Relationship between serostatus disclosure and categories of HIV transmission risk in men who have sex with men living with HIV

Abstract

In prior studies, the relationship between serostatus disclosure and reduced HIV transmission risk has been mixed. The demonstration of a clear connection may be restricted by three main methodological limitations. This study evaluates the relationship between (1) more refined measures of serostatus disclosure and (2) eight categories of HIV transmission risk (lowest to highest risk) among men who have sex with men (MSM) living with HIV, while (3) considering a number of control variables. Results demonstrate that disclosure is more likely in sexual encounters involving no intercourse or involving protected and unprotected anal intercourse with HIV-positive partners than unprotected insertive anal intercourse with HIV-negative/unknown status partners. Additionally, substance use prior to sexual encounters is less likely in lower risk categories than the highest risk category. Results of this study are important to the design of future studies, prevention, and intervention programs for MSM and to the methods used to evaluate their effectiveness.

Keywords

HIV risk behaviors HIV/AIDS men who have sex with men serostatus disclosure.

Introduction

Serostatus disclosure is a public health priority in the United States, especially among men who have sex with men (MSM) as they are greatly overrepresented in the HIV/AIDS statistics.¹ Currently, 33 U.S. states and two territories have passed criminal laws requiring people living with HIV to disclose their status to sexual partners and/or needle-sharing partners.² Disclosure can reduce the risk of secondary transmission of the HIV virus among MSM.^3,4 Further, while serostatus disclosure can be both stressful and dangerous,⁵ it can also positively impact physical and mental health in MSM living with HIV.^6,7

However, evidence for a causal role played by disclosure in HIV transmission risk reduction has been mixed, partly because disclosure does not consistently lead to safer sex (e.g. Hightow-Weidman et al.⁸ and St de Lore et al.⁹). The demonstration of a clear connection between disclosure and sexual risk may be restricted by three main methodological limitations specific to the constructs.¹⁰ First, some studies of serostatus disclosure only include data specific to one partner and a single sexual encounter while other studies consider multiple partners and sexual encounters. Disclosure studies often also lack a measure of disclosure that takes into account when disclosure occurred in relation to unprotected sex (i.e. before, during, after sex).

Second, while researchers often operationalize sexual risk in MSM as unprotected anal intercourse (UAI), a variety of other seroadaptive strategies have been reported. These strategies include abstinence from penetrative sex (i.e. mutual masturbation), limiting contact to oral sex, serosorting (i.e. choosing only HIV-positive partners), and seropositioning (i.e. in discordant couples, the HIV-positive partner as receptive partner only).^11–13

Third, some studies on serostatus disclosure lack the consideration of important control variables associated with disclosure and sexual behavior such as characteristics of partners including partnership type (exclusive versus nonexclusive) and partner serostatus.^14,15 Further, characteristics of MSM living with HIV including age, race, ethnicity, sexual orientation, viral suppression, and time since diagnosis have been linked to disclosure and sexual behavior.^10,16 Finally, contextual characteristics including location where sex occurs and substance use have also shown a relationship with disclosure and sexual behavior.^10,17,18

Regarding the latter contextual characteristics, substance use is of particular concern because it is higher among MSM compared to heterosexual men.^19,20 Additionally, there is a well-established link between substance use and sexual behaviors that increase HIV transmission risk among MSM.^18,20–23 For example, substance use affects decision-making capabilities, lowers inhibitions, and is associated with an increased likelihood of engaging in UAI.^21,24–27 As UAI increases, so do the odds of HIV transmission and seroconversion.²⁸

With these limitations of prior HIV disclosure research in mind, the current study examines the relationship between serostatus disclosure and risky sexual behaviors among MSM living with HIV by focusing on three prior methodological limitations. Specifically, this study utilizes more refined measures of (1) serostatus disclosure (considering before, during, after sex and within the context of up to five sexual partners in the past 30 days), (2) HIV transmission risk using a transmission risk index with eight categories that range from lowest to highest risk, and (3) controls for variables that are known to be associated with HIV disclosure and sexual behavior.

