Abstract
Objectives. We assessed the efficacy of an HIV behavioral intervention adapted for Black men who have sex with men (MSM).
Methods. We conducted serial cross-sectional surveys, 1 baseline measurement followed by initiation of an intervention and 3 follow-up measurements, among Black MSM in 3 North Carolina cities over 1 year.
Results. We observed significant decreases in unprotected receptive anal intercourse at 4 months (by 23.8%, n=287) and 8 months (by 24.7%, n=299), and in unprotected insertive anal intercourse (by 35.2%), unprotected receptive anal intercourse (by 44.1%), and any unprotected anal intercourse (by 31.8%) at 12 months (n=268). Additionally, at 12 months, the mean number of partners for unprotected receptive anal intercourse decreased by 40.5%. The mean number of episodes decreased by 53.0% for unprotected insertive anal intercourse, and by 56.8% for unprotected receptive anal intercourse. The percentage of respondents reporting always using condoms for insertive and receptive anal intercourse increased by 23.0% and 30.3%, respectively.
Conclusions. Adapting previously proven interventions designed for other MSM can significantly reduce HIV risk behaviors of Black MSM.
In the United States, high rates of HIV infection have been reported among Black men who have sex with men (MSM). From 2001 to 2004, rates of HIV infection remained higher among Black men than among men of other racial and ethnic populations, and 36% to 39% of new HIV diagnoses among MSM and MSM who inject drugs were among Black men.1 Surveillance data from 5 cities showed that Black MSM had the highest HIV prevalence (46%) among MSM, and that two thirds of HIV-seropositive Black MSM were unaware of their infection.2 In a 2000 study of young MSM, Black and multiethnic MSM were 9.1 times and Caribbean Black MSM were 10.2 times as likely to be infected with HIV as White men.3 In another study of MSM aged 23 to 29 years in 6 cities, 30% of Black MSM were infected with HIV compared with 7% of White MSM.4 A recent retrospective chart review in North Carolina found that 88% (49 of 56) of new HIV cases among men aged 18 to 30 years were among Black men and a majority reported MSM behavior.5
A recent literature review6 suggests that Black MSM are at elevated risk because of high rates of STDs that facilitate acquisition and transmission of HIV,7 less-frequent HIV testing,8 and unrecognized HIV.2,3,8 In another study, Black MSM who did not carry condoms and those who reported nonsupportive peer norms for condom use were more likely to engage in unprotected receptive anal intercourse.9 Despite the high risk of HIV infection among Black MSM, a systematic review of published HIV prevention interventions for MSM identified only 1 specifically targeted to Black MSM.10 In the absence of efficacious interventions for Black MSM, adapting and evaluating currently available evidence-based interventions designed for MSM12–19 may be an effective strategy.20
One such intervention is the Popular Opinion Leader (POL) intervention,15,22,23 a community-level intervention which seeks to increase safer-sex norms among members of a well-defined target population.15 Based on the diffusion of innovation theory,21 POL was designed for MSM and originally tested primarily among White MSM in 3 southern US cities.15 Opinion leaders are recruited and trained to have risk reduction conversations with their friends to increase healthy sex norms. Critical components of the POL intervention have been previously published.24 Early evaluations of POL showed decreases in unprotected anal intercourse of 15% to 29% from baseline levels.23 A larger-scale, randomized trial of POL found a 37% decrease in unprotected anal intercourse in 4 intervention cities, whereas a slight increase occurred in the 4 control cities.22
Several adaptations of the POL intervention have been employed with various populations other than Black MSM, such as young Latino migrant MSM,25 male sex workers,26 and women.27 It has been adapted for MSM in international settings such as London, England28; Glasgow, Scotland29; and Russia and Bulgaria.30,31 We measured the effectiveness of a POL intervention adapted specifically for young Black MSM aged 18 to 30 years in 3 North Carolina cities (Raleigh, Greensboro, and Charlotte).
METHODS
We collected data from December 2004 through December 2005 while evaluating HIV-prevention activities. The evaluation was conducted in 3 North Carolina cities that had nightclubs in which the target population could be accessed, recruited, and trained for the intervention. The evaluation used methods similar to those used by Kelly et al.,23 who evaluated the intervention based on pre- and posttest results in 3 cities with no control city.
Intervention Development and Procedures
In September 2004, we conducted focus groups with the target population as well as key informant interviews with stakeholders (such as bar and nightclub owners, community activists, and organizers of Black gay pride celebrations) in each city. Participants were asked to identify issues and challenges facing MSM, barriers to accessing prevention services, topics that prevention activities should address, and ideal ways of marketing intervention activities to Black MSM. Information gained through these interviews informed the adaptation of the intervention, social marketing materials, and assessments. Intervention sessions were adapted to include discussions about racism, homophobia, bisexuality, employment and poverty, and religion. Using role-play scenarios, opinion leaders learned how to deal with challenges facing Black MSM should these issues arise in their risk reduction conversations. We included a condom demonstration and created culturally relevant marketing materials, conversation starters, and a project logo.
Trained local prevention specialists used adapted ethnographic techniques to identify opinion leaders30 at local nightclubs frequented by the target population. Once sufficient numbers were recruited, local prevention specialists conducted four 2-hour sessions. Sessions covered the following topics: local and state epidemiology of HIV/AIDS and STDs, facts and myths about HIV/AIDS, and characteristics of an effective risk reduction conversation. Intervention participants were also given opportunities to role-play potential conversations that they could have with their friends and acquaintances. To ensure that learning objectives were met, participants were given tests measuring their knowledge before and after the intervention sessions. To compensate them for their time, opinion leaders received $100 in gift cards, marketing materials that contained the project logo, and safer-sex materials. All opinion leaders provided informed consent and signed confidentiality agreements. We sought to train 15% of the target population as opinion leaders because this is a core element of the intervention.15 Finally, the lead investigator and a majority of the local project staff were similar to the target population, based on race, gender, age, and sexual identity.
Outcome Monitoring Design
Based on available intervention literature,32,33 sample sizes were calculated to allow for an 8% decrease in unprotected anal intercourse. We determined that a single-group t test with a 0.05 2-sided significance level would have 85% power to detect the difference between the null hypothesis proportion and the alternative proportion when the sample size adjusted for the population size is 230.
