Human immunodeficiency virus (HIV) partner notification (PN) is the process by which health workers discuss with HIV-diagnosed clients the need to notify recent sex and drug-using partners of potential exposure, determine an appropriate manner for notification, and often assist in its execution.1 Partner notification is an effective method of diagnosing HIV-positive individuals, especially when supported by a health provider or trained professional.1–3 One systematic review of high-income countries demonstrated that 8% of all sex partners listed by index clients were successfully diagnosed with HIV through PN.2 A recent CDC MMWR further showed extensive partner services to be the most high-yield method of determining new HIV positives and linking them to subsequent care.3 Partner notification reduces HIV burden through diagnosis and subsequent changes in behavior,4 which are associated with reduced HIV/STI transmission5 and improved health outcomes through earlier linkage to treatment.6 The CDC, therefore, recommends that PN be offered to all recently HIV-diagnosed individuals.1
There are, however, inconsistencies in how clinicians,7,8 HIV testers and counselors,2,7 and health department workers7,9 facilitate PN. If clinicians do offer PN, they may use one of several techniques: they may discuss the importance of notifying partners with clients2,8 and suggest that clients self-refer their partners for testing, conduct some or all portions of PN themselves,2,8 or refer out to PN-providing programs.8,10 Also, many, but not all, HIV counselors refer index clients to PN-providing programs in state and local health departments or other programs tasked with PN.7 Finally, although health departments have traditionally overseen PN, often using disease intervention specialists,1 some jurisdictions contract out local health clinics or community-based organizations to facilitate these services.9 In Chicago, newly HIV diagnosed individuals may be offered PN by providers if the provider has the resources to do so, or if they have contracts with Chicago Department of Public Health. All providers are mandated to report new HIV cases to Chicago Department of Public Health, whose disease intervention specialist follow-up with any PN providers and the index clients, facilitating PN and linkage to care when necessary. Yet, HIV-positive clients may never be offered PN due to referral gaps between health workers, PN providers, and health departments7 as well as resource constraints.11 These lapses can result in delays to the testing, counseling, diagnosis, and treatment of recent seroconverters, enabling the continuation of potentially risky behavior and lack of linkage to clinical care.
Several studies address the cost effectiveness and efficacy of different PN strategies.2,3,10 Others discuss client and provider attitudes toward PN, or factors associated with the detail in information provided during PN interviews.7,12 Relatively few studies, however, address gaps in offering PN to recently diagnosed clients or the characteristics of clients who provide partner information in the first place. Even fewer studies focus on factors that impact PN service offering and participation among young black men who have sex with men (YBMSM), a population with disproportionate HIV prevalence and incidence rates.13 In this study, we identify factors associated with being offered PN and providing partner names in a Chicago-based cohort of YBMSM. Our goals are to elucidate gaps in the provision of PN and examine how PN participation can be improved in this population.
Study Population, Sampling, and Recruitment
Our data come from the first 2 waves of “uConnect” (years 2013–2015), a longitudinal population-based cohort study of YBMSM in Chicago.14 Study participants were recruited through respondent-driven sampling.15 A sample of 618 YBMSM was recruited by identifying seeds from diverse social spaces. Eligibility criteria required that all study participants: (1) have self-identified as African American or Black, (2) have been assigned male sex at birth, (3) be between the ages of 16 and 29 years at the time of recruitment, (4) report oral or anal sex with a male within the 24 months preceding the baseline interview, (5) have provided informed consent at the time of the interview, and (6) spent the majority of their time in the South Side Chicago or adjacent suburbs. Each participant was given instructions for recruiting up to 6 other study participants who met the inclusion criteria, and the recruitment protocol was repeated until the target sample size was achieved.14 Participants completed biobehavioral surveys and were tested for HIV at each wave. All study procedures were reviewed and approved by 2 institutional review boards: NORC at the University of Chicago and the University of Chicago Biological Sciences Division.
