Abstract
In the US, Latino MSM are disproportionately affected by HIV, yet there is a paucity of data for this risk group. To this end, we examined data on Latino and non-Latino white MSM who participated across six cities in a 2-year randomized behavioral intervention study—Project EXPLORE. At baseline, Latinos reported significantly more serodiscordant unprotected anal intercourse (SDUA) than non-Latinos. Longitudinal predictors of SDUA included marijuana, poppers, amphetamines and heavy drinking, as well as lower self-efficacy, poorer communication skills, weaker safe-sex norms and more enjoyment of risky sex. For HIV infection, Latinos had significantly higher seroconversion rate over follow-up than non-Latinos. Longitudinal predictors of seroconversion among Latinos included poppers and SDUA. Intervention effects did not significantly differ between Latino and non-Latinos. Findings support HIV intervention work with Latino MSM that includes skills training/counseling to address attitudes about safe sex and impact of substance use on HIV-risk behavior and acquisition.
Keywords: Hispanic/Latino, MSM, HIV prevention, AIDS/HIV, High-risk sexual behavior, Substance use
Introduction
Latinos, comprising approximately 15% of the population, are the largest minority group in the United States [1]. This ethnic group, however, is disproportionately impacted by HIV [2]. In 2007, Latinos accounted for 18% of those newly infected with HIV and 19% of people living with AIDS [2]. Amongst Latino men in the U.S. infected with HIV, male-to-male sexual contact is the most commonly reported transmission route: among Latino men diagnosed in 2007, 7% reported heterosexual contact risk, 11% injection drug use, and strikingly 60% of infections occurred in men reporting male-to-male sexual contact [2].
There is a paucity of data examining HIV prevention within Latinos in general and within Latino MSM in particular [3-6]. Variables identified as being associated with high-risk unprotected sexual behavior among Latino MSM include higher ethnic identification, age when sexual behavior was initiated (such that older sexual debut is associated with higher risk), sexual attraction, desire for acceptance from family or sexual partners, cultural messages (e.g., machismo), homophobic-related discrimination, drug and alcohol use, drug and alcohol use by sex partner, intentions and self-efficacy for safer sex, and poor communication about condom use [3-10]. Although informative, these studies are generally limited by cross sectional designs, geographic area, and lack of objective measurement of HIV status. Such limitations also exist in studies that compare Latino MSM to non-Latino white MSM [11, 12], including, due to small sample sizes, difficulties in having sufficient power to examine between-group differences. Although a limited amount of research has been done with this issue, much more is needed.
Data from the EXPLORE study, consisting of a sample of 4295 men across six metropolitan areas (i.e., Boston, Chicago, Denver, New York, San Francisco, Seattle), can begin to address some of these limitations. Accordingly, the present study examined: (1) differences between Latino MSM and non-Latino white MSM at baseline on factors believed to impact HIV risk—substance use, psychosocial factors and high-risk sexual behavior (i.e., SDUA), (2) predictors associated with SDUA within Latino MSM alone at baseline and over study follow-up, (3) comparison of HIV-seroconversion over study follow-up between Latino MSM and non-Latino white MSM; (4) predictors of HIV-seroconversion over study follow-up within Latino MSM alone, and (5) differential effect of the EXPLORE intervention for Latinos versus non-Latino white MSM [13]. The EXPLORE study has previously published findings of rates of HIV risk and HIV prevention intervention effects for a multi-site national MSM sample as a whole [14]. However, the current paper uniquely addresses both baseline (i.e., demographic, psychosocial, and HIV-risk behavior) and longitudinal data (i.e., SDUA; HIV infection) specific to Latino MSM who participated in the EXPLORE study.
Methods
Participants
The methods for the EXPLORE study have previously been described in greater detail [14-17]. Recruitment for the study occurred from January 1999 through February 2001 in six U.S. cities: Boston, Chicago, Denver, New York, San Francisco, and Seattle. Inclusion criteria for the EXPLORE study were men who were uninfected with HIV, were 16 years or older (although none of the sample was under age 18 at enrollment for the present analysis), had anal sex with another man during the past year, and had not been involved in a mutually monogamous relationship in the past 2 years with a male partner who was HIV-uninfected. Participants were excluded if they were not fluent in English. Men were randomized to receive a behavioral intervention versus standard risk reduction counseling. The experimental intervention, described in detail by Chesney et al. [16], consisted of 10 core counseling modules delivered within one-on-one counseling sessions.
Measures
Participants were asked to enroll in the study approximately 2 weeks after an initial HIV screening. After randomization, participants were evaluated every 6 months for up to 48 months (depending on the time for which a participant enrolled) for a behavioral assessment, HIV pre-test counseling, and blood specimen collection. Participation continued until the end of the study in July 2003. Data was collected using audio computer-assisted self-interviewing technology (ACASI). Additional details on recruitment, data collection, measures, and procedures are described elsewhere [14-17]. The following variables were included for the current study: demographic variables, beliefs about safer sex, perceived enjoyment of risky sex, substance use, symptoms of depression, history of childhood sexual abuse, sexual behavior, randomization status and HIV testing data over the course of the study.
Demographic Variables
Demographic variables included in this study included age (coded as less than 25, 26–30, 31–35, 36–40, greater than 40), education (less than high school, high school or GED, some college, after college), household annual income (less than $12000, $12000–$29999, $30000–$59999, more than $60000), employment status (full time, part time, unemployed, other), geographic location (Boston, Chicago, Denver, New York, San Francisco, Seattle) and ethnicity (Latino, non-Latino white).
