Abstract
Background
Men who have sex with men (MSM) and male-to-female transgender women (TW) are at increased risk of HIV and sexually transmitted infections (STIs). We evaluated factors associated with incidence of HIV, HSV-2, and chlamydia and gonorrhea (anal and pharyngeal).
Methods
We used data from the Comunidades Positivas trial with MSM/TW who have sex with men in Lima, Peru. Participants were asked about sexual risk behaviors and underwent HIV/STI testing at baseline and 9- and 18-month follow-ups. We used discrete time proportional hazards regression to calculate hazard ratios (HRs) for variables associated with incidence of each STI.
Results
Among 718 MSM/TW, HIV incidence was 3.6 cases per 100 person-years. HIV incidence was associated with having an incident STI (aHR 3.73). Unprotected receptive anal intercourse was associated with incident anal chlamydia (aHR 2.20). An increased number of sexual partners increased incident HSV-2 (aHR 3.15 for 6–14 partners and 3.97 for 15–46 partners compared to 0–2 partners). Risk of anal gonorrhea decreased with each sexually active year (aHR 0.94) and increased for unprotected compensated sex (aHR 2.36). Risk of pharyngeal gonorrhea also decreased with each year since sexual debut (aHR 0.95). Risk of anal chlamydia decreased with each sexually active year (aHR 0.96), risk increased with reports of unprotected sex work (aHR 1.61), and unprotected receptive anal sex (aHR 2.63). All aHRs have p-values < 0.05.
Conclusion
MSM/TW experience high incidence of HIV. Up-to-date prevalence and incidence information and identifying factors associated with infection can help develop a more effective combination prevention response.
Keywords: HIV, HSV-2, gonorrhea, chlamydia, MSM, transgender, Peru
Introduction
Up-to-date Sexually Transmitted Infection (STI) epidemiologic surveillance is necessary to develop and evaluate interventions. Men who have sex with men (MSM) and male-to-female transgender women (TW) reporting sexual contact with men and TW partners, have a significantly higher STI prevalence than any other populations in Peru. Their increased STI prevalence is well documented,1–5 but there are few incidence estimates6 and further analysis is needed to understand risk factors for infection. Estimates place Human Immunodeficiency Virus (HIV) seroprevalence at 13.9% for MSM overall1 and prevalence upwards of 30% among TW.7 The prevalence of many STIs is highest among MSM, such as herpes simplex virus type-2 (HSV-2) infection at 69.0%, recent syphilis at 10.5%, and urethral gonorrhea and/or chlamydia at 1.5%.8 Among Peruvian men from the general population, prevalences are 0.4% for HIV, 13.5% for HSV-2, 0.5% for syphilis, 4.2% for urethral chlamydia, and 0.1% for urethral gonorrhea.9 Investigations worldwide present similar findings of high HIV/STI seroprevalence among MSM and TW populations as compared to the general population.10,11 These elevated prevalence figures rationalize the focus of surveillance and prevention strategies specifically for MSM and TW.
In Peru, Lima and Callao remain the epicenter of a concentrated HIV epidemic with 77% of all new cases.12 It is estimated that 55% of all HIV incidence occurs among MSM13 who comprise approximately 6% of Peru’s population.5,9 Various factors influence the high STI prevalence in MSM and TW worldwide. These include biological factors (i.e. high transmissibility of HIV through condomless anal intercourse), behavioral (e.g. multiple sex partners, drug and alcohol use with sexual activity, frequent involvement in commercial sex work due to lack of other options), and structural (discrimination, violence, and limited access to healthcare, education, housing, or employment)14–18 Despite access to condoms, advances in antiretroviral therapy coverage, counseling promoting risk reduction, and access to STI screening, prevalence remains high among these most at risk groups.
Although STI prevalence has been well documented for Peru’s MSM, only two previous studies have assessed HIV and STI incidence, through case-control analysis and mathematical modeling to predict incidence.19,20 In our analysis we use the Comunidades Positivas study with data gathered from MSM and TW participants over 18 months.21 We assess incidence of HIV, HSV-2, anal and pharyngeal chlamydia and gonorrhea infections and associated risk factors in the MSM and TW population of Lima and Callao.
Methods
Design Overview
The Comunidades Positivas randomized control trial used a 2 × 2 factorial design testing a social/structural intervention (Positive Communities) and a biomedical intervention (participant-delivered partner STI treatment). In this trial, 24 barrios (low-income neighborhoods) in the region of Lima were randomized and assigned to one of four intervention conditions. Independent of intervention assignment, MSM and TW were recruited from each barrio and assessed for the outcomes of interest at baseline, 9 months, and 18 months. This assessment included a behavioral interview, serology for HIV-1, HSV-2, and nucleic acid amplification testing (NAAT) for chlamydia and gonorrhea in both pharyngeal and anal swabs.
The Comunidades Positivas intervention was not effective and therefore we have not explored the associations of the intervention arm with HIV/STI incidence,21 further results are forthcoming. All participants from both trial arms are included in the analyses presented here.
Inclusion Criteria
Participants included those biological males between the ages of 18 and 45 at baseline who reported at least one sexual encounter with a man in the past 12 months, reported a sexual preference for other men, lived or worked near the intervention area, showed willingness to participate in the study, and planned to stay in the community for the 18-month study period.
Participant Recruitment
Lower-income barrios with visible MSM and TW communities were selected and through a snowball technique potential participants were identified and invited to enroll in the study. Selection of- and participant recruitment in the first 16 barrios took place in the metropolitan area of Lima between March and May, 2008. In the remaining 8 barrios (4 within Metropolitan Lima and 4 in surrounding Lima provinces), these processes took place between September and December 2009.
Data Collection
Data for the baseline, 9-month, and 18-month assessments were gathered in rented storefronts or apartments with interviews and collected specimens. All participants completed a behavioral survey and went through pre-test counseling for HIV/STI with a trained counselor as required by Peruvian law. Following counseling a trained phlebotomist drew a 10 ml blood sample and collected a pharyngeal swab. Self-obtained rectal swab samples were also collected. STI and HIV results were provided within two weeks of initial visit along with post-test counseling. Treatment was offered to participants with STI signs and symptoms or laboratory diagnosed bacterial STIs. Given the intervention’s inclusion of participant-delivered partner STI treatment, this was provided to half of the participants, while the other half were referred for partner treatment to local health clinics, the standard of care. Newly diagnosed HIV cases were referred to the Peruvian HIV antiretroviral treatment program.