The primary outcomes were: (1) being offered PN and (2) providing partner names during PN interviews. The following questions were administered to HIV-positive respondents at baseline and follow-up: (1) “After you tested positive for HIV, were you asked by someone from your health care provider (HCP) to give the names of your sex or drug use partners so they could be notified that they may have been exposed to HIV?” (2) “After you tested positive for HIV, were you asked by someone from the city health department (CHD) to give the names of your sex or drug use partners so they could be notified that they may have been exposed to HIV?” We operationalized having been offered PN if participants who answered “Yes” to both of these questions given evidence that being offered services by both increases uptake.16 Providing partner names was assessed by asking the following question to HIV-positive respondents who were offered PN: “Did you give the names of any of your partners when asked?” The term contact tracing may also be applied to identifying new infectious disease cases through targeted interventions on index clients’ social and sexual networks.17 Though the terms PN and contact tracing are sometimes used interchangeably, we will exclusively use PN in this article, because this term is most used by the health departments and clinics providing these services.
Potential Explanatory Variables
We selected variables from previous analyses among YBMSM that may be associated with care engagement.14,18 These included: ever having been to jail, self-reported syphilis status, marijuana use (stratified into 3 categories—never, intermittent; less than daily use; heavy, daily use), having a mother figure, socializing with other MSM (categorized as those who rarely or never did and those who reported more than rarely socializing with All-MSM groups), support of same-sex marriage laws, use of poppers or volatile nitrates in the past 12 months, having a conversation with an HIV outreach worker in the past 12 months, closeness to the gay and black communities (categorized on a 5-point Likert-type scale), frequency of visiting clubs or bars to socialize with men (stratified by those who reported going less than once a month and those who reported going once a month or more), and the frequency of attending ball events to meet or socialize with other men (stratified by those who reported attending less than once a month and those who reported attending once a month or more frequently). Relationship status was stratified by those who identified as being in a relationship and those who did not. The number of sex partners in the last 6 months was stratified by those with more than 2 partners and those without. Sexual orientation was stratified by those who identified as gay and those who did not. Time since HIV diagnosis was stratified by those who reported diagnosis within 12 months before the interview and those who reported diagnosis 1 year or more previously. Additionally, demographic variables reporting age (stratified by those who were younger than 25 years and those who were 25 years and older) and student status (those who identified as students and those who did not) were included.
Weighted logistic regression was used to evaluate factors significantly associated with each outcome. These variables are listed in Table 1. Models were weighted using the Gile's sequential sampling estimator which enables population-level inference19 and is recommended when the population size is known (or can be estimated).20 The weights were computed using the respondent-driven sampling package21 in R. Bivariate analyses were conducted to shortlist variables for the multivariable model; all variables that were significantly associated with each outcome (at the 0.10 level of significance) were included in the multivariable model. The final parameter list was obtained by applying the “best subsets variable selection” (based on Akaike’s information criterion) to the shortlisted parameters for each outcome.22 The regression analyses were performed in Stata 14.0.
The number of participants who were offered PN and provided partner names is presented in Figure 1. Of the analytic sample obtained from the uConnect cohort (n = 618), 30.3% (n = 187) individuals identified as having tested positive for HIV in either wave 1 or wave 2. Of these individuals, 71.7% (n = 134) reported being offered PN. Of those, 40.3% (n = 54) reported being offered PN by both the CHD and their HCP, whereas 59.7% (n = 80) reported only being offered by one source. Of participants offered PN by only 1 source, 13.7% (n = 11) were offered PN by the CHD, and 86.3% (n = 69) by their HCP. Sociodemographic and behavioral characteristics of the sample are presented in Table 1, stratified by whether participants were offered PN and if they provided partner names when offered.
Our final weighted multivariable model assessing significant correlates of being offered PN included 11 variables (see Table 2). Because of incomplete survey responses, 2 (1.08%) participants were not included in the final model. The following variables showed significant correlations with being offered PN in the multivariable model: having ever been to jail (aOR, 2.49; 95% CI, 1.38–4.49); having used poppers (volatile nitrates) in the past year (aOR, 2.88; 95% CI, 1.20–6.94); and having a conversation with an HIV outreach worker in the past year (aOR, 2.68; 95% CI, 1.25–5.77).