Attitudes Associated with High-risk Sexual Behavior
Four areas related to high-risk sexual behavior were examined: self-efficacy for adopting safer sexual behaviors, poor communication skills regarding safer sexual practices, social norms regarding safe sex, and perceived enjoyment of risk-related sexual behaviors [14, 16]. Of these, three (i.e., self-efficacy, communication skills, social norms) are scales consisting of 9, 6, and 5 items, respectively, that were rated on a 6-point scale. Factor-based scales were created by summing scores for each item and rescaling scores to 0–100. As in prior studies with this cohort, scores were dichotomized at the mid-point of the range (i.e., 50) and used to create variables that reflected disagreement with safe sex behaviors, wherein “1” would correspond to lower self-efficacy, poorer communication skills, and weaker perception that social norms favor safe sex.
In addition, participants were asked questions related to their perceived enjoyment of risk-related sexual behavior. Two questions assessed enjoyment of unprotected insertive and receptive anal sex. Responses were recorded on a 4-point scale ranging from “dislike very much” to “enjoy very much” that were dichotomized for analysis by combining responses that indicated enjoyment (i.e., enjoy somewhat; enjoy very much) and dislike (i.e., dislike somewhat, dislike very much) of the sexual behavior.
Substance Use
Drug use was assessed for the prior 6 months and collected on a 5-point scale (never; less than 1 week; 1–2 days a week; 3–6 days a week; and every day), with responses categorized as “never used” or “ever used” [14, 15]. The following non-prescription drug categories were assessed: (1) marijuana or hashish; (2) poppers or inhaled nitrates (including ampules); (3) crack or rock cocaine (smoked) or cocaine (snorted or sniffed); (4) amphetamines, such as speed, crystal or crank (swallowed, snorted or smoked); (5) hallucinogens, such as PCP, Special K, angel dust, acid, LSD, mushrooms, or Ecstasy; and (6) injection drugs. Alcohol use was assessed for the prior 6 months and categorized as “never” (no drinks in past month), “light” (3 or fewer drinks/day though drinking less than 3 days/week), “moderate” (4–5 drinks/day and drinking less than 3 days/week; or 1–5 drinks/day and drinking 3–6 days/week; or 1–3 drinks/day and drinking 7 days/week) and “heavy” (more than 4 drinks/day and drinking 7 days/week; or drinking more than 6 drinks/day).
Depression Symptoms
Symptoms of depression were measured using a shortened version of the Center for Epidemiologic Studies Depression scale (CES-D)/National Institute of Mental Health scale (correlation with full CES-D = 0.92) [18]. This measure asked respondents to rate how often during the past week [never/rarely (scale score = 1), sometimes (1–2 days; scale score = 2), often (3–4 days; scale score = 3), mostly or always (5–7 days; scale score = 4)] they experienced the following: felt like they could not shake off the blues even with help from their family or friends, had trouble keeping their minds on what they were doing, felt that everything they did was an effort, had trouble sleeping, felt lonely, felt sad or felt like they just “couldn’t get going.” This seven-item scale has been validated [19]. However, because a cut-off point has not been established for this shortened version, in the current study—as was done in another, similarly structured EXPLORE analysis from our group [20]—the additive score was constructed, divided into quartiles, and dichotomized as “having depressive symptoms” if a participant’s score was in the top three quartiles (e.g., a score of 12.25 or higher) and “not having depressive symptoms” if the score was in the lowest quartile.
History of Childhood Sexual Abuse
Participants were asked the following: (1) “before you were 13 years old, did you have any sexual experience with someone who was five years or more older than you?”; and (2) “between the time you turned 13 and your 17th birthday, did you have any sexual experiences with someone who was ten years or more older than you?” Responses were used to create a dichotomous variable so that if a participant positively endorsed either of these two questions he was considered to have a history of childhood sexual abuse [14, 16, 20].
High-risk Sexual Behavior
Participants were assessed every 6 months for the previous 6 months using standard questions that asked about insertive and receptive anal sex, with and without condoms, and by partner status (HIV positive, HIV negative, HIV unknown status). In this study we use any serodiscordant unprotected anal (SDUA) sex at baseline and over study follow-up, defined as self-reported insertive or receptive anal sex without a condom with a partner whose HIV status was either known to be positive or whose HIV status was unknown.
HIV Sero-status
Blood samples were collected every 6 months and assessed for HIV sero-status using an enzyme-linked immunosorbent assay. Samples found to be reactive were retested and those repeatedly found reactive confirmed through Western blot assay or immunofluorescence assay.
Data Analysis
SAS statistical software was used to conduct these analyses. Data were collected at the baseline visit and then every 6 months for a period of up to 48 months.
Comparison of Demographics and Risk Analysis at Baseline
Latino and non-Latino white MSM were compared at baseline on demographic variables and factors believed to impact HIV risk—substance use, psychosocial factors, and high-risk sexual behavior (i.e., SDUA). Differences in distribution of demographic variables were assessed using the Chi-square test. Association of ethnicity (i.e., Latino or non-Latino white) with risk factors (i.e., substance use, psychosocial factors, SDUA) was assessed using logistic regression. Models were adjusted for significant demographic factors: age, education and study enrollment site.
Predictors of SDUA Among LATINO MSM at Baseline and Follow-up
Logistic regression procedures were used to assess predictors of SDUA sexual behavior within Latino MSM at baseline and during follow-up. Baseline predictors that were statistically significant at the P < 0.05 alpha level, after adjusting for study site, were entered in a multivariable model. These analyses were replicated using longitudinal follow-up data. Using GEE models [21, 22], univariate longitudinal predictors of SDUA sexual behavior that were statistically significant at the P < 0.05 alpha level, after adjusting for study site and randomization arm, were entered in a multivariable model. GEE equations account for the within subject correlations of the repeated measures across time.
Comparison of HIV-Seroconversion RATES at Follow-up
HIV infection rates and confidence intervals amongst Latino and non-Latino white MSM were computed using Poisson regression. Cox proportional hazards was used to compare infection rates between the two groups, adjusted for study site and randomization arm.
Predictors of HIV Infection Among Latino MSM at Follow-up
Among the Latino MSM subset, Cox proportional hazard regression modeling [23] was adopted to model the influence of baseline predictors on HIV seroincidence over the course of study follow-up. Statistically significant predictors of HIV infection at the P < 0.05 alpha level were entered in a final multivariable Cox proportional hazard regression model and variable elimination using backward selection was used to select a final model. All models were adjusted for study site and randomization arm.
Comparison of EXPLORE Intervention Effect
To explore whether Latino status was an effect modifier of the EXPLORE intervention effect (i.e., did the intervention work better, similar or worse for Latino MSM compared to their non-Latino white MSM counterparts), a Cox proportional hazard model was used to estimate the hazard ratio of the intervention effect on HIV infection within Latino and non-Latino white subgroups.
Results
Of the 4295 participants enrolled with baseline assessment of race/ethnicity data, 3764 (88%) are included in this study because they self-identified as either Latino (652/3764 = 17%) or non-Latino white (3112/3764 = 83%).
Comparing Latinos to Non-Latino Whites: Demographic, Substance Use, and Psychosocial Variables
Table 1 provides descriptive information on characteristics of the sample by ethnicity. When comparing Latinos to non-Latino whites at their baseline assessment visit, Latinos were significantly more likely to: be younger, have lower educational attainment, report lower annual household income, and be unemployed. Table 2 shows that Latinos, compared to non-Latino whites, were less likely to report moderate alcohol use (AOR = 0.66; 95% CI: 0.47–0.92), using marijuana (AOR = 0.81; 95% CI: 0.67–0.97), using hallucinogens (AOR = 0.68; 95% CI: 0.55–0.84), and using amphetamines (AOR = 0.69; 95% CI: 0.52–0.91), after adjusting for age, education and study site; whereas Latinos were more likely to report using injection drugs (AOR = 1.46; 95% CI: 1.10–1.94) and having a history of childhood sexual abuse (AOR = 2.44; 95% CI: 2.03–2.93). Additionally, in comparing Latinos to non-Latino whites, Latinos were significantly less likely to perceive that they would enjoy unprotected insertive or receptive anal sex (AOR = 0.79; 95% CI: 0.64–0.97), after adjusting for age, education and study site.
Table 1.
Demographic characteristics of study participants by Latino ethnicity
| Demographic Characteristics | Latino (n = 652) n (%) |
Non-Latino white (n = 3112) n (%) |
Test statistic |
|---|---|---|---|
| Age | χ2 = 137.9** (df = 4) | ||
| <25 | 209 (32.1) | 467 (15.0) | |
| 26–30 | 159 (24.4) | 642 (20.6) | |
| 31–35 | 133 (20.4) | 674 (21.7) | |
| 36–40 | 84 (12.9) | 577 (18.5) | |
| >40 | 67 (10.3) | 752 (24.2) | |
| Education | χ2 = 130.7** (df = 4) | ||
| Less than high school | 24 (3.7) | 30 (1.0) | |
| High school or GED | 88 (13.5) | 204 (6.6) | |
| Some college | 226 (34.7) | 723 (23.3) | |
| College | 213 (32.7) | 1167 (37.5) | |
| After college | 101 (15.5) | 986 (31.7) | |
| Household annual income, US$ | χ2 = 149.8** (df = 3) | ||
| <12000 | 138 (21.3) | 323 (10.4) | |
| 12000–29999 | 246 (37.9) | 743 (23.9) | |
| 30000–59999 | 207 (31.9) | 1276 (41.1) | |
| ≥60000 | 58 (8.9) | 766 (24.7) | |
| Employment status | χ2 = 32.3** (df = 3) | ||
| Full time | 453 (69.5) | 2414 (77.6) | |
| Part time | 87 (13.3) | 273 (8.8) | |
| Unemployed | 90 (13.8) | 274 (8.8) | |
| Other | 22 (3.4) | 151 (4.9) | |
| Geographic location | χ2 = 153.0** (df = 5) | ||
| Seattle | 71 (10.9) | 592 (19.0) | |
| Boston | 78 (12.0) | 590 (19.0) | |
| Chicago | 71 (10.9) | 467 (15.0) | |
| Denver | 110 (16.9) | 563 (18.1) | |
| New York | 207 (31.8) | 395 (12.7) | |
| San Francisco | 115 (17.6) | 505 (16.2) |
P<0.01
Table 2.
Substance Use and Psychosocial characteristics by Latino ethnicity
| Latino | Non-Latino white |
Latino compared to Non-Latino white |
||||
|---|---|---|---|---|---|---|
| n (%) | n (%) | Unadjusted odds ratio |
95% CI | Adjusted odds ratioa |
95% CI | |
| Alcohol and substance use | ||||||
| Alcohol use | ||||||
| None | 68 (10.5) | 298 (9.6) | 1.00 | – | 1.00 | – |
| Light | 298 (46.1) | 1437 (46.3) | 0.91 | (0.68, 1.22) | 0.87 | (0.63, 1.19) |
| Moderate | 177 (27.4) | 1057 (34.1) | 0.73* | (0.54, 1.00) | 0.66* | (0.47, 0.92) |
| Heavy | 103 (15.9) | 311 (10.0) | 1.45* | (1.03, 2.05) | 1.13 | (0.78, 1.64) |
| Marijuana use | 304 (47.1) | 1444 (46.5) | 1.03 | (0.87, 1.22) | 0.81* | (0.67, 0.97) |
| Popper use | 220 (34.1) | 1205 (38.8) | 0.82* | (0.69, 0.98) | 0.98 | (0.81, 1.19) |
| Crack use | 24 (3.7) | 117 (3.8) | 0.99 | (0.63, 1.54) | 0.64 | (0.40, 1.04) |
| Amphetamines use | 76 (11.8) | 401 (12.9) | 0.90 | (0.69, 1.17) | 0.69** | (0.52, 0.91) |
| Hallucinogens use | 159 (24.6) | 757 (24.4) | 1.01 | (0.83, 1.23) | 0.68** | (0.55, 0.84) |
| Injection drug use | 84 (13.0) | 276 (8.9) | 1.53 | (1.18, 1.99) | 1.46** | (1.10, 1.94) |
|
| ||||||
| Psychosocial factors related to HIV risk taking | ||||||
| Depression | 328 (51.0) | 1423 (45.9) | 1.23.* | (1.04, 1.46) | 1.00 | (0.83, 1.20) |
| Low self-efficacy for adopting safer sexual behaviors | 99 (15.6) | 477 (15.8) | 0.99 | (0.78, 1.25) | 1.06 | (0.82, 1.36) |
| Poor communication skills regarding safer sex practices | 219 (35.6) | 1018 (34.9) | 1.03 | (0.86, 1.24) | 1.12 | (0.93, 1.36) |
| Weaker safe-sex norms | 108 (17.1) | 535 (18.0) | 0.94 | (0.75, 1.18) | 0.95 | (0.74, 1.21) |
| History of childhood sexual abuse | 378 (58.3) | 1090 (35.1) | 2.60** | (2.19, 3.09) | 2.44** | (2.03, 2.93) |
| Greater perceived enjoyment of risky sex | 469 (72.9) | 2480 (80.3) | 0.66** | (0.54, 0.80) | 0.79** | (0.64, 0.97) |
Adjusted for study site, age, and education
P<0.05
P<0.01
Comparing Latinos to Non-Latino Whites: Serodiscordant Unprotected Anal Sex (SDUA; Data Not Shown)
In baseline univariate models, adjusting for age, education and site, Latino MSM were at significant increased odds of engaging in SDUA compared to non-Latino whites (AOR = 1.20; 95% CI: 1.00–1.44).
Baseline Predictors of SDUA Among Latino MSM Only
In univariate models adjusting for study site, statistically significant predictors of a higher odds of engaging in SDUA at baseline among Latino participants included: use of marijuana, use of poppers, amphetamine use, use of hallucinogens, depressive symptoms, lower self-efficacy for adopting safer sexual behaviors, poorer communication skills regarding safer sex practices, weaker safe-sex norms, a history of childhood sexual abuse, and greater perceived enjoyment of risky sex (see Table 3). In a multivariable model adjusting for study site, variables that remained statistically significant independent predictors of SDUA included: low self-efficacy for adopting safer sexual behaviors (AOR = 3.10; 95% CI: 1.71–5.61), poorer communication skills regarding safer sex practices (AOR = 1.69; 95% CI: 1.13–2.53), and greater perceived enjoyment of risky sex (AOR = 2.17; 95% CI: 1.45–3.26) (see Table 3).
Table 3.
Univariate and multivariable predictors of serodiscordant unprotected anal (SDUA) sex among Latino MSM at baseline and over study follow-up (n = 652)
| Baseline |
Longitudinal |
|||||||
|---|---|---|---|---|---|---|---|---|
| Univariate adjusted ORsa |
95% CI | Multivariable adjusted ORsa |
95% CI | Univariate adjusted ORsb |
95% CI | Multivariable adjusted ORsb |
95% CI | |
| Alcohol and substance use | ||||||||
| Alcohol use | ||||||||
| None | 1.00 | – | 1.00 | – | ||||
| Light | 1.07 | (0.63, 1.82) | 1.56** | (1.14, 2.12) | 1.02 | (0.71, 1.46) | ||
| Moderate | 1.27 | (0.72, 2.25) | 2.10** | (1.51, 2.91) | 1.17 | (0.80, 1.72) | ||
| Heavy | 1.21 | (0.65, 2.27) | 2.96** | (2.02, 4.36) | 1.56* | (1.03, 2.36) | ||
| Marijuana use | 1.57** | (1.14, 2.17) | 1.27 | (0.87, 1.87) | 1.65** | (1.37, 1.98) | 1.32* | (1.04, 1.68) |
| Popper use | 1.61** | 1.15, 2.25) | 1.18 | (0.80, 1.75) | 2.35** | (1.93, 2.87) | 1.91** | (1.50, 2.44) |
| Crack use | 1.71 | (0.71, 4.11) | 1.46* | (1.01, 2.11) | 1.12 | (0.69, 1.81) | ||
| Amphetamines use | 1.78* | (1.07, 2.96) | 0.94 | (0.48, 1.81) | 2.32** | (1.82, 2.96) | 1.98** | (1.44, 2.73) |
| Hallucinogens use | 1.80** | (1.23, 2.60) | 1.45 | (0.88, 2.39) | 1.45** | (1.16, 1.81) | 0.94 | (0.70, 1.26) |
| Injection drug use | 0.95 | (0.59, 1.52) | 1.01 | (0.73, 1.38) | ||||
|
| ||||||||
| Psychosocial factors related to HIV risk taking | ||||||||
| Depression | 1.39* | (1.02, 1.91) | 1.00 | (0.69, 1.45) | 1.49** | (1.26, 1.76) | 1.22 | (0.98, 1.50) |
| Low self-efficacy for adopting safer sexual behaviors |
4.37** | (2.59, 7.38) | 3.10** | (1.71, 5.61) | 2.54** | (1.93, 3.33) | 1.80** | (1.27, 2.55) |
| Poor communication skills regarding safer sex practices |
2.28** | (1.61, 3.23) | 1.69* | (1.13, 2.53) | 2.03** | (1.66, 2.48) | 1.88** | (1.49, 2.36) |
| Weaker safe-sex norms | 1.77* | (1.15, 2.75) | 1.33 | (0.81, 2.19) | 1.36** | (1.10, 1.68) | 1.39* | (1.05, 1.86) |
| History of childhood sexual abuse |
1.46* | (1.06, 2.01) | 1.30 | (0.90, 1.88) | 1.22 | (0.97, 1.55) | ||
| Greater perceived enjoyment of risky sex |
3.11** | (2.13, 4.52) | 2.17** | (1.45, 3.26) | 1.87** | (1.51, 2.33) | 2.17** | (1.66, 2.83) |
Adjusted for study site
Adjusted for study site and randomization arm
P<0.05
P<0.01
Longitudinal Predictors of SDUA Among Latino MSM Only
In univariate models adjusting for study site and randomization arm, statistically significant predictors of a higher odds of engaging in SDUA longitudinally among Latino participants included: light, moderate and heavy alcohol use, use of marijuana, use of poppers, crack use, amphetamine use, use of hallucinogens, depressive symptoms, lower self-efficacy for adopting safer sexual behaviors, poorer communication skills regarding safer sex practices, weaker safe-sex norms, and more perceived enjoyment of risky sex (see Table 3). In a multivariable model adjusting for study site and randomization arm, variables that remained statistically significant independent predictors of an increased odds of engaging in SDUA over study follow-up included: heavy alcohol use (AOR = 1.56; 95% CI: 1.03–2.36), marijuana use (AOR = 1.32; 95% CI: 1.04–1.68), popper use (AOR = 1.91; 95% CI: 1.50–2.44), amphetamine use (AOR = 1.98; 95% CI: 1.44–2.73), lower self-efficacy for adopting safer sexual behaviors (AOR = 1.80; 95% CI: 1.27–2.55), poorer communication skills regarding safer sex practices (AOR = 1.88; 95% CI: 1.49–2.36), weaker safe-sex norms (AOR = 1.39; 95% CI: 1.05–1.86), and more perceived enjoyment of risky sex (AOR = 2.17; 95% CI: 1.66–2.83) (see Table 3).
Comparing Latinos to Non-Latino Whites: HIV-seroconversion (Data Not Shown in a Table)
There were 259 HIV-1 infections in the EXPLORE study, with an overall rate of 2.1 per 100 person-years [17]. Over the course of the study 8.0% of Latinos (52/652) and 5.5% of non-Latino whites (171/3112) tested positive for HIV. Latinos had an overall rate of 2.99 per 100 person-years (95% CI: 2.23–3.92), whereas non-Latino whites had an overall rate of 1.90 per 100 person-years (95% CI: 1.63–2.21). Adjusting for study site and randomization arm, Latino MSM had a significantly higher overall rate of HIV infection over study follow-up than their non-Latino white peers (AHR = 1.44; 95% CI: 1.05-1.99).
Longitudinal Predictors of HIV-seroconversion Among Latino MSM Only
Among Latino participants, within univariate analyses adjusting for study site and randomization arm, a higher probability of becoming infected with HIV over study follow-up were related to: use of marijuana, use of poppers, amphetamine use, use of hallucinogens, and sexually engaging in SDUA. In a multivariable model, adjusting for study site and randomization arm, variables that remained statistically significant independent predictors of becoming infected with HIV over study follow-up among Latino participants included: popper use (AHR = 1.97; 95% CI: 1.08–3.59) and engaging sexually in SDUA (AHR = 1.94; 95% CI: 1.02–3.66) (see Table 4).
Table 4.
Longitudinal univariate and multivariable predictors of HIV infection among Latino MSM only over study follow-up (n = 652)
| Univariate adjusted HRsa |
95% CI | Multivariable adjusted HRsa |
95% CI | |
|---|---|---|---|---|
| Alcohol and substance use | ||||
| Alcohol use | ||||
| None | 1.00 | – | ||
| Light | 3.20 | (0.75, 13.59) | ||
| Moderate | 2.91 | (0.65, 13.02) | ||
| Heavy | 4.28 | (0.94, 19.57) | ||
| Marijuana use | 1.81* | (1.02, 3.23) | (dropped) | |
| Popper use | 2.35** | (1.35, 4.09) | 1.97* | (1.08, 3.59) |
| Crack use | 0.61 | (0.08, 4.53) | ||
| Amphetamines use | 2.16* | (1.07, 4.35) | (dropped) | |
| Hallucinogens use | 2.36** | (1.33, 4.18) | (dropped) | |
| Injection drug use | 1.43 | (0.66, 3.10) | ||
|
| ||||
| Psychosocial factors related to HIV risk taking | ||||
| Depression | 1.69 | (0.96, 3.00) | ||
| Low self-efficacy for adopting safer sexual behaviors | 1.68 | (0.87, 3.25) | ||
| Poor communication skills regarding safer sex practices | 1.21 | (0.67, 2.18) | ||
| Weaker safe-sex norms | 1.70 | (0.87, 3.33) | ||
| History of childhood sexual abuse | 1.06 | (0.60, 1.86) | ||
| Greater perceived enjoyment of risky sex | 1.75 | (0.84, 3.65) | ||
|
| ||||
| Sexual risk taking | ||||
| SDUA | 2.31** | (1.24, 4.33) | 1.94* | (1.02, 3.66) |
Adjusted for study site and randomization arm
P<0.05
P<0.01
Comparing Latinos to non-Latino Whites: Intervention Effect (Data Not Shown)
It was previously reported that the EXPLORE intervention did not significantly reduce the risk of HIV infection: a modest 18% (95% CI: −5%, 36%) reduction in risk of HIV infection was observed [17]. In an exploratory analysis, we computed the EXPLORE intervention effect within the subset of the Latino group of MSM and within the non-Latino whites. We observed a 13% reduction in the Latino men (95% CI: −51%, 50%) and a 19% reduction in the non-Latino whites (95% CI: −10%, 40%), giving no suggestion that the reductions observed were different in the Latino subgroup.
Discussion
In this large multi-site sample of MSM in the U.S., crosssectionally, Latino MSM reported significantly higher rates of SDUA compared to non-Latino white MSM, and were more likely to acquire HIV during the study period than non-Latino white MSM. These findings highlight the continuing need for prevention interventions for Latino MSM.
