Abstract
Objectives. We examined the characteristics of young Internet-using men who have sex with men (MSM) and risks associated with seeking sex online, offline, or through both strategies.
Methods. Data were obtained from MSM aged 18 to 24 years who completed a 45-minute online survey regarding sex and Internet use in the preceding 3 months.
Results. Significantly more Internet-using MSM who had met sexual partners both online and offline (43%) reported unprotected anal intercourse than did those who had met sexual partners exclusively online (29%) or offline (34%). MSM who met sexual partners exclusively offline reported the fewest partners but the greatest proportion of partnerships involving unprotected anal intercourse (49%). Meeting sexual partners both online and offline (odds ratio [OR]=3.38–58.42) and being drunk (OR=1.57) or high (OR=2.24) increased the odds of having more sexual partners. The same factors increased the odds of having unprotected anal intercourse (online and offline sexual partners, OR=1.60; being drunk, OR=1.43; being high, OR=1.61).
Conclusions. Risky sexual behavior was prevalent among all of the study subgroups. Our findings suggest that online sex seeking is associated with greater numbers of sexual partners but neither promotes nor discourages unprotected anal intercourse. Regardless of where sexual partners met, being drunk and high were significant risks for unprotected anal intercourse.
Recent reports show rising rates of HIV among men who have sex with men (MSM),1–4 including young adult MSM.5,6 In 2003, male–male sexual contact accounted for 74% of HIV diagnoses among young male adults between the ages of 13 and 24 years in the United States, and the estimated annual number of HIV cases among young men rose from 1763 in 1999 to 2443 in 2003.7 Many younger adults are unaware of their HIV-seropositive status and inadvertently infect their sexual partners.8,9 Rising HIV rates have led to calls for intensified prevention efforts targeting young people7,10 and a better understanding of contributing factors.
Well-recognized covariates of unprotected anal intercourse among MSM include alcohol and drug use11,12 and a heightened state of sexual arousal.13,14 Recently, the emergence of online sex seeking among MSM has led researchers to speculate that Internet use might be fueling HIV risk behaviors.15,16 A higher percentage of gay and bisexual men (approximately 40% across studies17) than of heterosexual men (4.4%–52%18–20) go online to meet sexual partners, and epidemiological investigations have traced the spread of HIV and other sexually transmitted infections among MSM who met online.21,22
In a recent meta-analysis of studies involving Internet-using MSM, the authors concluded that online sex seekers, especially HIV-seropositive MSM, were more likely than were non-Internet sex seekers to engage in unprotected anal intercourse,17 the behavior most likely to transmit HIV. However, studies of Internet users have not adequately examined differences in the type and frequency of behaviors with partners met online and offline.23 Thus, it remains unclear whether the unique features of the Internet promote risky behavior beyond those associated with non-Internet liaisons.17
Although youths are “early adopters” of new technologies such as the Internet,24 few studies24–27 have examined the characteristics of younger users or the Internet’s impact on their sexual behaviors. In a sample of young MSM recruited through a community-based health center,25 sex with a partner met online was common (42%), and, consistent with other studies,28 meeting sexual partners through the Internet was associated with a number of demographic (e.g., younger age, non-Black race) and drug use factors.
To our knowledge, no published studies have specifically examined the sexual risk behavior of young Internet-using MSM drawn from an exclusively online sample. In previous reports, Internet users often have been treated as a single homogeneous population; differences in duration, frequency, and other characteristics of use have not been taken into account.15,28 In the present study, we compared the sexual activities of Internet-using MSM aged 18 to 24 years who met their sexual partners exclusively online, exclusively offline, or via both strategies. Our primary research questions were as follows: do the 3 MSM groups have different sociodemographic and behavioral characteristics, and is the method used to meet sexual partners associated with numbers of partners and likelihood of unprotected anal intercourse?
We used survey data collected in the development of an online sexual risk reduction intervention. Its conceptual framework was the sexual health model,29 according to which risk reduction is promoted through comprehensive, culturally specific sexuality education.
METHODS
Eligibility and Recruitment
In our study, we included data from a subset of men aged 18 to 24 years who had completed an online survey and reported having engaged in sexual relations with a man in the preceding 3 months. Initially, all MSM who were US residents (and aged at least 18 years) were eligible to participate. Ethnicity and race eligibility criteria and banner advertisements were adjusted to oversample Black, Hispanic/ Latino, Asian, and “other” participants of color (i.e., American Indian and multiracial) with the goal of recruiting up to 750 men from each group.
Participants were recruited during 3 months in 2005 via banner advertisements placed on Gay.com, one of the most highly subscribed gay Web sites in the United States. (When the study began, the site had approximately 5 million registered members and 6 million unique monthly visitors [M. Latham, senior client services manager, PlanetOut Inc, oral communication, December 2002].) Registrants could view profiles, access chat rooms, browse articles at no cost, and pay for additional functionality. Initially $10 was offered for survey completion, with remuneration later raised to $20 to accelerate recruitment.
Procedures
Participants clicking on a study banner advertisement were transported to a study Web site secured with 128-bit encryption. Prospective participants viewed a welcome page with an overview of procedures and information about the study and staff. Participants responded to 7 questions to determine their eligibility (Table 1 ▶).
