Abstract
Objectives
Few studies have been conducted in Africa to assess prevalence of sexually transmitted infections (STIs) and risk factors among men who have sex with men (MSM). We report findings from the first behavioural survey to include STI testing among MSM in Kampala, Uganda.
Methods
Respondent-driven sampling (RDS) was used to recruit MSM for a biobehavioural survey. Eligible participants were men who reported anal sex with another man in the previous 3 months, were 18 years or older, and resided in Kampala. Information was collected on demographics, sexual behaviour, alcohol and drug use, and STI symptoms. Blood, urine and rectal specimens were tested for syphilis, HIV, rectal and urethral gonorrhoea, and chlamydia. Analyses weighted for RDS were conducted to assess associations with STI diagnosis.
Results
A total of 295 MSM participated in the survey. Almost half (weighted percentage: 47.3%) reported STI symptoms in the last 6 months and 12.9% tested HIV-positive. Prevalence of non-HIV STI was 13.5%; syphilis prevalence was 9.0%. Adjusting for age and education, STI was associated with HIV (adjusted OR (AOR)=3.46, 95% CI 1.03 to 11.64), alcohol use before sex (AOR=4.99, 95% CI 1.86 to 13.38) and having sold sex in the last 3 months (AOR=3.17, 95% CI 1.25 to 8.07), and inversely associated with having anonymous sex partners (AOR=0.20, 95% CI 0.07 to 0.61).
Conclusions
We observed high levels of self-reported STI symptoms and STI prevalence associated with alcohol use and HIV among MSM in Kampala. Public health interventions supporting MSM are needed to address STI risk and facilitate access to diagnosis and treatment services.
Introduction
Sexually transmitted infections (STIs) such as syphilis, gonorrhoea and chlamydia cause significant global morbidity and mortality and have been associated with increased risk of HIV transmission.1 It is estimated that 340 million new cases of curable STIs occur every year among men and women aged 15–49 years throughout the world. Of these, an estimated 4 million new cases of syphilis occur in sub-Saharan Africa, a third of the 12 million new syphilis cases estimated worldwide.2 Similarly, of 92 million new cases of chlamydia and 62 million new cases of gonorrhoea globally, sub-Saharan Africa accounted for 16 million (17.4%) and 17 million (27.4%), respectively. The 2011 Uganda AIDS Indicator Survey reported a syphilis prevalence of 1.8% among men in the general population aged 15–49 years. The same survey found that 20.6% of sexually active men reported having had a STI or STI symptoms in the previous year.3
Men who have sex with men (MSM) have been studied as a key population at higher risk for HIV and STIs; however, few studies have been conducted among MSM in Africa. Homosexuality and same-sex sexual behaviours are illegal and highly stigmatised in Uganda, presenting substantial challenges to public health surveillance and intervention efforts among the MSM population. In 2004, a study was conducted in Kampala that demonstrated the existence of a local MSM population and observed sexual risk behaviours in this group.4 The study found high rates of unprotected receptive anal intercourse and commercial sex, with low perception of risk for HIV infection; however, no biological testing was conducted to assess prevalence of HIVor STIs.
A follow-up study conducted in 2008–2009 was the first behavioural survey to include STI and HIV testing among MSM in Kampala.5 The objective of this study was to establish a periodic HIV surveillance system that would inform and evaluate the impact of interventions for MSM and other key populations. We present here findings on STI prevalence and associated risk behaviours from this first round of biobehavioural surveillance among MSM.
Methods
Respondent-driven sampling (RDS) was used to recruit MSM in Kampala, Uganda between May 2008 and April 2009. RDS is a type of chain referral sampling that adjusts for recruitment patterns and social network size to arrive at population-level estimates.6,7 Eligible participants were men who reported having anal sex with another man in the previous 3 months, were aged 18 years or older, spoke English or Luganda, and resided in greater Kampala.
We started sampling with eight diverse seeds purposively selected by age, HIV status and geographical location (Division) in Kampala; all were socially well networked, defined as having extensive social networks and being able to distribute coupons to a large number of peers. However, rates of recruitment were low throughout the study. Study recruitment was interrupted twice due to arrests of MSM in Kampala during the study period. In addition, six seeds were added in the course of the survey to improve sampling rates. A total of 455 coupons (26.7%) were redeemed out of 1706 coupons issued. Of these, 281 recruits and 14 seeds were eligible and participated in the survey. The longest recruitment wave was 11, and equilibrium for syphilis was reached after wave 4. The average degree size was 11.3.