Methods

Participants

Data for this study were taken from the baseline observation of a randomized controlled trial of a disclosure intervention that was conducted between 2009 and 2014 in two U.S. metropolitan areas in the Southeast and Midwest. The intervention sought to assist MSM living with HIV in deciding whether and how to disclose their serostatus to casual sexual partners. Participants were recruited through partnerships with local AIDS service organizations, local HIV-related venues and forums, advertisements in local media, and venues where solicitation for casual sex is more likely to occur (e.g. bathhouses).

Eligibility criteria included MSM living with HIV, being age 18 or older, English speaking, and reporting being sexually active in the prior three months with at least one man, and interested in learning more about serostatus disclosure to sexual partners. The study protocol was approved by the Institutional Review Boards of the Ohio State University for the Midwest area and the University of South Florida for the Southeast area. Written consent was obtained prior to the onset of the study from all eligible men who agreed to participate.

Data collection

Prior to randomization, participants completed a baseline questionnaire administered using audio computer-assisted self-interviewing (ACASI). Data were collected on person-level characteristics (i.e. items and scales asked only once for each participant) and encounter-level characteristics (i.e. repeated measures for each participant). Person-level items included measures of demographic characteristics; encounter-level information included sexual activity and disclosure (i.e. occurring during the most recent five sexual encounters in the 30 days prior to baseline).

Measures

HIV transmission risk

To improve the sensitivity of the transmission risk measure, a modified version of the HIV transmission risk scale utilized by Osmond et al.²⁹ was created. A transmission risk index, from lowest to highest transmission risk, was constructed from participant responses to nine items concerning sexual behaviors at each encounter. MSM were asked to report whether the partner was HIV-positive (i.e. Yes, No, I don’t know). Responses were recoded for serosorting behavior where 1 = Yes and 0 = No/I don’t know. For the purpose of creating a hierarchical transmission risk index, no and I don’t know responses were combined since they include partners that are both at greatest risk of HIV transmission. Similar coding decisions can be found in previous studies examining sexual behavior among HIV-positive men (e.g. Lauby et al.³⁰ and Purcell et al.³¹), and this coding decision should be kept in mind when interpreting the findings.

Participants also indicated if the encounter included oral or anal intercourse, what sexual positions were used during anal intercourse (i.e. insertive, receptive), and whether or not condoms were consistently used in each position. Where multiple risk behaviors were reported (e.g. both insertive and receptive anal intercourse), a single risk category was assigned to each encounter based on the highest risk activity to the receptive partner reported during that encounter. The eight categories in the index, arranged from lowest risk to highest risk of transmission, are presented in Table 1.

Table 1.

Descriptive statistics of serostatus disclosure and HIV transmission risk on the encounter level (N = 907).

	Total		Disclosed
	f	%	f	%
Disclosed at or before encounter			653	72.00
HIV transmission risk index ranging from lowest to highest risk category
1. No anal or oral intercourse	90	9.92	68	75.56
2. Oral intercourse only	215	23.70	147	68.37
3. Protected anal intercourse (PAI), 100% condom use, HIV+ partner	63	6.95	50	79.37
4. Unprotected anal intercourse (UAI), not 100% condom use, HIV+ partner	235	25.91	219	93.19
5. Protected receptive anal intercourse (PRAI), HIV-/unknown status insertive partner	85	9.37	57	67.06
6. Protected insertive anal intercourse (PIAI), HIV-/unknown status receptive partner	67	7.39	28	41.79
7. Unprotected receptive anal intercourse (URAI), HIV-/unknown insertive status partner	82	9.04	49	59.76
8. Unprotected insertive anal intercourse (UIAI), HIV-/unknown receptive status partner	70	7.72	35	50.00

Disclosure

Participant responses to three items were used to code serostatus disclosure in each sexual encounter. Participants were asked if they disclosed to their partner prior to this encounter, and if so, whether disclosure was before, during, or after sex. Disclosure was coded as occurring in situations where the participants disclosed before sex or prior to the encounter. Participants were also asked to indicate the method of disclosure (1 = direct disclosure [e.g. told partner point-blank about seropositive status; asked if partner saw social media profile that showed seropositive status] and 0 = indirect disclosure [e.g. verbal hinting; leaving out HIV materials or medication]). To account for potential lack of understanding resulting from indirect disclosure, disclosure was further limited to those encounters where participants told their partner directly.

Control variables

Two types of control variables were included – partner/encounter variables and participant variables. For partner/encounter variables, participants were asked to provide detailed information on their sexual partners (up to five) in the last 30 days. Participants reported the type of relationship they had with each partner (1 = main partners, 0 = other partners). Data regarding the location in which the encounter occurred were also obtained (1 = higher risk locations [i.e. bathhouse, public sex environment], 0 = lower risk locations). Data on substance use at each encounter were obtained based on participants’ answers regarding the use of alcohol or drugs before having sex (1 = Yes, 0 = No).

Regarding participant variables, age was recorded in years. Time since diagnosis (in years) was computed as the difference between the date of the assessment and the date of first diagnosis. Race was dichotomously coded (1 = minority race, 0 = nonminority race) as was participant ethnicity (1 = Hispanic/Latino, 0 = non-Hispanic/Latino). Participants were also asked to report on their current relationship status (1 = partnered/committed, 0 = single). Sexual orientation was coded 1 = gay and 0 = bisexual. Only one participant identified as heterosexual but reported bisexual behaviors and, thus, he was grouped into the bisexual category. Participants also provided self-reports of current viral load, which was dichotomized into viral suppression (1 = no viral suppression; viral load >200, 0 = viral suppression; viral load undetectable or <200).

Data analytic plan

Data were analyzed using SAS 9.4 (SAS Institute, Cary, NC, USA). Descriptive statistics were conducted for all variables included in the analyses to examine the data for normality and outliers. To examine the relationship between serostatus disclosure and the eight categories of risky sexual behaviors, a series of logistic regression models with correlated data (encounters nested within individuals) using generalized linear mixed models were conducted (SAS PROC GLIMMIX). Although conducting a series of logistic regression models is appropriate and common, we acknowledge that it can increase the risk for a Type I error. Insertive UAI with HIV-negative or unknown serostatus receptive partners was selected as the reference category against which to compare each of the other categories because the highest risk transmission category was of greatest interest in this study.

Results

MSM demographic characteristics

The average age of participants (N = 249) was 42.0 years (SD = 10.90). The average time since diagnosis was 10.21 years (SD = 8.22). Most MSM identified their race as White/Caucasian (n = 126, 50.60%) and their ethnicity as non-Hispanic (n = 226, 90.76%). Self-reported sexual orientation was mostly gay (n = 198, 79.52%). Most participants reported being single (n = 172, 69.08%). Additionally, the majority of participants reported viral suppression (n = 166, 66.67%).

Serostatus disclosure and HIV transmission risk index

MSM reported 907 sexual encounters. Table 1 shows descriptive statistics of serostatus disclosure and the HIV transmission risk index on the encounter level. Disclosure at or before the encounter was reported by 72.00% of MSM. Review of the HIV transmission risk index showed that roughly one-quarter of participants engaged in oral intercourse only (23.70%) and in UAI with an HIV-positive partner (25.91%). Between 6.95 and 9.92% of total encounters fell into one of the other risk categories. Descriptive results also confirmed that the percentage of encounters involving disclosure ranged from 41.79% (protected insertive anal intercourse [PIAI] with HIV-negative/unknown status receptive partner) to 93.19% (UAI with HIV-positive partner).

Table 2 displays the results of the correlated logistic regression analyses examining the relationship between disclosure and the eight risky sexual behavior categories. The results display the probability that HIV transmission risk was at a lower risk versus the highest risk category. Thus, odds ratios greater or equal to 1 indicate that disclosure is more likely in the lower than in the highest risk category. Conversely, odds ratios lower than 1 indicate that disclosure is less likely in the lower than in the highest risk category. Results showed that disclosure was 3.67 times more likely in no anal/oral encounters, 2.53 more likely in oral only encounters, 5.32 times more likely in PAI with HIV-positive partner encounters, and 23.40 times more likely in UAI with HIV-positive partner encounters than in unprotected insertive anal intercourse (UIAI) with HIV-negative or unknown receptive partner encounters.

Table 2.

Relationship between serostatus disclosure and HIV transmission risk categories.^a

	Category 1	Category 2	Category 3	Category 4	Category 5	Category 6	Category 7
	No anal/oral	Oral only	PAI, +partner	UAI, +partner	PRAI, −partner	PIAI, −partner	URAI, −partner
	OR	OR	OR	OR	OR	OR	OR
	(95% CL)	(95% CL)	(95% CL)	(95% CL)	(95% CL)	(95% CL)	(95% CL)
Encounter level
Disclosed	3.67*	2. 53*	5.32*	23.40***	1.86	0.71	2.70
Disclosed	(1.31, 10.31)	(1.00, 6.41)	(1.19, 23.75)	(7.43, 73.75)	(0.51, 6.85)	(0.22, 2.32)	(0.79, 9.20)
Main partner	0.80	0.64	0.60	0.93	0.42	0.17*	0.78
Main partner	(0.17, 3.69)	(0.18, 2.30)	(0.08, 4.55)	(0.25, 3.54)	(0.06, 2.74)	(0.03, 0.86)	(0.12, 4.97)
Higher risk location	0.45	0.88	0.98	0.27	0.18	0.54	0.57
Higher risk location	(0.09, 2.34)	(0.26, 2.94)	(0.11, 8.66)	(0.05, 1.42)	(0.02, 1.64)	(0.08, 3.63)	(0.09, 3.58)
Substance use before sex	0.34*	0.31**	0.39	0.19***	0.29*	0.48	0.49
Substance use before sex	(0.12, 0.99)	(0.14, 0.73)	(0.11, 1.39)	(0.08, 0.50)	(0.09, 0.96)	(0.16, 1.43)	(0.15, 1.56)
Person level
Age	1.02	1.01	0.97	0.98	0.96	0.99	1.02
Age	(0.95, 1.10)	(0.95, 1.07)	(0.88, 1.08)	(0.92, 1.05)	(0.88, 1.03)	(0.91, 1.09)	(0.93, 1.11)
Time since treatment	1.03	1.02	1.01	1.06	0.95	0.97	1.03
Time since treatment	(0.95, 1.12)	(0.95, 1.09)	(0.89, 1.15)	(0.98, 1.16)	(0.86, 1.07)	(0.87, 1.08)	(0.93, 1.15)
Minority race	1.04	0.68	5.77*	1.24	2.29	4.10*	2.25
Minority race	(0.30, 3.63)	(0.23, 1.97)	(1.14, 29.33)	(0.43, 3.59)	(0.58, 9.01)	(1.01, 16.65)	(0.44, 11.39)
Hispanic ethnicity	2.14	0.51	2.12	1.42	0.96	0.51	0.46
Hispanic ethnicity	(0.31, 15.01)	(0.08, 3.15)	(0.25, 18.12)	(0.22, 9.08)	(0.10, 9.32)	(0.04, 6.30)	(0.04, 5.35)
Gay	0.59	3.01	2.87	1.29	3.74	0.56	5.70
Gay	(0.15, 2.39)	(0.86, 10.58)	(0.47, 17.63)	(0.37, 4.46)	(0.72, 19.35)	(0.14, 2.29)	(0.88, 36.92)
Partnered	1.84	1.51	0.36	0.97	0.38	0.80	0.38
Partnered	(0.48, 7.09)	(0.49, 4.68)	(0.05, 2.42)	(0.25, 3.80)	(0.06, 2.20)	(0.16, 3.96)	(0.06, 2.45)
No viral suppression	1.85	1.37	1.67	1.33	1.10	2.16	1.44
No viral suppression	(0.51, 6.75)	(0.45, 4.17)	(0.33, 8.33)	(0.42, 4.22)	(0.25, 4.71)	(0.52, 8.93)	(0.31, 6.60)

CL: confidence limits; OR: odds ratio.

Modeled the probability that HIV transmission risk was at a lower risk category versus the highest risk category (category 8; UIAI with HIV-/unknown status receptive partner).

***p < .001; **p < .01; *p < .05.

Regarding the control variables, substance use before sex was the most consistent variable to show a significant relationship with risky sexual behaviors. Substance use was less likely in no anal/oral encounters and oral sex only encounters relative to its use before UIAI with HIV-negative/unknown receptive partner encounters. In encounters where intercourse occurred, substance use was less likely in UAI with HIV-positive partner encounters and protected receptive anal intercourse encounters with HIV-negative insertive partners, relative to its use in UIAI with HIV-negative/unknown receptive partners. Additionally, MSM who identified as a minority race were more likely to engage in PAI with HIV-positive partners and in PIAI with HIV-negative/unknown partners compared to UIAI encounters with HIV-negative/unknown receptive partners. Finally, MSM in main partner relationships were less likely to report PIAI encounters with HIV-negative/unknown partners than UIAI with HIV-negative/unknown receptive partners.

Discussion

The main purpose of this study was to examine the relationship between refined measures of serostatus disclosure and more nuanced measures of HIV transmission risk (ranging from lowest to highest risk) among MSM living with HIV, controlling for a number of variables. Disclosure at or prior to an encounter was more likely in sexual encounters that involved nonpenetrative sex as well as higher risk categories including PAI and UAI with HIV-positive partners than in the highest transmission risk category of UIAI with HIV-negative/unknown serostatus receptive partners.

Among lower risk encounters, it may be that disclosure of serostatus results in abstinence of penetrative sex as a way to reduce HIV transmission risk. O’Connell et al.³ found a 45% reduction in HIV transmission risks among MSM living with HIV when disclosure occurred prior to sexual encounters. It also suggests that disclosure might be an effective risk reduction strategy,^4,10 at least for some MSM. Among penetrative encounters with HIV-positive partners, disclosure also appears to play an important role, particularly for UAI encounters. Considering that UAI with HIV-positive partners was the largest reported category of penetrative sex, results further suggest that MSM are using serosorting as a risk reduction strategy.^3,11 Prior research has shown that the majority of HIV-positive MSM engage in UAI with seroconcordant partners while also expressing a lack of concern about HIV reinfection.³²

The absence of significant relationships between disclosure and penetrative encounters with HIV-negative/unknown status partners warrants further investigation. It may be that MSM are using a variety of seroadaptive strategies such as seropositioning to reduce the risk of HIV transmission,³³ although the effectiveness of these strategies is mixed. It is also possible that nondisclosure is used as a strategy for avoiding perceived negative outcomes associated with disclosure (e.g. abuse, violence, rejection, stigmatization).^34–36 Prevention and intervention programs should focus on providing information to MSM on the facts and myths of sexual behaviors as they relate to HIV transmission risks, pros and cons of disclosure, and on different strategies for practicing safer sex.

Among the control variables, only substance use before sex showed the most consistent significant relationship with HIV transmission risk. Across risk categories, substance use was more likely among MSM engaging in the highest transmission risk category (UIAI with HIV-negative/unknown partners) than in risk categories with no penetration or those involving HIV-positive partner encounters. Similar findings regarding the connection between substance use and risky sexual behaviors have also been confirmed in prior research.^{18,20,21,37–39}

Additionally, this link is not surprising considering that substance use is well known to interfere with decision-making surrounding sexual behaviors and inhibitions and, thus, increase engagement in UAI.^21,24–26 Prevention and intervention programs should also include topics surrounding substance use, risky sexual behaviors, and risk of HIV transmission. This is especially important in light of the fact that substance use is higher among MSM than other men,^19,20 is linked to HIV transmission risks,^19,20 and MSM are overrepresented in the HIV/AIDS statistics.^1,2

Limitations of the current study need to be considered when interpreting the findings. One limitation of this study was the nonrepresentative sample. Future studies should strive to replicate these findings with national random samples of MSM from diverse backgrounds. Further, MSM living with HIV comprise a population that is doubly stigmatized and nondisclosure of HIV serostatus is often considered a criminal offense. Therefore, MSM living with HIV who agree to participate in a study of disclosure may engage in sexual behavior that is different from those who do not participate.

The reliance on self-reported data from the participants’ last five sexual encounters was also a limitation. Though considerable effort was expended to ensure the reliability of these data (e.g. use of ACASI), participants may recall details of these encounters incorrectly or underreport sensitive behaviors. Finally, MSM in our study self-reported a higher percentage of viral suppression (66.7%) compared to those found in the general MSM population (51.0%).² As a result, MSM in our study may have reported risk-taking and disclosure behaviors that differ from other MSM.

Despite these limitations, the current study adds to the body of research on the relationship between disclosure and sexual behavior using more nuanced transmission risk categories and more refined measures of disclosure. More studies are needed to determine the best way to characterize transmission risk as well as the best scaling strategies to evaluate the effect of disclosure and other factors on HIV transmission risk. Future studies should also explore alternative category construction schemes and compare the reliability, validity, and discriminating power associated with data obtained using different categories.

Footnotes

Acknowledgments

We would like to thank the men who participated in this study.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by funding from the National Institute of Mental Health (R01MH082639) to the first author.

References

Centers for Disease Control and Prevention (CDC). The nation’s approach to HIV prevention for gay and bisexual men, https://www.cdc.gov/nchhstp/newsroom/docs/factsheets/approach-to-hiv-prevention-508.pdf (2016, accessed 2 October 2017).

Centers for Disease Control and Prevention (CDC). HIV among gay and bisexual men, https://www.cdc.gov/hiv/group/msm/index.html (2017, accessed 2 October 2017)

O’Connell

Reed

Serovich

JM.

The efficacy of serostatus disclosure for HIV transmission risk reduction. AIDS Behav 2015; 19: 283–290.

Serovich

Reed

Grafsky

et al . An intervention to assist men who have sex with men disclose their serostatus to casual sexual partners: results from a pilot study. AIDS Educ Prev 2009; 21: 207–219.

Bird

JDP

Voisin

DR.

“You’re an open target to be abused”: a qualitative study of stigma and HIV self-disclosure among Black men who have sex with men.

Am J Public Health 2013; 103: 2193–2199.

Gaskins

Foster

Sowell

et al . Reasons for HIV disclosure and non-disclosure: an exploratory study of rural African American men. Issues Mental Health Nurs 2011; 32: 367–373.

Parsons

VanOra

Missildine

et al . Positive and negative consequences of HIV disclosure among seropositive injection drug users. AIDS Educ Prev 2004; 16: 459–475.

Hightow-Weidman

Phillips

G II

Outlaw

et al . Patterns of HIV disclosure and condom use among HIV-infected young racial/ethnic minority men who have sex with men. AIDS Behav 2013; 17: 360–368.

St de Lore

Thiede

Cheadle

et al . HIV disclosure and subsequent sexual behaviors among men who have sex with men who meet online. J Homosex 2012; 59: 592–609.

10.

Simoni

Pantalone

DW.

HIV disclosure and safer sex. In: Kalichman

(ed) Positive prevention: reducing HIV transmission among people living with HIV/AIDS. New York: Kluwer Academic/Plenum, 2005; pp. 65–98.

11.

Eaton

Kalichman

O’Connell

et al . A strategy for selecting sexual partners believed to pose little/no risks for HIV: serosorting and its implications for HIV transmission. AIDS Care 2009; 21: 1279–1288.

12.

McFarland

Chen

Raymond

et al . HIV seroadaptation among individuals, within sexual dyads, and by sexual episodes, men who have sex with men, San Francisco, 2008. AIDS Care 2011; 23: 261–268.

13.

Philip

Donnell

et al . Serosorting is associated with a decreased risk of HIV seroconversion in the EXPLORE study cohort. PLoS One 2010; 5: e12662.

14.

Przybyla

Golin

Widman

et al . Serostatus disclosure to sexual partners among people living with HIV: examining the roles of partner characteristics and stigma. AIDS Care 2013; 25: 566–572.

15.

van den Boom

Stolte

Sandfort

et al . Serosorting and sexual risk behaviour according to different casual partnership types among MSM: the study of one-night stands and sex buddies. AIDS Care 2012; 24: 167–173.

16.

Przybyla

Golin

Widman

et al . Examining the role of serostatus disclosure on unprotected sex among people living with HIV. AIDS Patient Care STDs 2014; 28: 677–684.

17.

Fisher

Ramchand

Bana

et al . Risk behaviors among HIV-positive gay and bisexual men at party-oriented vacations. J Stud Alcohol Drugs 2013; 71: 158–167.

18.

Hatfield

Horvath

Jacoby

et al . Comparison of substance use and risky sexual behavior among a diverse sample of urban, HIV-positive men who have sex with men. J Addict Dis 2009; 28: 208–218.

19.

Balan

Carballo-Dieguez

Dolezal

et al . High prevalence of substance use among men who have sex with men in Buenos Aires, Argentina: implications for HIV risk behavior. AIDS Behav 2013; 17: 1296–1304.

20.

Dirks

Esser

Borgmann

et al . Substance use and sexual risk behavior among HIV-positive men who have sex with men in specialized out-patient clinics. HIV Med 2012; 13: 533–540.

21.

Heath

Lanoyse

Maisto

SA.

The role of alcohol and substance use in risky sexual behavior among older men who have sex with men: a review and critique of the current literature.

AIDS Behav 2012; 16: 578–589.

22.

Kahler

Wray

Pantalone

et al . Daily associations between alcohol use and unprotected anal sex among heavy drinking HIV-positive men who have sex with men. AIDS Behav 2015; 19: 422–430.

23.

Purcell

Parsons

Halkitis

et al . Substance use and sexual transmission risk behavior of HIV-positive men who have sex with men. J Subst Abuse 2001; 13: 185–200.

24.

Darrow

Biersteker

Giess

et al . Risky sexual behaviors associated with recreational drug use among men who have sex with men in an international resort area: challenges and opportunities. J Urban Health 2005; 82: 601–609.

25.

Ostrow

Plankey

Cox

et al . Specific sex drug combinations contribute to the majority of recent HIV seroconversions among MSM in the MACS. J Acquir Immune Defic Syndr 2009; 51: 349–355.

26.

Parsons

Kutnick

Halkitis

et al . Sexual risk behaviors and substance use among alcohol abusing HIV-positive men who have sex with men. J Psychoactive Drugs 2005; 37: 27–36.

27.

Woolf

Maisto

SA.

Alcohol use and risk of HIV infection among men who have sex with men.

AIDS Behav 2009; 13: 757–782.

28.

Patel

Borkowf

Brooks

et al . Estimating per-act HIV transmission risk: a systematic review. AIDS 2014; 28: 1509–1519.

29.

Osmond

Pollack

Paul

et al . Changes in prevalence of HIV infection and sexual risk behavior in men who have sex with men in San Francisco: 1997–2002. Am J Public Health 2007; 97: 1677–1683.

30.

Lauby

Millett

LaPollo

et al . Sexual risk behaviors of HIV-positive, HIV-negative, and serostatus-unknown Black men who have sex with men and women. Arch Sex Behav 2008; 37: 708–719.

31.

Purcell

Mizuno

Metsch

et al . Unprotected sexual behavior among heterosexual HIV-positive injection drug using men: associations by partner type and partner serostatus. J Urban Health 2006; 83: 656–668.

32.

Halkitis

Green

Remien

et al . Seroconcordant sexual partnerings of HIV-seropositive men who have sex with men. AIDS 2005; 19: S77–S86.

33.

Cassels

Katz

DA.

Seroadaptation among men who have sex with men: emerging research themes.

Curr HIV/AIDS Rep 2013; 10: 305–313.

34.

Brown

Serovich

Kimberly

JA.

Depressive symptoms, substance use and partner violence victimization associated with HIV disclosure among men who have sex with men.

AIDS Behav 2016; 20: 184–192.

35.

Deribe

Woldemichael

Wondafrash

et al . Disclosure experience and associated factors among HIV positive men and women clinical service users in Southwest Ethiopia. BMC Public Health 2008; 8: 81.

36.

Stutterheim

Bos

AER

Pryor

et al . Psychological and social correlates of HIV status disclosure: the significance of stigma visibility. AIDS Educ Prev 2011; 23: 382–392.

37.

Drumright

Gorbach

Little

et al . Associations between substance use, erectile dysfunction medication and recent HIV infection among men who have sex with men. AIDS Behav 2009; 13: 328–336.

38.

Spindler

Scheer

Chen

et al . Viagra, methamphetamine, and HIV risk: results from a probability sample of MSM, San Francisco. Sex Transm Dis 2007; 34: 586–591.

39.

Shoptaw

Reback

CJ.

Methamphetamine use and infectious disease-related behaviors in men who have sex with men: implications for interventions.

Addiction 2007; 102: 130–135.