Four equally spaced cross-sectional surveys were conducted during the 1-year study period. Respondents were recruited separately for each wave. As such, there was no effort to recruit the same individuals at subsequent waves. Each intervention city was approximately 60 miles apart, had at least 1 nightclub that catered to Black MSM, and had high rates of HIV infection among Black MSM. After the baseline assessment in each intervention city, we conducted interviews in the same venues in which we had recruited opinion leaders.
During 2 consecutive weekends for each assessment, trained interviewers recruited convenience samples of Black MSM in each city. Men were approached as they entered the nightclub. Interviewers explained the study and obtained informed consent. For initial screening, eligible respondents were (1) self-identifying Black or African American, (2) biologically male, (3) aged 18 to 30 years, and (4) living in or visiting the study areas since December 1, 2004. Because respondents may be uncomfortable disclosing specifics about their sexual behavior during an interview,34–36 men were asked generally whether they had had sex with any male or female partners in the previous year. In self-administered assessments using handheld computers, men were later reassessed for their same-sex activities.
Men were ineligible if they appeared to be intoxicated or had already completed the assessment during the same wave. The assessment was programmed to include a brief tutorial and system checks to improve data reliability.37,38 Of the men screened for eligibility (N=1481), 80.4% (n=1190) reported having had oral or anal intercourse with a man in the past year and were included in this analysis. On average, the self-assessment took 10.4 minutes (SD=3.2 minutes) to complete. Those who completed the interview received a $20 gift card as compensation for their time. HIV prevention information and condoms were made available.
Independent Variables
Unless otherwise specified, men were asked about their behaviors in the 2 months prior to assessment. City of interview, age, education (high school or less, some college, and college or more), employment status, and sexual identity (gay or homosexual, straight or heterosexual, bisexual, do not identify or label myself, and other) were assessed. Men who did not identify as gay or homosexual were described as non-gay identified. The men were asked whether they had any sexual contact (oral, vaginal, or anal) with female partners in the previous 2 months as well as in their lifetime. Age of respondents was recoded into 4 categories: 18 to 20, 21 to 22, 23 to 25, and 26 to 30 years. Respondents were also asked whether they had been incarcerated during the 2 months prior to the interview.
Respondents were asked the number of times they had been tested for HIV over their lifetime. This variable was dichotomized as ever or never tested for HIV. Those who reported ever having been tested were further dichotomized into those tested in the past 3 months versus those tested more than 3 months ago or who were not sure of their testing date. Men who tested at least once were asked the status of their last HIV test. HIV-testing histories and results were categorized as positive, negative, or other (never tested, unknown, don’t know, or refused). Respondents were also asked if they had been diagnosed with an STD in the past 2 months.
Exposure to the intervention was assessed by asking respondents how many times they had seen the project logo. This variable was dichotomized to indicate whether respondents had seen the logo or not. Additionally, respondents were asked if they had been trained as an opinion leader in the project.
Outcome Variables
Respondents were asked the number of times they engaged in protected and unprotected anal intercourse with men and the number of male partners with whom they engaged in protected and unprotected anal intercourse. Separate questions were asked concerning insertive and receptive anal intercourse. A series of 3 dichotomous outcomes were created to indicate whether the respondent had engaged in (1) unprotected insertive anal intercourse, (2) unprotected receptive anal intercourse, or (3) any unprotected anal intercourse with men, meaning either unprotected insertive anal intercourse or unprotected receptive anal intercourse.
The number of partners for unprotected anal intercourse was measured as 2 count variables: number of male partners for unprotected insertive anal intercourse and number of male partners for unprotected receptive anal intercourse. The number of episodes of unprotected anal intercourse with men was also available from the survey as 2 count variables: number of episodes of unprotected insertive anal intercourse with men and number of episodes of unprotected receptive anal intercourse with men.
Finally, the percentage of times that condoms were used during anal intercourse was calculated for respondents reporting insertive or receptive anal intercourse. Responses were recoded as always, sometimes, or never using condoms.
Statistical Methods
Data from the serial cross-sectional surveys were analyzed in 2 ways. First, results of each follow-up wave were compared with results at baseline. Second, the linear trend across the entire study period was examined to estimate the average change per wave. Logistic regression was used for the 3 dichotomous outcomes: unprotected insertive anal intercourse, unprotected receptive anal intercourse, and unprotected anal intercourse. Generalized estimating equations with the negative binomial distribution and log link were applied to the 4 count outcomes: number of partners for unprotected sex (insertive and receptive) and number of episodes of unprotected sex (insertive and receptive). Proportional odds models were employed for the 3-level measure of percentage condom use. Changes relative to baseline are presented for statistically significant results. These were calculated between the baseline and follow-up.
Multivariable analyses were used to control for all variables associated either with wave (the independent variable) or with the dependent variable of interest.
RESULTS
Demographics
Approximately 300 men responded at each follow-up wave for a total of 1190 responses of men who reported oral or anal intercourse with a man in the past year (Table 1 ▶). Responses were approximately equal across each city with 31.4% from Greensboro, 34.2% from Raleigh, and 34.4% from Charlotte. The mean age of respondents was 23 years (SD=3.3 years). Two thirds had at least some college education, and 79% were employed.