Laboratory Methods
HIV antibody status was determined using Genetic Systems HIV-1/HIV-2 EIA (BIO-RAD Laboratories, Redmond, WA). Positive results were confirmed with GenScreen HIV-1 Western Blot (BIO-RAD Laboratories, Redmond, WA). HSV-2 antibody status was determined by type-specific enzyme immunoassay HerpeSelect 2 ELISA IgG (Focus Diagnostics, Cypress, CA). Chlamydia and gonorrhea were determined in pharyngeal and rectal secretions using Aptima Combo 2 Assay for chlamydia trachomatis (CT) and/or Neisseria gonorrhea (NG) (GenProbe, San Diego, CA). For quality control (QC) purposes, 10% of samples were sent to the San Francisco Department of Public Health Laboratory (San Francisco, CA) for HIV, HSV-2, and CT/NG QC testing.
Anal and pharyngeal rather than urethral samples of gonorrhea and chlamydia were measured in this study because participant prevalence of urethral infection at baseline was too low to justify continued measurement.
Protection of Human Subjects
The study was approved by the Institutional Review Boards of the University of California, Los Angeles, and the Universidad Peruana Cayetano Heredia. Data were collected from eligible participants who gave their written informed consent to participate in the study. Study implementation was overseen by an independent Data and Safety Monitoring Board.
Measures Used
Sexual risk behavior including the number of sex partners and insertive and receptive anal sex were based on aggregate counts of these behaviors with the participants’ last three sex partners in the previous 6 months. Participants were determined to have unmet basic needs if they reported any month in the past year where they did not have money to cover their basic needs. Sexual partnerships were coded as stable if the participant self-reported a sex partner as being a live-in partner or boyfriend rather than a causal or commercial partner. Number of sex partners in the last year was split into quintiles.
The incident STI variable was coded as positive if during the course of the 18 month study period the participant had an incident infection of HSV-2, anal gonorrhea, or anal chlamydia. HIV infection was not included in this variable.
The CAGE scale was based on four questions asking about Cutting-down alcohol consumption, Annoyance with criticism of one’s drinking, feelings of Guilt about their drinking, and needing to have a drink as a morning Eye-opener. The questions were dichotomized into ever and never, and participants who answered ever to two or more questions were coded as CAGE-positive. In this population this scale has a Cronbach’s alpha of 0.75.22
Statistical Analysis
Variables associated with seroconversion to positivity for HIV or HSV-2 and infection with gonorrhea or chlamydia, were modeled utilizing Discrete Time Proportional Hazards regression for bivariate and multivariate analyses. For calculation of HIV and HSV-2 incidence rates and analyses of factors associated with these infections, baseline seropositives were excluded separately for each infection. We did not exclude patients with baseline positivity for gonorrhea and chlamydia because they received treatment for these infections and once treated, they returned to the pool of susceptibles. However, we took the potential lack of independence between multiple positives for gonorrhea and chlamydia into account by adjusting the standard errors by participant using the cluster command. Given the overlap between the social and sexual networks of MSM and transwomen in the neighborhoods understudy, our analyses have looked at transgender identity as an exposure, but analysis was not conducted separately for MSM and transgender women. In the multivariate models, we included all variables that had a bivariate p-value less than or equal to 0.10 or those variables considered important a priori regardless of bivariate p-value, such as number of years since sexual debut. All analyses were conducted using Stata 12.1 (College Station, TX).
Results
Population Characteristics
A total of 510 MSM and 208 TW participants were enrolled at baseline, at the 9-month follow-up 619 participated (86.2%), and the 18-month follow-up 574 MSM and TW (79.9%) participated. Our study participants were primarily low income, between the ages of 25 to 35 (43.8%) with an average age of 29.5 years; the majority originated from the metropolitan region of Lima (53.6%), were primarily homosexual or gay identified (64.4%), and had a secondary education (62.6%). The majority of participants reported more difficultly using condoms as a result of alcohol or drug use (60.2%). Those who engaged in compensated sex (54.8%) were more likely to report using condoms with clients rather than not using condoms (55.2% versus 44.8%). The majority of study participants did not engage in insertive anal sex (74.6%); conversely, most (91.1%) engaged in receptive anal sex. Those who engaged in receptive anal sex were more likely not to use condoms (56.1% versus 35.1%) (see Table 1).
Table 1.