We then examined predictors of SDUA in Latinos crosssectionally and longitudinally, as well as predictors of acquiring HIV within the Latino MSM in this trial. Multivariable predictors of SDUA at baseline included low self-efficacy for adopting safer sexual behaviors, poor communication regarding safe sex practices, and higher perceived enjoyment of risky sex. In longitudinal analyses, additional predictors of SDUA included weaker safe-sex norms and substance use (i.e., marijuana, poppers, amphetamines, heavy alcohol). Predictors of acquiring HIV for Latino MSM in the sample included using poppers and SDUA sexual behavior. Many of these variables are part of several theoretical models of HIV risk taking such as social cognitive theory, [24, 25] social action theory [26] and the AIDS Risk Reduction model [27]. Accordingly, some of the frequently seen predictors of HIV risk reduction seen in general population samples of MSM may apply to Latino MSM as well. The saliency of such predictors, however, may differ from that of non-Latino white MSM. Additional study is needed of potential personal characteristics and situational factors that may enhance the power of existing social cognitive and health psychology models to predict HIV risk behaviors among Latino MSM [28, 29].
Latino MSM are members of a high-risk group that have been generally underserved within HIV prevention and intervention research [28, 30-32]. A deficit of information about HIV risk patterns of Latino MSM has impeded the development of culturally appropriate HIV prevention interventions for this group [3, 14, 31, 32]. For example, there has been only one reported randomized control trial that addressed an HIV prevention intervention directed specifically at Latino MSM [33]. Furthermore, a recent review on HIV prevention interventions for high-risk sexual behavior suggested that interventions focused on MSM in general are less effective for Latino MSM [34]. The EXPLORE intervention was not meant to address HIV prevention within Latino MSM and, in line with this, our findings failed to indicate that Latino MSM benefited differentially from the intervention than did non-Latino whites.
There are some limitations of the current study to be noted. The selection of high-risk MSM limits generalizability to a wider population of MSM. In addition, although there was a large sample of Latino MSM in this study compared to the extant research, EXPLORE was not set up as a study of cultural and related contextual influences on HIV sexual risk taking and seroconversion. Accordingly, there were a number of important cultural factors—such as acculturation, language, country of birth, immigration status, and length of time in the U.S.—that might be important for HIV prevention efforts with Latinos but which were not assessed [35, 36]. Inclusion of such Latino cultural factors may be beneficial in developing HIV prevention interventions research for use with Latino MSM [3, 4, 35, 36]. Context or situational-related variables should also be studied given findings that, among Latino MSM, unprotected anal intercourse has been associated with factors such as difficult sexual situations [29], sexual attraction, discrimination [8], as well as perceived closeness to the sexual partner, seroconcordance, and concern about sexually transmitted diseases with that partner [28]. Finally, measurement of substance use may have impacted findings in that participants were asked to recall substance use for the past 6 months and use was measured as a binary (i.e., yes, no) variable.
Based on these findings, future studies should examine the contextual reasons that factors associated with safe sex behavior—self-efficacy, communication skills, safe-sex norms—were significantly related to SDUA among Latino MSM yet these factors were not significantly different between the two MSM groups. Similarly, greater enjoyment of high-risk sexual behavior was significantly related to SDUA even though Latinos were less likely to report this than non-Latino white MSM. Study is also needed on the roles of depression and childhood sexual abuse on HIV-related risk behavior within Latino MSM, as this and other studies [8, 28, 37] have provided conflicting findings as to their role. Additionally, studies should also include a strengths-based perspective to identify the factors that aide Latino MSM in practicing safe sexual behavior and avoiding HIV acquisition [30]. Based on longitudinal rather than solely cross-sectional data, our findings provide stronger indication of the causal relationships between predictors and our outcomes of interest.
The current findings support examining potential topics for HIV intervention work with Latino MSM that include training and counseling to address perceived ability to engage in safe sex behavior, enjoyment of high-risk sexual behavior, and the impact of substance use on HIV acquisition. Additional research is needed to examine substance use and sexual relationships in Latino MSM, and it would likely be useful to integrate substance use counseling with HIV prevention messages among Latino MSM who use substances. Findings also suggest the need for HIV prevention intervention research that considers the differential needs of Latino MSM and how the interaction of ethnicity and sexual orientation may impact HIV risk behavior and infection.
Acknowledgments
We gratefully acknowledge the contributions of the EXPLORE study participants and the entire EXPLORE Study Team. Protocol Co-Chairs, Sites and Principal Investigators are listed below. For a full listing of members of the EXPLORE study team please see The EXPLORE Study Team 2004. Protocol co-chairs: Beryl Koblin, Margaret Chesney and Thomas Coates. Boston’s Fenway Community Health Center and the Latin American Health Institute: Kenneth Mayer (Site Principal Investigator) and Team. Chicago’s Howard Brown Community Health Center: David McKirnan (Site Principal Investigator) and Team. Denver Public Health: Franklyn Judson (Site Principal Investigator) and Team. New York Blood Center: Beryl Koblin (Site Principal Investigator) and Team. San Francisco Department of Public Health: Susan Buchbinder (Site Principal Investigator), Grant Colfax (Site Co-Principal Investigator), and Team. Seattle’s University of Washington: Connie Celum (Site Principal Investigator) and Team. We are also grateful for the support we received from Abt Associates, Inc., Center for AIDS Prevention Studies, Statistical Center for HIV/AIDS Research and Prevention and the Central Laboratory. This work was supported by the HIV Network for Prevention Trials and sponsored by the US National Institute of Allergy and Infectious Diseases and the National Institute on Alcohol Abuse and Alcoholism, of the National Institutes of Health, Department of Health and Human Services, through contract N01 AI35176 with Abt Associates Inc; contract N01 AI45200 with the Fred Hutchinson Cancer Research Center; and subcontracts with the Denver Public Health, the Fenway Community Health Center, the Howard Brown Health Center, the New York Blood Center, the Public Health Foundation Inc., and the University of Washington. In addition, this work was supported by the HIV Prevention Trials Network and sponsored by the National Institute of Allergy and Infectious Diseases, the National Institute of Child Health and Human Development, the National Institute on Drug Abuse, the National Institute of Mental Health, and the Office of AIDS Research, of the National Institutes of Health, US Dept of Health and Human Services, through a cooperative agreement with Family Health International (cooperative agreement 5 U01 AI46749) with a sub-sequent subcontract to Abt Associates Inc. with subcontracts to the Howard Brown Health Center and Denver Public Health; cooperative agreement U01 AI48040 to the Fenway Community Health Center, cooperative agreement U01 AI48016 to Columbia University (including a subagreement with the New York Blood Center); and cooperative agreement U01 AI47981 to the University of Washington; and cooperative agreement U01 AI47995 to the University of California, San Francisco. Finally, funding for Andres Bedoya’s work on this manuscript was supported by Grant Number R01DA018603 from the National Institute on Drug Abuse.
