of 9

HIV Infection among Men Who Have Sex with Men in Kampala, Uganda–A Respondent Driven Sampling Survey

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
HIV Infection among Men Who Have Sex with Men in Kampala, Uganda–A Respondent Driven Sampling Survey
  HIV Infection among Men Who Have Sex with Men inKampala, Uganda–A Respondent Driven SamplingSurvey Wolfgang Hladik  1,2 * , Joseph Barker 1 , John M. Ssenkusu 1 , Alex Opio 3 , Jordan W. Tappero 1 , Avi Hakim 4 ,David Serwadda 5 , for the Crane Survey Group 1 Division of Global HIV/AIDS, Center for Global Health, Centers for Disease Control and Prevention (CDC), Entebbe, Uganda,  2 Department of Clinical Epidemiology,Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands,  3 Ministry of Health, Kampala, Uganda,  4 Division of Global HIV/AIDS, Center for GlobalHealth, Centers for Disease Control and Prevention (CDC), Atlanta, Georgia, United States of America,  5 School of Public Health, Makerere University, Kampala, Uganda Abstract Background:   Uganda’s generalized HIV epidemic is well described, including an estimated adult male HIV prevalence inKampala of 4.5%, but no data are available on the prevalence of and risk factors for HIV infection among men who have sexwith men (MSM). Methodology/Principal Findings:   From May 2008 to February 2009, we used respondent-driven sampling to recruit MSM $ 18 years old in Kampala who reported anal sex with another man in the previous three months. We collecteddemographic and HIV-related behavioral data through audio computer-assisted self-administered interviews. Laboratorytesting included biomarkers for HIV and other sexually transmitted infections. We obtained population estimates adjustedfor the non-random sampling frame using RDSAT and STATA. 300 MSM were surveyed over 11 waves; median age was 25years (interquartile range, 21–29 years). Overall HIV prevalence was 13.7% (95% confidence interval [CI] 7.9%–20.1%), andwas higher among MSM $ 25 years (22.4%) than among MSM aged 18–24 years (3.9%, odds ratio [OR] 5.69, 95% CI 2.02–16.02). In multivariate analysis, MSM $ 25 years (adjusted OR [aOR] 4.32, 95% CI 1.33–13.98) and those reporting ever havingbeen exposed to homophobic abuse (verbal, moral, sexual, or physical abuse; aOR 5.38, 95% CI 1.95–14.79) weresignificantly more likely to be HIV infected. Conclusions/Significance:   MSM in Kampala are at substantially higher risk for HIV than the general adult male population.MSM reporting a lifetime history of homophobic abuse are at increased risk of being HIV infected. Legal challenges andstigma must be overcome to provide access to tailored HIV prevention and care services. Citation:  Hladik W, Barker J, Ssenkusu JM, Opio A, Tappero JW, et al. (2012) HIV Infection among Men Who Have Sex with Men in Kampala, Uganda–ARespondent Driven Sampling Survey. PLoS ONE 7(5): e38143. doi:10.1371/journal.pone.0038143 Editor:  Andres G. Lescano, U.S. Naval Medical Research Unit Six (NAMRU-6), Peru Received  March 9, 2011;  Accepted  May 2, 2012;  Published  May 31, 2012This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone forany lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication. Funding:  This study was funded by the US President’s Emergency Plan for AIDS Relief. The funders had no role in study design, data collection and analysis,decision to publish, or preparation of the manuscript. Competing Interests:  The authors have declared that no competing interests exist.* E-mail: wfh3@cdc.gov Introduction Men who have sex with men (MSM) are a recognized high-risk group for infection with human immunodeficiency virus (HIV) inthe industrialized world. MSM are also well described in selectlow- and medium-income countries such as Brazil and Thailand[1]. Although HIV epidemics have severely affected large parts of  Africa for decades, only recently has there been attention to itsspread among African MSM outside of South Africa [2,3,4,5]. Even in the generalized HIV epidemics of sub-Saharan Africa,MSM are three to four times more likely to be HIV infected thanthe general adult population [6]. MSM in Africa frequently reportbi-sexual, often concurrent, relationships, making their femalepartners an important, and