Associations of Consistent Condom Use Among Men Who Have Sex with Men in Abuja,Nigeria

Abstract

The objective of the study was to characterize factors associated with consistent condom use among men who had sex with men (MSM) in Abuja, Nigeria. A convenience sample consisting of 297 MSM was recruited during 2008 using a combination of peer referral and venue-based sampling. Descriptive statistics with chi square and t-test were used for demographic, sexual identity, and practices variables. Univariate and multivariate logistic regressions were used to identify factors associated with consistent condom use with male partners in the past 6 months. Approximately more than half (53%, n=155/290) reported always using condoms with male partner in the past 6 months and 43% (n=95/219) reported always using condoms with female partners in the past 6 months. In all, 11% (n=16/144) reported always engaging in safe sex defined as always using condoms with both male and female partners and always using a water-based condom compatible lubricant with male partners in the past 6 months. Independent associations with consistent condom use with male partners in the past 6 months were knowledge of at least one sexually transmitted infection (STI) that can be transmitted through unprotected anal intercourse (OR 2.47, 95% CI: 1.27–4.83, p<0.01) and having been tested for HIV (OR 2.40, 95% CI: 1.27–4.54, p<0.01). MSM who had been HIV tested at least once were more likely to use condoms consistently during anal intercourse in multivariate analyses. In addition, STI knowledge was also associated with consistent condom use during anal intercourse implying that interventions targeting high-risk practices are effective as HIV prevention for this high-risk group. Future directions include intervention research to determine the appropriate package of services for MSM in Nigeria. In addition, implementation science evaluations of how best to operationalize combination HIV prevention interventions for MSM given the criminalization and stigmatization of same-sex practices are crucial.

Introduction

W ith an estimated 140 million people, Nigeria has the largest population in Sub-Saharan Africa (SSA).¹ The HIV prevalence among reproductive age adults in Nigeria has been estimated to be approximately 4.1%.² However, given the large population base of Nigeria, this prevalence translates to a midpoint estimate of 2.9 million infections among reproductive age adults in 2009, up from 2.4 million in 2001.³ Thus, Nigeria is second only to South Africa in the absolute number of people living with HIV/AIDS and carries approximately 10% of the global burden of HIV.³

Emerging studies continue to highlight the disproportionate HIV burden among specific populations including female sex workers, injecting drug users, and men who have sex with men (MSM).^3,4 Studies of MSM have been completed in several African countries including countries of North, South, West, and East Africa and consistently demonstrate high rates of HIV among samples of young men.⁴ While the primary drivers of the HIV epidemic in Southern and Eastern Africa are heterosexual and vertical transmission, epidemiological assessments of MSM demonstrate that African HIV epidemics are complex in nature, similar to the description of the HIV epidemic in other regions.^3,5 The epidemics of West Africa appear to be more concentrated among these risk groups than what is observed in Southern and Eastern Africa.^5

–9

In Nigeria specifically, population-based estimates have demonstrated high rates of HIV among MSM. In the 2010 IBBSS study, the HIV prevalence was 17.2% (n=266/1545) with significant variability by region ranging from 2.4% in Cross River State, 3.3% in Oyo, 8.3% in Kano, 15.8% in Lagos, 16.2% in Kaduna, to 37.6% in the Federal Capital Territory.² In the IBBSS from 2007, independent associations of HIV among MSM in Lagos included being older than 25 years (aOR 5.6, 95% CI: 2.2–14.1), recent receptive anal intercourse (aOR 4.2, 95% CI: 1.5–12.3), and feeling at risk for HIV (aOR 4.4, 95% CI: 1.6–2.1).⁶ The 2010 IBBSS HIV prevalence rate is more than four times higher among MSM than that of the general population.^3,6 This comparison is also a conservative estimate of the heightened risk of MSM given that women account for nearly 60% of people living with HIV/AIDS in Nigeria and prevalence rates are lower among men.³ In addition, 7.5–14% of new HIV infections in 2010 were estimated to be occurring among MSM in Nigeria; this is equal to or higher than among other most at-risk populations for HIV including sex workers, who contributed 2.6–5.2% of new HIV infections, and intravenous drug users (IDUs), who contributed 1.5–13% of new HIV infections.⁷ These data suggest that 11.6–32.2% of the globe's second largest HIV epidemic is attributable to three minority populations.