Characteristics of baseline seropositive and seronegative MSM and TW in Lima, Peru, 2009–2011
| Variables and Population Statistics | Total (713) N (%) |
HIV+ at baseline (127, 17.8%) N (%) |
HIV- at baseline (581, 81.5%) N (%) |
p-value | |
|---|---|---|---|---|---|
| Age (quintiles) | |||||
| 18–20 | 82 (11.5) | 5 (3.8) | 76 (11.6) | <0.01 | |
| 21–24 | 146 (20.5) | 27 (20.5) | 198 (30.1) | ||
| 25–35 | 312 (43.8) | 59 (44.7) | 252 (38.4) | ||
| 36–45 | 173 (24.3) | 41 (31.1) | 131 (19.9) | ||
| Sexuality and Gender Identity1 | |||||
| Homosexual | 459 (64.4) | 83 (6.6) | 372 (64.7) | 0.84 | |
| Bisexual | 40 (5.6) | 8 (6.3) | 31 (5.4) | ||
| Transgender | 207 (29.0) | 35 (27.8) | 172 (29.9) | ||
| Educational level | |||||
| Less than secondary | 66 (9.3) | 13 (10.2) | 52 (9.0) | 0.79 | |
| Complete secondary | 446 (62.6) | 73 (57.5) | 370 (63.7) | ||
| Post-secondary | 201 (28.2) | 41 (32.3) | 159 (27.4) | ||
| Employment Status | |||||
| Employed | 390 (54.7) | 64 (61.0) | 325 (62.5) | 0.77 | |
| Unemployed | 238 (33.4) | 41 (39.0) | 195 (37.5) | ||
| Use of alcohol and drugs made condom use during sex more difficult in last 3 months | |||||
| Never | 429 (60.2) | 81 (63.8) | 346 (59.9) | 0.41 | |
| Sometimes/Always | 281 (39.4) | 46 (36.2) | 232 (40.1) | ||
| Use of condoms with compensated sex | |||||
| No compensated sex | 321 (45.0) | 50 (39.4) | 267 (46.0) | 0.13 | |
| Always | 216 (30.3) | 37 (29.1) | 179 (30.9) | ||
| Sometimes/Never | 175 (24.5) | 40 (31.5) | 134 (23.1) | ||
| Lack money in last 12 months | |||||
| Never | 74 (10.4) | 8 (6.3) | 65 (11.2) | 0.20 | |
| Rarely | 216 (30.3) | 35 (27.6) | 179 (30.8) | ||
| Sometimes | 301 (42.2) | 57 (44.9) | 244 (42.0) | ||
| Very often | 122 (17.1) | 27 (21.3) | 93 (16.0) | ||
| # of sex partners in 1 year (quintiles) | |||||
| 0–2 | 143 (20.1) | 28 (22.0) | 114 (19.7) | 0.84 | |
| 3–5 | 146 (20.5) | 25 (19.7) | 119 (20.5) | ||
| 6–14 | 160 (22.4) | 29 (22.8) | 123 (21.2) | ||
| 15–46 | 131 (18.4) | 25 (19.7) | 109 (18.8) | ||
| 47+ | 133 (18.7) | 20 (15.7) | 115 (19.8) | ||
| Insertive anal sex in last 6 months2 | |||||
| Never | 532 (74.6) | 92 (72.4) | 434 (75.3) | 0.07 | |
| Always protected | 65 (9.1) | 18 (14.2) | 46 (8.0) | ||
| At least once unprotected | 116 (16.3) | 17 (13.4) | 96 (16.7) | ||
| Receptive anal sex in last 6 months2 | |||||
| Never | 62 (8.9) | 14 (11.1) | 46 (8.2) | <0.01 | |
| Always protected | 245 (35.1) | 59 (46.8) | 183 (32.5) | ||
| At least once unprotected | 392 (56.1) | 53 (42.1) | 334 (59.3) | ||
| Persons with a CAGE score < 23 | 242 (33.9) | 42 (37.5) | 197 (34.9) | 0.59 | |
| Persons with a CAGE score 2+ | 444 (62.3) | 70 (62.5) | 368 (65.1) | ||
| Years since sexual debut active (Mean ± StDev) | 15.3 ± 8.1 | 17.5 ± 7.2 | 14.9 ± 8.2 | <0.01 | |
The concepts of gender identity and sexuality are often conflated in Lima's LGBT community (i.e. transwomen when asked their gender identity may respond that they are gay)
Some of the HIV positives at baseline are known positives and report less unprotected anal sex
CAGE is the acronym for a 4 question screen for alcoholism, a 2+ CAGE score is indicative of a drinking problem
Participants who were HIV seropositive at baseline had had, on average, more years of sexual activity than those who were seronegative, 17.5 vs. 14.9 years (P < 0.01). HIV infection was not associated with region of origin, sexual identity, and education levels at baseline. Seronegative persons were more likely to engage in condomless receptive anal sex, 59.3%, than those seropositive, 42.1% (P < 0.01).
Table 2 displays HIV and STI prevalence at baseline as well as the incidence of these infections over the 18 month follow-up. Baseline HIV seroprevalence was 17.9%, 70% of whom did not know their infection status. HIV incidence during the study was 3.6 per 100 person-years. Incidence rates were calculated for each 9-month follow-up period, baseline to 9 months and 9 to 18 months. Rates remained similar across the two periods for most STIs, HIV showed a slight decrease from 5.0 cases per 100 person-years in the first 9 months of follow-up to 2.3 cases per 100 person years in the second 9 months of follow-up.
Table 2.
Prevalence and Incidence (per 100 person-years) of HIV, HSV-2, anal and pharyngeal gonorrhea, and anal and pharyngeal chlamydia among MSM and TW, Lima, Peru 2009–2012
| MSM/TGW | MSM | TGW | ||||
|---|---|---|---|---|---|---|
| STIs | Baseline prevalence (95% CI) |
Incidence per 100 person- years (95% CI) |
Baseline prevalence (95% CI) |
Incidence per 100 person- years (95% CI) |
Baseline prevalence (95% CI) |
Incidence per 100 person- years (95% CI) |
| HIV | 17.9% (15.2–21.0%) |
3.6 (3.5–3.7) |
18.4% (15.1–22.1%) |
4.0 (3.9–4.1) |
16.9% (12.1–22.7%) |
2.3 (2.1–2.5) |
| HSV-2 | 69.7% (66.1–73.1%) |
17.2 (16.8–17.7) |
65.0% (60.6–69.3%) |
17.9 (17.3–18.4) |
80.7% (74.6–85.8%) |
12.2 (10.1–14.2) |
| Anal gonorrhea | 9.6% (7.5–12.0%) |
10.1 (10.0–10.2) |
8.6% (6.2–11.6%) |
8.0 (7.8–8.1) |
12.3% (8.1–17.6%) |
15.3 (14.9–15.7) |
| Pharyngeal gonorrhea | 6.5% (4.8–8.5%) |
7.8 (7.8–7.9) |
5.4% (3.6–8.0%) |
7.6 (7.5–7.8) |
9.7% (6.0–14.6%) |
8.6(8.3–8.9) |
| Anal chlamydia | 19.0% (16.1–22.1%) |
16.3 (16.2–16.4) |
19.3% (15.7–23.2%) |
16.1 (15.9–16.3) |
20.2% (14.9–26.4%) |
16.9 (16.5–17.4) |
| Pharyngeal chlamydia | 4.8% (3.3–6.6%) |
3.4 (3.3–3.4) |
4.1% (2.5–6.4%) |
3.4 (3.3–3.5) |
6.8% (3.7–11.1%) |
3.4 (3.2–3.6) |
Bivariate and Multivariate Analysis
Notable positive bivariate associations were found between HIV incidence and a lower number of years of sexual activity (HR 0.95, P = 0.04) and greater incidence of a composite STI outcome including HSV-2, anal gonorrhea, and/or anal chlamydia infection (HR 4.93, P < 0.01). In the multivariate model, HIV incidence was associated with incident HSV-2, anal gonorrhea, or anal chlamydia (aHR 3.73, P < 0.01). Multivariate analysis found that HSV-2 incidence was positively associated with having 6 to 14 and 15 to 46 sex partners in the last year (aHR 3.21, P = 0.04; aHR 3.97, P = 0.02) (see Table 3).