TABLE 1—
Survey Items and Response Options: Men Who Have Sex With Men (MSM) Online Survey, 2005
| Item | Response Options | |
| Gender a | What gender are you? | Male, female, male-to-female transgender, female-to-male transgender, intersexual |
| Race/ethnicity a | What ethnicity are you? What race are you? |
Hispanic or non-Hispanic; American Indian or Alaskan Native, Asian American, Black or African American, Native Hawaiian or other Pacific Islander, White, other |
| Agea | How old are you? | Open-ended response box |
| Sexual activity with mena | Have you had sex, meaning any kind of sexual contact, with at least 1 other man? | Yes or no |
| Residencea | Where do you live? | US (including any US territory) or outside US |
| Previously completed surveya | How many times have you filled out this survey in the past 2 months? | Open-ended number of times (0 to qualify) |
| Sexual orientation/behavior | In the last 3 years, in terms of my sexual orientation, I see myself as attracted to . . . In the last 3 years, in terms of my sexual behavior, I have had sex with . . . |
1 = only men, 4 = equally men and women, 7 = only women |
| Area of residence | How would you describe the town or community where you live? | Rural (farms and small towns under 5000 people), small town (5001 to 50 000 people), medium-sized city (50 001 to 200 000 people), suburb of a large city (more than 200 000 people), downtown area or central district of a large city (more than 200 000 people) |
| Education | How many years of school have you completed? | Open-ended response box |
| Religiosity | How religious are you? | 1 = not religious at all, 5 = very religious |
| Internet use | In what year did you first start using the Internet for any task? | Drop down menu of years |
| What kind of Internet connection are you using? | Dial-up telephone connection, high speed (e.g., cable, DSL), other (with open-ended response box), do not know | |
| Please think back over the last 7 days. Approximately how many hours did you spend actively using the Internet for the following activities? | Work or education activities, sex-related activities, personal activities (open-ended response boxes) | |
| HIV testing status | When was your most recent test for HIV? | Drop down menu of years; I have never been tested for HIV |
| Have you ever been diagnosed with HIV? | Yes, more than 12 months ago; Yes, I was diagnosed in the last 12 months; No, never | |
| Sexual activity (parallel questions asked for sexual activities with partners not met via the Internet) | In the last 3 months, how many persons have you met online and then in person for any kind of sex? | Open-ended response boxes (male, female, transgender) |
| In the last 3 months, how many persons have you met online and then in person with whom you had vaginal or anal sex? | Open-ended response boxes (male, female, transgender) | |
| In the last 3 months, how many persons have you met online and then in person with whom you had unprotected vaginal or anal sex? | Open-ended response boxes (male, female, transgender) | |
| Substance use (parallel questions asked for sexual activities with partners not met via the Internet) | In the last 3 months, with men you met online, have you had sex when you were drunk with alcohol? | Yes or no |
| In the last 3 months, with men you met online, have you had sex when you used other drugs (e.g., poppers, crystal methamphetamine, Ecstasy, marijuana)? | Yes or no | |
| Sexual arousal | During sex you had with your most recent partner, is the following statement accurate: I felt I needed to have sex so badly I didn’t care. | Yes or no |
aEligibility screening item.
Eligible respondents were invited to create a username and password for reentry to the survey Web site. Ineligible individuals were transported to another Web page that thanked them for their interest. A “cookie” was stored on their computer, and an Internet Protocol (IP) address was captured to help prevent them from using different eligibility information in an attempt to reenter the study.
Measures
The survey, adapted from a study of Internet use among Latino MSM, included 170 questions regarding Internet use and sexual attitudes and behaviors.30 Algorithms were used to ask participants a variable number of questions depending on their responses. Survey sections were presented randomly to minimize order effects. Although each question contained a “refuse to answer” option, participants were required to answer all questions. Participants who did not complete the survey within 24 hours after they had started it were sent an automated reminder. The mean online survey completion time was 45 minutes. Automated and manual deduplication and validation protocols were applied to ensure that each completed survey represented a unique respondent.31
Analyses
Stata was used in conducting the statistical analyses.32 Demographic characteristics (e.g., race/ethnicity and residence), patterns of Internet use, and known covariates of risky sexual behavior among adolescents and MSM (i.e., substance use and sexual arousal) were assessed. The sexual health model29 and previous research11–14,17, 25 guided our choice of covariates. To examine the relationship between Internet use and sexual behaviors, we grouped respondents into 1 of 3 mutually exclusive categories based on self-reported sexual encounters with men in the preceding 3 months: (1) online exclusively (met partners through the Internet), (2) offline exclusively (met partners through non-Internet venues), and (3) both online and offline (online–offline group).
For the purpose of our analyses, we recoded responses to several variables. Religiosity was coded as low (1), medium (2–4), or high (5). HIV testing status was coded as “never tested,” “tested in the past 1.5 years,” or “tested more than 1.5 years ago.” Number of hours of Internet use in the preceding 7 days was coded as 0 through 12, 13 through 24, 25 through 42, and 43 or more. Weekly number of hours of Internet use for sexual purposes was coded as 0, 1 through 3, 4 through 10, and 11 or more. Number of male sexual partners in the previous 3 months was coded as 1, 2 or 3, 4 through 6, and 7 or more.
In addition to calculating medians and means with respect to participants’ sexual activity and risk behavior in the preceding 3 months, we calculated the proportion of anal intercourse partners with whom participants had engaged in unsafe sex as another method of assessing risk. Among men who reported at least 1 male partner with whom they had engaged in anal intercourse during the preceding 3 months (n = 587), percentage of unprotected anal intercourse was calculated as follows: (number of unprotected anal intercourse male sexual partners /number of anal intercourse male sexual partners) × 100.
We used the χ2 test and analyses of variance to examine differences in subgroup characteristics, HIV-antibody testing, and Internet use. Because the distribution of responses to items focusing on sexual behavior was not normal even after attempts at transformation, we used the Mann–Whiney and Wilcoxon signed rank tests to estimate group differences. In addition, we conducted a hierarchical cross-sectional regression analysis to examine group differences in log-transformed percentages of unprotected anal intercourse.
Factors that were significant in the bivariate analyses were entered into 2 logistic regression analyses assessing the main outcome variables of interest: number of male sexual partners and frequency of unprotected anal intercourse in the preceding 3 months. Because number of male sexual partners was an ordinal variable, we used the gologit2 command34 to construct an unconstrained partial proportional odds model.33 Robust standard errors were used in fitting models to adjust for possible interclass correlations. The α level was set at .05.
RESULTS
Banner advertisements were displayed a total of 98790803 times during the study period, and 62 257 clicks were recorded. Screening of items showed that 7547 men were eligible for the study: 6076 consented to the procedures, 4580 began the survey, and 3037 finished the survey. The validation protocol showed that 2722 unique respondents provided complete data and formed our study sample: 979 were younger than 25 years, and 770 had been sexually active with men in the preceding 3 months. This latter group formed our study sample.
Sample Characteristics
Sample characteristics are shown in Table 2 ▶. The mean age of the men was 21.5 years (SD = 1.8); 30% were aged between 18 and 20 years. In comparison with the other 2 groups, the offline group spent significantly fewer hours online for sexual purposes (χ23 = 22.51; P < .01) or other reasons (χ23 = 13.91; P < .05).
