Share this article on:

Identifying the Correlates of Membership in HIV-Serodiscordant Partnerships in New York City

Braunstein, Sarah L. PhD, MPH; Udeagu, Chi-Chi MPH; Bocour, Angelica MPH; Renaud, Tamar MPH; Shepard, Colin W. MD

doi: 10.1097/OLQ.0000000000000007
Original Study

Background Identification and characterization of patients in HIV-serodiscordant partnerships can inform strategies to improve HIV prevention efforts for this group.

Methods We conducted a case-control analysis using New York City Department of Health and Mental Hygiene HIV surveillance and partner services (PS) data from July 2006 to July 2010. HIV-infected index patients reported and interviewed for PS who named 1 or more sex and/or syringe-sharing partner (n = 1309) and their sex partners notified by New York City Department of Health and Mental Hygiene with confirmed HIV serostatus (n = 1564) were selected for analysis. Index patients were classified into either serodiscordant or seroconcordant-positive partnerships based on the HIV serostatus of their partner(s). Multivariable regression analysis was conducted to examine the likelihood of membership in a serodiscordant partnership by a range of individual- and partnership-level variables.

Results Of the 1309 index patients, 624 (48%) were in HIV-serodiscordant partnerships. In multivariable analysis, the likelihood of serodiscordant partnership membership was slightly higher among women, individuals with unknown HIV transmission risk, and those with 2 to 3 named partners versus 1. Index patients claimed more partners than they named; for example, index patients who named 1 partner claimed an average of 2.3 partners in the past 12 months.

Conclusions Many HIV-infected patients who received PS were in HIV-serodiscordant partnerships, with characteristics indicating potential for HIV transmission. Our findings suggest several potential programmatic and policy needs, including enhanced linkage-to-care efforts for this population, especially HIV-infected individuals with uncontrolled viremia; ongoing PS for individuals with evidence of continuing exposure of others; and participation by patients and their serodiscordant, steady partners in local prevention interventions.

An analysis of New York City HIV surveillance and partner services data found a large proportion of HIV-serodiscordant partnerships, with characteristics that may increase likelihood of HIV transmission.

From the HIV Epidemiology and Field Services Program, New York City Department of Health and Mental Hygiene, New York City, NY

Acknowledgment: The authors are grateful to the HIV Field Services Unit and HIV Surveillance Unit staff for their ongoing dedication to curb HIV transmission in New York City.

Supported in part by the Centers for Disease Control and Prevention cooperative agreement nos. U62/CCU223460-06-1 and U62/CCU223460-05-4. The findings and conclusions in this article are those of the authors and do not necessarily reflect the views of the Centers for Disease Control and Prevention.

Conflicts of interest: The authors have no conflicts of interest to disclose.

Correspondence: Sarah L. Braunstein, PhD, MPH, New York City Department of Health and Mental Hygiene, 42-09 28th St, CN44, Queens, NY 11101. E-mail:

Received for publication March 22, 2013, and accepted June 3, 2013.

Years of in-depth behavioral research have revealed the limitations of approaches that focus solely on individual-level risk factors for understanding transmission dynamics of HIV and other sexually transmitted infections (STI).1–3 It is well understood that the impact of individuals’ sexual behavior on disease risk depends not only on his/her sexual behavior but also on the clinical, behavioral, and other characteristics of his/her partners.4 For example, frequent sexual contact with low-risk partners may be less risky (despite the high frequency) than infrequent exposure to high-risk or already-infected partners,5 especially if those HIV-infected partners are not virally suppressed.6

Network theory in public health promotes analysis of both individual- and network-level attributes to explain the distribution and concentration of disease in populations or geographic areas7,8 and has been applied widely in HIV/STI epidemiologic research.9,10 A main objective of this approach is to identify and characterize individuals or partnerships with potential to transmit STI, including HIV, to uninfected sex or syringe-sharing partners. Prevention efforts with serodiscordant couples include risk-reduction counseling, prevention with positives, treatment for the HIV-infected partner, and more recently antiretroviral therapy (ART) for preexposure prophylaxis for HIV infection among uninfected partners.11–14

The HIV Epidemiology and Field Services Program (HEFSP) of the New York City (NYC) Department of Health and Mental Hygiene (DOHMH) conducts surveillance on HIV/AIDS and initiates field investigations and interviews with patients newly diagnosed with HIV. Nearly 3500 people are newly diagnosed as having HIV and reported annually to the NYC DOHMH.15 The Field Services Unit (FSU) of HEFSP was established in 2006 to provide partner services (PS) to New Yorkers recently diagnosed with HIV and other persons with a known HIV diagnosis needing assistance with partner notification.