Survey procedures are described in detail elsewhere.3 In brief, screened, eligible and consented participants responded to a standardised audio-computer assisted interview. The survey collected information on demographics and sexual behaviour in the 3 months preceding the interview. Sexual behaviour with male and female partners was queried by primary, steady, casual, anonymous, one-night stand, sex worker and client partner types. The survey also collected data on alcohol and drug use, violence and STI symptoms.
Blood specimens were collected from participants to test for syphilis; urine and rectal specimens were collected to test for gonorrhoea and chlamydia. Only 4 of the 295 participants (1.4%) refused to provide rectal specimens, and no participants refused to provide urine. Antisyphilis IgG ELISA (Biotec Laboratories, Suffolk, UK) was used to screen plasma for lifetime exposure to Treponema pallidum (TP), and the Rapid Plasma Reagin Syfacard-R Test (Murex Biotech, Dartford, UK) was used to detect current TP infection. The presence of Neisseria gonorrhoeae (NG) and Chlamydia trachomatis (CT) DNA was tested in urine specimens and rectal swabs using Cobas Amplicor or Amplicor PCR (Roche Diagnostics, Branchburg, New Jersey, USA). A parallel testing algorithm using Vironostika HIV Uniform II plus O2 (bioMeriéux, Marcy l'Etoile, France) and Murex HIV Ag/Ab Combination (Abbott Laboratories, Abbott Park, Illinois, USA) was used to test for HIV antibodies. HIV 1/2 STAT-PAK rapid test (Inverness Medical, Princeton, New Jersey, USA) was used to resolve discordant results. Laboratory tests were conducted at the STD Unit of Mulago Hospital, Kampala and the US Centers for Disease Control and Prevention laboratory in Entebbe, Uganda. Participants were asked to return after 2 weeks to receive their results; 61% of participants returned for a second visit. During these visits, participants received counselling for all biomarkers, were provided treatment for STIs and, if HIV-positive, were referred to care and treatment providers.
Weighted univariate and bivariate analyses were conducted in RDS Analysis Tool (RDSAT) V.5.6 (http://www.respondentdrivensampling.org). Individual STI sampling weights were generated in RDSAT and exported to SAS V.9.2 (SAS Institute, Cary, North Carolina, USA) to perform additional bivariate and multivariate analyses. Missing data were dropped from analysis. Bivariate associations with STI diagnosis were assessed using survey-weighted Wald tests. Multivariate analysis was conducted using survey-weighted logistic regression with variables associated with positive STI diagnosis. Variables considered in the modelling process were those associated with STI diagnosis at a level of p≤0.2 in bivariate analysis. Other variables were included based on a conceptual framework incorporating previous knowledge of association with STI risk. The final model was adjusted for age and education and those variables that were significantly associated with STI diagnosis.
Survey participation was anonymous: informed consent was obtained verbally and no personal identifiers were collected.
Results
A total of 295 MSM were included in the analysis. The majority of participants were under 25 years of age: 19.8% were 18–20 years old and 42.9% were 21–25 years old (table 1). Most participants reported drinking alcohol every day or at least once a week (17.9% and 43.8%, respectively), and 47.2% reported alcohol use before sex. Nearly a third of MSM (32.0%) reported having had an anonymous sex partner in the past 6 months, and 29.5% had sold sex in the 3 months prior to the survey.
Table 1. Sociodemographic and behavioural variables and sexually transmitted infections (STIs) among men who have sex with men in Kampala, Uganda.