Contributor Information
C. Andres Bedoya, Harvard Medical School/Massachusetts General Hospital, Boston, MA, USA; The Fenway Institute, Fenway Health, Boston, MA, USA; MGH Behavioral Medicine Service, One Bowdoin Square, 7th Floor, Boston, MA 02114, USA.
Mathew J. Mimiaga, Harvard Medical School/Massachusetts General Hospital, Boston, MA, USA; The Fenway Institute, Fenway Health, Boston, MA, USA
Geetha Beauchamp, Statistical Center for HIV/AIDS Research and Prevention (SCHARP), Fred Hutchinson Cancer Research Center, Seattle, WA, USA.
Deborah Donnell, Statistical Center for HIV/AIDS Research and Prevention (SCHARP), Fred Hutchinson Cancer Research Center, Seattle, WA, USA.
Kenneth H. Mayer, The Fenway Institute, Fenway Health, Boston, MA, USA; Brown Medical School/Miriam Hospital, Providence, RI, USA
Steven A. Safren, Harvard Medical School/Massachusetts General Hospital, Boston, MA, USA; The Fenway Institute, Fenway Health, Boston, MA, USA
References
- 1.US Census Bureau . We the people: Hispanics in the United States. US Census Bureau; Washington: 2004. [Google Scholar]
- 2.Centers for Disease Control and Prevention (CDC) HIV/AIDS surveillance report, 2007. US Department of Health and Human Services, CDC; 2009. Report number: 19. [Google Scholar]
- 3.Díaz RM. Latino gay men and HIV: culture, sexuality, and risk behavior. Cult Divers Ethnic Minor Psychol. 1999;5(3):292–3. [Google Scholar]
- 4.Fernandez MI, Jacobs RJ, Warren JC, et al. Drug use and Hispanic men who have sex with men in South Florida: implications for intervention development. AIDS Educ Prev. 2009;21(5 Suppl):45–60. doi: 10.1521/aeap.2009.21.5_supp.45. [DOI] [PubMed] [Google Scholar]
- 5.Herbst JH, Kay LS, Passin WF, et al. A systematic review and meta-analysis of behavioral interventions to reduce HIV risk behaviors of Hispanics in the United States and Puerto Rico. AIDS Behav. 2007;11(1):25–47. doi: 10.1007/s10461-006-9151-1. [DOI] [PubMed] [Google Scholar]
- 6.Marín G. AIDS prevention among Hispanics: needs, risk behaviors, and cultural values. Public Health Rep. 1989;104(5):411–5. [PMC free article] [PubMed] [Google Scholar]
- 7.Fernandez MI, Perrino T, Collazo JB, et al. Surfing new territory: club-drug use and risky sex among Hispanic men who have sex with men recruited on the internet. Journal of Urban Health. 2005;82(1 Suppl):79–88. doi: 10.1093/jurban/jti027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jarama SL, Kennamer JD, Poppen PJ, et al. Psychosocial, behavioral, and cultural predictors of sexual risk for HIV infection among Latino men who have sex with men. AIDS Behav. 2005;9(4):513–23. doi: 10.1007/s10461-005-9022-1. [DOI] [PubMed] [Google Scholar]
- 9.Warren JC, Fernandez MI, Harper GW, et al. Predictors of unprotected sex among young sexually active African American, Hispanic, and white MSM: the importance of ethnicity and culture. AIDS Behav. 2008;12(3):459–68. doi: 10.1007/s10461-007-9291-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wilson PA, Diaz RM, Yoshikawa H, et al. Drug use, interpersonal attraction, and communication: situational factors as predictors of episodes of unprotected anal intercourse among Latino gay men. AIDS Behav. 2009;13(4):691–9. doi: 10.1007/s10461-008-9479-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Flores SA, Mansergh G, Marks G, et al. Gay identity-related factors and sexual risk among men who have sex with men in San Francisco. AIDS Educ Prev. 2009;21(2):91–103. doi: 10.1521/aeap.2009.21.2.91. [DOI] [PubMed] [Google Scholar]
- 12.Harawa NT, Greenland S, Bingham TA, et al. Associations of race/ethnicity with HIV prevalence and HIV-related behaviors among young men who have sex with men in 7 urban centers in the United States. J Acquir Immune Defic Syndr. 2004;35:526–36. doi: 10.1097/00126334-200404150-00011. [DOI] [PubMed] [Google Scholar]
- 13.Bedoya CA, Mimiaga MJ, Beauchamp G, et al. Predictors of HIV transmission risk behavior and seroconversion among Latino MSM in Project EXPLORE; 31st Annual meeting and scientific sessions of the Society of Behavioral Medicine; Seattle, WA. 2010; abstract. [Google Scholar]
- 14.Koblin BA, Chesney MA, Husnik MJ, et al. High-risk behaviors among men who have sex with men in 6 US cities: Baseline data from the EXPLORE study. Am J Public Health. 2003;93(6):926–32. doi: 10.2105/ajph.93.6.926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Koblin BA, Husnik MJ, Colfax G, et al. Risk factors for HIV infection among men who have sex with men. AIDS. 2006;20:731–9. doi: 10.1097/01.aids.0000216374.61442.55. [DOI] [PubMed] [Google Scholar]