TABLE 2—
Sociodemographic and Behavioral Characteristics of the Total Sample and Subgroups: Men Who Have Sex With Men (MSM) Online Survey, 2005
| Total (n = 770) | Online Only (n = 286) | Offline Only (n = 112) | Online—Offline (n = 372) | χ2 or Fa | |
| Race/ethnicity, no. (%) | 10.37 | ||||
| Asian | 124 (16) | 59 (21) | 16 (14) | 49 (13) | |
| Black | 130 (17) | 51 (18) | 20 (18) | 59 (16) | |
| Latino | 216 (28) | 71 (25) | 29 (26) | 116 (31) | |
| White | 202 (26) | 73 (26) | 33 (30) | 96 (26) | |
| Other | 98 (13) | 32 (11) | 14 (13) | 52 (14) | |
| Area of residence, no. (%) | 5.19 | ||||
| Rural | 40 (5) | 14 (5) | 9 (8) | 17 (5) | |
| Small town | 137 (18) | 55 (20) | 19 (17) | 63 (17) | |
| Medium city | 206 (27) | 81 (29) | 27 (24) | 98 (27) | |
| Suburb | 195 (26) | 71 (25) | 30 (27) | 94 (25) | |
| Urban | 185 (24) | 60 (21) | 27 (24) | 98 (27) | |
| Level of religiosity, no. (%) | 6.92 | ||||
| Low | 234 (30) | 82 (29) | 32 (29) | 120 (32) | |
| Medium | 480 (63) | 189 (67) | 73 (65) | 218 (59) | |
| High | 53 (7) | 13 (5) | 7 (6) | 33 (9) | |
| HIV positive, no. (%) | 12 (2) | 5 (2) | 3 (3) | 4 (1) | 1.51 |
| HIV testing status, no. (%) | 7.78 | ||||
| Never tested | 199 (26) | 90 (32) | 24 (21) | 85 (23) | |
| Tested in past 1.5 y | 465 (61) | 161 (56) | 70 (63) | 234 (63) | |
| Tested but not in past 1.5 y | 104 (14) | 35 (12) | 18 (16) | 51 (14) | |
| Type of Internet connection, no. (%) | 5.94 | ||||
| Dial up | 82 (11) | 29 (10) | 17 (16) | 36 (10) | |
| High speed | 667 (88) | 251 (89) | 90 (82) | 326 (88) | |
| Other | 13 (2) | 2 (<1) | 3 (3) | 8 (2) | |
| Hours of Internet use in past week, no. (%) | 13.91* | ||||
| 0–12 | 173 (23) | 66 (23) | 35 (31) | 72 (19) | |
| 13–24 | 202 (26) | 65 (23) | 35 (31) | 102 (27) | |
| 25–42 | 200 (26) | 74 (26) | 20 (18) | 106 (28) | |
| ≥ 43 | 193 (25) | 79 (28) | 22 (20) | 92 (25) | |
| Hours of Internet use for sex in past week, no. (%) | 22.51** | ||||
| 0 | 61 (8) | 25 (9) | 18 (16) | 18 (5) | |
| 1–3 | 230 (30) | 83 (29) | 41 (37) | 106 (28) | |
| 4–10 | 297 (39) | 108 (38) | 37 (33) | 152 (41) | |
| ≥ 11 | 182 (24) | 70 (24) | 16 (14) | 96 (26) | |
| Years of schooling, mean (SD) | 14.25 (2.30) | 14.27 (2.50) | 13.96 (2.27) | 14.33 (2.15) | 1.14 |
| Sexual orientation score,b mean (SD) | 1.77 (1.05) | 1.69 (0.98) | 1.80 (1.04) | 1.83 (1.10) | 1.56 |
| Sexual behavior score,b mean (SD) | 1.44 (0.96) | 1.26c (0.75) | 1.42 (0.97) | 1.58c (1.07) | 9.14*** |
| No. of years of Internet use, mean (SD) | 9.54 (2.49) | 9.55 (2.45) | 9.54 (2.68) | 9.53 (2.46) | 0.006 |
aχ2 for percentages; F for means.
bOn a 7-point Likert scale (1 = only men, 7 = only women).
cSignificant difference between groups.
* P < .05; **P < .01; ***P < .001.
Sexual Activity in the Preceding 3 Months
Seventy-six percent (587 of 770) of the men reported having engaged in anal intercourse in the previous 3 months, and 36% (278 of 770) reported having engaged in unprotected anal intercourse. Table 3 ▶ shows overall, online, and offline numbers of male sexual partners, anal intercourse partners, and unprotected anal intercourse partners. Overall, participants reported a median of 3 male sexual partners and 2 male anal intercourse partners. The respective mean and median numbers of unprotected anal intercourse sexual partners were 1.4 (SD= 6.9) and 0. In the case of all of the participants, higher median numbers of sexual partners, anal intercourse partners, and unprotected anal intercourse partners were met online than offline (P< .001).
TABLE 3—
Types of Sexual Partners in the Preceding 3 Months Among the Total Sample and Subgroups: Men Who Have Sex With Men (MSM) Online Survey, 2005
| Total Sample (n = 770) | Online Only (n = 286) | Offline Only (n = 112) | Online–Offline (n = 372) | |||||
| Median | Mean (SD) | Median | Mean (SD) | Median | Mean (SD) | Median | Mean (SD) | |
| All sexual partners | ||||||||
| Any sexual activity | 3.00 | 6.33 (10.74) | 2.00 | 4.23 (9.78) | 1.00 | 1.96 (1.91) | 5.00 | 9.26 (12.12) |
| Anal intercourse | 2.00 | 3.95 (9.39) | 1.00 | 2.45 (9.28) | 1.00 | 0.96 (1.09) | 3.00 | 5.96 (10.40) |
| Unprotected anal intercourse | 0 | 1.40 (6.87) | 0 | 1.25 (9.18) | 0 | 0.38 (0.57) | 0 | 1.81 (9.19) |
| Partners met online | ||||||||
| Any sexual activity | 2.00 | 4.18 (8.32) | 2.00 | 4.23 (9.78) | . . . | . . . | 3.00 | 5.41 (7.95) |
| Anal intercourse | 1.00 | 2.61 (7.55) | 1.00 | 2.45 (9.28) | . . . | . . . | 1.00 | 3.51 (7.00) |
| Unprotected anal intercourse | 0 | 0.97 (6.27) | 0 | 1.25 (9.18) | . . . | . . . | 0 | 1.04 (4.06) |
| Partners met offline | ||||||||
| Any sexual activity | 1.00 | 2.15 (4.40) | . . . | . . . | 1.00 | 1.96 (1.91) | 2.00 | 3.85 (5.70) |
| Anal intercourse | 0 | 1.34 (3.44) | . . . | . . . | 1.00 | 0.96 (1.09) | 1.00 | 2.45 (4.65) |
| Unprotected anal intercourse | 0 | 0.43 (1.71) | . . . | . . . | 0 | 0.38 (0.57) | 0 | 0.77 (2.38) |
Note. Significant group differences calculated with Mann–Whitney U test (P < .005). For all sexual partners, the online–offline group reported significantly more of any sexual activity, anal intercourse, and unprotected anal intercourse than the online only or offline only groups. The online only group reported more of any sexual activity and anal intercourse than the offline only group. For partners met online, the online–offline group reported significantly more of any sexual activity and anal intercourse than the online only group. For partners met offline, the online–offline group reported significantly more of any sexual activity, anal intercourse, and unprotected anal intercourse than the offline only group.