| n (unweighted) | Sample % (unweighted) | Estimated population proportion % (weighted) | 95% CI | ||
|---|---|---|---|---|---|
| Age (years) | |||||
| 18–20 | 52 | 18.1 | 19.8 | 12.9 to 27.4 | |
| 21–25 | 111 | 38.5 | 42.9 | 32.9 to 49.0 | |
| 26–30 | 70 | 24.3 | 22.9 | 17.0 to 32.0 | |
| >30 | 55 | 19.1 | 14.5 | 9.6 to 21.6 | |
| Education | |||||
| None | 22 | 7.8 | 7.9 | 4.5 to 11.9 | |
| 1–7 years | 72 | 25.4 | 28.3 | 20.3 to 36.0 | |
| 8–11 years | 58 | 20.4 | 23.7 | 17.2 to 31.4 | |
| 12 or more years | 131 | 46.5 | 40.2 | 31.3 to 49.5 | |
| Drink alcohol, past 30 days | |||||
| About every day | 53 | 18.5 | 17.9 | 11.1 to 25.8 | |
| At least once a week | 124 | 43.4 | 43.8 | 36.8 to 51.2 | |
| Less than once a week | 24 | 8.4 | 10.2 | 5.1 to 16.5 | |
| None | 85 | 29.7 | 28.1 | 21.4 to 35.2 | |
| Sexual orientation | |||||
| Gay/homosexual | 162 | 56.3 | 54.2 | 47.5 to 62.6 | |
| Bisexual | 113 | 39.2 | 38.8 | 30.8 to 45.4 | |
| Straight/heterosexual | 13 | 4.5 | 7.0 | 2.6 to 12.9 | |
| Alcohol use before sex | |||||
| Yes | 123 | 43.2 | 47.2 | 38.5 to 56.0 | |
| No | |||||
| Had anonymous sex partner in the past 3 months | |||||
| Yes | 107 | 37.3 | 32.0 | 24.8 to 40.1 | |
| No | 180 | 62.7 | 68.0 | 59.9 to 75.2 | |
| Ever sold sex | |||||
| Yes | 128 | 44.4 | 39.3 | 29.5 to 46.2 | |
| No | 160 | 55.6 | 60.7 | 53.8 to 70.5 | |
| Sold sex in the past 3 months | |||||
| Yes | 94 | 31.8 | 29.5 | 21.9 to 35.8 | |
| No | 179 | 60.5 | 70.5 | 64.2 to 78.1 | |
| Used condom during insertive anal sex in the past 3 months | |||||
| Yes | 173 | 65.8 | 66.5 | 58.2 to 74.3 | |
| No | 90 | 34.2 | 33.5 | 25.7 to 41.8 | |
| Used condom during receptive anal sex in the past 3 months | |||||
| Yes | 140 | 60.1 | 62.4 | 53.7 to 70.1 | |
| No | 93 | 39.9 | 37.6 | 29.9 to 46.3 | |
| STI symptoms in the past 12 months | |||||
| Genital discharge | |||||
| Yes | 86 | 30.6 | 29.5 | 21.9 to 37.3 | |
| No | 195 | 69.4 | 70.5 | 62.7 to 78.1 | |
| Genital ulcers | |||||
| Yes | 93 | 33.6 | 31.6 | 23.6 to 39.3 | |
| No | 184 | 66.4 | 68.4 | 60.7 to 76.4 | |
| Anal ulcers | |||||
| Yes | 79 | 28.7 | 25.0 | 18.7 to 32.6 | |
| No | 196 | 71.3 | 75.0 | 67.4 to 81.3 | |
| Anal discharge | |||||
| Yes | 45 | 16.3 | 12.6 | 8.3 to 18.1 | |
| No | 232 | 83.8 | 87.4 | 81.9 to 91.7 | |
| Anal warts | |||||
| Yes | 57 | 21.4 | 16.6 | 11.1 to 22.2 | |
| No | 210 | 78.7 | 83.4 | 77.8 to 88.9 | |
| Any STI symptom | |||||
| Yes | 143 | 51.8 | 47.3 | 39.3 to 56.0 | |
| No | 133 | 48.2 | 52.7 | 44.0 to 60.7 | |
| Stopped having sex with symptoms | |||||
| Yes | 73 | 54.1 | 61.2 | 49.8 to 73.1 | |
| No | 62 | 45.9 | 38.8 | 26.9 to 50.2 | |
| Lab results | |||||
| Syphilis | |||||
| Positive | 30 | 10.3 | 9.0 | 5.5 to 13.3 | |
| Negative | 260 | 89.7 | 91.0 | 86.7 to 94.5 | |
| Rectal chlamydia | |||||
| Positive | 3 | 1.1 | 1.1 | 0 to 2.7 | |
| Negative | 283 | 99.0 | 98.9 | 97.3 to 100.0 | |
| Urethral chlamydia | |||||
| Positive | 3 | 1.0 | 1.2 | 0 to 2.9 | |
| Negative | 285 | 99.0 | 98.8 | 97.1 to 100.0 | |
| Rectal gonorrhoea | |||||
| Positive | 5 | 1.8 | 1.8 | 0.2 to 4.2 | |
| Negative | 281 | 98.3 | 98.2 | 95.8 to 99.8 | |
| Urethral gonorrhoea | |||||
| Positive | 4 | 1.4 | 1.4 | 0.2 to 3.0 | |
| Negative | 284 | 98.6 | 98.6 | 97.0 to 99.8 | |
| Any STI | |||||
| Positive | 42 | 14.8 | 13.5 | 9.2 to 19.3 | |
| Negative | 242 | 85.2 | 86.5 | 80.7 to 90.8 | |
| HIV serostatus | |||||
| Positive | 39 | 13.5 | 12.9 | 7.2 to 19.5 | |
| Negative | 251 | 86.6 | 87.1 | 80.5 to 92.8 | |
A total of 47.3% reported any genital or anal STI symptoms in the last 6 months (table 1). Self-reported genital discharge and ulcer symptoms were common (29.5% and 31.6%, respectively). In addition, 25.0% reported anal ulcers and 12.6% reported anal discharge in the past 6 months. The majority of those reporting symptoms stopped having sex during symptom occurrence (61.2%).