- 16.Chesney MA, Koblin BA, Barresi PJ, et al. An individually tailored intervention for HIV prevention: baseline data from the EXPLORE study. Am J Public Health. 2003;93(6):933–8. doi: 10.2105/ajph.93.6.933. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.The EXPLORE Study Team Effects of a behavioural intervention to reduce acquisition of HIV infection among men who have sex with men: the EXPLORE randomised controlled study. Lancet. 2004;364(9428):41–50. doi: 10.1016/S0140-6736(04)16588-4. [DOI] [PubMed] [Google Scholar]
- 18.Mirowsky J, Ross CE. Age and depression. J Health Soc Behav. 1992;33(3):187–205. [PubMed] [Google Scholar]
- 19.Ross CE, Mirowsky J. Sex differences in the effect of education on depression: resource multiplication or resource substitution? Soc Sci Med. 2006;63(5):1400–13. doi: 10.1016/j.socscimed.2006.03.013. [DOI] [PubMed] [Google Scholar]
- 20.Mimiaga MJ, Noonan E, Donnell D, et al. Childhood sexual abuse is highly associated with HIV risk-taking behavior and infection among MSM in the EXPLORE study. J Acquir Immune Defic Syndr. 2009;51(3):340–8. doi: 10.1097/QAI.0b013e3181a24b38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika. 1986;73:13–22. [Google Scholar]
- 22.Hardin JW, Hilbe JM. Generalized estimating equations. Chapman and Hall/CRC Press; Boca Raton: 2003. [Google Scholar]
- 23.Therneau TM, Grambsch PM. Modeling survival data: extending the Cox model. Springer; New York: 2000. [Google Scholar]
- 24.Bandura A. Social cognitive theory and exercise of control over HIV infection. In: DiClemente RJ, Peterson JL, editors. Preventing AIDS: theories and methods of behavioral interventions. Plenum Press; New York: 1994. pp. 25–60. [Google Scholar]
- 25.Wulfert E, Wan CK, Backus CA. Gay men’s safer sex behavior: an integration of three models. J Behav Med. 1996;19(4):345–66. doi: 10.1007/BF01904761. [DOI] [PubMed] [Google Scholar]
- 26.Ewart CK. Social action theory for a public health psychology. Am Psychol. 1991;46(9):931–46. doi: 10.1037//0003-066x.46.9.931. [DOI] [PubMed] [Google Scholar]
- 27.Catania JA, Kegeles SM, Coates TJ. Towards an understanding of risk behavior: an AIDS risk reduction model (ARRM) Health Educ Q. 1990;17(1):53–72. doi: 10.1177/109019819001700107. [DOI] [PubMed] [Google Scholar]
- 28.Zea MC, Reisen CA, Poppen PJ, Bianchi Ft. Unprotected anal intercourse among immigrant Latino MSM: the role of characteristics of the person and the sexual encounter. AIDS Behav. 2009;13(4):700–15. doi: 10.1007/s10461-008-9488-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Díaz RM, Ayala G, Bein E. Sexual risk as an outcome of social oppression: data from a probability sample of Latino gay men in three U.S. cities. Cult Divers Ethnic Minor Psychol. 2004;10(3):255–67. doi: 10.1037/1099-9809.10.3.255. [DOI] [PubMed] [Google Scholar]
- 30.Díaz RM, Ayala G. Social discrimination and Health: the case of Latino gay men and HIV risk. The Policy Institute of the National Gay and Lesbian Task Force; Jul 12, 2001. http://www.caps.ucsf.edu/projects/HLS/pdf/DiazEng.pdf. [Google Scholar]
- 31.Gonzalez JS, Hendriksen ES, Collins EM, Duran RE, Safren SA. Latinos and HIV/AIDS: examining factors related to disparity and identifying opportunities for psychosocial intervention research. AIDS Behav. 2009;13(3):582–602. doi: 10.1007/s10461-008-9402-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Darbes LA, Kennedy GE, Peersman G, Zohrabyan L, Rutherford GW. Systematic review of HIV behavioral prevention research in Latinos. Cochrane Collaborative Review Group on HIV/AIDS; San Francisco: 2002. [Google Scholar]
- 33.Carballo-Dieguez A, Dolezal C, Leu CS, et al. A randomized controlled trial to test an HIV-prevention intervention for Latino gay and bisexual men: lessons learned. AIDS Care. 2005;17(3):314–28. doi: 10.1080/09540120512331314303. [DOI] [PubMed] [Google Scholar]
- 34.Johnson WD, Diaz RM, Flanders WD, et al. Behavioral interventions to reduce risk for sexual transmission of HIV among men who have sex with men. Cochrane Database Syst Rev. 2008;(3) doi: 10.1002/14651858.CD001230.pub2. Art. No.: CD001230. doi:10.1002/14651858.CD001230.pub2. [DOI] [PubMed] [Google Scholar]
- 35.Zea MC, Reisen CA, Diaz RM. Methodological issues in research on sexual behavior with Latino gay and bisexual men. Am J Community Psychol. 2003;31(3-4):281–91. doi: 10.1023/a:1023962805064. [DOI] [PubMed] [Google Scholar]
- 36.Marín BV, Gómez CA. Latinos and HIV: Cultural issues in AIDS prevention. HIV InSite and University of California; San Francisco: Nov, 1998. http://hivinsite.ucsf.edu/InSite?page=pr-rr-03. [Google Scholar]
- 37.Arreola SG, Neilands TB, Díaz R. Childhood sexual abuse and the sociocultural context of sexual risk among adult Latino gay and bisexual men. Am J Public Health. 2009;99:S432–8. doi: 10.2105/AJPH.2008.138925. [DOI] [PMC free article] [PubMed] [Google Scholar]