There is limited information characterizing effective HIV prevention programs for MSM in Africa. In Nigeria, there are two active prevention programs supported by the United States Agency for International Development (USAID) and the Center for Disease Control (CDC): the Men's Network Nigeria by the Population Council and the Integrated MSM HIV Prevention Program by the Heartland Alliance for Human Needs and Human Rights. Driven by social stigma, MSM represent a hard-to-reach population in Nigeria. Same-sex practices are criminalized, as in many SSA countries, and can be punished with prison, and in some states of Nigeria even with the death penalty.⁸

Knowledge of the risk of HIV transmission with anal sex and having been counseled and tested for HIV have been shown to be associated with lower frequency of unprotected anal intercourse (UAI) among MSM in African settings.^9,10 Moreover, the multisite IpREX study has demonstrated the efficacy of antiviral oral chemoprophylaxis for MSM including a site in Africa.¹¹ Data characterizing condom and water-based condom-compatible lubricant use, bisexual partnerships and concurrency, and limited self-disclosure of high-risk status of MSM to health care professionals are needed to inform the selection of appropriate biomedical interventions for evaluation as part of combination HIV prevention interventions for MSM including biomedical, behavioral, and structural components.¹²

This study was implemented in partnership between the Center for Right to Health, an NGO working for access to health care for vulnerable groups, and Alliance Rights Nigeria, an LGBTI NGO focused on addressing the needs of MSM in Abuja. The objective of the study was to characterize factors associated with consistent condom use and estimate water-based lubricant use among MSM in Abuja, Nigeria.

Materials and Methods

Study population and accrual methods

The study was conducted during July–August 2008 in Abuja, Nigeria with a convenience sample of 297 MSM. Inclusion criteria included men aged 18–65 years old, living in Abuja, who self-reported ever having had anal intercourse with another man. The sample represented a convenience sample recruited through a combination of peer referral and venue-based sampling. For peer referral, trained peer educators recruited participants through their social and sexual networks. Venue-based recruitment was completed using a mobile clinic with direction from an MSM community-based organization that had insight into MSM social venues including nightclubs. It was not feasible to build a sampling frame of these venues as they are not permanent MSM venues; rather, venues are sporadically used as social venues for MSM given the high level of social stigma.

Survey administration

Participants answered a structured survey face to face with a trained peer in a safe and private environment to ensure confidentiality and safety. All participants signed a written informed consent. Participants who wanted HIV/syphilis testing and syndromic treatment for sexually transmitted infections (STIs) were provided this free of charge in a safe, discrete setting through a mobile health clinic. If tested positive, clients were referred to an HIV/STI clinic for HIV care and treatment services.

Key informant interviews helped develop the structured survey instrument that was then piloted to ensure cultural sensitivity. The survey explored sociodemographic characteristics, sexual identity and practices, condom use during the past 6 months, gender and number of partners, knowledge of HIV/STIs, signs and symptoms of STIs, HIV testing, and substance use.

Sample size

Sample size estimates were based on a conservative estimate of regular condom use during anal sex with other men among MSM in Nigeria of approximately 50%.⁶ Behavioral interventions targeting MSM have demonstrated that interventions can increase reported condom use by approximately 16.5% in all risk categories of MSM.¹² Assuming that those who have been exposed to HIV interventions will have a 15% increase in reported consistent condom use, with 80% power and an alpha of 0.05, the minimum necessary sample size required a minimum of 170 participants. While this sample represented a convenience sample, we multiplied our minimum estimated sample size by 1.5, which was 255, and then rounded up aiming for 300 MSM; we eventually accrued 297 MSM who met study inclusion and exclusion criteria.