At its core, HIV PS involves notifying partners of their exposure to HIV and testing notified partners. HIV PS has traditionally been viewed as a final step in the HIV testing process: a person is tested for HIV and if he/she tests positive, PS is provided and its outcomes are recorded (eg, number of new diagnoses per persons interviewed). However, particularly for index patients in serodiscordant partnerships, PS may be more fruitfully viewed as a starting point for intervention, for example, as a mechanism for linking the index patient with HIV care and supporting his or her retention in care, to ultimately achieve viral suppression.

Through a joint analysis of surveillance and available HIV PS interview data, we aimed to estimate the magnitude of serodiscordant partnerships among persons interviewed for PS in NYC between 2006 and 2010 and to characterize index patients in such partnerships, to deepen understanding of the local contributors to HIV transmission and identify avenues for prevention and early intervention.

Back to Top | Article Outline


Data Sources

Data for the analysis were derived from 2 main sources: Field Services Unit’s PS database and the HIV surveillance registry.

The FSU interviews HIV-infected patients for partner elicitation with a deliberate focus on recently diagnosed patients (including those diagnosed with acute HIV infection) and patients previously diagnosed but in need of HIV PS, including pregnant and women individuals newly diagnosed with an STI. Field Services Unit activities are centered on patients seeking HIV care at facilities with an established relationship to DOHMH for the purposes of increasing provision of PS to HIV-infected patients.16 Index patients sought for interview are identified by review of laboratory reports by FSU staff or provider referral at participating FSU facilities, or via new case reports to HIV surveillance. By the end of 2011, FSU partner facilities included 55 clinical facilities and their affiliates located throughout NYC, tuberculosis clinics citywide, and NYC jails. Together, these facilities are responsible for approximately 40% of new HIV diagnoses in NYC and the ongoing care of approximately 20,000 persons living with HIV/AIDS. Index patients who are successfully located are interviewed for partners and other information, including sociodemographics, risk behaviors, and HIV medical care status. Field Services Unit disease intervention specialists locate and notify partners of exposure to HIV, offer HIV testing and risk-reduction education, and assist with follow-up medical care if HIV infected. All case and partner investigations and outcomes are recorded using a standardized paper questionnaire.

The NYC HIV surveillance registry (the Registry) is a population-based registry of all persons diagnosed with AIDS (since 1981) or HIV infection (since 2000) and reported to the NYC DOHMH according to standard Centers for Disease Control and Prevention case definitions.17 The Registry contains demographic, HIV transmission risk and clinical information on HIV-diagnosed persons, as well as all Western blot tests, viral load tests, and CD4 counts reportable under New York State law.18 HIV transmission risk categories followed an expanded local definition.15

Back to Top | Article Outline

Analysis Design

This case-control analysis used retrospective data from FSU interviews and the Registry over a 4-year period, July 1, 2006, to July 1, 2010. The primary aim was to classify a sample of FSU index patients into 1 of 2 partnership types based on the HIV serostatus of the sex partners they named. The unit of analysis was the HIV-infected index patient reported to FSU during the 4-year period of interest. Index patients included in the sample were interviewed by FSU and named at least 1 sex and/or syringe-sharing partner. Partners were named by an index patient, were notified, and had a confirmed positive or negative HIV test result.

Back to Top | Article Outline

Partnership Type Assignment

Using information on the HIV serostatus of named partners, HIV-infected index patients were classified into 2 partnership types for analysis. “Serodiscordant partnerships” included index patients who named at least 1 HIV-negative partner regardless of the serostatus of other named partners. “Concordant-positive partnerships” included index patients who named only partner(s) who were HIV positive.

Back to Top | Article Outline

Statistical Analysis

Descriptive analyses were performed to characterize the patient sample and compare it with New Yorkers diagnosed with HIV in 1985–2010. Variables were created to capture attributes of index patients’ partnerships by relating index patient characteristics to those of their partners. For example, an index patient’s race/ethnicity was compared with his/her partners’ race/ethnicity. If the 2 were different, the index patient was assigned a “1” indicating race disassortativity, or mixing by race, with his/her partner(s). Age disassortativity was defined as a difference of 5 or more years between the index patient and at least 1 of the partners’ ages. Index patients classified into partnership types were then compared on a range of individual-level and partnership variables. χ 2 Tests were used to identify statistically significant differences at P < 0.05 for categorical variables.

Regression analyses were performed to examine the association between individual-level and partnership variables and partnership type. Variables from the bivariable analysis that were statistically significant (at P < 0.05) were added to an age-adjusted multivariable regression model of the likelihood of membership in a serodiscordant partnership. Adjusted prevalence ratios (aPRs) and 95% confidence intervals (CIs) were generated.

Back to Top | Article Outline


Patient Sample

In total, 5119 index patients were reported to FSU from 2006 to 2010, of whom 4154 were interviewed (81%; Fig. 1). Among those interviewed, 3887 index patients (76% of total reported) named a total of 4792 partners and 267 did not name partners. The final analytic sample comprises 1309 index patients who named 1564 sex and/or needle-sharing partners for whom HIV serostatus was ascertained through matching with the Registry or by self-report confirmed through medical record review by FSU staff, provider report, or HIV testing performed by FSU staff.

Figure 1

Figure 1

Back to Top | Article Outline

Partnership Type Assignment

Among the 1309 index patients, 624 (48%) were classified into serodiscordant partnerships and 685 (52%) were classified into concordant-positive partnerships. Within serodiscordant partnerships, there were 87 index patients with mixed positive and negative partnerships (with 16 of the positive partners being diagnosed by FSU after notification) and 537 index patients whose named partners were all HIV negative. Within concordant-positive partnerships, there were 111 index patients with at least 1 partner diagnosed with HIV by FSU after notification, and the remaining 574 index patients named all previously diagnosed partners. There were discrepancies in general between the number of partners claimed (past 12 months) by index patients and the partners named (last 2 years) who were included in this analysis. For example, among index patients who named 1 partner, the number of partners claimed in the past 12 months ranged from 0 to 104 (median, 1; mean, 2.3), with no difference by partnership type.

Back to Top | Article Outline

Patient Sociodemographic Characteristics

Tables 1 and 2 present selected individual- and partnership-level characteristics of the overall analytic sample of 1309 FSU index patients. Most patients in the sample were aged 20 to 39 years at diagnosis (56%), male (61%), and black (58%) or Hispanic (38%). More than one fifth (22%) had been diagnosed concurrently with HIV/AIDS (AIDS diagnosis within 31 days of HIV diagnosis). Nearly half (45%) of patients had heterosexual HIV transmission risk; more than one third (35%) were foreign-born; and 12% reported an STI diagnosis in the past 12 months. Index patients’ partnerships were frequently mixed by NYC borough (58%) and age (59%), although less often by race/ethnicity (33%). Most partnerships comprised 1 index patient and 1 named partner (86%).





Compared with other NYC HIV-diagnosed persons during the same period, a higher proportion of FSU patients in this analysis were young (32% vs. 26% aged 20–29 years), female (39% vs. 25%), and black or Hispanic (96% vs. 79%). In addition, sampled patients were more likely to report a history of homelessness (14% vs. 2%) or incarceration (14% vs. 9%) and to report heterosexual HIV transmission risk (45% vs. 22%).

Back to Top | Article Outline

Comparison of Partnership Types

Several individual-level characteristics differed significantly by partnership type (Table 1). Compared with index patients in concordant-positive partnerships, index patients in serodiscordant partnerships were more likely to be female (P < 0.01), more likely to have unknown HIV transmission risk compared with men who have sex with men (MSM; P < 0.001), and more likely to report substance use (either injection or noninjection drug use or binge drinking of ≥5 drinks on 1 occasion) in the past year (P = 0.03). Both male and female index patients in serodiscordant partnerships claimed a higher number of (male) sex partners in the past 12 months compared with those in concordant-positive partnerships. There were no significant differences by age, race/ethnicity, history of exchange sex, or acute versus chronic HIV infection at diagnosis or in the distribution of CD4 cell count within 3 months of HIV diagnosis.

Several partnership characteristics also differed between the 2 partnership types (Table 2). Compared with concordant-positive partnerships, serodiscordant partnerships had less same-sex behavior (P < 0.001) and more mixing by NYC borough (P < 0.001). Index patients in serodiscordant partnerships named more partners than those in concordant-positive partnerships. In fact, most persons in both types of partnerships were in a dyad in which index patients named only 1 partner (77% of serodiscordant partnerships, 94% of concordant-positive partnerships).

Back to Top | Article Outline

Correlates of Membership in an HIV-Serodiscordant Partnership

In an age-adjusted multivariable model, several individual-level and partnership factors were independently associated with membership in a serodiscordant partnership (Table 3). The likelihood of membership in a serodiscordant partnership was slightly higher among women (aPR, 1.07; 95% CI, 1.01–1.13) and among those with unknown HIV transmission risk (aPR, 1.10; 1.04–1.17). Index patients in serodiscordant partnerships were somewhat more likely to identify and name 2 to 3 partners versus 1 (aPR, 1.10; 1.04–1.17). Finally, the likelihood of serodiscordant partnership membership was somewhat higher among index patients whose partner(s) lived in a different NYC borough (aPR, 1.05; 1.01–1.09).



To further explore the finding that female index patients were more likely to be classified in serodiscordant partnerships, we compared women by partnership type. Compared with women classified in a seroconcordant-positive partnerships, women classified in serodiscordant partnerships had both claimed more sex partners in the past 12 months (P < 0.0001) and named more sex partners for inclusion in this analysis, were younger at HIV diagnosis (P = 0.03) and at FSU interview (P = 0.04), more frequently had perinatal or unknown HIV transmission risk (P = 0.02), and were more likely to report drug or alcohol use in the past 12 months (P = 0.01) and have partnerships that were mixed by NYC borough (P < 0.0001; data not shown).

Back to Top | Article Outline


In this retrospective analysis of combined PS and HIV surveillance data in NYC, we found a large proportion of serodiscordant partnerships among interviewed index patients and their named sex or syringe-sharing partners. Serodiscordant partnerships present important opportunities for secondary intervention to prevent ongoing transmission. Discrepancies between the number of partners claimed and number named by index patients underscore the potential underestimation of serodiscordant partnerships in such an analysis that relied on data from named and tested partners.

Index patients and their partners in serodiscordant partnerships had several characteristics that may increase the likelihood of HIV transmission. For example, index patients in these partnerships were more likely to report multiple recent sex partners. This is expected given persons who name larger numbers of partners are likely to name a negative partner; persons with fewer named partners may be in a married or cohabitating partnership, with both partners already HIV infected. Furthermore, patients in serodiscordant partnerships were more likely to have been diagnosed concurrently with HIV and AIDS, and many of these patients already had evidence of CD4 decline at HIV diagnosis. Late HIV diagnosis represents a missed opportunity for early treatment and control of viral load, and increases the likelihood that these patients were infectious to partners before diagnosis.19 Particular emphasis should be placed on early and frequent testing for the HIV-negative partner in a serodiscordant couple. In addition, a high proportion of patients in serodiscordant partnerships reported past-year drug use and/or binge drinking, factors that have been linked to HIV risk behaviors.20

The increased likelihood of female patients being in serodiscordant partnerships warrants further exploration. One explanation for this finding is that women may have motivation to name certain partners over others (eg, current partners over past partners from whom they may have acquired HIV infection, or HIV-negative over HIV-positive partners if they fear negative consequences of disclosure). Whatever the reason for the finding, it suggests potential benefit from including women as a target of secondary HIV prevention efforts alongside traditional high-risk groups such as MSM.

Serodiscordant partnerships were also geographically diverse, suggesting potential for these at-risk partnerships to act as bridges between low- and high-prevalence neighborhoods. Other geospatial analyses of HIV distribution in smaller US cities have shown that people tend to find sex partners who live nearby.21 However, in a well-connected, large, urban environment such as NYC, there may be a greater potential for partnerships to span neighborhoods.

The finding that index patients in serodiscordant partnerships were more likely to have unknown HIV transmission risk than MSM risk likely reflects undercounting of heterosexuals in this analysis. Heterosexuals are presumed to comprise a large proportion of those with unidentified risk because of particularly stringent criteria for classifying a person as having heterosexual HIV transmission risk in the surveillance registry.15

This analysis also has implications for HIV prevention and treatment interventions and policy recommendations. First, as demonstrated by several recent studies,13,14 ART is effective in preventing most HIV transmission within serodiscordant partnerships. The one quarter of index patients in serodiscordant partnerships in this sample with a CD4 count within 3 months of HIV diagnosis between 350 and 500 cells/µL would not have been considered eligible for ART under treatment guidelines active during the study period. However, current treatment guidelines recommend treatment for all HIV-infected persons regardless of CD4 cell count and specifically for those at risk for transmitting HIV to partners.22 Identification of persons in serodiscordant partnerships, as one component of a comprehensive test-and-treat strategy, is thus critical from a prevention standpoint. Index patients and their serodiscordant, steady partners would be optimal target groups for local interventions involving preexposure prophylaxis,23–25 vaginal microbicides,26 and prevention with positives,27 including assisting HIV-infected patients to disclose their serostatus to negative partner(s).