often unwitting, bridging population[7]. Although reliable MSM population size estimates in Africa arelargely lacking, the high HIV prevalence observed in severalstudies would suffice to account for a significant proportion of allnew HIV infections in many of these generalized epidemics [4].Despite this, tailored research for MSM in Africa, especially inHIV prevention, remains an unmet priority [4]. Illegality of homosexual behavior, human rights abuses, and severe stigma addto the specifics’ of this continent’s HIV epidemic among MSM[2,8]. As a result, many national HIV policies, AIDS controlprograms, and HIV services either ignore or inadequately addressthe specific needs of this most-at-risk population (MARP) in Africa[8]. In Uganda too, public health programming is also impededdue to the illegality of homosexual behavior and the relatedwidespread stigma [6]. An anti-homosexuality bill recently re-proposed in the Ugandan parliament prescribes life-long impris-onment and possibly, including for HIV-infected MSM, the deathpenalty [9].Uganda faces a mature, generalized HIV epidemic with anestimated prevalence among the general adult population of 6.4%[10]. HIV infection is more frequent in urban (10.1%) than in PLoS ONE | www.plosone.org 1 May 2012 | Volume 7 | Issue 5 | e38143  rural (5.7%) Ugandan adults, and is higher among female (7.5%)than male adults (5.0%) [10]. Reliable data on the burden of disease among populations at higher risk for HIV are scarce. Arecent survey estimated the HIV prevalence among female sexworkers (FSW) in Kampala as 33% and that of clients and spousesof FSW as 18% [11]. An additional high risk group in Ugandaincludes mobile men with money, such as fishermen among whoma recent survey found an HIV prevalence of 16.4% [12].For many hard-to-reach populations, including MSM, respon-dent-driven sampling (RDS) can be an effective sampling tool thatfacilitates population-based, weighted estimates. [13,14,15]. RDS- based MSM surveys have been conducted in numerous countries[16], including in Africa. A survey conducted in 2004 in Kampalaconfirmed the existence of an indigenous MSM population andthe presence of HIV-related risk behaviors, including high rates of commercial or transactional sex [17]. We report here on a morerecent survey, the Crane Survey, that aimed at estimating theburden of HIV and STI in select groups assumed to be atincreased risk for HIV. While our survey among Kampala MSMincluded for the first time testing for both HIV and STI, we reporthere on the HIV-related findings; separate analyses examine non-HIV domains (including STIs and behavioral outcomes) in moredetail. Methods Ethics statement The survey protocol and consent procedures were approved bythe Uganda Virus Research Institute’s institutional review board,the Uganda National Council of Science and Technology, and theCDC. The survey was conducted anonymously; informed consentwas obtained verbally; no personal identifiers were collected. Survey design This survey was cross-sectional and used RDS. The survey ispart of the ‘‘Crane Survey’’, a joint activity by MakerereUniversity, the Ministry of Health (MoH), and the US Centersfor Disease Control and Prevention (CDC). Prior to the survey,formative research was carried out over several weeks throughapproximately a dozen key informant interviews using semi-structured interview guides or open discussions to inform sampling design, seed identification, social connectedness among MSM inKampala, number and location of survey offices, protection of privacy, compensation, data and biological measures, language,and other procedures. Key findings included that RDS wasrecommended and preferred over time-location sampling, that asingle survey office would suffice, that anonymity was paramount,and that biomarker results needed to be returned and addressed.We conducted the survey between May 2008 and April 2009. Setting The survey was carried out in Kampala, Uganda’s capital andlargest city with approximately 1.8 million residents. Study population Inclusion criteria were male sex, age  $ 18 years, residence ingreater Kampala, and self-reported anal sex with another man inthe preceding 3 months. Exclusion criteria included