Table 3.
Bivariate and multivariate associations with HIV and HSV-2 infection
| HIV | HSV-2 | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Variables | Crude HR |
95% CI | P | Adjusted HR |
95% CI | P | Crude HR |
95% CI | P | Adjusted HR |
95% CI | P | |
| Years since sexual debut | 0.95 | 0.91–1.00 | 0.04 | 0.96 | 0.92–1.01 | 0.13 | 1.05 | 1.00–1.10 | 0.05 | 1.06 | 1.00–1.12 | 0.06 | |
| Sexual/Gender identity | |||||||||||||
| Gay/Bisexual | Ref | Ref | Ref | Ref | |||||||||
| Travesti | 0.59 | 0.22–1.57 | 0.29 | 0.47 | 0.16–1.35 | 0.16 | 0.70 | 0.29–1.69 | 0.43 | 0.74 | 0.28–1.98 | 0.55 | |
| Use of condoms during paid sex | |||||||||||||
| No paid sex | Ref | Ref | |||||||||||
| Always | 0.95 | 0.41–2.20 | 0.90 | 0.74 | 0.34–1.59 | 0.44 | |||||||
| Sometimes/Never | 0.65 | 0.19–2.29 | 0.51 | 1.52 | 0.73–3.16 | 0.26 | |||||||
| Alcohol/drugs make condom use more difficult, last 3 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Sometimes/Always | 1.15 | 0.52–2.53 | 0.73 | 1.62 | 0.89–2.95 | 0.11 | |||||||
| Lack money, last 12 months | |||||||||||||
| Never | Ref | Ref | Ref | ||||||||||
| Rarely | 1.69 | 0.38–7.57 | 0.50 | 0.41 | 0.16–1.05 | 0.06 | 0.39 | 0.15–1.01 | 0.05 | ||||
| Sometimes | 1.32 | 0.29–5.91 | 0.72 | 0.64 | 0.27–1.52 | 0.31 | 0.56 | 0.24–1.34 | 0.19 | ||||
| Very often | 1.89 | 0.32–11.10 | 0.48 | 0.85 | 0.27–2.64 | 0.78 | 0.73 | 0.21–2.59 | 0.63 | ||||
| CAGE positive | 1.10 | 0.46–2.64 | 0.83 | 1.46 | 0.81–2.65 | 0.21 | |||||||
| No. of sex partners, last 12 months (quintiles) | |||||||||||||
| 0–2 | Ref | Ref | Ref | Ref | |||||||||
| 3–5 | 2.45 | 0.48–12.48 | 0.28 | 2.08 | 0.41–10.72 | 0.38 | 1.62 | 0.47–5.61 | 0.45 | 1.70 | 0.47–6.06 | 0.42 | |
| 6–14 | 3.98 | 0.85–18.59 | 0.08 | 3.26 | 0.67–15.84 | 0.14 | 3.15 | 1.01–9.82 | 0.05 | 3.21 | 1.03–9.97 | 0.04 | |
| 15–46 | 3.14 | 0.64–15.30 | 0.16 | 1.99 | 0.37–10.61 | 0.42 | 3.65 | 1.15–11.65 | 0.03 | 3.97 | 1.27–12.38 | 0.02 | |
| 47+ | 1.95 | 0.36–10.71 | 0.44 | 1.69 | 0.30–9.68 | 0.55 | 1.48 | 0.36–6.06 | 0.59 | 1.67 | 0.39–7.14 | 0.49 | |
| Insertive anal sex, last 6 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Always protected | 0.88 | 0.21–3.75 | 0.87 | 0.69 | 0.25–1.87 | 0.47 | |||||||
| At least once unprotected | 0.67 | 0.16–2.85 | 0.59 | 0.94 | 0.39–2.27 | 0.89 | |||||||
| Receptive anal sex, last 6 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Always protected | 1.37 | 0.39–4.85 | 0.63 | 1.14 | 0.50–2.55 | 0.76 | |||||||
| At least once unprotected | 0.62 | 0.15–2.55 | 0.51 | 0.70 | 0.28–1.72 | 0.43 | |||||||
| Has a stable partner | |||||||||||||
| No | Ref | Ref | |||||||||||
| Yes | 0.59 | 0.27–1.28 | 0.18 | 1.34 | 0.72–2.50 | 0.35 | |||||||
| Incident HSV-2/Anal GC/Anal CT | |||||||||||||
| No | Ref | Ref | |||||||||||
| Yes | 4.93 | 2.27–19.73 | <0.01 | 3.73 | 1.59–8.74 | <0.01 | |||||||
In multivariate analysis, incidence of anal gonorrhea was associated with a lower number of years of sexual activity (aHR 0.94, P < 0.01), and with TW identity (aHR 1.68, P = 0.04), engaging in condomless compensated sex (aHR 2.36, P = 0.01), having 15 to 46 sex partners (aHR 3.31, P = 0.01) and engaging in receptive anal intercourse with condoms (aHR 2.76, P = 0.05). Pharyngeal gonorrhea was negatively associated with number of years of sexual activity (aHR 0.95, P < 0.01) (see Table 4).
Table 4.