Overall, 13.5% of participants were diagnosed with syphilis, rectal or urethral chlamydia, or rectal or urethral gonorrhoea (table 1). Prevalence of syphilis was the highest of the three infections (9.0%), while prevalence of chlamydia and gonorrhoea (rectal or urethral) ranged from 1.1% to 1.8%. Prevalence of HIV was 12.9%.
Table 2 shows weighted bivariate associations between select risk factors and laboratory-confirmed STI diagnosis of syphilis, gonorrhoea, or chlamydia. Reported alcohol use before sex was positively associated with STI diagnosis (OR=2.99). None of the STI symptoms reported in the last 6 months were significantly associated with STI diagnosis. HIV infection was positively associated with STI diagnosis (OR=2.76).
Table 2. Sociodemographic and behavioral variables by sexually transmitted infection (STI) diagnosis among men who have sex with men in Kampala, Uganda.
| Any STI diagnosis | No STI diagnosis | OR | p Value | |||
|---|---|---|---|---|---|---|
|
|
|
|||||
| n | Estimated population proportion % (95% CI) | n | Estimated population proportion % (95% CI) | |||
| Age (years) | ||||||
| 18–20 | 8 | 18.4 (7.6 to 30.8) | 43 | 18.3 (11.0 to 24.5) | Ref. | |
| 21–25 | 10 | 38.9 (21.3 to 56.2) | 99 | 43.3 (34.6 to 51.6) | 0.64 (0.20 to 2.07) | 0.452 |
| 26–30 | 11 | 20.3 (8.7 to 34.3) | 57 | 24.5 (17.8 to 33.6) | 0.78 (0.25 to 2.47) | 0.670 |
| >30 | 13 | 22.3 (8.4 to 40.0) | 41 | 13.9 (8.2 to 20.4) | 1.49 (0.47 to 4.67) | 0.495 |
| Years of schooling | ||||||
| None | 6 | 9.6 (2.0 to 18.9) | 16 | 8.0 (4.0 to 12.2) | Ref. | |
| 1–7 years | 9 | 27.7 (11.6 to 47.3) | 59 | 27.0 (17.1 to 34.5) | 0.91 (0.23 to 3.61) | 0.898 |
| 8–11 years | 6 | 19.5 (4.4 to 34.9) | 52 | 24.4 (16.8 to 32.5) | 0.64 (0.15 to 2.72) | 0.547 |
| 12 or more years | 20 | 43.2 (26.0 to 62.8) | 110 | 40.5 (32.5 to 52.8) | 0.85 (0.26 to 2.84) | 0.794 |
| Alcohol use before sex | ||||||
| Yes | 26 | 68.3 (50.5 to 85.8) | 94 | 42.7 (33.6 to 52.1) | 2.99 (1.28 to 6.96) | 0.011 |
| No | 15 | 31.7 (14.2 to 49.5) | 144 | 57.3 (47.9 to 66.4) | Ref. | |
| Had primary sex partner in the past 3 months | ||||||
| Yes | 39 | 89.5 (79.0 to 100.0) | 198 | 79.3 (72.2 to 86.1) | 2.58 (0.70 to 9.44) | 0.153 |
| No | 3 | 10.5 (0 to 21.0) | 42 | 20.7 (13.9 to 27.8) | Ref. | |
| Had anonymous sex partner in the past 3 months | ||||||
| Yes | 14 | 19.5 (7.6 to 34.5) | 90 | 34.3 (25.2 to 42.4) | 0.53 (0.22 to 1.26) | 0.151 |
| No | 27 | 80.5 (65.6 to 92.4) | 150 | 65.7 (57.6 to 74.8) | Ref. | |
| Ever sold sex | ||||||
| Yes | 26 | 50.7 (36.3 to 71.6) | 97 | 34.8 (26.5 to 43.4) | 1.22 (0.53 to 2.81) | 0.634 |
| No | 15 | 49.3 (28.4 to 63.7) | 125 | 65.2 (56.6 to 73.5) | Ref. | |
| Sold sex in the past 3 months | ||||||
| Yes | 17 | 35.2 (19.1 to 55.4) | 79 | 28.3 (19.2 to 33.9) | 1.57 (0.66 to 3.74) | 0.306 |
| No | 24 | 64.8 (44.6 to 80.9) | 161 | 71.7 (66.1 to 80.8) | Ref. | |
| Used condom during insertive anal sex in the past 3 months | ||||||
| Yes | 25 | 64.8 (48.5 to 85.8) | 137 | 64.6 (57.4 to 74.9) | 1.01 (0.39 to 2.63) | 0.983 |
| No | 10 | 35.2 (14.2 to 51.5) | 77 | 35.4 (25.1 to 42.6) | Ref. | |
| Used condom during receptive anal sex in the past 3 months | ||||||
| Yes | 19 | 60.7 (39.2 to 79.2) | 111 | 60.9 (52.2 to 70.3) | 0.97 (0.40 to 2.39) | 0.951 |
| No | 12 | 39.3 (20.8 to 60.8) | 77 | 39.1 (29.7 to 47.8) | ||
| STI symptoms in the past 12 months | ||||||
| Genital discharge | ||||||
| Yes | 19 | 42.3 (24.8 to 63.2) | 65 | 27.2 (19.3 to 35.0) | 2.19 (0.95 to 5.04) | 0.067 |
| No | 22 | 57.7 (36.8 to 75.2) | 169 | 72.8 (65.0 to 80.7) | Ref. | |
| Genital ulcers | ||||||
| Yes | 19 | 40.4 (21.5 to 59.8) | 70 | 29.1 (21.0 to 37.4) | 1.59 (0.69 to 3.67) | 0.282 |
| No | 21 | 59.6 (40.2 to 78.5) | 161 | 70.9 (62.6 to 79.0) | Ref. | |
| Anal ulcers | ||||||
| Yes | 15 | 33.6 (17.0 to 50.3) | 61 | 22.8 (15.6 to 30.3) | 1.44 (0.61 to 3.44) | 0.406 |
| No | 25 | 66.4 (49.7 to 83.0) | 168 | 77.2 (69.8 to 84.4) | Ref. | |
| Anal discharge | ||||||
| Yes | 9 | 21.4 (7.9 to 35.3) | 34 | 10.7 (6.1 to 16.1) | 2.09 (0.80 to 5.48) | 0.135 |
| No | 31 | 78.6 (64.8 to 92.1) | 197 | 89.3 (83.9 to 93.9) | Ref. | |
| Anal warts | ||||||
| Yes | 11 | 26.7 (12.1 to 41.6) | 45 | 15.3 (9.7 to 20.9) | 1.70 (0.66 to 4.39) | 0.274 |
| No | 28 | 73.3 (58.4 to 87.9) | 177 | 84.7 (79.1 to 90.3) | Ref. | |
| Any STI symptom | ||||||
| Yes | 26 | 38.0 (20.4 to 57.7) | 113 | 56.5 (47.3 to 65.1) | 1.94 (0.81 to 4.70) | 0.140 |
| No | 15 | 62.0 (42.4 to 79.6) | 116 | 43.5 (34.9 to 52.7) | Ref. | |
| HIV status | ||||||
| Positive | 10 | 27.6 (10.4 to 45.1) | 29 | 13.2 (6.6 to 19.8) | 2.76 (0.98 to 7.80) | 0.055 |
| Negative | 32 | 72.4 (54.9 to 89.6) | 213 | 86.8 (80.3 to 93.4) | Ref. | |
Results of the multivariate analysis are shown in table 3. Adjusting for age and education, STI diagnosis remained positively associated with alcohol use before sex (adjusted OR (AOR)=4.99), selling sex in the last 3 months (AOR=3.17) and HIV infection (AOR=3.46). Having anonymous partners in the last 3 months was inversely associated with STI diagnosis (AOR=0.20).