couponreceipt from a stranger and language barriers. Sampling RDS methodology is well described elsewhere [13,14,15] and represents an advanced version of chain referral sampling. Weinitiated sampling with eight seeds purposively selected by age,HIV status, and geographic location in Kampala; all were sociallywell networked. After seeds began their recruiting efforts,candidate recruits presented the coupons they had received tothe single survey office in Old Kampala, near the city center.Sampling was interrupted twice; first as a result of the arrest of three local lesbian, gay, bisexual, and transgender (LGBT) activistsat the PEPFAR Implementer’s Meeting in Kampala in early June2008 [18]. Sampling remained at a low level until it rebounded in August, but was again affected in September 2008 by arrests of alleged homosexuals [19]. A total of six additional seeds wereadded during the survey. However, following the second round of arrests, sampling rates remained low until we stopped sampling inMarch 2009, close to the survey’s scheduled end.We issued 1,706 coupons and redeemed 455 of these. Thenumber of coupons issued per recruit ranged from two at thesurvey’s beginning to six after the arrests and slump in sampling. Atotal of 300 eligible recruits (n=286) and seeds (n=14)participated in the survey; the remainder were ineligible, mostlybecause they did not report anal sex with men in the last 3 months.The longest recruitment wave was 11; equilibrium for HIVserostatus was reached after wave number 2, and that fornationality, age, and condom use at last sex was reached afterwave number 3, 4, and 4, respectively, the  computed   design effect(using actual survey data) for HIV infection was 2.3. Survey office procedures Candidate participants were screened for eligibility face-to-faceby survey staff. Following a briefing about specific interview terms,such as the definition of sexual intercourse, frequency of sex,partner types, or commercial sex and a short computer-basedtutorial about audio-computer-assisted self-interviewing (ACASI),enrolled participants underwent a standardized interview using Questionnaire Design Studio (QDS v2.5) software (NOVA,Bethesda, Maryland, USA). A small number of recruits preferreda computer-assisted personal interview with trained staff using thesame QDS instrument. Following the interview and pre-testcounseling, recruits provided venous blood and urine samples, andhad rectal swabs collected. At the end of the first visit, recruitsreceived instructions and coupons for peer recruitment.Recruits were scheduled to return to the survey office two weekslater. Survey staff post-test counseled recruits for all biomarkersmeasured, provided treatment for non-viral sexually transmittedinfections (STI) according to World Health Organization andMOH guidelines, and referred HIV-positive recruits to health careproviders with whom we had made arrangements prior to surveystart. Recruits who failed to return to the survey office could not bereached by survey staff as no personal identifiers were collected.Recruits who returned to the survey office when not all biomarkerresults were yet available were asked (and compensated for) toreturn for an additional visit. At both visits to the survey office, we compensated recruits fortheir time and transport costs (US $ 3.00), and, at the return visit,recruitment efforts (US $ 1.00 per successfully recruited eligiblepeer). At the time of the survey, US  $ 3.00 could purchase threekilograms of sugar. Data measures The main interview’s key domains included demographics, life-time sexual characteristics, sexual behaviors in the last threemonths, sexual violence, and STDs. Condom use was examinedboth quantitatively (number of protected sex acts divided by all sexacts in last three months), as well as qualitatively (condom use everor never, condom use at last sex). Lifetime exposure tohomophobic abuse was measured through the question ‘‘  Did you HIV in Kampala MSMPLoS ONE | www.plosone.org 2 May 2012 | Volume 7 | Issue 5 | e38143  Table 1.  