TABLE 1—
Sample Characteristics, by Study Wave: Popular Opinion Leader Intervention Adapted for Black Men Who Have Sex With Men, North Carolina, December 2004 to December 2005
| Wave 1 (Baseline) | Wave 2 | Wave 3 | Wave 4 | Overall | |
| Sample, no. | 295 | 296 | 317 | 282 | 1190 |
| City,*** no. (%) | |||||
| Raleigh | 111 (37.6) | 100 (33.8) | 116 (36.6) | 80 (28.4) | 407 (34.2) |
| Greensboro | 93 (31.5) | 107 (36.2) | 91 (28.7) | 83 (29.4) | 374 (31.4) |
| Charlotte | 91 (30.9) | 89 (30.1) | 110 (34.7) | 119 (42.2) | 409 (34.4) |
| Age, y, mean (SD) | 22.8 (3.14) | 22.6 (3.32) | 23.0 (3.49) | 22.6 (3.29) | 22.8 (3.32) |
| Age group, y, no. (%) | |||||
| ≤ 20 | 78 (26.4) | 97 (32.8) | 84 (26.5) | 92 (32.6) | 351 (29.5) |
| 21–22 | 79 (26.8) | 66 (22.3) | 77 (24.3) | 64 (22.7) | 286 (24.0) |
| 23–25 | 84 (28.5) | 76 (25.7) | 77 (24.3) | 73 (25.9) | 310 (26.1) |
| > 25 | 54 (18.3) | 57 (19.3) | 79 (24.9) | 53 (18.8) | 243 (20.4) |
| Education, No. (%) | |||||
| High school or less | 91 (31.1) | 105 (36.1) | 109 (35.5) | 88 (31.9) | 393 (33.7) |
| Some college | 122 (41.6) | 115 (39.5) | 126 (41.0) | 124 (44.9) | 487 (41.7) |
| College or more | 80 (27.3) | 71 (24.4) | 72 (23.5) | 64 (23.2) | 287 (24.6) |
| Employed,** no. (%) | 220 (75.1) | 224 (75.7) | 264 (84.1) | 227 (80.5) | 935 (78.9) |
| HIV testing, no. (%) | |||||
| Ever tested for HIV | 265 (90.8) | 260 (89.0) | 281 (90.1) | 242 (88.6) | 1048 (89.7) |
| Tested within 3 months prior to survey | 88 (29.8) | 111 (37.5) | 121 (38.2) | 95 (33.7) | 415 (34.9) |
| Tested more than 3 months prior to survey | 155 (52.5) | 130 (43.9) | 140 (44.2) | 123 (43.6) | 548 (46.1) |
| HIV status, no. (%) | |||||
| Positive | 14 (4.8) | 14 (4.7) | 17 (5.4) | 8 (2.8) | 53 (4.5) |
| Negative | 242 (82.0) | 236 (79.7) | 260 (82.0) | 223 (79.1) | 961 (80.8) |
| Never tested/unknown/don’t know/refused to answer | 39 (13.2) | 46 (15.5) | 40 (12.6) | 51 (18.1) | 176 (14.8) |
| Diagnosed with STD,a no./total (%) | 8/292 (2.7) | 7/291 (2.4) | 13/316 (4.1) | 9/282 (3.2) | 37/1181 (3.1) |
| Nongay identified,a no. (%) | 139/293 (47.4) | 130/295 (44.1) | 118/313 (37.7) | 121/281 (43.1) | 508/1182 (43.0) |
| Sexual intercourse with a female,a no./total (%) | |||||
| Ever had it | 179/294 (60.9) | 176/296 (59.5) | 201/316 (63.6) | 178/282 (63.1) | 734/1188 (61.8) |
| Had it in past 2 months | 44/295 (14.9) | 58/296 (19.6) | 53/317 (16.7) | 45/282 (16.0) | 200/1190 (16.8) |
| Ever been in jail,a no./total (%) | 22/293 (7.5) | 15/293 (5.1) | 15/311 (4.8) | 12/276 (4.4) | 64/1173 (5.5) |
| Popular opinion leader,a† no./total (%) | 11/292 (3.8) | 38/292 (13.0) | 54/310 (17.4) | 39/279 (14.0) | 142/1173 (12.1) |
| Seen logo,a† no./total (%) | 60/292 (20.6) | 164/289 (56.8) | 196/307 (63.8) | 171/276 (62.0) | 591/1164 (50.8) |
| Completed similar previous test, no. (%) | . . . | 74 (25.3) | 81 (25.6) | 74 (26.3) | 229 (25.7) |
Note. STD = sexually transmitted disease.
aChange in sample numbers because of missing data.
** P < .05; ***P < .01; †P < .001.
Forty-three percent of the sample was non-gay identified, and 62% reported lifetime sexual intercourse with a female. However, only 17% reported recent sexual intercourse with a female. Over 5% of the sample reported incarceration in the past 2 months. Ninety percent of the sample reported ever testing for HIV with 35% having been tested in the 3 months prior to assessment. Of the entire sample, over 4% reported testing HIV positive and 15% had unknown results.
After the implementation of the intervention, 15% (131 of 881) of respondents also reported being trained as an opinion leader, and 61% (531 of 872) of respondents reported having seen the project logo. Also in waves 2 through 4, 25% to 27% of respondents reported completing a similar handheld survey approximately 4 months earlier when the previous assessment was being conducted.
Decrease in Unprotected Anal Sex
At baseline, 29.3% of respondents reported unprotected insertive anal intercourse, 32.4% reported unprotected receptive anal intercourse, and 42.1% reported any unprotected anal intercourse (Table 2 ▶). In adjusted analyses, significant decreases were observed in unprotected insertive anal intercourse from wave 1 (baseline) to 4, in unprotected receptive anal intercourse from wave 1 to waves 2 and 4, and in any unprotected anal intercourse from wave 1 to waves 2 and 4. Unadjusted results for each type of unprotected sex were similar to the adjusted results. Additionally, a significant decrease was observed in unprotected receptive anal intercourse from wave 1 to 3. In terms of relative decrease, unprotected insertive anal intercourse decreased by 35.2% from wave 1 to 4 ([29.3%–19.0%]/29.3%). From wave 1, unprotected receptive anal intercourse decreased by 23.8% at wave 2, by 24.7% at wave 3, and by 44.1% at wave 4. Unprotected anal intercourse decreased by 31.8% from wave 1 to 4. The adjusted and unadjusted linear trends per wave were significant for each of the 3 dichotomous outcomes.