Final models for name provision during PN included 12 variables (see Table 3). Due to incomplete survey responses, 18 (13.4%) participants were not included in the final model. An analysis of this missing data showed no significant difference between those missing and those included in the model (Supplemental Table 1 https://links.lww.com/OLQ/A303). We found that opposing same sex marriage laws (aOR, 0.068; 95% CI, 0.007–0.65) was significantly associated with a decreased odds of providing partner names. Marijuana use, both at the intermittent (aOR, 7.26; 95% CI, 1.75–30.07) and heavy (aOR, 11.47; 95% CI, 2.57–51.22) levels of use were significantly positively associated with the outcome. Being offered PN by both the CHD and HCP (aOR, 8.36; 95% CI, 2.73–25.62) also showed a significant positive association, relative to partner name provision during PN offered by only an HCP. There were too few participants (n = 11) offered PN by only the CHD to draw a meaningful comparison with this group.
This article contributes to limited literature on the factors associated with HIV PN for YBMSM, or participation in PN among those who are offered. We found that HIV-diagnosed individuals who were previously jailed were more likely to be offered PN by both health providers and health department workers, but were not significantly more likely to provide partner names if offered PN. Respondents who reported volatile nitrate/popper use or talking with an HIV outreach worker at least once in the past 12 months were significantly more likely to be offered PN. Intermittent and heavy marijuana users were significantly more likely to provide partner names for PN than never-users. Participants who did not support same-sex marriage were significantly less likely to provide partner names. Finally, respondents who were offered PN by both CHD and health provider affiliates were significantly more likely to provide partner names than those offered by only one source.
The significant association between being in jail and being offered PN might be a consequence of expanding HIV testing and linkage in correctional facilities, developed in response to high rates of HIV among criminal justice involved populations.23,24 This finding is consistent with a national study demonstrating that over half of newly diagnosed black men in jail and prison were referred to PN.23 Yet our results also show that HIV diagnosed individuals who were previously jailed were no more likely to provide partner names when offered PN than those who were not jailed. There are many barriers to improving the HIV care continuum in correctional facilities including HIV stigma, fear of status disclosure and ensuing discrimination, retribution, and danger to personal safety.24 To increase partner information provision during PN, correctional facilities should consider methods of reducing PN stigma and increasing patient confidentiality, such as cross-training their HIV service providers25 and implementing opt-out testing.26 Though vital to improve these services, it is troubling that correctional facilities serve as reasonable YBMSM health promotion venues, suggesting the need to reduce YBMSM correctional involvement via legal reform, increased social services, and other means.
Use of marijuana is also significantly associated with providing partner names during PN interviews. Past studies point to marijuana’s mixed effects on HIV care engagement, with some correlating marijuana use with antiretroviral nonadherence,27 and others finding no such relationship.28 Yet marijuana use may improve partner information provision by reducing mental barriers caused by the stress and anxiety often associated with a new HIV diagnoses. Marijuana use can reduce posttraumatic stress symptom severity29 and provide relief to HIV-positive individuals as well as relieve stress, anxiety, and depression.30 Users who experience marijuana's stress-relieving effects may experience less stress and anxiety and be more willing to provide partner information during PN interviews. Considering the national rise in marijuana use among YBMSM over the past decade,31 additional research must be completed to evaluate its impact on PN participation and partner information provision.
Several factors explain why respondents offered PN by both health provider and CHD workers are more likely to provide partner names than those offered by only one source. Partner name and contact information provision is largely influenced by the type of worker offering these services.7,32 Although respondents in 1 study were most willing to provide partner information when offered PN by their doctors (64%), 84% of respondents indicated they would be somewhat or very likely to provide such information with at least 1 type of health or social service provider.32 This result suggests those offered PN by multiple sources had increased opportunities of finding PN services they deemed acceptable, improving their participation rates. The frequency of PN offering may also contribute to disparities in partner information provision. Given that repeated exposures to viewpoints increase viewpoint adoption,16 respondents who are repeatedly exposed to PN and offered these services might view PN participation more favorably, increasing their rates of partner name provision. In addition, we noted that being offered PN by both sources has a strong, positive correlation with providing partner names. Considering these findings, CHDs could coordinate with health providers to offer PN from multiple sources to recently diagnosed individuals. For example, in Chicago, the health department is funding some clinics to provide PN in populations that overlap with health department PN. This can depart from conventional teaching that multiple contacts by different sources can confuse clients.