Eighty-six percent (320 of 372) of the online–offline group, 69% (197 of 286) of the online group, and 63% (70 of 112) of the offline group reported having engaged in anal intercourse in the previous 3 months (χ22 = 39.86; P < .001; data not shown). A higher percentage of online–offline group members (43%) than offline (34%) and online (29%) group members reported having engaged in unprotected anal intercourse during the preceding 3 months (χ22 = 14.70; P < .001).
Group differences in median numbers of male sexual partners were found. Men in the online–offline group reported significantly more partners than men in the online (any sexual partners: U= 4007.5; anal intercourse partners: U= 8720; unprotected anal intercourse partners: U= 18 038; P< .001 for each) and offline (any sexual partners: U= 24894.5; anal intercourse partners: U= 31823; unprotected anal intercourse partners: U= 43 618.5) groups (P< .001 for each). The online group reported more male sexual partners (U= 11092) and anal intercourse partners (U= 13059.5) than did the offline group (P< .005); there were no significant differences in numbers of unprotected anal intercourse partners.
Unprotected Anal Intercourse in the Preceding 3 Months
On average, participants reported having engaged in unprotected anal intercourse with one third of their anal intercourse partners. In the total sample, the percentage of unprotected anal intercourse as a proportion of all sexual partners did not differ according to whether partners had been met online (30.4%) or offline (36.3%; b = 0.042; SE = 0.048; P = .38). However, a significant group effect was detected, with a higher percentage of unprotected anal intercourse in the offline group (48.9%) than in the online (31.1%) or online–offline (30.7%) group (χ22 = 6.30; P < .05).
Substance Use and Sexual Arousal
A significantly larger percentage of participants in the online–offline group than the online group reported being drunk (18% vs 34%; χ21 = 22.64; P < .001) or high (11% vs 21%; χ21 = 13.62; P < .001) during sex in the preceding 3 months with partners met online. Likewise, a larger percentage of men in the online–offline group than in the offline group reported being high during sex with partners met offline (21% vs 9%; χ21 = 8.03; P=.005). Similar percentages of offline (39%) and online–offline (42%) group members reported having been drunk during sex with a partner met offline.
With respect to sexual arousal, 20% (149 of 754) of the participants indicated agreement with the statement, “I felt I needed to have sex so badly I didn’t care.” Significantly fewer members of the offline group (9%) than either the online (24%) or online–offline (20%) group (χ22 = 11.93; P = .003) indicated agreement with this statement.
Bivariate Analyses
Of the online–offline group, 71% reported having had 4 or more sexual partners in the preceding 3 months, as opposed to 30% of the online group and 13% of the offline group (χ23 = 262.53; P < .001). Having a greater number of male sexual partners was significantly associated with spending more time online for any purpose (χ23 = 23.85; P< .01) and specifically for sexual purposes (χ23 = 72.14; P < .001), as well as being drunk (χ23 = 41.97; P < .001) or high (χ23 = 68.28; P< .001) during sex, experiencing high levels of sexual arousal (χ23 = 9.41; P < .05), and meeting partners both online and offline (χ23 = 262.53; P < .001).
Engaging in unprotected anal intercourse was significantly associated with spending more time online in general (χ23 =14.81; P <.01) and specifically for sexual purposes (χ23 =17.88; P<.001), being drunk (χ21 =16.02; P < .001) or high (χ21 = 16.97; P < .001) during sex, and meeting partners both online and offline (χ22 = 14.70; P < .01). Frequency of sex with male partners and frequency of unprotected anal intercourse were not associated with race/ethnicity, area of residence, religiosity, HIV testing status, or age.
Mulivariate Analyses
Variables included in the model focusing on number of male sexual partners in the preceding 3 months (an ordinal outcome) are shown in Table 4 ▶. The overall model was significant (χ21 = 254.98; P < .001). With the exception of the online–offline group variable, other predictors did not violate the proportional odds assumption. Online–offline group membership significantly increased the odds of having higher numbers of sexual partners for each of the comparisons shown in Table 4 ▶. Members of the offline group reported fewer partners than did members of the online group. Men who spent 4 to 10 hours per week and 11 or more hours per week seeking sex online had significantly more sexual partners than did those who did not use the Internet for sexual purposes. Being drunk or high during sex also increased the odds of having more sexual partners.
TABLE 4—
Multivariate Analyses of Number of Male Sexual Partners and Frequency of Unprotected Anal Intercourse in the Preceding 3 Months: Men Who Have Sex With Men (MSM) Online Survey, 2005
| Odds Ratio (95% Confidence Interval) | |
| No. of sexual partners | |
| Online group (Ref) | 1.00 |
| Offline group | 0.41*** (0.25, 0.66) |
| Online–offline groupa | |
| 1 vs 2–3, 4–6, ≥7 | 58.42*** (18.00, 189.63) |
| 1, 2–3 vs 4–6, ≥7 | 5.19*** (3.64, 7.42) |
| 1, 2–3, 4–6 vs ≥7 | 3.38*** (2.27, 5.03) |
| Weekly hours of Internet use | |
| 0–12 (Ref) | 1.00 |
| 13–24 | 0.84 (0.57, 1.23) |
| 25–42 | 0.95 (0.62, 1.46) |
| ≥ 43 | 0.95 (0.59, 1.52) |
| Weekly hours of Internet use for sex | |
| 0 (Ref) | 1.00 |
| 1–3 | 0.94 (0.56, 1.58) |
| 4–10 | 2.19** (1.29, 3.72) |
| ≥ 11 | 2.99*** (1.63, 5.49) |
| High during sex | 2.24*** (1.52, 3.29) |
| Drunk during sex | 1.57** (1.16, 2.13) |
| Experienced intense sexual arousalb | 1.37 (0.95, 1.98) |
| Unprotected anal intercourse | |
| Group | |
| Online (Ref) | 1.00 |
| Offline | 1.42 (0.84, 2.39) |
| Online–offline | 1.60** (1.13, 2.27) |
| Weekly hours of Internet use | |
| 0–12 (Ref) | 1.00 |
| 13–24 | 0.79 (0.49, 1.26) |
| 25–42 | 1.38 (0.85, 2.24) |
| ≥ 43 | 0.87 (0.51, 1.48) |
| Weekly hours of Internet use for sex | |
| 0 (Ref) | 1.00 |
| 1–3 | 0.75 (0.40, 1.41) |
| 4–10 | 1.01 (0.55, 1.87) |
| ≥ 11 | 1.57 (0.80, 3.08) |
| High during sex | 1.61* (1.06, 2.44) |
| Drunk during sex | 1.43* (1.01, 2.01) |
| Experienced intense sexual arousalb | 1.32 (0.90, 1.93) |
Note. Robust estimates are reported.