TABLE 2—
Dichotomous Outcomes, Odds Ratios (ORs), and 95% Confidence Intervals (CIs) for Anal Intercourse With Male Partners, by Study Wave: Popular Opinion Leader Intervention Adapted for Black Men Who Have Sex With Men, North Carolina, December 2004 to December 2005
| Reported Behavior, % (No.) | Unadjusted OR (95% CI) | Adjusted OR (95% CI) | |
| Unprotected insertive anal intercourse with male partnersa | |||
| Sample, no. | 1144 | 1144 | 1132 |
| Wave | |||
| 1 (Ref) | 29.3 (83/283) | 1.00 | 1.00 |
| 2 | 25.9 (74/286) | 0.84 (0.58, 1.22) | 0.85 (0.60, 1.19) |
| 3 | 26.7 (82/307) | 0.88 (0.61, 1.26) | 0.87 (0.62, 1.22) |
| 4 | 19.0 (51/268) | 0.56*** (0.38, 0.84) | 0.58*** (0.40, 0.84) |
| Linear trend | . . . | 0.86** (0.75, 0.97) | 0.86*** (0.77, 0.96) |
| Unprotected receptive anal intercourse with male partnersb | |||
| Sample, no. | 1141 | 1141 | 1138 |
| Wave | |||
| 1 (Ref) | 32.4 (92/284) | 1.00 | 1.00 |
| 2 | 24.7 (71/287) | 0.69** (0.48, 0.99) | 0.68** (0.46, 1.00) |
| 3 | 24.4 (73/299) | 0.67** (0.47, 0.97) | 0.68* (0.46, 1.00) |
| 4 | 18.1 (49/271) | 0.46† (0.31, 0.69) | 0.46† (0.30, 0.70) |
| Linear trend | . . . | 0.79† (0.72, 0.88) | 0.79† (0.70, 0.91) |
| Any unprotected anal intercourse with male partnersa | |||
| Sample, no. | 1141 | 1141 | 1130 |
| Wave | |||
| 1 (Ref) | 42.1 (119/283) | 1.00 | 1.00 |
| 2 | 34.5 (99/287) | 0.73 (0.52, 1.02) | 0.73** (0.53, 1.00) |
| 3 | 36.0 (109/303) | 0.77 (0.56, 1.08) | 0.77 (0.57, 1.06) |
| 4 | 28.7 (77/268) | 0.56*** (0.39, 0.79) | 0.57† (0.41, 0.80) |
| Linear trend | . . . | 0.85*** (0.76, 0.95) | 0.85*** (0.77, 0.95) |
aOR adjusted for city where survey took place, employment status, and ever been to jail.
bOR adjusted for city where survey took place, employment status, and HIV status (positive, negative, or unknown, don’t know, refused, or never tested).
*P = .05; **P < .05; ***P < .01; †P < .001.
At baseline, respondents reported a mean of 1.15 episodes of unprotected insertive anal intercourse and 1.25 episodes of unprotected receptive anal intercourse (Table 3 ▶). In adjusted analyses, significant decreases were observed for the number of episodes from wave 1 to 4 for both unprotected insertive anal intercourse and unprotected receptive anal intercourse. Unadjusted results were similar to adjusted results. In terms of relative decrease, the mean number of episodes decreased by 53.0% for unprotected insertive anal intercourse and by 56.8% for unprotected receptive anal intercourse. The unadjusted linear trend was significant for episodes of unprotected receptive anal intercourse only; adjusted analyses for linear trends per wave were significant for both.
TABLE 3—
Count Outcomes, Rate Ratios (RRs), and 95% Confidence Intervals (CIs) for Unprotected Anal Intercourse With Male Partners, by Study Wave, Number of Partners, and Number of Episodes: Popular Opinion Leader Intervention Adapted for Black Men Who Have Sex With Men, North Carolina, December 2004 to December 2005
| Mean (95% CI) | Unadjusted RR (95% CI) | Adjusted RR (95% CI) | |
| Number of episodes of unprotected insertive anal intercourse with male partnersa | |||
| Sample, no. | 1144 | 1144 | 1132 |
| Wave | |||
| 1 (Ref) | 1.15 (0.72, 1.57) | 1.00 | 1.00 |
| 2 | 1.03 (0.66, 1.39) | 0.90 (0.56, 1.44) | 0.84 (0.52, 1.37) |
| 3 | 1.48 (0.70, 2.26) | 1.29 (0.81, 2.05) | 1.19 (0.73, 1.93) |
| 4 | 0.54 (0.35, 0.73) | 0.47*** (0.29, 0.77) | 0.45*** (0.27, 0.75) |
| Linear trend | 0.87 (0.73, 1.03) | 0.84** (0.70, 1.00) | |
| Number of episodes of unprotected receptive anal intercourse with male partnersb | |||
| Sample, no. | 1141 | 1141 | 1138 |
| Wave | |||
| 1 (Ref) | 1.25 (0.83, 1.68) | 1.00 | 1.00 |
| 2 | 0.98 (0.65, 1.30) | 0.78 (0.49, 1.25) | 0.73 (0.46, 1.18) |
| 3 | 1.18 (0.66, 1.70) | 0.94 (0.59, 1.50) | 0.97 (0.60, 1.57) |
| 4 | 0.54 (0.35, 0.72) | 0.43† (0.26, 0.70) | 0.42† (0.26, 0.70) |
| Linear trend | 0.81*** (0.69, 0.95) | 0.81*** (0.69, 0.96) | |
| Number of male partners for unprotected insertive anal intercoursea | |||
| Sample, no. | 1144 | 1144 | 1132 |
| Wave | |||
| 1 (Ref) | 0.40 (0.30, 0.50) | 1.00 | 1.00 |
| 2 | 0.43 (0.31, 0.54) | 1.07 (0.76, 1.50) | 1.09 (0.78, 1.53) |
| 3 | 0.36 (0.27, 0.45) | 0.90 (0.64, 1.26) | 0.85 (0.60, 1.20) |
| 4 | 0.30 (0.20, 0.41) | 0.76 (0.53, 1.09) | 0.79 (0.55, 1.13) |
| Linear trend | 0.91 (0.81, 1.02) | 0.91 (0.81, 1.02) | |
| Number of male partners for unprotected receptive anal intercourseb | |||
| Sample, no. | 1141 | 1141 | 1138 |
| Wave | |||
| 1 (Ref) | 0.42 (0.32, 0.52) | 1.00 | 1.00 |
| 2 | 0.39 (0.29, 0.48) | 0.92 (0.66, 1.27) | 0.89 (0.65, 1.23) |
| 3 | 0.34 (0.26, 0.42) | 0.81 (0.58, 1.12) | 0.84 (0.60, 1.16) |
| 4 | 0.25 (0.17, 0.33) | 0.59*** (0.42, 0.85) | 0.60*** (0.42, 0.86) |
| Linear trend | 0.85*** (0.76, 0.95) | 0.86*** (0.77, 0.96) | |
aRR adjusted for city where survey took place, employment, and ever been to jail.
bRR adjusted for city where survey took place, employment status, and HIV status (positive, negative, or unknown [don’t know, refused, or never tested]).