There are several limitations worth discussing. First, the PN information collected in uConnect is self-reported, lending to potential biases, and it is unclear in which direction these biases may operate. Second, the Chicago Department of Health often contracts out health clinics to conduct PN on its behalf. This may cause respondents to conflate HCP and CHD workers, or report both workers offered them PN when only one had. Third, because syphilis serostatus was associated with being offered PN, participants who were coinfected with HIV and syphilis may have experienced additional exposure to PN, influencing their likelihood to provide partner names. Fourth, as in most cross-sectional analyses, we were unable to determine causality, because we could not assess whether PN occurred before or after some of the participant characteristics that we assessed. Fifth, some significant results may be influenced by small sample sizes, such as the number of respondents who reported views on same sex marriage. Sixth, respondents may not have provided partner names because their sex partners were anonymous, though this is an issue common to most PN studies and services. Finally, this study assesses neither the effectiveness of the PN offered, nor the extent of partner information provided by PN participants. It may also be noteworthy that recent PN research3 demonstrates reduced efficacy in diagnosing new HIV cases compared with older data,2 potentially due to decreases in HIV unawares or changes in hookup patterns vis-à-vis dating applications. Yet, these same data show extensive PN and associated services to be more effective in diagnosing new HIV cases and facilitating linkage to care than all other testing methods,3 indicating changes to overall HIV testing efficacy rather than PN specifically.
Despite these limitations, our study provides important information on the factors associated with being offered PN and providing partner names. It demonstrates the importance of reducing HIV-associated stigma in correctional facilities, and it deepens our understanding of marijuana's effects on HIV care engagement, suggesting that future research should assess if marijuana use truly does encourage partner name provision during PN interviews. Finally, this article indicates the value of offering PN from more than one source in improving PN engagement rates. Together, these guidelines support increasing PN offering among HIV-diagnosed individuals and improving partner information collection during PN interviews.
1. Centers for Disease Control and Prevention (CDC). Recommendations for partner services programs for HIV infection, syphilis, gonorrhea, and chlamydial infection. MMWR Recomm Rep 2008; 57(RR-9):1–83; quiz CE81–84.
2. Hogben M, McNally T, McPheeters M, et al. The effectiveness of HIV partner counseling and referral services in increasing identification of HIV-positive individuals a systematic review. Am J Prev Med 2007; 33(2 Suppl):S89–S100.
3. Seth P, Wang G, Collins NT, et al.; Centers for Disease Control and Prevention (CDC). Identifying new positives and linkage to HIV medical care—23 testing site types, United States, 2013. MMWR Morb Mortal Wkly Rep 2015; 64:663–667.
4. Fox J, White PJ, Macdonald N, et al. Reductions in HIV transmission risk behaviour following diagnosis of primary HIV infection: A cohort of high-risk men who have sex with men. HIV Med 2009; 10:432–438.
5. Khanna A, Goodreau SM, Wohlfeiler D, et al. Individualized diagnosis interventions can add significant effectiveness in reducing human immunodeficiency virus incidence among men who have sex with men: Insights from Southern California. Ann Epidemiol 2015; 25:1–6.
6. 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.
7. Passin WF, Kim AS, Hutchinson AB, et al. A systematic review of HIV partner counseling and referral services: Client and provider attitudes, preferences, practices, and experiences. Sex Transm Dis 2006; 33:320–328.
8. St Lawrence JS, Montano DE, Kasprzyk D, et al. STD screening, testing, case reporting, and clinical and partner notification practices: A national survey of US physicians. Am J Public Health 2002; 92:1784–1788.
9. Katz DA, Hogben M, Dooley SW Jr, et al. Increasing public health partner services for human immunodeficiency virus: Results of a second national survey. Sex Transm Dis 2010; 37:469–475.
10. Golden MR, Hogben M, Potterat JJ, et al. HIV partner notification in the United States: A national survey of program coverage and outcomes. Sex Transm Dis 2004; 31:709–712.
11. Hoots BE, MacDonald PD, Hightow-Weidman LB, et al. Developing a predictive model to prioritize human immunodeficiency virus partner notification in North Carolina. Sex Transm Dis 2012; 39:65–71.
12. Edelman EJ, Gordon KS, Hogben M, et al. Sexual partner notification of HIV infection among a National United States-based sample of HIV-infected men. AIDS Behav 2014; 18:1898–1903.
13. CDC. HIV Among African American Gay and Bisexual Men. 2017; https://www.cdc.gov/hiv/group/msm/bmsm.html
. Accessed March 17, 2017.
14. Khanna AS, Michaels S, Skaathun B, et al. Preexposure prophylaxis awareness and use in a population-based sample of young black men who have sex with men. JAMA Intern Med 2016; 176:136–138.
15. Heckathorn DD. Respondent-driven sampling: A new approach to the study of hidden populations. Soc Probl 1997; 44:174–199.
16. Schulz-Hardt S, Giersiepena A, Mojzischb A. Preference-consistent information repetitions during discussion: Do they affect subsequent judgments and decisions? Journal of Experimental Social Psychology 2016; 64:41–49.
17. Eames KT, Keeling MJ. Contact tracing and disease control. Proc Biol Sci 2003; 270:2565–2571.
18. Schneider JA, Kozloski M, Michaels S, et al. Criminal justice involvement history is associated with better HIV care continuum metrics among a population-based sample of young black MSM. AIDS 2017; 31:159–165.
19. Gile K. Improved inference for respondent-driven sampling data with application to HIV prevalence estimation. J Am Stat Assoc 2011; 106:135–146.
20. Gile KJ, Johnston LG, Salganik MJ. Diagnostics for respondent-driven sampling. J R Stat Soc Ser A Stat Soc 2015; 178:241–269.
: Respondent-Driven Sampling, Version 0.7-8
[computer program]. 2012.
22. Lindsey C, Sheather S. Best subsets variable selection in nonnormal regression models. Stata J 2015; 15:1046–1059.
23. Seth P, Figueroa A, Wang G, et al. HIV testing, HIV positivity, and linkage and referral services in correctional facilities in the United States, 2009–2013. Sex Transm Dis 2015; 42:643–649.
24. Muessig KE, Rosen DL, Farel CE, et al. “Inside These Fences Is Our Own Little World”: Prison-based HIV testing and HIV-related stigma among incarcerated men and women. AIDS Educ Prev 2016; 28:103–116.
25. Robillard AG, Gallito-Zaparaniuk P, Arriola KJ, et al. Partners and processes in HIV services for inmates and ex-offenders. Facilitating collaboration and service delivery. Eval Rev 2003; 27:535–562.
26. Young SD, Monin B, Owens D. Opt-out testing for stigmatized diseases: A social psychological approach to understanding the potential effect of recommendations for routine HIV testing. Health Psychol 2009; 28:675–681.
27. Voisin DR, Quinn K, Kim DH, et al. A longitudinal analysis of antiretroviral adherence among young black men who have sex with men. J Adolesc Health 2017; 60:411–416.
28. Morgan E, Khanna AS, Skaathun B, et al. Marijuana use among young black men who have sex with men and the HIV care continuum: Findings from the uconnect cohort. Subst Use Misuse 2016; 51:1751–1759.
29. Bonn-Miller MO, Vujanovic AA, Feldner MT, et al. Posttraumatic stress symptom severity predicts marijuana use coping motives among traumatic event-exposed marijuana users. J Trauma Stress 2007; 20:577–586.
30. Prentiss D, Power R, Balmas G, et al. Patterns of marijuana use among patients with HIV/AIDS followed in a public health care setting. J Acquir Immune Defic Syndr 2004; 35:38–45.
31. Services UDoHaH. Results from the 2013 National Survey on Drug Use and Health: Summary of national findings. Substance Abuse and Mental Health Services Administration; 2014.
32. Golden MR, Hopkins SG, Morris M, et al. Support among persons infected with HIV for routine health department contact for HIV partner notification. J Acquir Immune Defic Syndr 2003; 32:196–202.