Factors associated with HIV testing among men who have sex with men in Western Kenya: a cross-sectional study

Abstract

HIV diagnosis is an important step in the HIV cascade of prevention and treatment. However, men who have sex with men in low- and middle-income countries have limited access to HIV care services. We examined factors associated with prior HIV testing among men who have sex with men in western Kenya. We recruited 95 men who have sex with men aged 18 years and older, and who reported at least one sexual contact with a man in the past 6 months; however, this analysis is restricted to 89 participants who completed questions on HIV testing. Logistic regression model was used to determine factors associated with HIV testing in the past one year. Results indicate that 23 (26%) had not been tested in the past 12 months. Bivariate analyses demonstrated that condomless anal sex (odds ratio = 3.29, 95% confidence interval = 1.18–9.17) and comfort with healthcare providers (odds ratio = 1.15, 95 % CI = 1.05–1.26) were associated with higher odds of HIV testing in the past 12 months. Experiencing social stigma was associated with lower odds of HIV testing in the last 12 months (odds ratio = 0.91, 95% confidence interval = 0.84–0.94). In multivariable models, social stigma remained significantly associated with lower odds of HIV testing in the last 12 months odds ratio = 0.90, 95% confidence interval = 0.82–0.99) after inclusion of sexual risk and individual level variables. Development of men who have sex with men–sensitive HIV-testing services, addressing stigma, and training healthcare workers to provide culturally sensitive services may assist in effectively engaging men who have sex with men in the HIV treatment cascade.

Keywords

HIV testing men who have sex with men stigma Kenya

Introduction

Despite the advancement of HIV prevention efforts, men who have sex with men (MSM) continue to bear a disproportionately heavy burden of HIV infection compared to the general adult population.^1,2 According to surveillance data in low- and middle-income countries, MSM are 19.3 times more likely to be living with HIV than the general population,³ yet, only 14% have access to HIV/AIDS prevention and treatment services.² Early linkage to antiretroviral therapy (ART) has been shown to be effective in reducing HIV transmission and mortality among those who are HIV-positive. Early diagnosis, prompt linkage and retention in care, receipt of ART, and viral suppression constitute key points along a comprehensive continuum of HIV care.⁴ However, more than 75% of MSM in low- and middle-income countries lack access to HIV testing services.² To achieve WHO (2011) recommendations of annual HIV testing among MSM and UNAIDS goal of creating an AIDS-free generation,⁵ it is important to expand counselling and testing services to MSM, especially among those residing in sub-Saharan Africa where HIV epidemic strikes hardest.

HIV prevalence among MSM in Kenya is almost four times higher than that in the general adult population.⁶ Latest statistics show that HIV prevalence among MSM was an estimated 18.2% in 2010,⁷ yet most HIV resources are directed to the heterosexual epidemic. In Kenya and as in other African countries except South Africa, sex acts between men are criminalized and carry a maximum penalty of 14 years imprisonment, but generally this is not actively enforced.⁸ However, stigma is a major issue for MSM due to conservative political and religious (Christian and Muslim) segments of Kenyan society and hostile public attitudes. Although the government has recently prioritized HIV prevention activities among MSM, the numbers among those who test for HIV are still very low.^6,9 Findings from the three regions of Kenya – Coast, Kisumu, and Nairobi – indicate that MSM are at higher risk for HIV infection and in need of targeted intervention.^9–12 A 2009 report by the Kenya National AIDS Control Council estimated that MSM accounted for 15.2% of new HIV infections.¹³ For example, estimates from coastal Kenya show HIV incidence of 8.6 per 100 person-years among MSM only and 5.6 per 100 person-years among men who have sex with men and women.¹² HIV prevalence among the MSM population in western Kenya is presently unknown despite the region having highest HIV epidemic compared to other parts of Kenya.⁶ Because MSM are more vulnerable to HIV infection, we may be unaware of an HIV epidemic among this sub-population.

Early initiation of ART therapy, a consequence of HIV testing, reduces rates of sexual transmission of HIV and mortality.¹⁴ It is estimated that new HIV infections could be reduced by 30% per year if all HIV-positive individuals were aware of their serological status.¹⁵ WHO (2011) recommends that all MSM test for HIV at least annually and those MSM who have multiple or anonymous partners or use illegal drugs should test at least every 3 or 6 months. National public policy in Kenya has recently taken a more proactive approach to HIV prevention and control for most-at-risk populations including recommendations to increase HIV testing among MSM.⁶ Our study aids these efforts by providing preliminary data from western Kenyan MSM.