aProportional odds assumption violated for online–offline group; odds ratios reported for odds of being in higher-outcome group.
bSexual arousal was measured by whether participants indicated agreement with the statement, “I felt I needed to have sex so badly I didn’t care.”
* P < .05; **P < .01; ***P < .001.
The same predictor variables were entered into a multivariate logistic model in which any unprotected anal intercourse in the preceding 3 months was the dichotomous outcome variable. The overall model was significant (χ21 =54.87; P <.001), and being drunk, high, and a member of the online–offline group predicted having unprotected anal intercourse.
DISCUSSION
This is one of the first empirical studies of MSM who were sampled online. As has been the case in other online surveys,10,35 we successfully reached a large and ethnically diverse group of young MSM who use the Internet. Although 10% of the US male population is aged 18 to 24 years,36 men in this age range represented more than one third of our survey respondents. This proportion is consistent with the results of international studies showing that younger MSM are more likely than older MSM to seek sex online and respond to online surveys.37,38
We grouped MSM into 3 categories according to the way they met their sexual partners. Despite similar demographic characteristics, their sexual behavior patterns differed. The largest was the online–offline group (i.e., those who met partners through both venues), whose members reported more sexual partners than did those in the other 2 groups (online exclusively and offline exclusively) and were more likely to have been intoxicated during sex at least once in the preceding 3 months. In addition, consistent with a previous study of adult MSM,39 the percentage of unprotected anal intercourse in the preceding 3 months was higher among men in the online–offline group (43%) than among men who met their sexual partners exclusively online or offline.
Having more ways to meet sexual partners (i.e., both online and offline) may create more opportunities for safe as well as unsafe sex. Other investigators have found that MSM who seek sex in multiple venues (e.g., bathhouses and public cruising areas) are at greater risk for engaging in unprotected anal intercourse with nonprimary partners than are men seeking sex in only 1 venue.40 Use of alcohol and other substances in offline settings such as bars can impair one’s judgment regarding risk reduction. Perhaps “sensation seeking”41,42 (i.e., the willingness to take risks to experience novel and intense sensations) or other personality characteristics43–45 not measured in our study can explain the high levels of unprotected anal intercourse and substance abuse in this group. For these reasons, young MSM who seek sex in multiple venues appear to be a particularly important population for prevention services, including online interventions.
Men in the online and offline groups reported meeting fewer sexual partners than did men in the online–offline group. The online group fell between the other 2 groups in terms of likelihood of having ever engaged in unprotected anal intercourse (29%) and likelihood of having engaged in unprotected anal intercourse (31%) in the preceding 3 months. Surprisingly, the percentage of unprotected anal intercourse was highest in the offline group (49%), even though this group’s members had the fewest anal intercourse partners and had numbers of unprotected anal intercourse partners similar to those of the online group.
Thus, in comparison with members of the online–offline group, who had high numbers of sexual and unprotected anal intercourse partners, members of the offline group had notably fewer partners but were at proportionally higher risk of unprotected anal intercourse. Meeting partners offline might adversely affect condom use because it allays concerns about transmission of HIV/AIDS or inhibits frank communication about risk reduction.
As in previous studies,28,46 online sex seeking was associated with higher numbers of male partners, probably because this strategy is an efficient way to make contact. Amount of time spent online on sexual activities predicted higher numbers of sexual partners but not an increased likelihood of unprotected anal intercourse. An earlier study similarly showed that MSM were more likely to use a condom with partners they had met online than with those they met offline (71% vs 57%),19 whereas another recent investigation revealed no difference in rates of unprotected anal intercourse.47
Our findings add to the growing evidence that the Internet facilitates meeting sexual partners without necessarily promoting unprotected anal intercourse. However, we cannot conclude that Internet use is protective, because online–offline group members reported that they engaged in unprotected anal intercourse more with partners met online than with those met offline. There is a need for further research on longitudinal changes in the amount of time men spend seeking sex online in relation to their sexual behaviors with sexual partners met online.
As with studies of young MSM sampled offline,48,49 we found an association between substance use and risky sexual behavior. Other investigators have noted high rates of club drug and methamphetamine use among Internet-using MSM and an association between drug use and risky sex.25,50,51 These observations are strong reminders that alcohol and drug use are powerful cofactors of disease transmission, especially when they operate in combination with other risk variables. Future studies of young MSM should examine the dynamics of substance use and sexual risk behavior in the context of specific sexual situations (i.e., event-level analyses).
A small number of the men in our sample self-identified as HIV seropositive, fewer than expected from the findings of the Young Men’s Survey, a venue-based survey conducted from 1994 through 1998 in 7 urban centers.52 The prevalence of HIV infection in that survey was 7.2% (range=2.2%–12.1%); 35% of the participating young MSM had not been tested previously, and only 18% of the HIV-positive men in the sample were aware of their serostatus.52 Similarly, the men in our sample may have been unaware of or misinformed about their serostatus. As with other online studies,47,53,54 we found suboptimal levels of HIV testing (i.e., 26% of the participants had not been tested) despite relatively high levels of unprotected anal intercourse. The Centers for Disease Control and Prevention estimate that one quarter of US residents who are HIV positive are unaware of their status and recommend that those who are “at high risk for HIV infection” be screened at least annually.55
Study Limitations
Various limitations of our study should be noted. First, the study’s cross-sectional design precludes causal inferences about the impact of our independent variables on the outcomes assessed. Second, our findings do not apply to MSM who were excluded from the study because they abstained from sex or did not use the Internet. Because all of the participants were sampled online, our offline group may have been smaller than is actually the case in the general population.
Third, although banner advertisements were randomly placed, participants were not randomly selected, and thus the extent to which they represent young Internet-using MSM as a whole is unknown. Providing a (modest) financial incentive for participation may have biased the sample toward young people in financial need. However, the 3-month incidence of unprotected anal intercourse (36%)—one of our primary outcomes—was within the range detected via a different time–place method used with young MSM in 6 US cities (i.e., 14%–39%),56 suggesting that the finding is reliable. Finally, although precautions were taken to detect and eliminate deception,31 we relied on self-reported data. Other research has shown that results associated with computer-assisted data collection are comparable to results derived from face-to-face interviews57,58 and that online self-reports of health information are valid.59
Implications
The Internet is a popular and easy way for MSM to meet sexual partners,17 and our findings suggest that online sex seeking neither promotes nor discourages unprotected anal intercourse. Future research and interventions should recognize different subgroups of young MSM who use the Internet rather than dichotomizing Internet users and nonusers. Nonrecognition of these groups obscures the different ways in which risk may be incurred, be it through relatively large numbers of safe-sex partners or lower numbers of unprotected anal intercourse partners.
There is a need and a demand for online health promotion60–62 and disease prevention services,22,37,63–66 and the Internet creates an opportunity to access large numbers of otherwise difficult-to-reach and vulnerable people.23 Existing online interventions include banner advertisements,22 risk reduction Web sites,67 chat room–based outreach efforts,68,69 and interactive Web-based educational and skill-building modules.70,71 Our results suggest that rather than focusing on the dangers of online sex seeking, Internet-based programs, similar to offline interventions, should encourage young MSM who are at risk to reduce their numbers of sexual partners, decrease the frequency at which they engage in unprotected anal intercourse, avoid alcohol and other substance use in sexual situations, and seek HIV testing.12,39,72
Acknowledgments
Funding for this study was provided by the National Institute of Mental Health (grant R01 MH063688-05).
We express our appreciation to the study participants for their time and effort devoted to this research.
Human Participant Protection This study was approved by the institutional review board of the University of Minnesota. Participants provided informed consent.
Peer Reviewed
Contributors K.J. Horvath wrote the article and completed the analyses. B. R. S. Rosser edited the article during its development. G. Remafedi originated the study and was involved in the writing of the article.
References
- 1.Lampinen TM, Ogilvie G, Chan K, et al. Sustained increase in HIV-1 incidence since 2000 among men who have sex with men in British Columbia, Canada. J Acquir Immune Defic Syndr. 2005;40:242–244. [DOI] [PubMed] [Google Scholar]
- 2.McFarland W, Chen S, Weide D, Kohn R, Klausner J. Gay Asian men in San Francisco follow the international trend: increases in rates of unprotected anal intercourse and sexually transmitted diseases, 1999–2002. AIDS Educ Prev. 2004;16:13–18. [DOI] [PubMed] [Google Scholar]
- 3.Murphy G, Charlett A, Jordan LF, Osner N, Gill ON, Parry JV. HIV incidence appears constant in men who have sex with men despite widespread use of effective antiretroviral therapy. AIDS. 2004;18:265–272. [DOI] [PubMed] [Google Scholar]
- 4.Trends in HIV/AIDS diagnosis—33 states, 2001–2004. MMWR Morb Mortal Wkly Rep. 2005; 54:1149–1153. [PubMed] [Google Scholar]
- 5.Chen SY, Weide D, McFarland W. Are the recent increases in sexual risk behavior among older or younger men who have sex with men? Answer: both. AIDS. 2003;17:942–943. [DOI] [PubMed] [Google Scholar]
- 6.Centers for Disease Control and Prevention. HIV/ AIDS among youth. Available at: http://www.cdc.gov/hiv/resources/factsheets/youth.htm. Accessed August 3, 2006.
- 7.Rangel MC, Gavin L, Reed C, Fowler MG, Lee LM. Epidemiology of HIV and AIDS among adolescents and young adults in the United States. J Adolesc Health. 2006;39:156–163. [DOI] [PubMed] [Google Scholar]
- 8.Niccolai LM, Farley TA, Ayoub MA, Mangus M, Kissinger PJ. HIV-infected persons’ knowledge of their sexual partners’ HIV status. AIDS Educ Prev. 2002;14: 183–189. [DOI] [PubMed] [Google Scholar]
- 9.MacKellar DA, Valleroy LA, Behel S, et al. Unintentional HIV exposures from young men who have sex with men who disclose being HIV-negative. AIDS. 2006;20:1637–1644. [DOI] [PubMed] [Google Scholar]
- 10.McFarlane M, Bull SS, Rietmeijer CA. Young adults on the Internet: risk behaviors for sexually transmitted diseases and HIV(1). J Adolesc Health. 2002;31:11–16. [DOI] [PubMed] [Google Scholar]
- 11.Chesson HW, Harrison P, Stall R. Changes in alcohol consumption and in sexually transmitted disease incidence rates in the United States: 1983–1998. J Stud Alcohol. 2003;64:623–630. [DOI] [PubMed] [Google Scholar]
- 12.Stall R, Mills TC, Williamson J, et al. Association of co-occurring psychosocial health problems and increased vulnerability to HIV/AIDS among urban men who have sex with men. Am J Public Health. 2003;93:939–942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bancroft J, Janssen E, Carnes L, Goodrich D, Strong D, Long SJ. Sexual activity and risk taking in young heterosexual men: the relevance of sexual arousability, mood, and sensation seeking. J Sex Res. 2004; 41:181–192. [DOI] [PubMed] [Google Scholar]