Bivariate and multivariate associations with anal and pharyngeal gonorrhea infection
| Anal Gonorrhea | Pharyngeal Gonorrhea | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Variables | Crude HR |
95% CI | P | Adjusted HR |
95% CI | P | Crude HR |
95% CI | P | Adjusted HR |
95% CI | P | |
| Years since sexual debut | 0.95 | 0.92–0.97 | <0.01 | 0.94 | 0.91–0.97 | <0.01 | 0.95 | 0.93–0.98 | <0.01 | 0.95 | 0.92–0.97 | <0.01 | |
| Sexual/Gender Identity | |||||||||||||
| Gay/Bisexual | Ref | Ref | Ref | Ref | |||||||||
| Travesti | 1.98 | 1.27–3.08 | <0.01 | 1.68 | 1.04–2.73 | 0.04 | 1.19 | 0.74–1.93 | 0.47 | 1.23 | 0.76–2.00 | 0.39 | |
| Use of condoms with paid sex | |||||||||||||
| No paid sex | Ref | Ref | Ref | ||||||||||
| Always | 1.89 | 1.16–3.01 | 0.01 | 1.43 | 0.78–2.65 | 0.25 | 1.27 | 0.76–2.12 | 0.37 | ||||
| Sometimes/Never | 2.64 | 1.54–4.52 | <0.01 | 2.36 | 1.26–4.42 | 0.01 | 1.07 | 0.54–2.11 | 0.84 | ||||
| Alcohol/drugs make condom use more difficult in last 3 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Sometimes/Always | 1.36 | 0.88–2.08 | 0.16 | 1.19 | 0.73–1.92 | 0.48 | |||||||
| Lack money in last 12 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Rarely | 1.22 | 0.56–2.68 | 0.62 | 2.07 | 0.74–5.80 | 0.17 | |||||||
| Sometimes | 1.23 | 0.57–2.65 | 0.60 | 1.89 | 0.67–5.29 | 0.23 | |||||||
| Very often | 1.45 | 0.58–3.67 | 0.43 | 1.76 | 0.50–6.23 | 0.38 | |||||||
| CAGE | 1.07 | 0.67–1.69 | 0.79 | 0.83 | 0.50–1.37 | 0.46 | |||||||
| # of sex partners in 1 year (quintiles) | |||||||||||||
| 0–2 | Ref | Ref | |||||||||||
| 3–5 | 3.80 | 1.46–9.90 | <0.01 | 2.34 | 0.98–5.60 | 0.06 | 0.87 | 0.41–1.85 | 0.72 | ||||
| 6–14 | 3.52 | 1.27–9.75 | 0.02 | 1.89 | 0.74–4.84 | 0.18 | 0.52 | 0.21–1.29 | 0.16 | ||||
| 15–46 | 6.95 | 2.67–18.10 | <0.01 | 3.31 | 1.34–8.18 | 0.01 | 1.36 | 0.66–2.78 | 0.40 | ||||
| 47+ | 4.17 | 1.55–11.22 | <0.01 | 1.74 | 0.67–4.54 | 0.26 | 1.26 | 0.60–2.66 | 0.54 | ||||
| Insertive anal sex in last 6 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Always protected | 1.05 | 0.48–2.29 | 0.89 | 0.62 | 0.25–1.53 | 0.29 | |||||||
| At least once unprotected | 0.67 | 0.29–1.55 | 0.35 | 0.91 | 0.42–1.97 | 0.82 | |||||||
| Receptive anal sex in last 6 months | |||||||||||||
| Never | Ref | Ref | Ref | ||||||||||
| Always protected | 1.90 | 0.90–4.02 | 0.09 | 2.76 | 1.02–7.51 | 0.05 | 1.83 | 0.78–4.31 | 0.16 | ||||
| At least once unprotected | 1.65 | 0.73–3.74 | 0.23 | 2.07 | 0.72–5.89 | 0.18 | 1.48 | 0.63–3.49 | 0.37 | ||||
| Do you have a stable partner? | |||||||||||||
| No | Ref | Ref | |||||||||||
| Yes | 1.20 | 0.77–1.88 | 0.42 | 1.03 | 0.65–1.63 | 0.90 | |||||||
Anal chlamydia was also associated with a lower number of years of sexual activity (aHR 0.96, P < 0.01), with engaging in condomless compensated sex (aHR 1.61, P = 0.05), and in anal sex with and without condoms (aHR 2.25, P= 0.04; aHR 2.63, P = 0.02) (see Table 5).
Table 5.