Despite improvements in policies that increasingly promote more frequent and universal HIV testing, MSM in Kenya may experience barriers to HIV testing. Factors such as low self-esteem, negative social consequences, lack of knowledge about HIV/AIDS, voluntary counselling and testing (VCT), ART, perceived risk, anxieties about positive test results, fears about being seen accessing HIV services, discrimination by healthcare providers, income, and sexual identity have been found to be associated with utilization of HIV testing in low- and middle-income countries.^16–21 Qualitative findings from Kisumu and Coastal Kenya revealed reasons for not accessing care that included discrimination, lack of privacy and confidentiality in healthcare facilities, and discomfort with healthcare provider.^22,23 In all of western Kenya, Kisumu is the only county that has provided data on MSM; thus, the purpose of this study is to document correlates of HIV testing among MSM in western Kenya. It is intended that the results of this study will inform interventions to expand HIV testing programs to include those who are at risk of HIV infection and who do not test frequently.

Methods

Study area

The study was based in Eldoret, the fifth largest city in Kenya with approximately 894,179 people. Eldoret is home to one of Kenya’s largest public universities and referral hospital, which hosts the USAID-Academic Model Providing Access to Healthcare (AMPATHplus) program. AMPATHplus is a broad teaching and treatment initiative implemented in 2002 by Moi Teaching and Referral Hospital (MTRH) and Moi University School of Medicine (MUSoM), in collaboration with Indiana University (Indianapolis, USA) and its partners.²⁴ AMPATHplus is the largest HIV clinical care program in East Southern Africa, providing services to over 150,000 patients.

Study design and population

This paper reports on findings from a parent study on health and HIV risk in MSM conducted from February to April 2014 in western Kenya. The parent study utilized both qualitative and quantitative approaches to collect data. Prior to data collection, mapping and description of the general context of MSM in western region was carried out. This mapping exercise provided information to guide targeted recruitment strategies. Quantitative surveys were administered to MSM. Only data from quantitative surveys are reported here.

Study sample

The population included men who are residents of Eldoret and neighbouring communities. The inclusion criteria for this sample were men who were (1) willing to participate; (2) aged ≥18 years; (3) reporting having anal or oral sex with a man within the previous six months; and (4) residing in Eldoret. Chain-link referrals Sampling technique²⁵ was used to enrol participants in the study. The sampling began with a set of initial participants (‘seeds’) recruited with the help of key informants and MSM who are out about their sexual orientation. The initial seeds were different from each other in terms of income, age, and occupation. These seeds initiated the expanding chain of referrals, where respondents from each link in the chain introduced the interviewers to 2–3 MSM known to them if those individuals agreed. In total, 95 participants met eligibility criteria, but 89 completed the survey.

Study procedures and measures

All data collection activities were facilitated by five research assistants who recruited participants, administered surveys, and managed data, under the supervision of study investigators. All research assistants had prior experience administering surveys with similar sensitive questions related to sexual behaviour and HIV/STI. In addition, they received intensive training on the study, confidentiality, mapping, and interviewing. Research assistants were also instructed on consenting and questionnaire administration. They were also expected to hold at least a college diploma, be fluent in English and Kiswahili, and be of good standing in the community and respectful of MSM. The standard questionnaire which lasted approximately 40 minutes was administered in a private and secure space depending on the participants’ preference. Participants received Kshs.150 ($1.5 USD) reimbursement at the end of each interview session. They were also informed about available free care for HIV at Moi Teaching and Referral Hospital (MTRH) and other healthcare facilities in Eldoret. All study procedures were approved by the Moi University College of Health Sciences and Moi Teaching and Referral Hospital Institutional Research and Ethics Committee (IREC).

The survey instrument was adapted from Behavioural Surveillance Surveys for populations at risk of HIV.²⁶ Additional questions were developed and included in the semi-structured interview to address local cultural issues in western Kenya. The survey assessed a variety of topics; for this paper, we focused on questions concerning HIV testing, background demographics, sexual risk behaviours, psychosocial factors, and healthcare seeking practices. Demographic information included age, education, ethnicity, income, marital status, and sexual identity.

To assess condom use with male partners, participants were asked ‘In the past six months when you had anal sex, did you or your male partner use a condom?’ Response options were dichotomized to ‘yes’ vs ‘no’ condomless anal sex. We also assessed participants’ sexual experiences with females; ‘Think about the last time you had sex with a female partner during the past 6 months, was a condom used?’ Response options were dichotomized to ‘yes’ vs ‘no’ condomless vaginal sex. To assess STI/STD infection, participants were asked if they had ever had any genital ulcer/sore or genital discharge, and responses were dichotomized to reflect any or no STI/STD infection history. To capture HIV knowledge of care clinics, participants were asked whether they had ever heard of HIV care clinics and responses were either ‘yes’ or ‘no.’ To assess our study outcome, which was HIV testing in the last 12 months, participants were asked, ‘Have you ever had an HIV test in the last 12 months?’ Response options included ‘yes’ and ‘no.’