Statistical analysis

Exploratory, descriptive and inferential data analysis was conducted using Stata 11.1 software.¹³ Descriptive statistics with chi-square and t-test were used for demographic, sexual identity, and practices variables. The predictor variables were dichotomized and a univariate binomial logistic regression and t-test were used to identify factors associated with consistent condom use with male partners in the past 6 months. Thirteen variables with a response rate of less than 60% were excluded from the logistic analysis. A collinearity test was performed and one variable with a variance inflation factor of >10 was excluded. In total, 18 variables were tested for association with consistent condom use. Backward stepwise elimination with a p-value set to 0.1, locking variables for education and age in the model due to being hypothesized confounders in the conceptual framework, was used to determine which variables to include in the multivariate model. Significant variables from the backward stepwise regression, age, and education were refitted in a binomial multivariate logistic regression as the final model, for which the Hosmer-Lemeshow test was performed to ensure goodness of fit. Tests for interaction between the predictors HIV knowledge, HIV test, and STI knowledge were performed with the likelihood ratio test. The stability of the model was confirmed by Akaike information criterion (AIC) values.¹⁴ Variables that are significantly (p<0.05) or moderately significantly (p<0.1) associated with the outcome, consistent condom use with male partners in the past 6 months, in the final multivariate logistic regression models are reported.

Ethical approval

Ethical approval was granted by the Federal Capital Territory Health Research Ethics Committee in Abuja as well as by the institutional review board of Johns Hopkins Bloomberg School of Public Health.

Results

Sociodemographic characteristics of the study population

The characteristics of the 297 MSM who participated in the study are described in Table 1. The median age of participants was 26 years old (range 18–45, n=295). Among the MSM in this study 95% (n=282/297) had a secondary school or higher educational level, 45% (n=134/297) had finished tertiary or vocational school, and 59% (n=175/297) had steady employment. The majority, 78% (n=222/283), were Christian and 17% (n=49/283) were Muslim. The three largest ethnic groups were all represented: Igbo 37% (n=107/289), Yoruba 18% (n=53/289), and Hausa 15% (n=44/289). Of these, 97% (n=287/297) self-reported being circumcised and 15% (n=43/294) reported belonging to an MSM community-based association. The majority 63% (n=189/297) self-reported their sexual identity as bisexual, 35% (n=106/297) reported being gay, 1% (n=4/297) reported being straight, and 0% (n=0/297) reported being transgender. Bisexual concurrency was defined as currently being in a sexual relationship with both at least one male and one female sexual partner, while bisexual partnerships were defined as having had sex with at least one man and one woman within the past 6 months. More than half 66% (n=179/271) of the MSM reported bisexual concurrency at the date of taking the survey, while 67% (n=179/268) reported bisexual partnerships.

Table 1.

Selected Sociodemographic Characteristics of Men Who Have Sex with Men in Abuja

Selected characteristics	Proportion	n
Age (n=295)
Mean/median (range)	26/26.05	(18–45)
Education
Primary school	5.1%	15/297
Secondary school	49.8%	148/297
Tertiary or vocational school	45.1%	134/297
Employment	58.9%	175/297
Religion
Christian	78.2%	222/283
Muslim	17.3%	49/283
Traditional	0.7%	2/283
No religion	3.5%	10/283
Ethnic group
Yoruba	18.3%	53/289
Igbo	37.0%	107/289
Hausa	15.2%	44/289
Other	29.4%	85/289
Self-disclosed sexual orientation
Heterosexual/straight	1.3%	4/297
Homosexual/gay	35.5%	106/297
Bisexual	63.2%	189/297
Transgender	0%	0/297
Disclosed sexual orientation to family and/or community	10.9%	32/293
Disclosed sexual orientation to any health care workers	28.3%	84/297
Self-reported circumcision	96.6%	287/297
Belong to MSM association	14.6%	43/294

High-risk sexual practices

Of the study participants, 28% (n=77/278) reported having had more than five sexual partners, male or female, in the past 6 months, 12% (n=32/273) had more than five male partners in the past 6 months, and 3% (n=4/146) reported more than five female partners in the past 6 months, as described in table 2. More than half, 66% (n=107/163), of MSM reported using a condom at last sex and 53% (n=155/290) reported always using condoms with male partners in the past 6 months. The majority, 64% (n=46/189), of MSM reported always using lubricant with a male partner. Of the type of lubricant used, 36% (n=94/264) reported using a water-based lubricant, 33% (n=87/264) reported using body creams, 17% (n=46/264) reported using petroleum jelly, 9% (n=23/264) reported using lubricant in the condom, 1% (n=3/264) reported using vaginal gel, and the remaining 4% (n=11/264) reported using other lubricants. A total of 86% (n=122/142) reported having a regular female partner, which was defined as having a wife or girlfriend. Almost half of the MSM (43%, n=95/219) reported always using condoms with female partners in the past 6 months. Furthermore, 23% (n=6/26) reported always using condoms with their wife, 64% (n=57/89) reported always using condoms with their girlfriend, and 80% (n=8/10) reported always using condoms with nonregular female partners. Using a Student's t-test, condom use with a wife was significantly lower compared to condom use with a girlfriend (p<0.05). Approximately 11% (n=16/144) reported always engaging in safe sex, defined as always using condoms with both male and female partners and always using a water-based lubricant with male partners in the past 6 months (Table 3). In all, 19% (n=54/289) reported having had a genital ulcer in the past 12 months.

Table 2.

Self-Reported HIV Risk Factors Among Men Who Have Sex with Men in Abuja

Characteristic	Proportion	n
No. of male partners in past 6 months (n=273)
Mean/median (range)	3.0/2	(0–20)
No. of female partners past 6 months (n=146)
Mean/median (range)	1.9/1	(0–20)
>5 sex partners (male and female) over the past 6 months	27.7%	77/278
Regular female partner (wife/girlfriend)	85.9%	122/142
Bisexual concurrency (current female and male sexual partner)	66.1%	179/271
Bisexual partner (have had female and male sex partner within the past 6 months)	66.8%	179/268
Knowledge that HIV/AIDS transmits through anal sex	71.9%	209/294
Knowledge of at least one STI that transmits through anal sex	52.6%	141/268
Self-reported ever tested for HIV	65.2%	180/276
Self-reported HIV⁺	9.5%	15/143
Self-reported HIV⁺ and ARV treatment	10.0%	1/10
Of those not reporting previous HIV test, % who would like an HIV test	54.0%	52/96
Genital ulcer in past 12 months	18.7%	54/289
Received money/gifts to have anal intercourse with another man	61.7%	169/274
Paid money/gifts to have anal intercourse with another man	44.5%	129/290
Had sex under influence of drugs	31.4%	90/287
Alcohol	59.3%	54/91
Injecting drug use in the previous 6 months	1.6%	2/128

ARV, antiretroviral.

Table 3.

Condom Use with Male and Female Sexual Partners Among Men Who Have Sex with Men in Abuja

Practice characteristic	Always	Sometimes/rarely	Never
Condom use with male partner in past 6 months	53.4% (155/290)	39.7% (115/290)	6.9% (20/290)
Condom use with female partner past 6 months	43.4% (95/219)	27.4% (60/219)	29.2% (64/219)
Water-based lubricant use with male partner	38% (66/173)	33% (27/83)	25% (1/4)
Condom use at last sex	Yes: 65.6% (107/163)	Don't remember: 12.9% (21/163)	No: 21.5% (35/163)
Always using condoms with male partner in past 6 months and always using condoms with female partner in past 6 months and always using water-based lubricant with male partner	11.1% (16/144)

Consistent condom use with male partners

In the univariate analysis, described in Table 4, the following variables were found to be significantly associated with consistent condom use with male partners: being employed (OR 1.86, 95% CI: 1.16–3.00), tertiary schooling (OR 2.67, 95% CI: 1.65–4.34), receiving money/gifts for casual sex (OR 0.53, 95% CI: 0.32–0.89), knowledge that HIV can be transmitted through unprotected anal intercourse (OR 3.19, 95% CI: 1.53–6.65), knowledge of at least one STI that can be transmitted through unprotected anal intercourse (OR 3.26, 95% CI: 1.96–5.43), and having been tested for HIV (OR 2.77, 95% CI: 1.64–4.65). After the multivariate adjustment for variables, being older than the median of 26 years (OR 1.55, 95% CI: 0.83–2.90), tertiary or higher education (OR 0.98, 95% CI: 0.51–1.92), knowledge of HIV transmission through unprotected anal intercourse (OR 1.88, 95% CI: 0.80–4.40), knowledge of at least one STI that can be transmitted through unprotected anal intercourse (OR 2.47, 95% CI: 1.27–4.83), and having been tested for HIV (OR 2.40, 95% CI: 1.27–4.54) were found.

Table 4.