Table 3. Factors associated with sexually transmitted infection among men who have sex with men in Kampala, Uganda (n=257).
| Unadjusted ORs (95% CI) | Adjusted ORs (95% CI) | |
|---|---|---|
| Age (years) | ||
| 18–20 | Ref. | Ref. |
| 21–25 | 0.64 (0.20 to 2.07) | 0.36 (0.11 to 1.15) |
| 26–30 | 0.78 (0.25 to 2.47) | 0.29 (0.07 to 1.24) |
| >30 | 1.49 (0.47 to 4.67) | 0.85 (0.22 to 3.22) |
| Years of schooling | ||
| None | Ref. | |
| 1–7 years | 0.91 (0.23 to 3.61) | 0.82 (0.16 to 4.20) |
| 8–11 years | 0.64 (0.15 to 2.72) | 0.52 (0.08 to 3.63) |
| 12 or more years | 0.85 (0.26 to 2.84) | 0.88 (0.18 to 4.42) |
| Had anonymous sex partner in the past 3 months | ||
| Yes | 0.53 (0.22 to 1.26) | 0.20 (0.07 to 0.61) |
| No | Ref. | Ref. |
| Alcohol use before sex | ||
| Yes | 2.99 (1.28 to 6.96) | 4.99 (1.86 to 13.38) |
| No | Ref. | Ref. |
| Sold sex in the past 3 months | ||
| Yes | 1.57 (0.66 to 3.74) | 3.17 (1.25 to 8.07) |
| No | Ref. | Ref. |
| HIV status | ||
| Positive | 2.76 (0.98 to 7.80) | 3.46 (1.03 to 11.64) |
| Negative | Ref. | Ref. |
Discussion
We found a high combined prevalence of STI diagnoses among MSM in Kampala, and very high rates of self-reported STI symptoms. Though few studies have been conducted to assess burden of STI among MSM populations in sub-Saharan Africa, our findings are consistent with levels of STI prevalence ranging from 2.6% to 23.7% found among MSM in Kenya, Tanzania and Senegal.8–12 Compared with the population in Uganda, prevalence of syphilis among MSM in this study was over four times that of the general male population aged 15–49 years in 2011.3 Overall, MSM in Kampala had a higher burden of STI than the general population.
MSM also reported higher levels of STI symptoms in the previous 12 months, two times the amount reported by men in the general population; 18.4% of men aged 15–49 years reported STI symptoms in the 12 months prior to the 2011 Uganda AIDS Indicator Survey.3 Though laboratory-confirmed diagnoses did not precisely match self-reported symptoms, some infections may have been missed due to the cross-sectional design of the survey. We also did not find significant associations between any self-reported symptom and STI diagnosis. Finally, a significant proportion of those with STI diagnoses did not report any symptoms, suggesting that syndromic surveillance and diagnostic approaches may not be effective in this population. Still, given the prevalence of STI and the frequency of self-reported STI symptoms in this population, STI prevention and treatment programmes as well as routine STI surveillance should specifically support MSM and other high-risk groups and provide guidance on recognising urethral and rectal STI symptoms. To detect asymptomatic infections, routine STI testing should be provided to MSM based on reported risk behaviours and healthcare workers should be trained to provide STI services in a non-stigmatising environment.13 MSM should be educated to seek treatment if they experience symptoms of STI, and messages regarding preventive behaviours to reduce transmission to partners should be reinforced. Given the association between STI and HIV infection, HIV care and treatment services should also provide comprehensive counselling and screening for STI.
Overall, regular alcohol use among MSM was common, and nearly half reported using alcohol before sex, which was significantly associated with STI. This association has been reported in other populations, including female sex workers and the general population.14–16 Though our study did not find significant interactions between unprotected sex and alcohol use, alcohol use has been linked to sex work and other risky sexual behaviours such as unprotected sex.17,18 The use of alcohol before sex may lead to loss of inhibition, thus increasing risk-taking behaviours. Alcohol may also increase susceptibility to HIV and other infections and worsen disease course by impairing immunity.19 Counselling messages for this population should incorporate harm reduction material regarding the risks associated with alcohol use with sex.
We found that participants who reported having anonymous sex partners in the 6 months before the survey were less likely to be diagnosed with a STI than those who did not have anonymous partners. It is possible that MSM recognise higher infection risk with anonymous partners and are more likely to practice preventive behaviours with these partners, such as using condoms.20 However, risk behaviours have also been found to be associated with having anonymous partners.21 Our study may have been limited in assessing confounders in the relationship between STI and having anonymous sex partners. This association should be examined further using qualitative and quantitative methodologies.