Social stigma was assessed with a 12-item measure that assessed occurrence of interpersonal negativity and discrimination related to being gay, including the experience of being mistreated (e.g. ‘Someone was unwilling to share a meal with you?’) and the experience of social avoidance by others because of sexual identity (e.g. ‘Someone was unwilling to be your neighbour?’). Responses were rated on a four-point frequency scale (‘0 = Never’ to ‘3 = Often’). A scale score was computed by summing up the scores across the twelve items. The hypothetical score ranged from 0 to 36 with higher scores representing greater perceived levels of social stigma. The scale demonstrated good internal consistency (Cronbach’s alpha = 0.90).

Participant experiences of self-stigma were assessed using seven items. Sample items included: ‘I hang with the wrong crowd because I am a MSM’ and ‘I should be ashamed of being MSM.’ Four-point Likert scale response options ranged from ‘strongly agree (3) to strongly disagree’ (0). All scores were then summed up to produce a scale of self-stigma, the hypothetical score ranged from 0 to 21 with higher scores indicating increasing levels of self-stigma. The scale demonstrated good internal consistency (Cronbach’s alpha = 0.78).

Participants completed a seven-item measure that assessed their comfort in accessing healthcare services. Sample items included ‘I feel comfortable discussing my sexual practices with the healthcare provider’ and ‘I feel safe discussing my sexual orientation with the healthcare provider.’ Responses ranged from agree strongly (3) to disagree strongly (0). A scale score was computed by summing up the scores across the seven items, the hypothetical score ranged from 0 to 27 with higher scores indicating greater levels of comfort with healthcare provider. The scale demonstrated good internal consistency (Cronbach’s alpha = 0.91).

Data analysis

This analysis is restricted to participants who completed questions on HIV testing (n = 89). Data were analysed using STATA v.10 (Stata Corp 2002; College Station, Texas, USA). The main outcome ‘HIV testing in the last 12 months’ was assessed as a binary variable. In the descriptive analysis, mean and standard deviations were estimated for continuous variables and proportions were calculated for categorical variables. ‘Ever HIV tested’ were defined as participants who reported undergoing HIV testing in the past 12 months and ‘not HIV tested’ were defined as those who reported no prior history of HIV testing in the last 12 months. In bivariate analysis, we used logistic regression models to estimate the odds of HIV testing in the past 12 months for individual variables which included socio-demographic, sexual risk, and psychosocial factors (social stigma, self-stigma. and comfort seeking care). Odds ratio (OR) with 95% confidence intervals (CI) were used to measure associations. A p value of 0.05 or less was considered as indicating a statistically significant association.

Hierarchical logistic set regression was used to estimate the odds of HIV testing during the past year. This approach is an extension of traditional logistic regression, but with sets of conceptually related variables entered in sequential steps. Significance tests, calculated as the difference in the –2 log likelihood of the successive models, assessed the impact of each additional set entered beyond the previously entered sets.²⁷ We used social epidemiologic framework from Poundstone et al.²⁸ to estimate the odds of HIV testing in the past 12 months for psychosocial factors (self-stigma, social stigma, and comfort seeking care) (model 1), the components of model 1 with sexual risk variables (ever having STI or STD symptoms and condomless anal sex (model 2), and finally the components of model 2 with individual variables (age, sexual identity, and education) (model 3). The goal was to examine whether the strength of psychosocial factors (social stigma, self-stigma, and comfort with healthcare provider) are weakened with further additions of other variables in the model. This is based on prior literature and our bivariate associations that have showed that psychosocial factors are significantly associated with HIV testing.

Results

Table 1 shows descriptive characteristics of the sample. Most participants were between the ages of 18–24 (51.69%) with 48.31% aged 25 and older. Participants were primarily single (67.9%), and more than half (59.74%) had monthly income of less than Kshs. 5000 ($50 USD). Fifty percent of the sample described themselves as ‘homosexual or gay,’ while 35.2% described themselves as ‘bisexual’ and 14.77% as ‘heterosexual.’ The majority in this sample had children (79.67%) and had at least a university or college degree (58.43%). In terms of sexual behaviour, 51.19% have had condomless anal sex, and 22.47% reported ever having STI/STD symptoms. Of the sample, 74.16% reported ever having been tested for HIV in the past one year.

Table 1.

Demographic characteristics and sexual behavior of the respondents (N = 89).

Characteristics	Mean (SD) or n (%)
Demographics
Age group in years
18–24	46 (51.69)
>25	43 (48.31)
Sexual orientation identity
Bisexual	31 (35.23)
Homosexual	44 (50.00)
Heterosexual	13 (14.77)
Income monthly
<5000	46 (59.74)
>5000	31 (40.26)
Relationship status
Single	55 (67.9)
Married	28 (33.73)
Ever been married to a woman
Yes	72 (80.9)
No	17 (19.1)
Have children
Yes	67 (79.76)
No	17 (20.24)
No. of children M (SD)	1.9 (0.28)
Education
=<Secondary	37 (41.57)
University or college	52 (58.43)
Sexual behaviors
STI/STD symptom
Yes	20 (22.47)
No	69 (77.53)
Condomless anal sex
Yes	43 (51.19)
No	41 (48.81)
Condomless vaginal sex
Yes	16 (31.37)
No	35 (68.63)
Knowledge of HIV care clinics
Yes	75 (87.21)
No	11 (12.79)
Ever having sex with a woman
Yes	70 (78.65)
No	No (21.35)
Structural barriers
Social stigma M (SD)	8.22 (1.67)
Self-stigma M (SD)	6.14 (0.36)
Comfort seeking care M (SD)	9.11 (0.63)
Tested for HIV in the past one year
Yes	66 (74.16)
No	23 (25.84)

STD: sexually transmitted disease; STI: sexually transmitted infections.

Results from bivariate analyses are presented in Table 2. Of the 13 variables, five variables were associated with HIV testing in the past 12 months. Experiences of social stigma (OR = 0.91, 95%CI = 0.84–0.94) was associated with lower odds of HIV testing in the past 12 months. However, comfort with healthcare providers (OR = 1.15, 95% CI = 1.05–1.26) and condomless anal sex (OR = 3.29, 95% CI = 1.18-9.17) were significantly associated with greater odds of HIV testing in the past 12 months. Additionally, ever having STI/STD symptoms (OR = 3.93, 95% CI 0.84–18.52) and higher educational level (OR = 2.28, 95% CI = 0.87–5.97) were trending toward statistical significance.

Table 2.

Relationships between demographic demographics, sexual risk indicators, psychosocial factors, and HIV testing (N = 89).

Variable	Ever HIV tested n = 66	Not HIV tested n = 23	OR (95% CI)	p
Age groups in years
18–24 (ref)	46 (51.69	34 (51.52)	1.00
>25	43 (48.31)	12 (52.17)	1.02 (0.39–2.65)	0.95
Sexual orientation identity
Bisexual/heterosexual (ref)	35 (53.03)	9 (40.91)	1.00
Homosexual	31 (46.97)	13 (59.09)	0.61 (0.23–1.63	0.32
Income monthly (Kshs.)
<5000 (ref)	33 (55.93)	13 (72.22)	1.00
>5000	26 (44.07)	5 (27.78)	2.04 (0.64–6.48)	0.22
Relationship status
Single (ref)	41 (66.13	14 (73.68)	1.00
Married	21 (33.87)	5 (26.32)	1.43 (0.46–4.52	0.53
Education level
=<Secondary (ref)	24 (36.36)	13 (56.52)	1.00
University or college	42 (63.64)	10 (43.4)	2.28 (0.87–5.97)	0.09
STI/STD symptoms
Yes	18 (27.27)	2 (8.70)	3.93 (0.84–18.52)	0.06
No (ref)	48 (72.73)	21 (91.30)
Condomless anal sex
Yes	36 (59.02)	7 (30.43)	3.29 (1.18–9.17)	0.01
No (ref)	25 (40.98)	16 (69.57)
Condomless vaginal sex
Yes	13 (36.11)	3 (20.00)	2.26 (0.54–9.51)	0.25
No (ref)	23 (63.89)	12 (80.00)	1.00
Knowledge of HIV care clinics
Yes	57 (89.06)	18 (81.82)	1.81 (0.47–6.89)	0.38
No (ref)	7 (10.94)	4 (18.18)	1.00
Ever having sex with a woman
Yes	53 (80.30)	17 (73.91)	2.26 (0.54–9.51)	0.26
No (ref)	13 (19.70)	6 (26.09)	1.00
Social stigma M (SD)	7 (0.89)	11.74 (1.12)	0.91 (0.84–0.94)	0.005
Self-stigma M (SD)	5.8 (0.44)	7.10 (0.65)	0.90 (0.78–1.03)	0.1
Comfort seeking care M (SD)	10.26 (0.71)	5.83 (1.10)	1.15 (1.05–1.26)	0.001

STD: sexually transmitted disease; STI: sexually transmitted infections.

Table 3 shows results from hierarchical logistic regression analysis with the outcome of ever testing for HIV in the past 12 months. In model 1 containing psychosocial factors, social stigma (adjusted OR [AOR] = 0.92, 95% CI = 0.86–0.99) was associated with lower odds of HIV testing in the past year, whereas comfort with healthcare providers (AOR = 1.14 95% CI = 1.03–1.27) was associated with greater odds of HIV testing in the past year. In model 2, which included the psychosocial and sexual risk variables, social stigma (AOR = 0.92, 95% CI = 0.84–0.99) remained significantly associated with lower odds of HIV testing in the past year but comfort with healthcare provider was no longer significant. In our final model 3, including all model 1 and model 2 variables in addition to participant characteristics, only social stigma (AOR = 0.92, 95% CI = 0.84–0.99) continued to be significantly associated with lower odds of HIV testing in the past 12 months. Also, in our final model, higher education level (AOR = 3.72, 95% CI = 1.02–13.48) was significantly associated with higher odds of HIV testing.

Table 3.

Results of hierarchical logistic regression: variable associations with HIV testing in the past 12 months among MSM (N = 89).

Variable	Model 1	Model 2	Model 3
Variable	AOR (95% CI)	AOR (95% CI)	AOR (95% CI)
Psychosocial factors
Social stigma	0.92 (0.86–0.99)^a	0.92 (0.84–0.99)^a	0.90 (0.82–0.99)^a
Self-stigma	0.92 (0.79–1.06)	0.96 (0.82 1.12)	0.96 (0.81–1.14)
Comfort seeking care	1.14 (1.03–1.27)^b	1.09 (0.98–1.23)	1.11 (0.99–1.25)
Sexual risk factors
STI/STD symptoms
Yes		3.57 (0.60–21.13)	4.09 (0.63–26.8)
No (ref)		1.00	1.00
Condomless anal sex
Yes		1.91 (0.60–6.05)	1.56 (0.42–5.73)
No (ref)		1.00	1.00
Individual factors
Age
>31			0.82 (0.22–3.06)
18–30			1.00
Sexual identity
Homosexual			0.86 (0.22–3.33)
Heterosexual & bisexual (ref)			1.00
Education
University or college			3.72 (1.02–13.48)^a
Secondary (ref)			1.00

AOR: adjusted odds ratio; CI: confidence interval; STD: sexually transmitted disease; STI: sexually transmitted infections.

p < 0.05.

p < 0.01.

Discussion

We found that 74% of Kenyan MSM analysed in the survey had ever tested for HIV in the past 12 months. This proportion is similar to that of the recent national figure⁶ and the integrated Bio-Behavioural survey (IBBS) conducted in Nairobi and Mombasa in 2011. Although this number is high, our sample might not be generalizable to all MSM in western Kenya. However, we believe this proportion could be increased to achieve Kenya National AIDS Strategic plan (NACC) goal of expanding HIV testing and counselling services by 80% to all adults,²⁹ given the advancement of HIV testing technologies and revision of testing policies. However, more substantial improvements are necessary to achieve the WHO (2011) recommendation of at least annual HIV testing among MSM.

Our study also found that experiencing social stigma was consistently associated with lower odds of HIV testing in the past 12 months, over and above all other variables assessed in this survey. The relationship between social stigma and HIV testing among MSM corresponds with earlier work done in both high-income and low-middle income countries.^20,30–35 A study in Atlanta in the USA found that HIV-negative black MSM with greater stigma had longer gaps in time since their last HIV appointment. ³⁶ Similarly, recent qualitative findings from South Africa by Sandfort et al.³⁷ showed that anticipated stigma from healthcare providers was linked to HIV testing; men stated that discrimination from healthcare workers hindered many MSM from seeking HIV testing services. Current findings and earlier studies demonstrate that MSM in sub-Saharan Africa experience multiple types of stigma.^3,38 Stigma, which is defined as a negative attribute that represents failure to meet social expectations,³⁹ has been found to be associated with lower access to condoms, basic healthcare services, as well as HIV treatment for HIV-positive MSM in sub-Saharan Africa.^40,41 Previous qualitative research with MSM in coastal Kenya has reported on the internalization of stigma among MSM, contributing to low levels of self-worth and self-efficacy.⁴² Given Kenyan criminalization of same-sex practices and that many Kenyan harbour stigmatizing attitudes and beliefs against those infected with HIV,⁴³ HIV-positive MSM may suffer twofold stigma related to their sexuality and their HIV-status, which may compromise their ability to seek HIV treatment and care services.