Associations of Consistent Condom Use During Anal Intercourse Among Men Who Have Sex with Men in Abuja

Variable	Univariate	Multivariate
Age >26	1.17 (0.73–1.87)	1.55 (0.83–2.90)
Higher education, tertiary or vocational school	2.67 (1.65–4.34)^**	0.98 (0.51–1.92)
Regular female partner (wife/girlfriend)	1.15 (0.32–4.18)
Circumcised	1.47 (0.39–5.61)
> 5 male sexual partners in the past 6 months	0.63 (0.298–1.32)
Self-identified sexuality as bisexual compared to gay	1.12 (0.70–1.82)
Bisexual concurrency, current female and male sexual partner	1.13 (0.68–1.87)
Bisexual partnership, have had female and male sexual partner within the past 6 months	0.45 (0.09–2.38)
Open to community regarding sexual identity	0.90 (0.47–1.72)
Disclosed sexual orientation to medical personnel	1.35 (0.81–2.25)
Received money/gifts to have casual sex with a man	0.53 (0.32–0.89)^*
Paid money/gifts to have casual sex with a man	1.36 (0.85–2.18)
Knowledge that HIV can be contracted through anal sex	3.19 (1.53–6.65)^**	1.88 (0.80–4.40)
Always using water-based lubricant with male partner	0.76 (0.46–1.27)
Have had a genital ulcer/sore in the past 12 months	0.65 (0.36–1.19)
Belongs to MSM association	1.25 (0.65–2.40)
Self-reported ever tested for HIV	2.77 (1.64–4.65)^**	2.40 (1.27–4.54)^**
Knowledge that STIs can be transmitted through anal sex	3.26 (1.96–5.43)^**	2.47 (1.27–4.83)^**

p-value<0.05.

p-value<0.01.

MSM, men who have sex with men; STI, sexually transmitted infection.

When assessing independent significant associations with consistent condom use with male partners within the past 6 months, knowledge of at least one STI that can be transmitted through unprotected anal intercourse (OR 2.47, 95% CI: 1.27–4.83) and having been tested for HIV (OR 2.40, 95% CI: 1.27–4.54) remained significantly associated.

Discussion

This is the first study known to the authors that describes the association between consistent condom usage during anal sex and HIV-related sexual risk practices among MSM in Abuja, Nigeria. It was completed to complement the existing epidemiological data from the Nigeria IBBSS and to inform future prevention efforts.

Just over one-tenth of the sample reported always practicing safe sex as defined by always using condoms with both male and female partners (Table 3) and always using water-based lubricants with male partners in the past 6 months. The rate of MSM reporting safe sexual practices in Nigeria was higher than what has been reported in studies of MSM in other settings.¹⁵ However, reports of unprotected anal intercourse and having had a genital ulcer in the past 12 months were high and over two-thirds of the sample reported not using water-based lubricants during anal sex. Non-condom-compatible lubricants contribute to condom breakage and thus potentiate the risk of HIV acquisition and transmission during anal intercourse. Further qualitative studies are needed to characterize the reasons for the use of non-condom-compatible products during anal intercourse in terms of delineating whether this results from lack of knowledge, availability, or both.

Encouragingly, MSM who had been HIV tested at least once were more likely to use condoms consistently during anal intercourse in multivariate analyses. These data are consistent with previous data highlighting the value of HIV Testing and Counseling (HTC) in East Africa in decreasing high-risk practices among MSM.^9,10 HTC services are an important component of combination HIV preventive interventions for MSM.¹⁵ HTC provides an important opportunity for interaction with the health care system facilitating the delivery of counseling related to safe sex practices among MSM as well as initiation of antiretroviral treatment (ART).¹⁶ Given that information concerning risk reduction during anal intercourse is not widely disseminated with traditional media, HTC for MSM represents a unique opportunity for the delivery of this information and should be included in HTC guidelines for health workers in Nigeria.

In addition, STI knowledge was also associated with consistent condom use during anal intercourse. These data suggest that behavioral HIV interventions addressing HIV/STI knowledge are effective in decreasing high-risk practices for HIV among MSM in Abuja. These data further indicate that relatively few MSM have reported their sexual orientation to a health care worker, likely for fear of adverse outcomes related to this disclosure¹⁷ Furthermore, the data highlight the missed opportunity for disseminating information concerning HIV prevention for MSM through the health care system; this opportunity should be addressed in a safe way ensuring confidentiality for MSM. Structural interventions to increase the clinical and cultural competency of health care workers is likely an important component of combination HIV preventive interventions.¹²

More than half of the MSM in this study reported transactional sex with another man by receiving money/gifts in exchange for anal intercourse. While there was no biological outcome here, transactional sex among MSM in Africa has been demonstrated to be a risk factor for being HIV positive.^18
–20 There was no association observed with decreased consistent condom usage among men reporting transactional sex. However, this could be because this study did not individually assess sexual practices by partner type. Given the high probability of HIV transmission during anal intercourse among men, high levels of sexual partnerships such as observed with transactional sex further potentiates the risk of HIV transmission.²¹ Addressing unprotected anal intercourse during these sexual partnerships is vital.