MSM characteristics, Kampala, Uganda, 2008/9. Numerator unweighted(N=295) Sample %Estimated populationproportion* % (N=281) 95% C.I.Race African 276 93.9 94.0 91.1–97.4Other 18 6.1 6.0 2.6–8.9 Nationality Ugandan 267 91.4 93.9 90.4–97.2Other 25 8.6 6.1 2.8–9.6 Religion Catholic 123 42.0 39.2 31.7–47.0Protestant 73 24.9 29.0 21.8–36.8Moslem 45 15.4 17.1 10.9–23.3Other religion 41 14.0 11.6 7.5–15.9None 11 3.8 3.1 0.6–7.4 Age (years) 18–24 143 48.5 49.7 39.2–56.525 +  152 51.5 50.3 43.5–60.8 Schooling (years) 0–6 71 24.6 24.9 17.4–32.57 +  218 75.4 75.1 67.5–82.6 Occupation Student 52 17.8 18.5 11.1–24.7Unemployed 49 16.8 16.0 10.6–22.0Employed 191 65.4 65.5 57.7–74.6 Marital status (with women) Never married 194 66.0 70.2 61.9–77.6Married 57 19.4 13.2 8.8–19.1Previously married 43 14.6 16.5 10.5–22.7 Current marital status Yes 57 19.4 13.3 9.0–19.3No 237 80.6 86.7 80.7–91.0 Heterosexual exposure Ever sex with a woman 226 76.6 77.6 70.7–83.5Ever fathered children 94 32.2 29.6 21.8–37.1 Living with a female sex partner Yes 53 18.0 16.4 11.2–22.0No 241 82.0 83.6 78.0–88.8 Circumcision status Circumcised 145 49.2 44.1 35.5–53.6Uncircumcised 150 50.8 55.9 46.4–64.5 Ever tested for HIV Yes 128 45.1 43.4 36.5–52.2No 156 54.9 56.6 47.5–63.5 No. sex partners in last 6 months , =10 partners 133 48 52.7 44.4–63.811–24 partners 57 20.6 23.0 15.1–29.3 . =25 partners 87 31.4 24.3 16.8–31.5 STIs Yes 40 13.9 12.4 8.2–17.8No 248 86.1 87.6 82.2–91.8 HIV in Kampala MSMPLoS ONE | www.plosone.org 3 May 2012 | Volume 7 | Issue 5 | e38143  ever suffer any violence or abuse because you have sex with other men?  ’’.Respondents who affirmed such abuse, where probed about thetype of abuse, including moral abuse (isolation or exclusion), verbal abuse (threats or insults), mistreatment, or having beensubjected to physical or sexual violence. Respondents were alsoprobed about blackmail (‘‘ Have you ever been blackmailed by someone because you have sex with other men?  ’’) and rape (‘‘ Were you ever forced tohave sex against your will?  ’’). Other data measures included alcoholuse, drug use (including injection drug use), as well as buying sex(defined as paying for sex with money, goods or services) or selling sex (defined as giving sex in exchange for money, goods orservices). Laboratory measures Laboratory testing was performed off-site at the STD ReferenceLaboratory based at Mulago Hospital, Kampala; a small numberof tests were also performed at the CDC laboratory in Entebbe,Uganda. Blood specimens were stored at 2 to 8 u C at the surveyoffice and transported twice daily to the laboratory. Testing forantibodies against HIV was performed through a parallel testing algorithm using Vironostika H  HIV Uniform II plus O2 (bioMer-ie´ux, Marcy l’Etoile, France) and Murex H  HIV Ag/Ab Combi-nation (Abbott Laboratories, Abbott Park, Illinois, U.S.A.);discordant results were resolved through the use of HIV 1/2STAT-PAK rapid test (Inverness Medical, Princeton, New Jersey,U.S.A.). All recruits enrolled into the survey accepted HIV testing.Plasma was also tested for  Treponema pallidum  (TP) infection, using the Anti-syphilis IgG ELISA (Biotec Laboratories, Suffolk, UK) forscreening and, if reactive, the Rapid Plasma Reagin Syfacard-RTest (Murex Biotech, Dartford, UK) to detect current TPinfection. Urine specimens and rectal swabs were tested for thepresence of   Neisseria gonnorhea   (NG) and  Chlamydia trachomatis   (CT)DNA (Cobas Amplicor or Amplicor PCR, Roche Diagnostics,Branchburg, New Jersey, U.S.A.). Data management and analysis For sample size calculations, we assumed an HIV prevalence of 14%, 95% confidence intervals (CI) of 10.4%–18.3% (approxi-mately twice that of urban men in general [14] ), and a designeffect of 2 [15]. Aiming for an effective sample size of 300, weadjusted the target sample size to 600.Survey events (enrollments, recruiter-recruitee links, couponsnumbers issued, unique codes, etc.) were tracked with an in-housedeveloped software. Interview data were checked for errors andinconsistencies, and cleaned after importation from QDS intoStatistical Analysis Software - SAS v9.2 (SAS Institute, Gary,North Carolina).Of the 300 eligible respondents, 5 provided little interview dataand were excluded from this analysis. Our principal outcome of interest was HIV infection; predictor variables included demo-graphics, sexual orientation and