, Ms Katrina Jones, Ms Mackenzie Nowell, and Ms Manuela Sifuentes of GWU School of Public Health and Health Services (SPHHS); Dr Amy Lansky, Dr Elizabeth DiNenno, Ms Amy Drake, and Ms Tricia Martin of CDC; the Washington Outreach Research Drive to Understand Prevention (WORD UP) Community Advisory Board Members; and Dr Marsha Martin, Mr Paul Cunningham, Ms Marie Sansone, and Mr Leo Rennie formerly of the District of Columbia DOH/HAA. This study could not have been conducted without the enormous support of Ms Angela Wood, Mr Ricardo Branic, and the Family and Medical Counseling Service Inc. (FMCS) management and staff. For their participation in and support of NHBS, the study team would like to acknowledge the participants of the study and the citizens of Washington, District of Columbia, without whom this study would not have been possible.
Manya Magnus, PhD, MPH – principal investigator and primary responsibility for study; conducted all analyses for manuscript, and holds the primary writing responsibility of manuscript.
Irene Kuo, PhD, MPH – Project Director/Coinvestigator; directed all day-to-day functions of study, integral to analyses and writing of manuscript.
Katharine Shelley, MPH – Data Manager; responsible for all data cleaning, uploading, and management, data analysis, and participated materially in writing and review of manuscript.
Anthony Rawls – Field Supervisor; directed all day-to-day functions of study in the field and participated materially in writing and review of manuscript.
James Peterson, EdD, MHS – conducted all formative and qualitative data collection and analysis and participated in writing and review of manuscript.
Luz Montanez – Field Interviewer; actively involved in all day-to-day interview functions of study and also reviewed the manuscript.
Tiffany West-Ojo, MSPH, MPH – Surveillance Division Bureau Chief, DC DOH co-PI and integral to oversight of study; reviewed the manuscript.
Shannon Hader, MD, MPH – Director of DOH/HAA; participated materially in writing and review of the manuscript.
Flora Hamilton, DSW – Executive Director of FMCS; participated in overseeing collaboration at FMCS and reviewed the manuscript.
Alan E. Greenberg, MD, MPH – Principal Investigator of Public Health-Academic Partnership between GWU and the District of Columbia DOH/HAA, primary overall oversight for NHBS within the Partnership; guidance for faculty and staff; participated materially in writing and review of the manuscript.
1. District of Columbia Department of Health, HIV/AIDS Administration. District of Columbia HIV/AIDS Epidemiology Annual Report 2007; 2007.
2. Levine S. Study calls HIV in D.C. a ‘Modern Epidemic’
. The Washington Post 2007; 2007.
3. Wilson DHD. ‘Know your epidemic, know your response’: a useful approach, if we get it right. Lancet 2008; 372:423–426.
4. Hall H, Song R, Rhodes P, Prejean J, An Q, Lee LM, et al
. Estimation of HIV incidence in the United States. JAMA 2008; 300:520–529.
5. Osborn J. The past present, and future of AIDS. JAMA 2008; 300:581–583.
6. Wilson P, Wright K, Isbell MT. Left behind: Black America: a neglected priority in the Global AIDS Epidemic
. A report of the Black AIDS Institute 2008; 2008.
7. Lansky A, Drake A, DiNenno E, Lee CW. HIV behavioral surveillance among the U.S. general population. Public Health Rep 2007; 122(Suppl 1):24–31.
8. Lansky A, MacKellar D, Gallagher KM, Lin LS, Sullivan PS, Onorato IM. Evaluation of the new National HIV Behavioral Surveillance System (NHBS) is imperative. Sex Transm Dis 2006; 33:272–273, author reply 4–5.
9. Lansky A, Sullivan PS, Gallagher KM, Fleming PL. HIV behavioral surveillance in the U.S.: a conceptual framework. Public Health Rep 2007; 122(Suppl 1):16–23.
10. MacKellar DA, Gallagher KM, Finlayson T, Sanchez T, Lansky A, Sullivan PS. Surveillance of HIV risk and prevention behaviors of men who have sex with men: a national application of venue-based, time-space sampling. Public Health Rep 2007; 122(Suppl 1):39–47.
11. Allen DRFT, Abdul-Quader A, Lansky A. The role of formative research in the National HIV Behavioral Surveillance System. Public Health Rep 2009; 124:26–33.
13. Abdul-Quader AS, Heckathorn DD, McKnight C, Bramson H, Nemeth C, Sabin K, et al
. Effectiveness of respondent-driven sampling for recruiting drug users in New York City: findings from a pilot study. J Urban Health 2006; 83:459–476.
14. Heckathorn D. Respondent-driven sampling: a new approach to the study of hidden populations. Soc Probl 1997; 44:174–199.
15. Heckathorn D. Respondent-driven sampling II: deriving valid population estimates from chain-referral samples of hidden populations. Soc Probl 2002; 49:11–34.
16. Heckathorn D. Extensions of respondent-driven sampling: analyzing continuous variables and controlling for differential recruitment. Sociological Methodol 2007; 37:151–207.
17. Lansky A, Mastro TD. Using respondent-driven sampling for behavioural surveillance: response to Scott. Int J Drug Policy 2008; 19:241–243.
18. McKnight C, Des Jarlais D, Bramson H, Tower L, Abdul-Quader AS, Nemeth C, Heckathorn D. Respondent-driven sampling in a study of drug users in New York City: notes from the field. J Urban Health 2006; 83(6 Suppl):i54–i59.
19. Ramirez-Valles J, Heckathorn DD, Vazquez R, Diaz RM, Campbell RT. From networks to populations: the development and application of respondent-driven sampling among IDUs and Latino gay men. AIDS Behav 2005; 9:387–402.
20. Stormer A, Tun W, Guli L, Harxhi A, Bodanovskaia Z, Yakovleva A, et al
. An analysis of respondent driven sampling with injection drug users (IDU) in Albania and the Russian Federation. J Urban Health 2006; 83(6 Suppl):i73–i82.
21. Deiss RG, Brouwer KC, Loza O, Lozada RM, Ramos R, Cruz MA, et al
. High-risk sexual and drug using behaviors among male injection drug users who have sex with men in 2 Mexico-US border cities. Sex Transm Dis 2008; 35:243–249.
23. Pisani E, Lazzari S, Walker N, Schwartlander B. HIV surveillance: a global perspective. J Acquir Immune Defic Syndr 2003; 32:S3–S11.
24. Joint United Nations Programme on HIV/AIDS (UNAIDS)/World Health Organization (WHO), Guidelines for second generation HIV surveillance; 2000.
25. Adimora AA, Schoenbach VJ. Contextual factors and the black-white disparity in heterosexual HIV transmission. Epidemiology 2002; 13:707–712.
26. Adimora AA, Schoenbach VJ, Bonas DM, Martinson FE, Donaldson KH, Stancil TR. Concurrent sexual partnerships among women in the United States. Epidemiology 2002; 13:320–327.
27. Adimora AA, Schoenbach VJ, Martinson F, Donaldson KH, Stancil TR, Fullilove RE. Concurrent sexual partnerships among African Americans in the rural south. Ann Epidemiol 2004; 14:155–160.
28. Adimora AA, Schoenbach VJ, Martinson FE, Donaldson KH, Stancil TR, Fullilove RE. Concurrent partnerships among rural African Americans with recently reported heterosexually transmitted HIV infection. J Acquir Immune Defic Syndr 2003; 34:423–429.
29. Forna FM, Fitzpatrick L, Adimora AA, McLellan-Lemal E, Leone P, Brooks JT, et al
. A case-control study of factors associated with HIV infection among black women. J Natl Med Assoc 2006; 98:1798–1804.
30. Adimora AA. Public health and the AIDS epidemic in North Carolina. N C Med J 1990; 51:571–572.
31. Adimora AA, Schoenbach VJ, Martinson FE, Donaldson KH, Fullilove RE, Aral SO. Social context of sexual relationships among rural African Americans. Sex Transm Dis 2001; 28:69–76.
32. Centers for Disease Control and Prevention (CDC). False-positive oral fluid rapid HIV tests: New York City, 2005–2008. MMWR Morb Mortal Wkly Rep
33. Abdul-Quader AS, Heckathorn DD, Sabin K, Saidel T. Implementation and analysis of respondent driven sampling: lessons learned from the field. J Urban Health 2006; 83(6 Suppl):i1–i5.
34. Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO). AIDS Epidemic Update 2007; 2007.
35. Adimora AA, Schoenbach VJ, Doherty IA. Concurrent sexual partnerships among men in the United States. Am J Public Health 2007; 97:2230–2237.
36. Adimora AA, Schoenbach VJ, Martinson FE, Coyne-Beasley T, Doherty I, Stancil TR, Fullilove RE. Heterosexually transmitted HIV infection among African Americans in North Carolina. J Acquir Immune Defic Syndr 2006; 41:616–623.
37. Magnani R, Sabin K, Saidel T, Heckathorn D. Review of sampling hard-to-reach and hidden populations for HIV surveillance. AIDS 2005; 19(Suppl 2):S67–S72.
38. Scott G. ‘They got their program, and I got mine’: a cautionary tale concerning the ethical implications of using respondent-driven sampling to study injection drug users. Int J Drug Policy 2008; 19:42–51.
This article has been cited 9 time(s).
Journal of Urban Health-Bulletin of the New York Academy of MedicineDiffering HIV Risks and Prevention Needs among Men and Women Injection Drug Users (IDU) in the District of ColumbiaJournal of Urban Health-Bulletin of the New York Academy of Medicine
AIDS and BehaviorHIV Prevalence and Risk Behaviours from Three Consecutive Surveys Among Men Who Have Multiple Female Sexual Partners in Cape TownAIDS and Behavior
Janac-Journal of the Association of Nurses in AIDS CareA Qualitative Description of Women's HIV Self-Management Techniques: Context, Strategies, and ConsiderationsJanac-Journal of the Association of Nurses in AIDS Care
Journal of Urban Health-Bulletin of the New York Academy of MedicineHIV Prevalence Overall and among High-HIV-Risk Behaviorally Defined Subgroups among Heterosexuals at Community-Based Venues in a Mid-Atlantic, US CityJournal of Urban Health-Bulletin of the New York Academy of Medicine
Clinical Infectious DiseasesDeveloping and Evaluating Comprehensive HIV Infection Control Strategies: Issues and ChallengesClinical Infectious Diseases
Revista De Saude Publica
Magnitude and trend of the AIDS epidemic in Brazilian cities, from 2002 to 2006
Revista De Saude Publica, 44(3):
Clinical Infectious DiseasesAddressing Research Priorities for Prevention of HIV Infection in the United StatesClinical Infectious Diseases
Health AffairsFighting HIV/AIDS In Washington, DCHealth Affairs
Journal of Nursing ScholarshipFocus on Increasing Treatment Self-Efficacy to Improve Human Immunodeficiency Virus Treatment AdherenceJournal of Nursing Scholarship
behavioral surveillance; HIV/AIDS; HIV/AIDS prevention
© 2009 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.