** P < .05; ***P < .01; †P < .001.
Decrease in Number of Unprotected Anal Sex Partners
At baseline, respondents reported a mean of 0.40 partners for unprotected insertive anal intercourse and 0.42 partners for unprotected receptive anal intercourse (Table 3 ▶). In adjusted analyses, a significant decrease was observed in the mean number of partners for unprotected receptive anal intercourse from wave 1 to 4. In terms of relative decrease, the mean number of partners for unprotected receptive anal intercourse decreased by 40.5%. Overall, the linear trend per wave was significant for partners for unprotected receptive anal intercourse. Although there was a decrease across time in partners for unprotected insertive anal intercourse, none of the changes were significant. Unadjusted results for each type of unprotected sex were similar to the adjusted results.
Increases in Condom Use During Anal Sex
At baseline, 54.7% of the respondents always used condoms and 32.0% sometimes used condoms during insertive anal intercourse. During receptive anal intercourse, 50.8% always used condoms and 36.1% sometimes used condoms (Table 4 ▶). In adjusted analyses, a significant increase was observed in the percentage of respondents reporting condom use during receptive anal intercourse from wave 1 to 4. Unadjusted results were similar to adjusted results except that the increase in condom use during insertive anal intercourse from wave 1 to 4 in the unadjusted analysis was significant. In terms of relative increases, the percentage of respondents who reported always using condoms during insertive anal intercourse increased by 23.0% at wave 4 and by 30.3% during receptive anal intercourse at wave 4.
TABLE 4—
Ordinal Outcomes, Odds Ratios (ORs), and 95% Confidence Intervals (CIs) for Condom Use for Anal Intercourse, by Study Wave: Popular Opinion Leader Intervention Adapted for Black Men Who Have Sex With Men, North Carolina, December 2004 to December 2005
| Reported Condom Use | OR (95% CI) | ||||
| Always, % (no.) | Sometimes, % (no.) | Never, % (no.) | Unadjusted | Adjusted | |
| Condom Use for Insertive Anal Sexa | |||||
| Sample, no. | 713 | 713 | 703 | ||
| Wave | |||||
| 1 (Ref) | 54.7 (99/181) | 32.0 (58/181) | 13.3 (24/181) | 1.00 | 1.00 |
| 2 | 61.9 (117/189) | 29.6 (56/189) | 8.5 (16/189) | 1.39 (0.93, 2.08) | 1.34 (0.89, 2.02) |
| 3 | 59.1 (114/193) | 31.1 (60/193) | 9.8 (19/193) | 1.23 (0.82, 1.83) | 1.21 (0.80, 1.81) |
| 4 | 67.3 (101/150) | 22.7 (34/150) | 10.0 (15/150) | 1.67** (1.08, 2.59) | 1.51 (0.97, 2.36) |
| Linear trend | 1.15** (1.01, 1.31) | 1.12 (0.97, 1.29) | |||
| Condom Use for Receptive Anal Sexb | |||||
| Sample, no. | 665 | 665 | 663 | ||
| Wave | |||||
| 1 (Ref) | 50.8 (93/183) | 36.1 (66/183) | 13.1 (24/183) | 1.00 | 1.00 |
| 2 | 58.3 (98/168) | 33.3 (56/168) | 8.3 (14/168) | 1.38 (0.92, 2.08) | 1.45 (0.96, 2.19) |
| 3 | 58.9 (103/175) | 29.7 (52/175) | 11.4 (20/175) | 1.34 (0.90, 2.01) | 1.34 (0.89, 2.02) |
| 4 | 66.2 (92/139) | 22.3 (31/139) | 11.5 (16/139) | 1.76** (1.13, 2.74) | 1.82*** (1.17, 2.85) |
| Linear trend | 1.18** (1.03, 1.36) | 1.19** (1.03, 1.37) | |||
aOR adjusted for city where survey took place, employment, ever been to jail, ever tested for HIV, and sex with a female partner in past 2 months.
bOR adjusted for city where survey took place, employment status, and HIV status (positive, negative, or unknown, [don’t know, refused, or never tested]).
** P < .05; ***P < .01.
The adjusted linear trend per wave was significant for condom use during receptive anal intercourse, and the unadjusted results for condom use were significant for both insertive and receptive anal sex.
DISCUSSION
These data demonstrate high levels of HIV risk among this sample; 42.1% reported unprotected anal intercourse in the 2 months prior to assessment. At the final assessment wave, there were significant decreases in the proportion reporting (1) unprotected anal intercourse, (2) the number of partners for unprotected receptive anal intercourse, and (3) the mean number of partners for and episodes of unprotected sex. There also were significant increases in respondents reporting consistent condom use. The decrease in unprotected anal intercourse observed at 12 months (35.2% for unprotected insertive anal intercourse, 44.1% for unprotected receptive anal intercourse, and 31.8% for any unprotected anal intercourse) are comparable to findings from a previous evaluation of POL.22 We believe our study is the first that suggests that adapting already-proven interventions developed for other MSM can reduce risk among Black MSM.
High levels of risk among this sample underscore the importance of designing and testing interventions specifically for Black MSM. Twenty-five years since the first reported AIDS case, there has yet to be an intervention for Black MSM that has been rigorously evaluated, demonstrated to be effective, and reported in the literature. The Centers for Disease Control and Prevention (CDC) is currently funding prevention research activities for Black MSM. The studies include an evaluation of HIV-testing strategies for identifying at-risk Black MSM unaware of their HIV status, and efficacy trials of interventions to reduce the HIV risk of Black MSM. Because these study data are not yet available, it is important to simultaneously adapt interventions with proven efficacy, particularly those designed for MSM and Blacks. To date, the CDC has identified 8 evidence-based interventions for MSM and 24 for Blacks.39–41 None of the identified interventions were specifically designed for or tested among Black MSM.