Bivariate and multivariate associations with anal and pharyngeal chlamydia infection
| Anal Chlamydia | Pharyngeal Chlamydia | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Variables | HR | 95% CI | P | aHR | 95% CI | P | HR | 95% CI | P | aHR | 95% CI | P | |
| Years since sexual debut | 0.96 | 0.94–0.98 | <0.01 | 0.96 | 0.93–0.98 | <0.01 | 0.96 | 0.91–1.00 | 0.06 | 0.96 | 0.91–1.00 | 0.07 | |
| Sexual/Gender Identity | |||||||||||||
| Gay/Bisexual | Ref | Ref | Ref | Ref | |||||||||
| Travesti | 1.08 | 0.76–1.54 | 0.65 | 1.08 | 0.75–1.57 | 0.67 | 1.03 | 0.48–2.18 | 0.94 | 1.05 | 0.49–2.25 | 0.89 | |
| Use of condoms with paid sex | |||||||||||||
| No paid sex | Ref | Ref | Ref | ||||||||||
| Always | 1.29 | 0.91–1.85 | 0.16 | 1.15 | 0.76–1.75 | 0.50 | 0.58 | 0.25–1.35 | 0.21 | ||||
| Sometimes/Never | 1.50 | 0.97–2.33 | 0.07 | 1.61 | 1.01–2.57 | 0.05 | 0.87 | 0.32–2.37 | 0.79 | ||||
| Alcohol/drugs make condom use more difficult in last 3 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Sometimes/Always | 1.21 | 0.85–1.71 | 0.29 | 1.14 | 0.53–2.43 | 0.73 | |||||||
| Lack money in last 12 months | |||||||||||||
| Never | Ref | Ref | |||||||||||
| Rarely | 1.07 | 0.59–1.94 | 0.82 | 0.59 | 0.15–2.32 | 0.45 | |||||||
| Sometimes | 0.87 | 0.48–1.59 | 0.67 | 0.84 | 0.23–3.12 | 0.80 | |||||||
| Very often | 0.85 | 0.37–1.92 | 0.69 | 1.47 | 0.33–6.45 | 0.61 | |||||||
| CAGE | 1.37 | 0.91–2.07 | 0.13 | 1.79 | 0.75–4.29 | 0.19 | |||||||
| # of sex partners in 1 year (quintiles) | |||||||||||||
| 0–2 | Ref | Ref | |||||||||||
| 3–5 | 1.25 | 0.71–2.21 | 0.44 | 1.25 | 0.73–2.14 | 0.42 | 2.32 | 0.71–7.59 | 0.17 | ||||
| 6–14 | 1.29 | 0.73–2.32 | 0.38 | 1.12 | 0.63–1.99 | 0.71 | 1.33 | 0.36–4.96 | 0.67 | ||||
| 15–46 | 1.86 | 1.07–3.24 | 0.03 | 1.22 | 0.69–2.18 | 0.49 | 1.20 | 0.30–4.81 | 0.80 | ||||
| 47+ | 1.43 | 0.82–2.49 | 0.20 | 1.03 | 0.57–1.85 | 0.93 | 2.24 | 0.67–7.45 | 0.19 | ||||
| Insertive anal sex in last 6 months | |||||||||||||
| Never | Ref | Ref | Ref | ||||||||||
| Always protected | 0.52 | 0.25–1.10 | 0.09 | 0.56 | 0.27–1.16 | 0.12 | 0.26 | 0.04–1.89 | 0.18 | ||||
| At least once unprotected | 1.20 | 0.74–1.95 | 0.47 | 1.31 | 0.76–2.24 | 0.33 | 0.55 | 0.13–2.32 | 0.41 | ||||
| Receptive anal sex in last 6 months | |||||||||||||
| Never | Ref | Ref | Ref | ||||||||||
| Always protected | 1.63 | 0.90–2.94 | 0.11 | 2.25 | 1.04–4.86 | 0.04 | 1.25 | 0.35–4.40 | 0.73 | ||||
| At least once unprotected | 2.24 | 1.19–4.19 | 0.01 | 2.63 | 1.20–5.79 | 0.02 | 1.71 | 0.53–5.53 | 0.37 | ||||
| Do you have a stable partner? | |||||||||||||
| No | Ref | Ref | |||||||||||
| Yes | 0.96 | 0.71–1.29 | 0.77 | 0.93 | 0.46–1.87 | 0.84 | |||||||
aHR = Adjusted HR, adjusted for all variables listed in the model.
Discussion
To date the Comunidades Positivas study offers the longest follow-up, 18 months, for HIV/STI incidence in Lima’s MSM and TW population. Our findings of the high baseline prevalence and incidence rates of the STIs measured are supported by previous high prevalence findings in other studies4,5,23,24. This provides further validation of Lima’s lower-income MSM and TW as populations at the heart of a concentrated and active HIV epidemic and at high-risk for other STIs. Compared to previously reported incidence from 1998–200019 (3.5 per 100 person-years for HIV, and 10.4 per 100 person-years for HSV-2) and HIV incidence from 2007–2009 (3.9 per 100 person-years)6, our HIV rate was similar at 3.6 per 100 person-years, but our HSV-2 incidence was higher at 17.2 per 100 person-years. The similarity in HIV incidence with the previous cohorts shows the sustained risk of HIV infection among MSM/TW in Peru despite the availability of treatment, in line with data from this same cohort on low serostatus awareness among HIV positive MSM/TW, leading to late diagnosis and initiation of treatment25. Gonorrhea and chlamydia incidence rates were similar to the prevalence at baseline, as expected for treatable infections.
Incident HSV-2, anal gonorrhea, or anal chlamydia infection were risk factors for incident HIV infection. It is likely that factors which placed study participants at risk for STI were common between HIV, HSV-2, and anal gonorrhea and chlamydia. We excluded pharyngeal gonorrhea and chlamydia from the analysis because these sexual practices have a different route of infection and are likely associated with different risk factors as well. Although it has proven difficult to capture all potential behavioral – and, even less so, structural, risk factors for incident HIV/STI, by focusing on the proximal outcome of incident infection we demonstrate an association between infections by different STIs. This has important implications for HIV interventions because co-infection synergism or HIV infection capacitation by an existing STI may increase the likelihood of HIV infection. Pre- or co-existing STI can lead to lesions in the skin, allowing the HIV virus to more easily penetrate the skin barrier, or can cause inflammation facilitating HIV contact with immune cells.26–28
Interestingly, when comparing population statistics at baseline, the HIV-seronegative MSM population was more likely to engage in unprotected receptive anal sex than the HIV positive population. We may hypothesize that some HIV positive men who were aware of their status were taking greater precautions to prevent HIV transmission to their sex partners. However, not all participants had performed an HIV test within the last year of baseline; as such they would not have the information necessary to make decisions based on their serostatus. It is then likely that other dynamics, not directly expressed in individual behaviors, are contributing to this paradoxical phenomenon. For example, as related to sexual networks of the individual; some respondents may be in more at-risk sex networks and may therefore be more likely to contract HIV even with safer sex practices.29 Those who engage in riskier sex, such as sex workers or those having sex with sex workers, would be more likely to use condoms because of the perceived risk of this activity. Likewise they may even report condom use more frequently due to social desirability bias.
A higher number of years of sexual activity was the only protective factor for anal gonorrhea and chlamydia. We attribute this to the transient nature of bacterial infections, which can be cured with antibiotics. Experience with prior infections or general knowledge can allow individuals to recognize the infections or take precautions to prevent infection. Significant risk factors for anal gonorrhea included TW identity, condomless compensated sex, having a greater number of sex partners, and engaging in receptive anal sex with condoms. STI prevalence is typically elevated within the TW population compared to the already high levels among homosexual and bisexual men. As a result of the aforementioned personal, socioeconomic, and structural conditions it is reasonable that TW identity associates as a risk factor for anal gonorrhea. Any form of sexual activity with or without protection can increase the likelihood of STI acquisition, and the risk likely rises when engaging in compensated sex with clients.30 If a client engages in compensated unprotected sex with a sex worker, it is likely that they have engaged in similar behaviors with other gay or transgender individuals, including sex workers.