The majority of the MSM in this study report bisexual activity as well as bisexually concurrent partnerships. This is consistent with what has been observed in other settings among African MSM.²² HIV rates tend to be lower among bisexually active African MSM likely because these men tend to be the insertive partner during anal intercourse, which is associated with less risks of acquisition than being the receptive partner.^18,23 However, these practices also indicate that epidemics of HIV among MSM are not isolated from that of the general population. Prevention programs should be comprehensive and address all populations at risk including MSM. Separately, HIV prevention programs should address all high-risk sexual practices, including decreasing rates of unprotected anal intercourse, during development and implementation.²⁴

There are several limitations with the methods of this study. The study sample represents a convenience sample with mixed accrual methods including peer referral or snowball sampling and venue-based sampling. Thus, generalizability or the ability to extrapolate these findings to all MSM in Nigeria is limited. External validity is further limited given that this study was completed in an urban center. In addition, this was a cross-sectional assessment as both the outcome of consistent condom use and predictors were assessed simultaneously. Consequently, these results do not reflect causality. There are sampling methods such as respondent-driven sampling (RDS) that harness unbiased estimates of disease outcomes in the absence of a population sampling frame. Given the limited budget associated with this work, RDS was not possible, though the ability to accrue MSM quickly suggests that RDS is a feasible option for future studies in Abuja. The study was completed in direct collaboration with an MSM community group and employed members of the MSM community at all stages of the work including the design, instrument development, and implementation to ensure the safety of the participants and to minimize social desirability bias.

There is now a consensus that HIV prevention programs for all populations, including MSM, must be multidimensional, including behavioral, biomedical, and structural components.^25,26 Data continue to emerge demonstrating the efficacy of biomedical interventions such as oral antiviral preexposure prophylaxis and treatment as prevention.^11,27 The data presented here provide insight into the benefits of HTC and the need for scale-up of services and structural interventions to facilitate access to HTC for MSM. Future directions include intervention research to determine the appropriate package of services for MSM in Nigeria. In addition, implementation science evaluations of how best to operationalize combination HIV prevention interventions for MSM given the criminalization and stigmatization of same-sex practices are crucial.

Footnotes

Acknowledgments

We gratefully acknowledge the courage and hard work of the Nigerian LGBTI community for effectively mobilizing to disseminate messages about this study, and thank study participants for partaking in this project. The authors would like to thank the study staff and interviewers who worked on this project at personal risk including disclosure of sexual orientation to their communities.

This research project was funded by a grant from the Foundation for AIDS Research (amfAR) to the Center for Right to Health based in Abuja, Nigeria.

Author Disclosure Statement

No competing financial interests exist.

References

World

. Gross National Income per Capita 2008, Atlas Method and PPP. World Bank: Geneva, 2009.

Nigeria FMoHF. Nigeria Integrated Biological and Behavioural Surveillance Survey. 2010.

UNAIDS: Report on the Global AIDS Epidemic 2010. United Nations: Geneva, 2010.

Smith

, Tapsoba

, Peshu

, Sanders

, Jaffe

. Men who have sex with men and HIV/AIDS in sub-Saharan Africa. Lancet, 2009; 374,9687:416–422.

Baral

, Sifakis

, Cleghorn

, Beyrer

. Elevated risk for HIV infection among men who have sex with men in low- and middle-income countries 2000–2006: A systematic review. PLoS Med., 2007; 4,12:e339.

Merrigan

, Azeez

, Afolabi

, Chabikuli

, Onyekwena

, Eluwa

et al. HIV prevalence and risk behaviours among men having sex with men in Nigeria. Sex Transmit Infect, 2011; 87,1:65–70.

UNAIDS WB. New HIV Infections by mode of transmission in West Africa: A Multi- Country Analysis. 2010.

Ilga. State-Sponsored Homophobia. A World Survey of Laws Prohibiting Same Sex Activity between Consenting Adults. Brussels: Belgium, 2007.

Kajubi

, Kamya

, Raymond

, Chen

, Rutherford

, Mandel

et al. Gay and bisexual men in Kampala, Uganda. AIDS Behav, 2008; 12,3:492–504.

10.

Geibel

, Tapsoba

, Kellerman

. HIV Vulnerability of Men Who Have Sex with Men in Developing Countries: Horizons Studies, 2001–2008Population Council2010.

11.

Grant

, Lama

, Anderson

, McMahan

, Liu

, Vargas

et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med, 2010; 363,27:287–299.

12.

Beyrer

, Wirtz

, Walker

, Johns

, Sifakis

, Baral

. The Global HIV Epidemics among Men Who Have Sex with Men. Epidemiology, Prevention, Access to Care and Human Rights. World Bank Publications: Washington, DC, 2011.

13.

StataCorp. Stata Statistical Software: Release 11.1. College Station, TX: StataCorp LP, 2010.

14.

Wagenmakers

, Farrell

. AIC model selection using Akaike weights. Psychon Bull Rev, 2004; 11,1:192–196.

15.

Baral

, Adams

, Lebona

, Kaibe

, Letsie

, Tshehlo

et al. A cross-sectional assessment of population demographics, HIV risks and human rights contexts among men who have sex with men in Lesotho. J Int AIDS Soc, 2011; 14:36.

16.

Holtgrave

, McGuire

. Impact of counseling in voluntary counseling and testing programs for persons at risk for or living with HIV Infection. Clin Infect Dis, 2007; 45,s4:S240–S243.

17.

Fay

, Baral

, Trapence

, Motimedi

, Umar

, Iipinge

et al. Stigma, health care access, and HIV knowledge among men who have sex with men in Malawi, Namibia, and Botswana. AIDS Behav, 2011; 15,6:1088–1097.

18.

Baral

, Trapence

, Motimedi

, Umar

, Iipinge

, Dausab

et al. HIV prevalence, risks for HIV infection, and human rights among men who have sex with men (MSM) in Malawi, Namibia, and Botswana. PLoS One, 2009; 4,3:e4997.

19.

Geibel

, van der Elst

, King'ola

, Luchters

, Davies

, Getambu

et al. 'Are you on the market?': A capture-recapture enumeration of men who sell sex to men in and around Mombasa, Kenya. AIDS (London, England), 2007; 21,10:1349–1354.

20.

Sanders

, Graham

, Okuku

, van der Elst

, Muhaari

, Davies

et al. HIV-1 infection in high risk men who have sex with men in Mombasa, Kenya. AIDS (London, England), 2007; 21,18:2513–2520.

21.

Baggaley

, White

, Boily

. HIV transmission risk through anal intercourse: Systematic review, meta-analysis and implications for HIV prevention. Int J Epidemiol, 2010; 39,4:1048–1063.

22.

Beyrer

, Trapence

, Motimedi

, Umar

, Iipinge

, Dausab

et al.

Bisexual concurrency, bisexual partnerships, and HIV among Southern African men who have sex with men (MSM)

Sex Transmit Infect, 2010; 86,4:323–327.

23.

Lane

, Raymond

, Dladla

, Rasethe

, Struthers

, McFarland

et al. High HIV prevalence among men who have sex with men in Soweto, South Africa: Results from the Soweto Men's Study. AIDS Behav, 2011; 15,3:626–634.

24.

Veldhuijzen

, Ingabire

, Luchters

, Bosire

, Braunstein

, Chersich

et al. Anal intercourse among female sex workers in East Africa is associated with other high-risk behaviours for HIV. Sex Health, 2011; 8,2:251–254.

25.

Merson

, Padian

, Coates

, Gupta

, Bertozzi

, Piot

et al. Combination HIV prevention. Lancet, 2008; 372,9652:1805–1806.

26.

PEPFAR. Technical Guidance on combination HIV prevention among MSM. USAID 2011: Washington, DC.

27.

Das

, Chu

, Santos

, Scheer

, Vittinghoff

, McFarland

et al. Decreases in community viral load are accompanied by reductions in new HIV infections in San Francisco. PLoS One, 2010; 5,6:e11068.