Adapting these interventions will not be without challenges. For example, some adaptations of POL failed to produce significant effects compared with other evaluations.15,22,23,26–30 Kelly cites several reasons for this discrepancy.24 Flowers et al.29 used health educators rather than opinion leaders to conduct risk-reduction conversations. Elford et al.28 trained only a small cadre of peers (1.3%), far short of the 15% specified as a core element of POL (although POL has been shown to be efficacious with as little as 8% of the target population being trained as opinion leaders).22,24 In our study, 11% of the target population was trained as opinion leaders, although 15% of respondents reported becoming opinion leaders after the implementation of the intervention. Therefore, maintaining fidelity to an intervention’s core elements is important for successfully adapting interventions and was the case for our current adaptation.
A component of the intervention was to dispel myths about HIV/AIDS, including those that contribute to conspiracy beliefs. Such beliefs have been associated with inconsistent condom use among Black men.42 Hutchinson et al. recommend that interventions for Black MSM be designed to address conspiracy beliefs and existing interventions be culturally adapted.43
As we did in our study, interventionists should assess community and agency needs and challenges,44,45 and establish appropriate linkages between researchers, the target population, and community-based agencies46 prior to designing, implementing, or evaluating prevention strategies. Demographic similarities between the target population, lead investigators, and project staff may have been important to this intervention’s success.47 Therefore, trained individuals who are similar to the target population on key characteristics (race, gender, age, and sexual identity) should be given lead roles in designing, implementing, and evaluating HIV prevention activities for Black MSM. Employing these strategies will help ensure that adapted interventions are culturally relevant, meet the needs of community-based organizations, maintain fidelity to the original intervention, and dispel myths about HIV/AIDS.
Increased attention must be given to Black MSM in prevention research to understand their decisionmaking for condom use. During our study, the percentage of respondents inconsistently using condoms decreased by 30.3% and 38.2% for insertive and receptive anal intercourse, respectively. Approximately 20% of Black MSM in this sample had unknown or positive HIV test results. Given high levels of undiagnosed HIV infection and high-risk social and sexual networks of Black MSM,48 additional studies are warranted to understand why some, and especially those who rely on perceived serosorting practices, continue to engage in risky sex. Thoroughly understanding this will inform the design, implementation, evaluation, and future adaptations of already proven interventions for Black MSM.
Limitations
This study had several limitations. First, the study did not include a control group. Therefore, it is impossible to know whether the observed changes would have occurred without the presence of the intervention. Second, it is possible that these results are not generalizable in locales in which HIV prevention information and resources are more abundant and MSM communities are more visible. Third, these results are based on convenience samples recruited at nightclubs, and MSM who attend nightclubs may be more likely to engage in high-risk behaviors.5 Associations have been found between alcohol consumption, drug use (which may be likely in nightclubs), and unprotected anal intercourse.49,50 It is possible that very-high-risk Black MSM may be overrepresented in this sample. Black MSM who do not go to nightclubs are likely to be underrepresented. Therefore, our findings are not generalizable to all Black MSM. Fourth, the study relied on self-reported behaviors. Some Black MSM may have felt uncomfortable disclosing risky behaviors, under-reporting some behaviors and overreporting others. However, we attempted to limit this bias through the use of self-administered surveys on handheld computers. Research shows that respondents more readily report risky behaviors using computer technology rather than answering questions face to face.51,52 Nonetheless, the presented results are promising and encouraging. Further investigations of adapted interventions, particularly those with rigorous evaluations, will bolster these findings.
Conclusions
We believe this study is the first to provide evidence suggesting that adapting proven interventions for Black MSM can potentially significantly reduce risky behavior for HIV transmission and acquisition. Our adapted intervention involved the community and addressed important communal and cultural concerns of the target population. It is critical that interventions are designed and tested specifically for Black MSM. However, while primary intervention research is being conducted, available efficacious interventions should be adapted to address cultural concerns and realities for Black MSM. Ultimately, these strategies will be conducive to increasing available prevention interventions and to reducing the HIV risk of Black MSM.
Acknowledgments
This activity was supported by the Centers for Disease Control and Prevention and awarded to North Carolina Division of Public Health (grant U62/CCU423507-02).
The authors would like to acknowledge the following individuals for their contributions to the development and implementation of this study: Jacqueline Clymore, Erica Dunbar, Carolyn Guenther-Grey, H. Mac McCraw, Ann O’Leary, John Peebles, Tomas Rodriguez, Ron Stall, Craig Studer, Richard Wolitski and the North Carolina Men’s Health Initiative Study Team: Devin Baez, Corey Barr, Debra Bost, Kristina Clay-James, Bernard Davis, Jesse Duncan, Evelyn Foust, Phyllis Gray, Mohammed Griffin, Anthony Hannah, Nicholas Helms, Robert Hinton, Ron Jackson, Guy Jenkins, Wayne Johnson, David Jolly, Kenneth Jones, Steven Little, Jeffery Love, Timika Meekins-Terrell, Sonji Pass, Stanley Phillip, Randy Rogers, LaVerne Reid, LaHoma Romocki, Albert Sanders, Michael Scott, Christopher Turner, Terry Wang, Christopher Watson, and Omar Whiteside. We would also like to thank Cynthia Lyles for her critical review of this article. Most important, the authors thank the study participants.
Note. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
Human Participant Protection This study was reviewed by the Centers for Disease Control and Prevention and local human subjects review boards and determined to be exempt from review by the North Carolina Department of Health and Human Services institutional review board.
Peer Reviewed
Contributors K.T. Jones led the writing of this article and oversaw adaptation and field data-collection activities. P. Gray and D. Bost supervised field data collection and monitored intervention activities. Y.O. Whiteside assisted in the adaptation of the intervention and led the implementation and facilitation of intervention activities at 1 project site. T. Wang and W.D. Johnson led statistical analyses activities and helped with writing the article and interpreting results. E. Dunbar and E. Foust assisted with monitoring the funded activities.