As in other prior studies,^34,44,45 comfort with healthcare providers was linked to HIV testing in the past 12 months. For instance, qualitative findings from Coastal and Kisumu Kenya showed that MSM were more likely to seek care that they perceived to be friendly, non-judgmental, confidential, and responsive to individual needs.^22,23 Similarly, Lane and colleagues⁴⁶ reported that verbal abuse targeting MSM from healthcare workers barred MSM in South Africa from seeking care. From this study and earlier work, it is obvious that distrust of the healthcare system, in combination with social stigma, creates a barrier to HIV testing among MSM. Training of healthcare providers to ensure culturally and clinically competent care is important. For instance, a study in Coastal Kenya evaluated a two-day training about homosexuality and HIV prevention targeting Kenyan healthcare workers.⁴⁷ After 3-months post-training, participants were found to have reductions in homophobic attitudes and improvements in knowledge. Similar training may be applied in western Kenya, although infrastructural and cultural issues that differentiate western versus coastal Kenya must be considered.

Our study also yielded an association between HIV testing in the past 12 months and having a higher education attainment, which corresponds with earlier studies conducted in Ethiopia⁴⁸and recently in Zambia.⁴⁹ Although we did not include knowledge about HIV in this analysis, intervention outreach efforts should particularly target Kenyan MSM of lower socio-economic status in order to improve their awareness of and engagement in HIV-testing programs. Also, in this study, we found that condomless anal sex was associated with recent HIV testing, contrary to findings by Scott and team.⁵⁰ Findings in this study might suggest that HIV testing reflects perceived vulnerability among MSM who engage in condomless sex.

Although this study is the first to document HIV testing among MSM in western Kenya, there are also limitations that are important to highlight. Our use of convenience sampling may have created a selection bias for MSM who are connected to gay communities. There may also be selection bias for MSM who are particularly motivated to participate. However, this bias is likely to underestimate social stigma as the men who participated in the study disclosed their sexual practices to the research team. Another limiting factor was the use of non-random sampling, which may limit the external validity of the study findings. Also, these results may be subject to recall bias and social desirability; for instance, participants may have been more likely to overestimate HIV testing. In addition, given our population of interest, our sample size is small, which may affect the internal validity of our study findings; however, it is important to note that this is a hard-to-reach population given criminalization of same sex practice in Kenya. Although our data showed that many MSM had children, we did not examine stigma in the context of social norms, gender roles, and societal expectations, which in turn influence health-seeking practices of MSM. Further research is necessary to better characterize this association. Lastly, we did not collect information around human rights violations or violence affecting MSM, which may likely limit HIV testing in contexts where same sex practices are criminalized; these issues should likely be considered in future studies. Despite these limitations, these data are among the first to describe HIV-related stigma among MSM, a population with increasing rates of HIV infection in Kenya.

Conclusion

This study is one of the first to examine correlates of HIV testing behaviour among western Kenya MSM, highlighting the role of stigma and comfort with providers as factors that might influence HIV testing. Many of our findings validate and corroborate findings from other research examining the relationship between social and structural factors and HIV prevention behaviours among MSM. Our study stressed concerns about mounting environment of fear, including fear of seeking healthcare services faced by MSM in settings that criminalize same-sex practices. The results of this study have several implications to prevent continued heavy burden of HIV among MSM in Kenya. First, strategies to reach MSM to engage in HIV testing and care in highly stigmatized environments are necessary to reduce time to HIV diagnosis and treatment. As recommended by WHO, decriminalization of same sex behaviour is key to address the high burden of HIV among MSM. Second, efforts to train healthcare providers about HIV testing and counselling with MSM are needed to improve the safety and accessibility of services for MSM clients. Third, broader social and policy-level efforts are needed to address general stigma toward MSM. Future research and prevention efforts should explore the role of peers, healthcare providers, and social support in addressing social stigma and access to HIV prevention including testing services among MSM and other sexual minority population.

Footnotes

Acknowledgment

We wish to gratefully acknowledge all the study participants for taking part in the study. We also thank all the data entry personnel and administrative staff for the project facilitation and data management.

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 work was supported by the Moi University and VLIR-UOS University Development Cooperation and National Institute of Health (NIH) grant, Award Number NIH-NICHD R24-HD077976. Manuscript preparation was supported by grants U24 AA022000, P01 AA019072, and T32DA016184 from the National Institutes of Health.

References

Joint United Nations Programme on HIV/AIDS (UNAIDS). The Gap Report. UNAIDS, Geneva, Switzerland, 2014. Available at: http://www.unaids.org/en/resources/documents/2014/20140716_UNAIDS_gap_report (accessed 10 June 2015).

WHO. Global update on HIV treatment 2013: results, impact and opportunities. Geneva, Switzerland, 2013. 2014.

Baral

Sifakis

Cleghorn

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

Gardner

McLees

Steiner

, et al. The spectrum of engagement in HIV care and its relevance to test-and-treat strategies for prevention of HIV infection. Clin Infect Dis 2011; 52: 793–800.

Piot P, Abdool Karim SS, Hecht R, et al. Defeating AIDS–advancing global health. Lancet 2015; 386(9989): 171–218.

NACC. Kenya AIDS Response Progress Report. Nairobi: National AIDS Control Council, 2014.

IOM. Integrated Biological and Behavioural Surveillance Survey among Migrant Female Sex Workers in Nairobi, Kenya, 2010.

Government of Kenya. Laws of Kenya, Penal Code, chapter 63, sections 162–165. National Council for law reporting 2012.

NASCOP. Kenya most at risk populations size estimate consensus. Nairobi: National AIDS and STI Control Programme, Government of Kenya, 2013.

10.

Geibel

Tun

Tapsoba

, et al. HIV vulnerability of men who have sex with men in developing countries: Horizons studies, 2001–2008. Public Health Rep 2010; 125: 316–324.

11.

Okall DO, et al. Men who have sex with men in Kisumu, Kenya: support group membership and knowledge of HIV-risk factors. Cult Health Sex 2013; 15(8): 968–-80.

12.

Sanders

Okuku

Smith

, et al. High HIV-1 incidence, correlates of HIV-1 acquisition, and high viral loads following seroconversion among MSM. AIDS 2013; 27: 437–446.

13.

Gelmon

. Kenya HIV prevention response and modes of transmission analysis, Nairobi, Kenya: National AIDS Control Council, 2009.

14.

Kitahata

Gange

Abraham

, et al. Effect of early versus deferred antiretroviral therapy for HIV on survival. N Engl J Med 2009; 360: 1815–1826.

15.

Branson

. The future of HIV testing. J Acquir Immune Defic Syndr 2010; 55(Suppl 2): S102–S105.

16.

Chakrapani

Newman

Shunmugam

, et al. Structural violence against Kothi-identified men who have sex with men in Chennai, India: a qualitative investigation. AIDS Educ Prevent 2007; 19: 346–364.

17.

Onyango-Ouma

Birungi

Geibel

. Engaging men who have sex with men in operations research in Kenya. Cult Health Sexual 2009; 11: 827–839.

18.

Sullivan

Carballo-Dieguez

Coates

, et al. Successes and challenges of HIV prevention in men who have sex with men. Lancet 2012; 380: 388–399.

19.

Wirtz

Kamba

Jumbe

, et al. A qualitative assessment of health seeking practices among and provision practices for men who have sex with men in Malawi. BMC Int Health Human Rights 2014; 14: 20–20.

20.

Pyun T, Santos GM, Arreola S, et al. Internalized homophobia and reduced HIV testing among men who have sex with men in China. Asia-Pacific journal of public health / Asia-Pacific Academic Consortium for Public Health 2014; 26: 118–125.

21.

Smith

Tapsoba

Peshu

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

22.

van der Elst EM, Gichuru E, Omar A, et al. Experiences of Kenyan healthcare workers providing services to men who have sex with men: qualitative findings from a sensitivity training programme. J Int AIDS Soc 2013; 16(Suppl 3): 18741.

23.

Okall

Ondenge

Nyambura

, et al. Men who have sex with men in Kisumu, Kenya: comfort in accessing health services and willingness to participate in HIV prevention studies. J Homosexual 2014; 61: 1712–1726.

24.

Einterz

Kimaiyo

Mengech

, et al. Responding to the HIV pandemic: the power of an academic medical partnership. Acad Med 2007; 82: 812–818.

25.

Penrod

Preston

Cain

, et al. A discussion of chain referral as a method of sampling hard-to-reach populations. J Transcult Nurs 2003; 14: 100–107.

26.

Amon J, Brown T, Hogle J, et al. Behavioral Surveillance Surveys (BSS): Guidelines for repeated behavioral surveys in populations at risk of HIV; Weighting in MultiStage Sampling. FHI, USAIDS & DFID, 2000.

27.

Hosmer

Lemeshow

. Applied regression analysis, New York: John Willey, 1989.

28.

Poundstone

Strathdee

Celentano

. The social epidemiology of human immunodeficiency virus/acquired immunodeficiency syndrome. Epidemiol Rev 2004; 26: 22–35.

29.

NACC. Kenya National AIDS Strategic Plan. 2009.

30.

Arreola S, Santos GM, Beck J, et al. Sexual stigma, criminalization, investment, and access to HIV services among men who have sex with men worldwide. AIDS Behav 2015; 19(2): 227–234.

31.

Eaton

Driffin

Kegler

, et al. The role of stigma and medical mistrust in the routine health care engagement of black men who have sex with men. Am J Public Health 2015; 105: e75–82.

32.

Dowshen N, Binns HJ and Garofalo R. Experiences of HIV-related stigma among young men who have sex with men. AIDS Patient Care STDS 2009; 23(5): 371–376.

33.

Okal

Luchters

Geibel

, et al. Social context, sexual risk perceptions and stigma: HIV vulnerability among male sex workers in Mombasa, Kenya. Culture Health Sexual 2009; 11: 811–826.

34.

Fay

Baral

Trapence

, 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: 1088–1097.

35.

Schwartz

Nowak

Orazulike

, et al. The immediate effect of the Same-Sex Marriage Prohibition Act on stigma, discrimination, and engagement on HIV prevention and treatment services in men who have sex with men in Nigeria: analysis of prospective data from the TRUST cohort. Lancet HIV 2015; 2(7): e299–e306.

36.

Reilly

Neaigus

Jenness

, et al. Factors associated with recent HIV testing among men who have sex with men in New York City. AIDS Behav 2014; 18: 297–304.

37.

Sandfort

Knox

Collier

, et al. HIV testing practices of South African township MSM in the era of expanded access to ART. AIDS Behavior 2015; 19: 561–574.

38.

van Griensven

de Lind van Wijngaarden

Baral

, et al. The global epidemic of HIV infection among men who have sex with men. Curr Opin HIV AIDS 2009; 4: 300–307.

39.

Goffman E. Stigma: Notes on a spoiled identity. Jenkins, JH & Carpenter, 1963.

40.

Jeffries

WLIV

Marks

Lauby

, et al. Homophobia is associated with sexual behavior that increases risk of acquiring and transmitting HIV infection among Black men who have sex with men. AIDS Behav 2013; 17: 1442–1453.

41.

Musheke

Ntalasha

Gari

, et al. A systematic review of qualitative findings on factors enabling and deterring uptake of HIV testing in Sub-Saharan Africa. BMC Public Health 2013; 13: 220–220.

42.

Midoun M, Shangani S, Mbete B, et al. How intersectional constructions of sexuality, culture, and masculinity shape identities and sexual decision-making among men who have sex with men in coastal Kenya. Culture, health & sexuality. Epub ahead of print 9 November 2015. DOI: 10.1080/13691058.2015.1102326.

43.

Kenya Ministry of Health, ICF International. Kenya Demographic and Health Survey. Calverton, MD, 2009.

44.

Musheke

Bond

Merten

. Deterrents to HIV-patient initiation of antiretroviral therapy in urban Lusaka, Zambia: a qualitative study. AIDS Patient Care STDs 2013; 27: 231–241.

45.

Dijkstra

van der Elst

Micheni

, et al. Emerging themes for sensitivity training modules of African healthcare workers attending to men who have sex with men: a systematic review. Int Health 2015; 7: 151–162.

46.

Lane

Mogale

Struthers

, et al. “They see you as a different thing”: the experiences of men who have sex with men with healthcare workers in South African township communities. Sex Transm Infect 2008; 84: 430–433.

47.

van der Elst

Smith

Gichuru

, et al. Men who have sex with men sensitivity training reduces homoprejudice and increases knowledge among Kenyan healthcare providers in coastal Kenya. J Int AIDS Soc 2013; 16(4Suppl 3): 18748–18748.

48.

Leta

Sandoy

Fylkesnes

. Factors affecting voluntary HIV counselling and testing among men in Ethiopia: a cross-sectional survey. BMC Public Health 2012; 12: 438–438.

49.

Hensen

Lewis

Schaap

, et al. Factors Associated with HIV-Testing and Acceptance of an Offer of Home-Based Testing by Men in Rural Zambia. AIDS Behav 2014; 19: 492–504.

50.

Scott

Pollack

Rebchook

, et al. Peer social support is associated with recent HIV testing among young black men who have sex with men. AIDS Behav 2014; 18: 913–920.