Protected and unprotected anal sex were risk factors for anal chlamydia. As it would be expected, engaging in anal sex with or without protection increases the likelihood of contracting an anal STI. As with anal gonorrhea, not using condoms during compensated sex would increase the likelihood of exposure to chlamydia.
There were several limitations to this study. The 9 month gaps between follow-ups of study participants present challenges in terms of capturing curable STIs which may have been contracted but were subsequently treated within this period. Additionally, measurement of risk behaviors focused on a recall window of 6 months, which does not cover the entire 9-month time between visits; this could miss risk behavior associated with STI incidence during the follow-up period. However, the shorter recall window was included to achieve more accurate recollection among the study participants. The 80% retention of study participants over the course of 18 months of the study may have produced follow-up dropout bias; however the only statistically significant difference between those who remained until the second follow-up and those who did not was related to the number of years since sexual debut. Dropouts were an average of 13.6 years from their sexual debut, compared to 15.8 years among those who remained in the study (P < 0.01). Although this study includes an ethnographically identified population and not a random sample, study ethnographers identified and recruited all eligible MSM/TW from the included barrios. The conclusions of this study should be interpreted as representing MSM/TW residents of low-income barrios in the province of Lima. An additional limitation is the multiple comparisons conducted; we have tried to focus our conclusions only on strong associations and not on those with nominal significance.
Conclusions
The high prevalence and incidence rates of HIV/STIs measured in the Comunidades Positivas study highlight the high risk among low income MSM and TW in Lima, their role as an important target population at the heart of a concentrated HIV epidemic, and the need for effective, community-relevant combination interventions. As STI co-infection is a notable risk factor for HIV infection, justifying HIV screening, prevention, and treatment interventions which focus on other STIs such as HSV-2, gonorrhea, and chlamydia. Hence, interventions will need to focus on prevention and treatment of HIV and these STIs in MSM and TW.
The lack of association of behavioral factors with incident HIV infection signals that our surveillance tools and measures may have been unable to capture dynamics that, for example, may affect the probability of having sex with infected partners. Beyond instructing participants to test frequently and always use condoms, HIV interventions need to further broaden the intervention perspective from the individual to the community and focus on screening and treating at-risk sex networks. More accurate surveillance tools need to be utilized; for example, not only focus on the last three partners, but rather capture the risk factors for those who have a substantially larger sex network.
Our study outlines the factors associated with incident infection; these should be taken into account when designing and implementing future interventions. Through continued focus on combination prevention and treatment of HIV infection, informed by up-to-date population surveillance and significant risk factors for infection, we can improve HIV/STI prevention interventions in Peru’s MSM and TW populations.
Acknowledgments
The first author would like to thank the Peru Fulbright Commission for the support provided to carry out his research. This study was funded by National Institutes of Health NIH NIMH R01 MH078752 (PI Coates). Research reported in this publication was supported by NIH Research Training Grant # R25TW009343 funded by the Fogarty International Center, Office of Behavioral and Social Sciences Research, Office of Research on Women’s Health, Office of AIDS Research, National Institute of Mental Health, and National Institute on Drug Abuse, as well as the University of California Global Health Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the University of California Global Health Institute.
Sources of Funding: This study was conducted with NIMH NIH funding R01MH078752. RC also received support from Fulbright and KAK received support from T32MH080634, GloCal R25TW009343, and K01MH098698.
Footnotes
Conflicts of Interest: None
References
- 1.Beyrer C, Baral SD, Walker D, Wirtz AL, Johns B, Sifakis F. The expanding epidemics of HIV type 1 among men who have sex with men in low-and middle-income countries: diversity and consistency. Epidemiologic reviews. 2010;32(1):137–151. doi: 10.1093/epirev/mxq011. [DOI] [PubMed] [Google Scholar]
- 2.Sanchez J, Lama JR, Kusunoki L, et al. HIV-1, sexually transmitted infections, and sexual behavior trends among men who have sex with men in Lima, Peru. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2007;44(5):578–585. doi: 10.1097/QAI.0b013e318033ff82. [DOI] [PubMed] [Google Scholar]
- 3.Perez-Brumer AG, Konda KA, Salvatierra HJ, et al. Prevalence of HIV, STIs, and Risk Behaviors in a Cross-Sectional Community-and Clinic-Based Sample of Men Who Have Sex with Men (MSM) in Lima, Peru. PloS one. 2013;8(4):e59072. doi: 10.1371/journal.pone.0059072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Snowden JM, Konda KA, Leon SR, et al. Recent syphilis infection prevalence and risk factors among male low-income populations in coastal Peruvian cities. Sexually transmitted diseases. 2010;37(2):75. doi: 10.1097/OLQ.0b013e3181c03434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Clark JL, Caceres CF, Lescano AG, et al. Prevalence of same-sex sexual behavior and associated characteristics among low-income urban males in Peru. PloS one. 2007;2(8):e778. doi: 10.1371/journal.pone.0000778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Grant RM, Anderson PL, McMahan V, et al. Uptake of pre-exposure prophylaxis, sexual practices, and HIV incidence in men and transgender women who have sex with men: a cohort study. The Lancet. Infectious diseases. 2014;14(9):820–829. doi: 10.1016/S1473-3099(14)70847-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Silva-Santisteban A, Raymond HF, Salazar X, et al. Understanding the HIV/AIDS epidemic in transgender women of Lima, Peru: results from a sero-epidemiologic study using respondent driven sampling. AIDS and Behavior. 2012;16(4):872–881. doi: 10.1007/s10461-011-0053-5. [DOI] [PubMed] [Google Scholar]
- 8.Clark JL, Lescano AG, Konda KA, et al. Syndromic management and STI control in urban Peru. PloS one. 2009;4(9):e7201. doi: 10.1371/journal.pone.0007201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cárcamo CP, Campos PE, García PJ, Hughes JP, Garnett GP, Holmes KK. Prevalences of sexually transmitted infections in young adults and female sex workers in Peru: a national population-based survey. The Lancet Infectious Diseases. 2012 doi: 10.1016/S1473-3099(12)70144-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Baral S, Sifakis F, Cleghorn F, Beyrer C. 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(12):e339. doi: 10.1371/journal.pmed.0040339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jaffe HW, Valdiserri RO, De Cock KM. The reemerging HIV/AIDS epidemic in men who have sex with men. JAMA: the journal of the American Medical Association. 2007;298(20):2412–2414. doi: 10.1001/jama.298.20.2412. [DOI] [PubMed] [Google Scholar]
- 12.Ugarte Ubillúz O, Arce Rodríguez M. Informe nacional sobre los progresos realizados en la aplicación del UNGASS. Período: enero 2008-diciembre 2009. Perú. 2010 [Google Scholar]
- 13.UNAIDS. AIDS epidemic update: December 2009. WHO Regional Office Europe. 2009
- 14.Koblin BA, Husnik MJ, Colfax G, et al. Risk factors for HIV infection among men who have sex with men. Aids. 2006;20(5):731–739. doi: 10.1097/01.aids.0000216374.61442.55. [DOI] [PubMed] [Google Scholar]
- 15.Bautista C, Sanchez J, Montano S, et al. Seroprevalence of and risk factors for HIV-1 infection among South American men who have sex with men. Sexually Transmitted Infections. 2004;80(6):498–504. doi: 10.1136/sti.2004.013094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Salvatierra HJ, Segura ER, Konda K, et al. HIV/STD risk behaviors and social discrimination in MSM gay, MSM no gay and travestis in Lima, Peru; Paper presented at: XVII International AIDS Conference; Mexico City, Mexico: 2008. [Google Scholar]
- 17.Campuzano G, et al. Building Identity While Managing Disadvantage: Peruvian Transgender Issue. Institute of Development Studies. 2008 [Google Scholar]
- 18.Pollock L, Konda K, Klausner J, et al. The impact of homophobia among MSM and transgender women in Lima, Peru; XVIII International AIDS Conference; Vienna, Austria: 2010. [Google Scholar]
- 19.Sanchez J, Lama JR, Peinado J, et al. High HIV and ulcerative sexually transmitted infection incidence estimates among men who have sex with men in Peru: awaiting for an effective preventive intervention. Journal of acquired immune deficiency syndromes (1999) 2009;51(Suppl 1):S47. doi: 10.1097/QAI.0b013e3181a2671d. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Alarcón JO, Pun M, Gutiérrez C, et al. Estimation and analysis of HIV incidence in the adult population in Peru: results of application of the MoT mathematical model. Revista Peruana de Medicina Experimental y Salud Pública. 2012;29(4):452–460. doi: 10.1590/s1726-46342012000400006. [DOI] [PubMed] [Google Scholar]
- 21.Silva-Santisteban A, Konda K, Leon S, et al. P4. 122 Effectiveness of Communidades Positivas: A Randomised Community-Level Combination HIV Prevention Intervention For Men Who Have Sex with Men in Peru. Sexually Transmitted Infections. 2013;89(Suppl 1):A325–A326. [Google Scholar]
- 22.Deiss RG, Clark JL, Konda KA, et al. Problem drinking is associated with increased prevalence of sexual risk behaviors among men who have sex with men (MSM) in Lima, Peru. Drug and alcohol dependence. 2013;132(1):134–139. doi: 10.1016/j.drugalcdep.2013.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Tabet S, Sanchez J, Lama J, et al. HIV, syphilis and heterosexual bridging among Peruvian men who have sex with men. Aids. 2002;16(9):1271–1277. doi: 10.1097/00002030-200206140-00010. [DOI] [PubMed] [Google Scholar]
- 24.Cáceres CF, Konda KA, Salazar X, et al. New populations at high risk of HIV/STIs in low-income, urban coastal Peru. AIDS and Behavior. 2008;12(4):544–551. doi: 10.1007/s10461-007-9348-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Caceres CF, Konda K, Silva-Santisteban A, et al. The Continuum of HIV Care in Peru - Where Are We Now? Key Lessons from an Estimation in the Context of Very Limited Data. AIDS Res. Hum. Retroviruses. 2014;30(Suppl 1) [Google Scholar]
- 26.Nusbaum MR, Wallace RR, Slatt LM, Kondrad EC. Sexually transmitted infections and increased risk of co-infection with human immunodeficiency virus. JAOA: Journal of the American Osteopathic Association. 2004;104(12):527–535. [PubMed] [Google Scholar]
- 27.Kaul R, Nagelkerke NJ, Kimani J, et al. Prevalent herpes simplex virus type 2 infection is associated with altered vaginal flora and an increased susceptibility to multiple sexually transmitted infections. Journal of Infectious Diseases. 2007;196(11):1692–1697. doi: 10.1086/522006. [DOI] [PubMed] [Google Scholar]
- 28.Hayes R, Watson-Jones D, Celum C, van de Wijgert J, Wasserheit J. Treatment of sexually transmitted infections for HIV prevention: end of the road or new beginning? AIDS (London, England) 2010;24(0 4) doi: 10.1097/01.aids.0000390704.35642.47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Adimora AA, Schoenbach VJ. The New Public Health and STD/HIV Prevention. Springer; 2013. Social determinants of sexual networks, partnership formation, and sexually transmitted infections; pp. 13–31. [Google Scholar]
- 30.Cambou M, Perez-Brumer A, Segura E, et al. O07. 5 Unprotected Sex Among High-Risk Partners: Associations Between Relational Characteristics of Last Sexual Partner and Unprotected Anal Intercourse (UAI) Among Men Who Have Sex with Men (MSM) and Transgender Women (TGW) in Lima, Peru. Sexually Transmitted Infections. 2013;89(Suppl 1):A39–A40. [Google Scholar]