References
- 1.Centers for Disease Control and Prevention. Racial/Ethnic Disparities in Diagnoses of HIV/AIDS–33 States, 2001–2004. MMWR Morb Mortal Wkly Rep. 2006;55(5):121–125. [PubMed] [Google Scholar]
- 2.Centers for Disease Control and Prevention. HIV prevalence, unrecognized infection, and HIV testing among men who have sex with men—five U.S. cities, June 2004–April 2005. MMWR Morb Mortal Wkly Rep. 2005;54:597–601. [PubMed] [Google Scholar]
- 3.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;35:526–536. [DOI] [PubMed] [Google Scholar]
- 4.Valleroy L, Secura G, MacKellar D, Behel S. High HIV and risk behavior prevalence among 23- to 29-year-old men who have sex with men in 6 US cities. Paper presented at: 8th Conference on Retroviruses and Opportunistic Infections; February 4–8, 2001; Chicago, IL. Abstract no. 211.
- 5.Centers for Disease Control and Prevention. HIV transmission among Black college student and non-student men who have sex with men—North Carolina, 2003. MMWR Morb Mortal Wkly Rep. 2004;53: 731–734. [PubMed] [Google Scholar]
- 6.Millett GA, Peterson JL, Wolitski RJ, Stall R. Greater risk for HIV infection of Black men who have sex with men: a critical literature review. Am J Public Health. 2006;96:1007–1019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2001. Atlanta, GA: Centers for Disease Control and Prevention; 2002.
- 8.Centers for Disease Control and Prevention. Unrecognized HIV infection, risk behaviors, and perceptions of risk among young Black men who have sex with men—six U.S. cities, 1994–1998. MMWR Morb Mortal Wkly Rep. 2002;51:733–736. [PubMed] [Google Scholar]
- 9.Hart T, Peterson JL, Community Intervention Trial for Youth Study Team. Predictors of risky sexual behavior among young African American men who have sex with men. Am J Public Health. 2004;94:1122–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Johnson WD, Holtgrave DR, McClellan WM, Flanders WD, Hill AN, Goodman M. HIV prevention research for men who have sex with men: a seven-year update. AIDS Educ Prev. 2005;17:568–589. [DOI] [PubMed] [Google Scholar]
- 11.Peterson JL, Coates TJ, Catania J, et al. Evaluation of an HIV risk reduction intervention among African-American homosexual and bisexual men. AIDS. 1996; 10:319–325. [DOI] [PubMed] [Google Scholar]
- 12.Centers for Disease Control and Prevention AIDS Community Demonstration Projects Research Group. Community-level HIV intervention in five cities: final outcome data from the CDC AIDS Community Demonstration Projects. Am J Public Health. 1999;89: 336–345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Dilley J, Woods W, Sabatino J, et al. Changing sexual behavior among gay male repeat testers for HIV: a randomized, controlled trial of a single-session intervention. J Acquir Immune Defic Syndr. 2002;30: 177–186. [DOI] [PubMed] [Google Scholar]
- 14.Kegeles SM, Hays RB, Coates TJ. The MPowerment Project: a community-level HIV prevention intervention for young gay men. Am J Public Health. 1996; 86:1129–1136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kelly JA, St Lawrence JS, Diaz YE, et al. HIV risk behavior reduction following intervention with key opinion leaders of population: an experimental analysis. Am J Public Health. 1991;81:168–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kelly JA, St Lawrence JS, Hood HV, Brasfield TL. Behavioral intervention to reduce AIDS risk activities. J Consult Clin Psychol. 1989;57:60–67. [DOI] [PubMed] [Google Scholar]
- 17.Koblin B, Chesney M, Coates T. 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; 364:41–50. [DOI] [PubMed] [Google Scholar]
- 18.Valdiserri RO, Lyter DW, Leviton LC, Callahan CM, Kingsley LA, Rinaldo CR. AIDS prevention in homosexual and bisexual men: results of a randomized trial evaluating two risk reduction interventions. AIDS. 1989;3:21–26. [PubMed] [Google Scholar]
- 19.Wolitski R, Gomez C, Parsons J. Effects of a peer-led behavioral intervention to reduce HIV transmission and promote serostatus disclosure among HIV-seropositive gay and bisexual men. AIDS. 2005;19(suppl. 1): S99–109. [DOI] [PubMed] [Google Scholar]
- 20.McKleroy VS, Galbraith J, Cummings B, et al. Adapting evidence-based behavioral interventions for new settings and target populations. AIDS Educ Prev. 2006;18(4 suppl A):59–73. [DOI] [PubMed] [Google Scholar]
- 21.Rogers EM. Diffusion of Innovations. New York, NY: Free Press; 1983.
