The incidence of human immunodeficiency virus (HIV) infection has significantly increased among black men who have sex with men (MSM) in the United States, and young black MSM have been disproportionately affected.1 HIV-infected black MSM are also less likely to engage in HIV care2 and achieve viral suppression3 than MSM of other races/ethnicities. Engaging in care and achieving viral suppression is a multistep process that starts with diagnosis. Diagnosing persons unaware of their HIV status traditionally has been a critical component of HIV partner services, but partner services also provide an important opportunity to reengage HIV-infected partners in medical care. One approach for partner services involves contacting partners of persons with newly diagnosed HIV infection and using sexual and social network and molecular phylogenetic data to improve the continuum of HIV care among black MSM. To evaluate the effectiveness of that approach, results from a prospective partner services study conducted in North Carolina were examined, and one of the partner networks identified through this study was evaluated in depth. Overall, partner services were provided to 30 black, HIV-infected MSM who named 95 sex partners and social contacts, of whom 39 (41%) previously had been diagnosed with HIV infection. The partner network evaluation demonstrated that HIV-infected and HIV-negative partners were frequently in the same network, and that the majority of HIV-infected partners were already aware of their diagnosis but had not achieved viral suppression. Using partner services to ensure that HIV-infected partners are linked to care and treatment might reduce HIV transmission and might improve outcomes along the continuum of care.
Partner services include a broad array of medical (e.g., linkage to HIV medical care and treatment), prevention (e.g., education and counseling to reduce further HIV transmission), and psychosocial services provided to persons diagnosed with HIV infection and their partners. One critical function of partner services is partner notification, a process routinely used in the prevention and control of sexually transmitted diseases (STDs), including HIV. Persons infected with HIV are interviewed to elicit information about their partners (both sexual and needle-sharing) and social contacts who can then be confidentially notified of their possible exposure to or potential risk for HIV infection.4 Partner notification, as conducted by a health department, is a network-based approach to HIV prevention and treatment. Having information about networks with active HIV transmission provides an opportunity to interrupt chains of transmission.5 Contacting partners within a potential HIV transmission network allows public health practitioners to diagnose HIV-infected persons who are unaware of their status, help HIV-infected partners engage or reengage in medical care, and refer at-risk but HIV-negative partners for HIV prevention services.
Screening Targeted Populations to Interrupt Ongoing Chains of HIV Transmission with Enhanced Partner Notification (STOP) is a prospective study evaluating acute HIV infection diagnosis linked to partner services at 12 HIV testing sites in North Carolina; New York, New York; and San Francisco, California.6 Participants were screened with a rapid HIV enzyme immunoassay (IA). Reactive results were confirmed with a Food and Drug Administration (FDA)-approved HIV-1/HIV-2 antibody differentiation assay (Multispot HIV-1/HIV-2 Rapid Test [Multispot], Bio-Rad Laboratories). Specimens that were negative by the rapid IA were screened for acute HIV infection with a fourth-generation combination antigen/antibody IA (Architect HIV Ag/Ab Combo assay, Abbott Diagnostics) and an HIV-1 RNA test (Aptima HIV-1 RNA qualitative assay, Gen-Probe; 80 rapid negative specimens were pooled for this testing). Repeatedly reactive Architect or Aptima results were confirmed with a second HIV-1 RNA test (m2000 RealTime HIV-1 quantitative assay, Abbott Diagnostics). Based on this testing, HIV-infected participants were diagnosed with either 1) acute HIV infection (HIV rapid test negative but HIV-1 RNA detectable); 2) new, established HIV infection (HIV rapid test reactive and not previously diagnosed); or 3) previously diagnosed HIV infection (HIV rapid test reactive but previously diagnosed).
For partner services, HIV-infected participants (index patients) were offered notification services. Contact information was elicited for sex partners from the previous 3 months for index patients with acute HIV infection and the previous 12 months for index patients with established or previously diagnosed HIV infection. In addition, contact information was elicited for social contacts considered by the index patient to be at high risk for HIV infection (i.e., those who would benefit from an HIV test). Health department personnel trained as disease intervention specialists contacted sex partners and social contacts and used Internet-based communication (e.g., e-mail and social network messaging) and text messaging when available. Sex partners and social contacts were offered HIV testing. HIV status was defined as 1) previously diagnosed HIV infection, 2) newly diagnosed HIV infection, 3) HIV-negative (HIV testing during partner services was negative), or 4) HIV status unknown (could not be located or refused testing). HIV polymerase (pol) gene sequences of newly diagnosed HIV-infected persons were analyzed with standard phylogenetic techniques7 to provide further insight into HIV transmission networks. Specifically, persons were considered to form a cluster when their HIV pol sequences were genetically very similar (>97% of aligned nucleotides were identical) and there was high statistical support in phylogenetic analyses (bootstrapping >99% and posterior probabilities =1.0) to suggest the sequences were highly related compared with local controls. This analysis was limited to a subset of black MSM tested in the STOP study in North Carolina for whom HIV-1 pol sequences were available.
During September 2011 - December 2012, partner notification services were provided to 30 black MSM (median age = 23 years) who had a reactive HIV test result and an available HIV-1 pol sequence in the STOP study in North Carolina (45 black MSM who had a reactive HIV test result in the STOP study, but without an HIV-1 pol sequence, were excluded from this analysis). The 30 index patients named 95 persons (74 sex partners and 21 social contacts), of whom 39 (41%) previously had been diagnosed with HIV infection, including 14 who had been diagnosed within the most recent year and 17 who were aged <25 years. An additional 29 (31%) of the 95 named sex partners and social contacts accepted an HIV test, and two sex partners (7% of tested and 3% of all sex partners) were newly diagnosed with HIV infection. Of the remaining sex partners and social contacts, eight refused HIV testing, eight refused any partner services counseling, and 11 could not be located. Most sex partners and social contacts were male (98%) and black (81%), with a median age of 26 years (Table). Sex partners were not more likely to be HIV-infected compared with social contacts (p=0.49), and regular (defined as having sex at least weekly) sex partners were not more likely to be HIV-infected compared with nonregular (having sex less than weekly) sex partners (p=0.16). Considering sex partners only, 18 (60%) of the 30 index patients had at least one HIV-infected sex partner identified, and 12 of 17 index patients who named more than one sex partner had both HIV-infected and HIV-negative sex partners.
HIV pol gene sequences were available for the 30 index patients, but not for their 95 sex partners and social contacts. Although none of the 30 index patients named another index patient as a sex partner, phylogenetic analyses identified four highly supported clusters involving eight (27%) index patients (two men per cluster). The sexual network and molecular phylogenetic data were combined for each of these four clusters. Based on data collected during April 2012 - April 2013, the largest of the resulting networks included 23 black MSM connected by 20 sexual relationships, one social contact, and one molecular phylogenetic link (Figure). Overall, 15 (65%) were HIV-infected, six (26%) tested HIV-negative, one refused HIV testing, and one could not be located. A majority of men in this network were young (aged <25 years), but age-disparate sexual relationships were also represented, and the oldest person in the network was named by four persons (no other member of the network was named by more than two persons). Among nine partners with previously diagnosed HIV infection at the time of the investigation, eight (89%) had been diagnosed within the previous 2 years, but only two were in HIV medical care and only one had achieved viral suppression. Partner services facilitated linkage to care for nine of the HIV-infected partners who were out-of-care, and five additional men achieved viral suppression by August 2013, including the person named by four other persons. Of the six HIV-negative men, five had previously been tested for HIV, but only one had been tested within the last year.