Furthermore, efforts should continue to link newly diagnosed individuals with HIV medical care, to reengage HIV-infected individuals who have fallen out of care, and, in general, to maximize patients’ engagement in all stages of the continuum of HIV care, which collectively will benefit the individual, the couple, and public health in general. Follow-up efforts could prioritize HIV-infected individuals with uncontrolled or sustained high HIV viremia28 after diagnosis, given their greater potential to transmit HIV to uninfected partners.29 HIV-related laboratory data from surveillance programs could be used to identify such individuals. This would require more open use of surveillance registries than has historically been practiced. However, HIV could borrow from models such as tuberculosis and syphilis programs that use registry data for active case management. Sustained funding for HIV care coordination programs is also critical for ensuring continuous engagement in care among at-risk persons.

Health department data could also be used as the basis for offering ongoing PS to HIV-infected persons with evidence of continuing exposure of others, for example, persons who are repeatedly named during PS interviews of newly diagnosed persons. Successful identification of such persons would hinge on collaboration between health departments and community clinicians. Clinicians should be encouraged to discuss sexual practices and risk with their patients at each visit and to request assistance with PS from the health department as needed.

Under current New York State regulations, HEFSP is only able to keep information on partners confirmed to be HIV negative at the time of notification for 3 years. Currently, HEFSP matches HIV-negative partners to the Registry 3 years after the report to assess change in HIV status.30 However, the ability to maintain partner information for a longer period would enable us not only to monitor serodiscordant relationships in the longer-term but also to identify patients who are repeatedly named by newly HIV-diagnosed persons and therefore pose real risks to ongoing transmission.

Our study has several potentially important limitations. The analysis was limited to index patients who reported and named partners whose HIV serostatus could be confirmed and to the partner(s) they chose to report; this resulted in a substantial loss of data, which may have decreased model effect sizes and introduced selection bias. The gap between claimed and named partners demonstrates the potential scope of the problem. Furthermore, findings related to the larger size and greater diversity of serodiscordant partnerships may reflect selection bias in that patients classified in serodiscordant partnerships could have named both negative and positive partners, whereas those in concordant-positive partnerships could only have named positive partners. In addition, FSU staff are located at large diagnosing facilities, mostly with colocated HIV medical care, largely in boroughs and neighborhoods with high HIV morbidity and mortality; furthermore, FSU participating facilities are located in predominately black and Hispanic neighborhoods, and so relatively few whites were included. Field Services Unit clients are therefore not representative of all new HIV diagnoses citywide. Risk data are self-reported, and so information bias is possible. Despite reaching statistical significance, effect estimates from the multivariable model were small. Finally, there were limited data available on potentially important variables, for example, length of partnership, frequency of sex, partner concurrency and partnership type, and more specific geospatial information about partnerships.

The joint efforts of HIV surveillance and PS programs present important opportunities to identify at-risk individuals and partnerships through routine PS and HIV testing activities. In particular, identification and characterization of persons in serodiscordant partnerships can inform the design of innovative public health interventions and suggest policy changes that will interrupt ongoing HIV transmission.

Back to Top | Article Outline


1. Doherty IA, Padian NS, Marlow C, et al. Determinants and consequences of sexual networks as they affect the spread of sexually transmitted infections. J Infect Dis 2005; 191: S42–S54.
2. Rothenberg R, Potterat JJ, Woodhouse DE. Personal risk taking and the spread of disease: Beyond core groups. J Infect Dis 1996; 174: S144–S149.
3. Thomas JC, Tucker MJ. The development and use of the concept of a sexually transmitted disease core. J Infect Dis 1996; 174: S134–S143.
4. Boily MC, Baggaley RF, Wang L, et al. Heterosexual risk of HIV-1 infection per sexual act: Systematic review and meta-analysis of observational studies. Lancet Infect Dis 2009; 9: 118–129.
5. Varghese B, Maher JE, Peterman TA, et al. Reducing the risk of sexual HIV transmission: Quantifying the per-act risk for HIV on the basis of choice of partner, sex act, and condom use. Sex Transm Dis 2002; 29: 38–43.
6. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med 2000; 342: 921–929.
7. Blanchard JF. Populations, pathogens, and epidemic phases: Closing the gap between theory and practice in the prevention of sexually transmitted diseases. Sex Transm Infect 2002; 78: i83–i88.
8. Rothenberg R. Maintenance of endemicity in urban environments: A hypothesis linking risk, network structure and geography. Sex Transm Infect 2007; 83: 10–15.
9. Hixson BA, Omer SB, del Rio C, et al. Spatial clustering of HIV prevalence in Atlanta, Georgia and population characteristics associated with case concentrations. J Urban Health 2011; 88: 129–141.
10. Jolly AM, Muth SQ, Wylie JL, et al. Sexual networks and sexually transmitted infections: A tale of two cities. J Urban Health 2001; 78: 433–444.
11. Friedman SR, Bolyard M, Sandoval M, et al. Relative prevalence of different sexually transmitted infections in HIV-discordant sexual partnerships: Data from a risk network study in a high-risk New York neighbourhood. Sex Transm Infect 2008; 84: 17–18.
12. Bunnell R, Mermin J, De Cock KM. HIV prevention for a threatened continent: Implementing positive prevention in Africa. JAMA 2006; 296: 855.
13. Baeten JM, Donnell D, Ndase P, et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women. N Engl J Med 2012; 367: 399–410.
14. Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med 2011; 365: 493–505.
15. New York City Department of Health and Mental Hygiene, HIV Epidemiology and Field Services Program. Semiannual Surveillance Report. Vol 6(2): Department of Health and Mental Hygiene, 2011.
16. Udeagu CC, Shah D, Shepard CW, et al. Impact of a New York City Health Department initiative to expand HIV partner services outside STD clinics. Public Health Rep 2012; 127: 107–114.
17. Centers for Disease Control and Prevention. Revised surveillance case definitions for HIV infection among adults, adolescents, and children aged <18 months and for HIV infection and AIDS among children aged 18 months to <13 years—United States. MMWR 2008; 57: 4–5.
18. State of New York Laws. HIV Testing and Counseling. Amendment to New York State Public Health Law Article 21, Amendment of Part 63 of Title 10, Codes, Rules and Regulations of the State of New York (HIV/AIDS Testing, Reporting and Confidentiality of HIV-Related Information). Chapter 308. Albany, NY: State of New York; 2010.
19. Girardi E, Sabin CA, Monforte AD. Late diagnosis of HIV infection: Epidemiological features, consequences and strategies to encourage earlier testing. J Acquir Immune Defic Syndr 2007; 46 (suppl 1): S3–S8.
20. Kalichman SC, Simbayi LC, Kaufman M, et al. Alcohol use and sexual risks for HIV/AIDS in sub-Saharan Africa: Systematic review of empirical findings. Prev Sci 2007; 8: 141–151.
21. Rothenberg R, Muth SQ, Malone S, et al. Social and geographic distance in HIV risk. Sex Transm Dis 2005; 32: 506–512.
22. Department of Health and Human Services. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. Washington, DC: Department of Health and Human Services; 2011: 1–161.
23. Grant RM, Lama JR, Anderson PL, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med 2010; 363: 2587–2599.
24. Thigpen MC, Kebaabetswe PM, Paxton LA, et al. Antiretroviral preexposure prophylaxis for heterosexual HIV transmission in Botswana. N Engl J Med 2012; 367: 423–434.
25. Celum C, Baeten JM. Tenofovir-based pre-exposure prophylaxis for HIV prevention: Evolving evidence. Curr Opin Infect Dis 2012; 25: 51–57.
26. Abdool Karim Q, Abdool Karim SS, Frohlich JA, et al. Effectiveness and safety of tenofovir gel, an antiretroviral microbicide, for the prevention of HIV infection in women. Science 2010; 329: 1168–1174.
27. Fisher JD, Smith LR, Lenz EM. Secondary prevention of HIV in the United States: Past, current, and future perspectives. J Acquir Immune Defic Syndr 2010; 55 (suppl 2): S106–S115.
28. Terzian AS, Bodach SD, Wiewel EW, et al. Novel use of surveillance data to detect HIV-infected persons with sustained high viral load and durable virologic suppression in New York City. PLoS One 2012; 7: e29679.
29. Donnell D, Baeten JM, Kiarie J, et al. Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: A prospective cohort analysis. Lancet 2010; 375: 2092–2098.
30. Shah D, Braunstein S, Terzian A, et al. Shepard C. Matching HIV Registry to Partner Services Data: Identifying New Infections among Partners HIV-Negative at Notification. 2011 National HIV Prevention Conference; Atlanta, GA; 2011.
© Copyright 2013 American Sexually Transmitted Diseases Association