- 22.Kelly JA, Murphy DA, Sikkema KJ, et al. Randomised, controlled, community-level HIV-prevention intervention for sexual-risk behaviour among homosexual men in US cities. Community HIV Prevention Research Collaborative. Lancet. 1997;350:1500–1505. [DOI] [PubMed] [Google Scholar]
- 23.Kelly JA, St Lawrence JS, Stevenson LY, et al. Community AIDS/HIV risk reduction: the effects of endorsements by popular people in three cities. Am J Public Health. 1992;82:1483–1489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kelly JA. Popular opinion leaders and HIV prevention peer education: resolving discrepant findings, and implications for the development of effective community programs. AIDS Care. 2004;16:139–150. [DOI] [PubMed] [Google Scholar]
- 25.Somerville G, Diaz S, Davis S, Coleman K, Taveras S. Adapting the Popular Opinion Leader intervention for Latino young migrant men who have sex with men. AIDS Educ Prev. 2006;18:137–148. [DOI] [PubMed] [Google Scholar]
- 26.Miller RL, Klotz D, Eckholdt HM. HIV prevention with male prostitutes and patrons of hustler bars: replication of an HIV prevention intervention. Am J Community Psychol. 1998;26:97–131. [DOI] [PubMed] [Google Scholar]
- 27.Sikkema KJ, Kelly JA, Winett RA, et al. Outcomes of a randomized community-level HIV prevention intervention for women living in 18 low-income housing developments. Am J Public Health. 2000;90:57–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Elford J, Bolding G, Sherr L. Peer education has no significant impact on HIV risk behaviours among gay men in London. AIDS. 2001;15:535–538. [DOI] [PubMed] [Google Scholar]
- 29.Flowers P, Hart GJ, Williamson LM, Frankis JS, Der GJ. Does bar-based, peer-led sexual health promotion have a community-level effect amongst gay men in Scotland? Int J STD AIDS. 2002;13:102–108. [DOI] [PubMed] [Google Scholar]
- 30.Kelly JA, Amirkhanian YA, Kabakchieva E, et al. Prevention of HIV and sexually transmitted diseases in high risk social networks of young Roma (Gypsy) men in Bulgaria: randomised controlled trial. BMJ. 2006; 333:1098–1101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Amirkhanian YA, Kelly JA, Kabakchieva E, McAuliffe TL, Vassileva S. Evaluation of a social network HIV prevention intervention program for young men who have sex with men in Russia and Bulgaria. AIDS Educ Prev. 2003;15:205–220. [DOI] [PubMed] [Google Scholar]
- 32.Guenther-Grey CA, Varnell S, Weiser JI, et al. Trends in sexual risk-taking among urban young men who have sex with men, 1999–2002. J Natl Med Assoc. 2005;97(suppl):38S–43S. [PMC free article] [PubMed] [Google Scholar]
- 33.Johnson WD, Hedges LV, Ramirez G, et al. HIV prevention research for men who have sex with men: a systematic review and meta-analysis. J Acquir Immune Defic Syndr. 2002;30(suppl 1):S118–S129. [PubMed] [Google Scholar]
- 34.Hightow LB, Leone PA, Macdonald PDM, McCoy SI, Sampson LA, Kaplan AH. Men who have sex with men and women: a unique risk group for HIV transmission on North Carolina College Campuses. Sex Transm Dis. 2006;33:585–593. [DOI] [PubMed] [Google Scholar]
- 35.O’Leary A, Jones KT. Bisexual men and heterosexual women: how big is the bridge? How can we know? [Letter]. Sex Transm Dis. 2006;33:594–595. [DOI] [PubMed] [Google Scholar]
- 36.Ross MW, Essien EJ, Williams ML, Fernandez-Esquer ME. Concordance between sexual behavior and sexual identity in street outreach samples of four racial/ethnic groups. Sex Transm Dis. 2003;30:110–113. [DOI] [PubMed] [Google Scholar]
- 37.Gravlee CC. Mobile computer-assisted personal interviewing with handheld computers: the Entryware System 3.0. Field Methods. 2002;14:322–336. [Google Scholar]
- 38.Lane SJ, Heddle NM, Arnold E, Walker I. A review of randomized controlled trials comparing the effectiveness of hand held computers with paper methods for data collection. BMC Med Inform Decis Mak. 2006;6:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Centers for Disease Control and Prevention. Compendium of HIV Prevention Interventions With Evidence of Effectiveness. Atlanta, GA: Centers for Disease Control and Prevention; 1999.
- 40.Kay LS, Crepaz N, Lyles CM. Update of the Compendium of HIV prevention interventions with evidence of effectiveness. Paper presented at: National HIV Prevention Conference; July 27–30, 2003; Atlanta, GA. Abstract no. TP-045.
- 41.Lyles C, Kay L, Crepaz N. Best-evidence interventions: findings from a systematic review of HIV behavioral interventions for US populations at high risk, 2000–2004. Am J Public Health. 2007;97:133–143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Klonoff EA, Landrine H. Do Blacks believe that HIV/AIDS is a government conspiracy against them? Prev Med. 1999;28:451–457. [DOI] [PubMed] [Google Scholar]
- 43.Hutchinson AB, Begley EB, Sullivan P, Clark HA, Boyett BC, Kellerman SC. Conspiracy beliefs and trust in information about HIV/AIDS among minority men who have sex with men [letter]. J Acquir Immune Defic Syndr. 2007;45:603–605. [DOI] [PubMed] [Google Scholar]
- 44.Gandelman A, DeSantis L, Rietmeijer C. Assessing community needs and agency capacity—an integral part of implementing effective evidence-based interventions. AIDS Educ Prev. 2006;18(suppl A):32–43. [DOI] [PubMed] [Google Scholar]
- 45.Hitt J, Robbins A, Galbraith J, et al. Adaptation and implementation of an evidence-based prevention counseling intervention in Texas. AIDS Educ Prev. 2006;18(4 suppl A):108–118. [DOI] [PubMed] [Google Scholar]
- 46.Eke A, Neumann M, Wilkes A, Jones P. Preparing effective behavioral interventions to be used by prevention providers: the role of researchers during HIV prevention research trials. AIDS Educ Prev. 2006;18 (4 suppl A):45–48. [DOI] [PubMed] [Google Scholar]
- 47.Durantini MR, Albarracin D, Mitchell AL, Earl NA, Gillette JC. Conceptualizing the influence of social agents of behavior change: a meta-analysis of the effectiveness of HIV-prevention interventionists for different groups. Psychol Bull. 2006;132:212–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Bingham TA, Harawa NT, Johnson DF, Secura GM, MacKellar DA, Valleroy LA. The effect of partner characteristics on HIV infection among African American men who have sex with men in the Young Men’s Survey, Los Angeles, 1999–2000. AIDS Educ Prev. 2003; 15:39–52. [DOI] [PubMed] [Google Scholar]
- 49.Celentano DD, Valleroy LA, Sifakis F, et al. Associations between substance use and sexual risk among very young men who have sex with men. Sex Transm Dis. 2006;33:265–271. [DOI] [PubMed] [Google Scholar]
- 50.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;93:926–932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Newman JC, Des Jarlais DC, Turner CF, Gribble J, Cooley P, Paone D. Differential effects of face-to-face and computer interview modes. Am J Public Health. 2002;92:294–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Williams ML, Freeman RC, Bowen AM, et al. Comparison of the reliability of self-reported drug use and sexual behaviors using computer-assisted versus face-to-face interviewing. AIDS Educ Prev. 2000;12: 199–213. [PubMed] [Google Scholar]