This study had a number of limitations. Levels of stigma and discrimination of MSM in Uganda are high, and harassment of MSM, while unrelated to this study, adversely affected study participation. Though RDS is designed to capture a representative sample of the target population, RDS is based on a number of assumptions that may not have been met because this discrimination may have biased recruitment towards a certain subgroup of MSM. Thus, the findings of this study may not represent the larger MSM population in Kampala. Given the limitations of the data, it is difficult to ascertain whether the assumptions of RDS, specifically that respondents recognise each other as members of the population, respondent networks are linked and form a single component, and the sampling fraction is small, were fulfilled.22 In addition, the survey design was cross-sectional, limiting our ability to determine causal associations between risk factors and STI diagnoses. While study findings on multivariate associations were significant, CIs were wide due to the small sample size. STI symptoms and other risk behaviours were self-reported for the previous 3 months, 6 months or 12 months, while laboratory-confirmed diagnoses of STIs only reflected infection at the time of testing. With regards to PCR testing for NG and CT, we cannot rule out the possibility of some false-positive results, though the estimated prevalence was relatively low. Further, the assay we used is not approved for use on rectal specimens without further confirmation of positive test results. Finally, syphilis and rectal and urethral infections of NG and CTwere combined for our outcome variable. Although these infections are all current and reflect bacterial illness, it is possible that the epidemiology of each type of infection differs from one another.
This first round of biobehavioural surveillance among MSM in Kampala demonstrated the feasibility of conducting surveys with biological testing and the public health need related to STI and HIV infection in this highly stigmatised key population. We observed high prevalence of syphilis, NG and CT, and found high levels of self-reported STI symptoms. Our findings support the need for public health interventions specific to MSM to address risk behaviours related to alcohol use, STI transmission, identification of STI through symptom screening and testing, STI surveillance, and STI treatment-seeking behaviours. Finally, contextual factors including cultural perceptions and government policy in Uganda should be addressed to improve public health outcomes in this population.
Key messages.
Levels of self-reported STI symptoms and STI prevalence among MSM were high.
STI among MSM was associated with HIV and risky sexual behaviours, including alcohol use before sex and selling sex.
Public health interventions are needed to address STI risk and facilitate access to diagnosis and treatment services among MSM.
Acknowledgments
Funding This research has been supported by the President's Emergency Plan for AIDS Relief (PEPFAR) through the Centers for Disease Control and Prevention under the terms of U2GH000466.
Footnotes
Contributors EJK conducted the statistical analysis and wrote the manuscript; WH and AO developed the study and coordinated study implementation; JB, GL, SS and DS were involved in study implementation and data collection; JMS was involved in data collection, data management and statistical analysis. All authors contributed to writing and editing the manuscript.
Disclaimer The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Competing interests None declared.
Ethics approval Science and Ethics Committee of the Uganda Virus Research Institute (GC 127), the Uganda National Council of Science and Technology (HS370) and the Associate Director for Science of the CDC.
Provenance and peer review Not commissioned; externally peer reviewed.
References
- 1.Fleming DT, Wasserheit JN. From epidemiological synergy to public health policy and practice: the contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex Trans Infect. 1999;75:3–17. doi: 10.1136/sti.75.1.3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.World Health Organization. Global Prevalence and incidence of selected curable sexually transmitted infections: overview and estimates. Geneva: World Health Organization; 2001. [Google Scholar]
- 3.Ministry of Health (MOH) [Uganda] and ICF International. Uganda AIDS Indicator Survey 2011. Calverton, MD: MOH and ICF International; 2011. [Google Scholar]
- 4.Kajubi P, Kamya MR, Raymond HF, et al. Gay and bisexual men in Kampala, Uganda. AIDS Behav. 2008;12:492–504. doi: 10.1007/s10461-007-9323-7. [DOI] [PubMed] [Google Scholar]
- 5.Hladik W, Barker J, Ssenkusu JM, et al. HIV infection among men who have sex with men in Kampala, Uganda—a respondent driven sampling survey. PLoS One. 2012;7:e38143. doi: 10.1371/journal.pone.0038143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Heckathorn D. Respondent-driven sampling II: deriving valid population estimates from chain-referral samples of hidden populations. Soc Probl. 2002;49:11–34. [Google Scholar]
- 7.Abdul-Quader AS, Heckathorn DD, Sabin K, et al. Implementation and analysis of respondent driven sampling: lessons learned from the field. J Urban Health. 2006;83(Suppl 6):i1–5. doi: 10.1007/s11524-006-9108-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sanders EJ, Thiong'o AN, Okuku HS, et al. High prevalence of Chlamydia trachomatis and Neisseria gonorrhoeae infections among HIV-1 negative men who have sex with men in coastal Kenya. Sex Transm Infect. 2010;86:440–1. doi: 10.1136/sti.2010.043224. [DOI] [PubMed] [Google Scholar]
- 9.Johnston LG, Holman A, Dahoma M, et al. HIV risk and the overlap of injecting drug use and high-risk sexual behaviours among men who have sex with men in Zanzibar (Unguja), Tanzania. Int J Drug Policy. 2010;21:485–92. doi: 10.1016/j.drugpo.2010.06.001. [DOI] [PubMed] [Google Scholar]
- 10.Wade AS, Larmarange J, Diop AK, et al. Reduction in risk-taking behaviors among MSM in Senegal between 2004 and 2007 and prevalence of HIV and other STIs. ELIHoS Project, ANRS 12139. AIDS Care. 2010;22:409–14. doi: 10.1080/09540120903253973. [DOI] [PubMed] [Google Scholar]
- 11.Ross MW, Nyoni J, Ahaneku HO, et al. High HIV seroprevalence, rectal STIs and risky sexual behaviour in men who have sex with men in Dar es Salaam and Tanga, Tanzania. BMJ Open. 2014;4:e006175. doi: 10.1136/bmjopen-2014-006175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Muraguri N, Tun W, Okal J, et al. HIV and STI prevalence and risk factors among male sex workers and other men who have sex with men in Nairobi, Kenya. JAIDS. 2015;68:91–6. doi: 10.1097/QAI.0000000000000368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Branger J, van der Meer JTM, van Ketel RJ, et al. High incidence of asymptomatic syphilis in HIV-infected MSM justifies routine screening. Sex Trans Dis. 2009;36:84–5. doi: 10.1097/OLQ.0b013e318186debb. [DOI] [PubMed] [Google Scholar]
- 14.Norris AH, Kitali AJ, Worby E. Alcohol and transactional sex: how risky is the mix? Soc Sci Med. 2009;69:1167–76. doi: 10.1016/j.socscimed.2009.07.015. [DOI] [PubMed] [Google Scholar]
- 15.Mansergh G, Flores S, Koblin B, et al. Alcohol and drug use in the context of anal sex and other factors associated with sexually transmitted infections: results from a multi-city study of high-risk men who have sex with men in the USA. Sex Transm Infect. 2008;84:509–11. doi: 10.1136/sti.2008.031807. [DOI] [PubMed] [Google Scholar]
- 16.Tumqesigye NM, Wanyenze RK, Greenfield TK. Intoxication before last sexual intercourse and HIV risk behavior among men and women in Uganda: evidence from a nationwide survey. Int J Alcohol Drug Res. 2012;1:17. doi: 10.7895/ijadr.v1i1.36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hahn JA, Woolf-King SE, Muyindike W. Adding fuel to the fire: Alcohol's effect on the HIV epidemic in sub-Saharan Africa. Curr HIV/AIDS Rep. 2011;8:172–80. doi: 10.1007/s11904-011-0088-2. [DOI] [PubMed] [Google Scholar]
- 18.Kiene SM, Subramanian SV. Event-level association between alcohol use and unprotected sex during last sex: evidence from population-based surveys in sub-Saharan Africa. BMC Public Health. 2013;13:583. doi: 10.1186/1471-2458-13-583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Shuper PA, Neman M, Kanteres F, et al. Causal considerations on alcohol and HIV/ AIDS—a systematic review. Alcohol Alcohol. 2010;45:159–66. doi: 10.1093/alcalc/agp091. [DOI] [PubMed] [Google Scholar]
- 20.Cambou MC, Perez-Brumer AG, Segura ER, et al. The risk of stable partnerships: associations between partnership characteristics and unprotected anal intercourse among men who have sex with men and transgender women recently diagnosed with HIV and/or STI in Lima, Peru. PLoS One. 2014;9:e102894. doi: 10.1371/journal.pone.0102894. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tieu HV, Nandi V, Frye V, et al. Concurrent partnerships and HIV risk among men who have sex with men in New York City. Sex Transm Dis. 2014;41:200–8. doi: 10.1097/OLQ.0000000000000090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lansky A, Drake A, Wejnert C, et al. Assessing the assumptions of respondent-driven sampling in the National HIV Behavioral Surveillance System among injecting drug users. The Open AIDS Journal. 2012;6(Suppl 1):77–82. doi: 10.2174/1874613601206010077. [DOI] [PMC free article] [PubMed] [Google Scholar]