- 14.Rosser BRS, Gobby JM, Carr WP. The unsafe sexual behavior of persons living with HIV/AIDS: an empirical approach to developing new HIV prevention interventions targeting HIV-positive persons. J Sex Educ Ther. 1999;24:18–28. [Google Scholar]
- 15.Elford J, Bolding G, Sherr L. Seeking sex on the Internet and sexual risk behaviour among gay men using London gyms. AIDS. 2001;15:1409–1415. [DOI] [PubMed] [Google Scholar]
- 16.Grov C. Barebacking websites: electronic environments for reducing or inducing HIV risk. AIDS Care. 2006;18:990–997. [DOI] [PubMed] [Google Scholar]
- 17.Liau A, Millett G, Marks G. Meta-analytic examination of online sex-seeking and sexual risk behavior among men who have sex with men. Sex Transm Dis. 2006;33:576–584. [DOI] [PubMed] [Google Scholar]
- 18.Kim AA, Kent C, McFarland W, Klausner JD. Cruising the Internet highway. J Acquir Immune Defic Syndr. 2001;28:89–93. [DOI] [PubMed] [Google Scholar]
- 19.Bull SS, McFarlane M, Reitmeijer C. HIV and sexually transmitted infection risk behaviors among men seeking sex with men on-line. Am J Public Health. 2001;91:988–999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Rietmeijer CA, Bull SS, McFarlane M, Landrigan Patnaik J, Douglas JM. Risks and benefits of the Internet for populations at risk for sexually transmitted infections (STIs): results of an STI clinic survey. Sex Transm Dis. 2003;30:15–19. [DOI] [PubMed] [Google Scholar]
- 21.Tashima KT, Alt EN, Harwell JI, Fiebich-Perez DK, Flanigan TP. Internet sex-seeking leads to acute HIV infection: a report of two cases. Int J STD AIDS. 2003; 14:285–286. [DOI] [PubMed] [Google Scholar]
- 22.Klausner JD, Levine DK, Kent CK. Internet-based site-specific interventions for syphilis prevention among gay and bisexual men. AIDS Care. 2004;16:964–970. [DOI] [PubMed] [Google Scholar]
- 23.Pequegnat W, Rosser BRS, Bowen AM, et al. Conducting Internet-based HIV/STD prevention survey research: considerations in design and evaluation. AIDS Behav. 2006;11:505–521. [DOI] [PubMed] [Google Scholar]
- 24.Skinner H, Biscope S, Poland B, Goldberg E. How adolescents use technology for health information: implications for health professionals from focus group studies. J Med Internet Res. 2003;5:e32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Garofalo R, Herrick A, Mustanski B, Donenberg GR. Online and at-risk: young men who have sex with men and the Internet. J Adolesc Health. 2006;38:104. [Google Scholar]
- 26.Liau AK, Khoo A, Ang PH. Factors influencing adolescents’ engagement in risky Internet behavior. Cyberpsychol Behav. 2005;8:513–520. [DOI] [PubMed] [Google Scholar]
- 27.Yang CK, Choe BM, Baity M, Lee JH, Cho JS. SCL-90-R and 16PF profiles of senior high school students with excessive Internet use. Can J Psychiatry. 2005;50:407–414. [DOI] [PubMed] [Google Scholar]
- 28.Benotsch EG, Kalichman S, Cage M. Men who have met sex partners via the Internet: prevalence, predictors, and implications for HIV prevention. Arch Sex Behav. 2002;31:177–183. [DOI] [PubMed] [Google Scholar]
- 29.Robinson BBE, Bockting WO, Rosser BRS, Miner M, Coleman E. The sexual health model: application of a sexological approach to HIV prevention. Health Educ Res. 2002;17:43–57. [DOI] [PubMed] [Google Scholar]
- 30.Ross MW, Rosser BR, Standon J, Konstan J. Characteristics of Latino men who have sex with men on the Internet who complete and drop out of an Internet-based sexual behavior survey. AIDS Educ Prev. 2004; 16:526–537. [DOI] [PubMed] [Google Scholar]
- 31.Konstan JA, Rosser BRS, Ross MW, Stanton J, Edwards WM. The story of subject naught: a cautionary but optimistic tale of Internet survey research. J Comp Mediated Commun. Published online June 23, 2006. doi:10.1111/j.1083-6101.2005.tb00248.x.
- 32.StataCorp. Stata Statistical Software: Release 9. College Station, TX: StataCorp LP; 2005.
- 33.Williams R. Generalized ordered logit/partial proportional odds models for ordinal dependent variables. Stata J. 2006;6:58–82. [Google Scholar]
- 34.Peterson B, Harrell FE. Partial proportional odds models for ordinal response variables. Appl Stat. 1990; 39:205–217. [Google Scholar]
- 35.Ross M, Rosser BRS, Coleman E, Mazin R. Misrepresentation on the Internet and in real life about sex and HIV: a study of Latino men who have sex with men. Cult Health Sex. 2006;8:133–144. [DOI] [PubMed] [Google Scholar]
- 36.Sex by age for the United States. Available at: http://factfinder.census.gov/home/saff/main.html?_langen. Accessed November 6, 2006.
- 37.Lau JT, Kim JH, Lau M, Tsui HY. Prevalence and risk behaviors of Chinese men who seek same-sex partners via the Internet in Hong Kong. AIDS Educ Prev. 2003;15:516–528. [DOI] [PubMed] [Google Scholar]
- 38.Tikkanen R, Ross MW. Technological tearoom trade: characteristics of Swedish men visiting gay Internet chat rooms. AIDS Educ Prev. 2003;15:122–132. [DOI] [PubMed] [Google Scholar]
- 39.Hirshfield S, Remien RH, Humberstone M, Walavaklar I, Chiasswon MA. Substance use and high-risk sex among men who have sex with men: a national online study in the USA. AIDS Care. 2004;16: 1036–1047. [DOI] [PubMed] [Google Scholar]
- 40.Binson D, Woods WJ, Pollack L, Paul J, Stall R, Catania JA. Differential HIV risk in bathhouses and public cruising areas. Am J Public Health. 2001;91: 1482–1486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Zuckerman M. Sensation Seeking and Risky Behavior. Washington, DC: American Psychological Association; 2007.
- 42.Horvath KJ, Beadnell B, Bowen AM. Sensation seeking as a moderator of Internet use on sexual risk-taking among men who have sex with men. Sex Res Soc Policy. 2006;3:77–90. [Google Scholar]
- 43.Chak K, Leung L. Shyness and locus of control as predictors of Internet addiction and Internet use. Cyberpsychol Behav. 2004;7:559–570. [DOI] [PubMed] [Google Scholar]
- 44.Cooper A, Morahan-Martin J, Mathy RM, Maheu M. Toward an increased understanding of user demographics in online sexual activities. J Sex Marital Ther. 2002;28:105–129. [DOI] [PubMed] [Google Scholar]
- 45.Modayil MV, Thompson AH, Varnhagen S, Wilson DR. Internet users’ prior psychological and social difficulties. Cyberpsychol Behav. 2003;6:585–590. [DOI] [PubMed] [Google Scholar]
- 46.Tikkanen R, Ross MW. Looking for sexual compatibility: experiences among Swedish men in visiting Internet gay chat rooms. Cyberpsychol Behav. 2000;3: 605–616. [Google Scholar]
- 47.Bolding G, Davis M, Hart G, Sherr L, Elford J. Gay men who look for sex on the Internet: is there more HIV/STI risk with online partners? AIDS. 2005; 19:961–968. [DOI] [PubMed] [Google Scholar]
- 48.Waldo CR, McFarland W, Katz MH, MacKellar D, Valleroy LA. Very young gay and bisexual men are at risk for HIV infection: the San Francisco Bay Area Young Men’s Survey II. J Acquir Immune Defic Syndr. 2000;24:168–174. [DOI] [PubMed] [Google Scholar]
- 49.Stueve A, O’Donnell L, Duran R, San Doval A, Geier J. Being high and taking sexual risks: findings from a multisite survey of urban young men who have sex with men. AIDS Educ Prev. 2002;14:482–495. [DOI] [PubMed] [Google Scholar]
- 50.Fernandez MI, Varga LM, Perrino T, et al. The Internet as recruitment tool for HIV studies: viable strategy for reaching at-risk Hispanic MSM in Miami? AIDS Care. 2004;16:953–963. [DOI] [PubMed] [Google Scholar]
- 51.Halkitis PN, Parsons JT, Stirratt MJ. A double epidemic: crystal methamphetamine drug use in relation to HIV transmission among gay men. J Homosex. 2001; 41:17–35. [DOI] [PubMed] [Google Scholar]
- 52.Valleroy LA, Mackellar DA, Karon JM, et al. HIV prevalence and associated risks in young men who have sex with men. JAMA. 2000;284:198–204. [DOI] [PubMed] [Google Scholar]
- 53.Gullette DL, Turner JG. Stages of change and condom use among an Internet sample of gay and bisexual men. J Assoc Nurses AIDS Care. 2004;15:27–37. [DOI] [PubMed] [Google Scholar]
- 54.Hospers HJ, Harterink P, Van Den HK, Veenstra J. Chatters on the Internet: a special target group for HIV prevention. AIDS Care. 2002;14:539–544. [DOI] [PubMed] [Google Scholar]
- 55.Centers for Disease Control and Prevention. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/rr5514a1.htm. Accessed January 14, 2008.
- 56.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. JAMA. 2005; 97:38S–43S. [PMC free article] [PubMed] [Google Scholar]
- 57.Whittier DK, Seeley S, St. Lawrence JS. A comparison of Web- with paper-based surveys of gay and bisexual men who vacationed in a gay resort community. AIDS Educ Prev. 2004;16:476–485. [DOI] [PubMed] [Google Scholar]
- 58.Ellen JM, Gurvey JE, Patsch L, Tschann J, Nanda JP, Catania J. A randomized comparison of A-CASI and phone interviews to assess STD/HUV-related risk behaviors in teens. J Adolesc Health. 2002;31:26–30. [DOI] [PubMed] [Google Scholar]
- 59.Gilbert GH, Rose JS, Shelton BJ. A prospective study of the validity of data on self-reported dental visits. Community Dent Oral Epidemiol. 2002;30: 352–362. [DOI] [PubMed] [Google Scholar]
- 60.Borzekowski DL, Rickert VI. Adolescent cyber-surfing for health information: a new resource that crosses barriers. Arch Pediatr Adolesc Med. 2001;155: 813–817. [DOI] [PubMed] [Google Scholar]
- 61.Goold PC, Ward M, Carlin EM. Can the Internet be used to improve sexual health awareness in Web-wise young people? J Fam Plann Reprod Health Care. 2003;29:28–30. [DOI] [PubMed] [Google Scholar]
- 62.Weerakoon P. E-learning in sexuality education. Med Teach. 2003;25:13–17. [DOI] [PubMed] [Google Scholar]
- 63.Bull SS, McFarlane M, King D. Barriers to STD/ HIV prevention on the Internet. Health Educ Res. 2001;16:661–670. [DOI] [PubMed] [Google Scholar]
- 64.Bensley RJ, Mercer N, Brusk JJ, et al. The ehealth behavior management model: a stage-based approach to behavior change and management. Prev Chronic Dis. 2004;1:A14. [PMC free article] [PubMed] [Google Scholar]
- 65.Davis M, Hart G, Bolding G, Sherr L, Elford J. E-dating, identity and HIV prevention: theorising sexualities, risk and network society. Sociol Health Illn. 2006;28:457–478. [DOI] [PubMed] [Google Scholar]
- 66.Flicker S, Goldberg E, Read S, et al. HIV-positive youths’ perspectives on the Internet and e-health. J Med Internet Res. 2004;6:e32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Noar SM, Clark A, Cole C, Lustria MLA. Review of interactive safer sex Web sites: practice and potential. Health Commun. 2006;20:233–241. [DOI] [PubMed] [Google Scholar]
- 68.Rhodes SD. Hookups or health promotion? An exploratory study of a chat room-based HIV prevention intervention for men who have sex with men. AIDS Educ Prev. 2004;16:315–327. [DOI] [PubMed] [Google Scholar]
- 69.Rhodes SD, Hergenrather KC, Yee LJ, Ramsey B. Comparing MSM in the southeastern United States who participated in an HIV prevention chat room-based outreach intervention and those who did not: how different are the baseline HIV-risk profiles? Health Educ Res. Published online April 5, 2007. doi:10. 1093/her/cym015. [DOI] [PubMed]
- 70.Bowen AM, Horvath KJ, Williams ML. A randomized control trial of Internet delivered HIV prevention targeting rural MSM. Health Educ Res. 2007;22: 120–127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Kok G, Harterink P, Vriens P, deZwart O, Hospers HJ. The gay cruise: developing a theory- and evidence-based Internet HIV-prevention intervention. Sex Res Soc Policy. 2006;3:52–67. [Google Scholar]
- 72.Koblin BA, Husnik MJ, Colfax G, et al. Risk factors of HIV infection among men who have sex with men. AIDS. 2006;20:731–739. [DOI] [PubMed] [Google Scholar]