experience, sexual behavior in the3 months preceding the interview, HIV testing (history), sexual violence, abuse, or blackmail, as well as laboratory markers of sexually transmitted infections and STD symptoms. Condom useover the last three months includes both male and female sexpartners (unless stated otherwise) and was computed as acontinuous variable with the denominator being the number of sex acts and the numerator being the number of sex acts protectedby condoms. We examined condom use as a categorical variable,indicating whether in the preceding three months condoms wereused for less than 33% of all sex acts, for 33%–66% of sex acts, orfor more than 66% of sex acts.We present weighted data except for continuous data;univariate analyses were conducted in RDSAT version 6.0.1.(www.respondentdrivensampling.org). Individual HIV sampling weights were generated in RDSAT and exported to STATA.Using HIV weights imported from RDSAT, logistic regression wasconducted for bivariate and multivariate analysis in STATA 10.0(Stata Corporation, College Station, Texas). For multivariateanalysis we employed backward elimination using predictor variables associated with HIV infection at a level of   P  # 0.2 inbivariate analysis. The final weighted model displays all predictor variables significantly associated with HIV infection at a level of  P  # 0.05. Table 1.  Cont. Numerator unweighted(N=295) Sample %Estimated populationproportion* % (N=281) 95% C.I.Nationality of last male sex partner Ugandan 225 79.2 81.1 75.5–87.2Other African country 33 11.6 10.9 6.1–15.8Outside Africa 23 8.1 6.7 3.3–11.0Don’t know 3 1.1 1.3 0–2.4 Sexual orientation Gay/homosexual 166 56.8 55.0 48.4–63.3Bisexual 113 38.7 37.6 29.4–44.1Heterosexual 13 4.5 7.4 3.0–13.1 Sexual attraction Mostly/only to men 208 71.0 68.8 62.0–76.7Equally to men and women 38 13.0 12.3 7.3–15.8Mostly/only to women 47 16.0 18.8 13.5–25.1Note: Denominators vary due to missing data (refuse to answer, don’t know).*) Estimated using RDSAT software.doi:10.1371/journal.pone.0038143.t001 HIV in Kampala MSMPLoS ONE | www.plosone.org 4 May 2012 | Volume 7 | Issue 5 | e38143  Table 2.  Bivariate associations with HIV status. Sample estimates Population estimatesHIV   (%) HIV   (%) Odds ratio 95% CI  p  Age (years) 18–24 137 (95.8) 6 (4.2) Ref - -25 +  117 (77.5) 34 (22.5) 5.69 2.02–16.02 0.001 Circumcision status Uncircumcised 131 (87.3) 19 (12.7) Ref - -Circumcised 123 (85.4) 21 (14.6) 0.84 0.35–2.01 0.690 No. years in school 0–6 64 (90.1) 7 (9.9) Ref - -7 +  185 (85.3) 32 (14.7) 1.55 0.56–4.34 0.401 Religion Catholic 105 (85.4) 18 (14.6) Ref - -Protestant 63 (87.5) 9 (12.5) 1.18 0.37–3.78 0.775Moslem 36 (80.0) 9 (20.0) 1.06 0.37–3.04 0.912Other religion 37 (90.2) 4 (9.8) 0.38 0.10–1.47 0.161None 11 (100) 0 (0) - - - Occupation Student 50 (96.2) 2 (3.8) Ref - -Unemployed 43 (87.8) 6 (12.2) 2.42 0.42–13.99 0.322Employed 158 (83.2) 32 (16.8) 4.03 0.86–18.87 0.077 Marital status (with women) Never married 168 (86.6) 26 (13.4) Ref - -Married 53 (94.6) 3 (5.4) 0.21 0.06–0.80 0.021Previously married 32 (74.4) 11 (25.6) 2.78 0.88–8.77 0.081 Alcohol consumption past 30 days None 81 (94.2) 5 (5.8) Ref - -Less than once a week 22 (91.7) 2 (8.3) 0.71 0.09–5.26 0.734At least once a week 108 (85.0) 19 (15.0) 1.45 0.48–4.37 0.504About every day 40 (75.5) 13 (24.5) 4.72 1.24–17.96 0.023 Illict drug consumption Never 174 (82.9) 36 (17.1) Ref - -Ever 80 (95.2) 4 (4.8) 0.14 0.04–0.54 0.004 Ever injected drugs Never 223 (85.1) 39 (14.9) Ref - -Ever 31 (96.9) 1 (3.1) 0.08 0.01–0.60 0.014 Condom use Never 67 (88.2) 9 (11.8) Ref - -Ever 179 (86.1) 29 (13.9) 2.00 0.75–5.32 0.167 Condom use (for sex acts in last 3 months) . 66% 91 (86.7) 14 (13.3) Ref - -33%–66% 50 (84.8) 9 (15.2) 2.28 0.71–7.36 0.167 , 33% 98 (88.3) 13 (11.7) 0.59 0.23–1.53 0.281 Lubricant use Ever 190 (84.1) 36 (15.9) Ref - -Never 55 (96.5) 2 (3.5) 0.12 0.02–0.67 0.015 Ever sex with women Never 55 (79.7) 14 (20.3) Ref - -Ever 199 (88.4) 26 (11.6) 0.65 0.26–1.61 0.354 History of blackmail Never 116 (92.1) 10 (7.9) Ref - - HIV in Kampala MSMPLoS ONE | www.plosone.org 5 May 2012 | Volume 7 | Issue 5 | e38143
Related Search
Related Docs
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks