Abstract

Chat room-based human immunodeficiency virus (HIV) prevention interventions are being implemented to reduce the risk of HIV exposure, infection and reinfection among men who have sex with men (MSM). However, little is known about how participants in chat room-based prevention interventions differ from their online non-participating peers. This analysis compared the baseline risk profiles of participants in an HIV prevention intervention (‘active recruitment’) to their chat room peers who did not participate in the intervention (‘passive recruitment’). Data were collected using an online brief risk assessment from MSM (N = 448) who were recruited within Internet chat rooms. Mean age was 30 years. Half self-identified as Black or African American, 29% as White and 64% as gay. Compared with participants, non-participants were more likely to report: spending higher mean number of hours in online chat rooms; using condoms inconsistently during anal intercourse with a man met online during the past 3 months; having had an sexually transmitted disease; being HIV seropositive; using methamphetamines during the past 30 days and using drugs to enhance sexual satisfaction during the past 30 days. Although risk among MSM who use chat rooms remains high, those at greater risk may be those who are less likely to engage in online HIV prevention interventions.

Introduction

Traditionally, human immunodeficiency virus (HIV) education, prevention and outreach efforts targeting men who have sex with men (MSM) have focused on physical spaces where these men meet including bars and dance clubs, bathhouses and social groups [1]. With the expansion of the Internet, the number of MSM who use hypertext markup language chat rooms for both social and sexual networking continues to increase [2, 3].

Because of the increased use of chat rooms and the evidence that online men are at increased risk for HIV exposure, infection and reinfection through their sexual risk behaviors [4–14], HIV prevention outreach interventions have been proposed and are being implemented to target men in MSM-oriented chat rooms [3, 15–17]. Online men have been identified as being at increased risk for HIV but difficult for researchers to access through traditional approaches, including individuals who have chosen to engage in unprotected sex (e.g. barebackers) and are young MSM; tourists and other mobile populations; transgender and sensation seekers [18–20]. However, little is known about how men who participate in chat room-based prevention interventions differ from those who do not participate.

This analysis compares the baseline demographic characteristics and HIV-related risk profiles of two groups of MSM who use chat rooms for social and sexual networking. Participants in the first group (‘active recruitment’) were recruited to participate in the HIV prevention outreach intervention, were given the web address of a risk assessment and were provided a unique password necessary for accessing the assessment. This occurred through private messages with a trained educator.

The comparison group (‘passive recruitment’) did not engage with the educator; rather, the educator solicited participation in the assessment by sending general messages to the public chat room that included all necessary information to access the Web site of the assessment. Access to the assessment for this group did not require a password.

Chat rooms defined

An Internet chat room is a channel of synchronous dialogue between computer users connected through a network of computers. The most common chat rooms use a client–server model with a centralized computer relaying messages sent to it from computers used by chatters who want to communicate with one another through written dialogue. These chatters log in via the Internet to a chat-server computer that hosts a specific chat room. Chatters can type messages that are transferred almost instantaneously by the server to the other chatters within the chat room. Thus, users are able to communicate to one another ‘in real time’ through what is referred to as an ‘instant message’ (IM). This speed differentiates chat-room dialogue from asynchronous computer-mediated communication methods such as electronic mail and newsgroups [3].

There are public chat rooms in which written dialogue is seen by all chatters who are logged on in the room. Depending on the host, up to 100 chatters may be in a chat room at one time. Chatters also may communicate privately to one another using IM, in which only the designated chatter receives the message. Private IM is often used for detailed discussions between chatters that may include getting to know one another and determining whether and how to meet in person for social and/or sexual contact.

Methods

This analysis is based on baseline data from Cyber-Based Education and Referral/Men for Men (CyBER/M4M), a chat room-based HIV prevention outreach intervention that was developed using community-based participatory research (CBPR). Briefly, CyBER/M4M was conceived and conducted by a CBPR partnership comprised of community members including gay men (some of whom had experience with seeking sexual partners online) and representatives from community-based organizations (CBOs); two acquired immune deficiency syndrome (AIDS) service organizations (ASOs); local public health departments; a local foundation supporting lesbian, gay, bisexual and transgender health; Wake Forest University Health Sciences; the University of North Carolina Center for AIDS Research; among others. The history of this community–university partnership and the application of CBPR in CyBER/M4M has been described previously [21].

This partnership serves as a catalyst for identifying priorities and approaches to meet locally identified HIV/AIDS prevention and care needs. Guided by partnership principles, including agreement on mission and sharing resources and decision-making responsibilities, the partnership has prioritized CBPR to promote a co-learning and empowering process that moves beyond communities ‘informing’ or offering ‘consultation’. This process increases the level of decision making by communities to allow negotiation and trade offs with traditional power holders [22] (e.g. university researchers) and community members and CBO representatives to increase authenticity of research methods and accuracy of findings.

Measures

Data were collected using a revised version of the assessment used in the partnership's previous assessment and intervention research [23]. Development and revision of the assessment was iterative with partners negotiating and agreeing on a version that was sufficient to collect necessary data yet concise to ensure completion by the target community. Much effort went into the process of developing the assessment in order to ensure its rapid completion. Community partners suggested that any assessment taking chatters >5 min was likely to provide inaccurate and incomplete responses given the chatters' competing motivations for being online.

The final assessment was comprised of 36 items, which were based on self-report, using pre-defined response options with binary, categorical or Likert-scale response options. Demographic characteristics were assessed, including age in years, gender, educational attainment, medical insurance and estimated annual income. Race/ethnicity was assessed using the item ‘How would you describe your race or ethnic background?’. Response options included ‘American Indian or Alaskan Native’, ‘Asian’, ‘African American or Black’, ‘Hispanic or Latino’, ‘Native Hawaiian or Other Pacific Islander’, ‘White’ and ‘other’. Participants who selected ‘other’ were asked to specify how they would describe their race or ethnic background.

Sexual orientation was assessed using the item ‘How would you describe your sexuality?’. Response options included ‘bisexual’, ‘gay’, ‘heterosexual or straight’, ‘transgender’ and ‘other’. Participants who selected ‘other’ were asked to specify how they would describe their sexuality.

Behaviors assessed included sexual behavior with men and women, numbers of sexual partners, engagement in oral sex and receptive anal and insertive anal intercourse, condom use during oral sex and receptive anal and insertive anal intercourse during the past 3 months and condom use with oral sex and receptive anal and insertive anal intercourse with men met online.

Each participant was asked a series of items to assess whether he had chatted online with another male whom he did not already know, met him in person and had oral or anal sex with him. Number of hours spent during the past 7 days in chat rooms designed for social and sexual networking of MSM also was measured.

History of sexually transmitted diseases (STDs), including gonorrhea, syphilis, chlamydia, herpes, hepatitis A, hepatitis B and genital warts; and HIV counseling, testing and diagnosis were assessed. Participants were asked whether they had ever used and to estimate the past 30-day frequency use of methamphetamine (crystal), cocaine, crack, heroin and ecstasy.

Lifetime and past 30-day frequency use of drugs to enhance sexual satisfaction (i.e. Viagra, Cialis and Levitra) and of amyl nitrite (poppers) also were assessed.

A final item assessed whether a participant had completed the assessment previously.

Participant recruitment

In summer 2004, two lay health educators were trained in the data collection and intervention protocols. The 16-h training was conducted in two 8-h sessions during a 2-week period. These educators were self-identified gay men, one self-identified as African American and the other as white. They were 28 and 24 years old, respectively, and were familiar with online social and sexual networking communities. They were particularly knowledgeable about the local MSM communities and subcommunities, including MSM who have female partners.

Recruitment occurred in five free chat rooms that facilitate social and sexual networking among MSM in central North Carolina; two were geared toward African American MSM. The other three rooms tended to attract a diversity of ethnicities and races. Because of an American Online Internet chat room for MSM was bombarded with profanity and death threats following media reports that health officials traced a syphilis outbreak to chatters who met there [24], the web addresses and the name of the ASOs are not published here in order to maintain the safety of the chatters and ensure maintenance of the trust the educators earned with them.

The educators were online in random 2-h shifts.

Recruitment of men participating in the HIV prevention outreach intervention: active recruitment

After entering a pre-determined chat room, the educator followed a standard protocol. He announced his purpose and availability to answer questions and provide education about HIV and AIDS through general messages in the public chat room. The number of messages the educator could send to the public chat room was limited because too aggressive of an online approach was likely to alienate the chatters or even result in expulsion of the educator from the chat room. CBPR partners were sensitive to respecting the chat room culture. Messages included ‘In the room to answer questions about HIV and AIDS’, ‘I can answer questions about HIV and AIDS’, ‘Want to get tested for HIV? I can help’ and ‘Need condoms? I can tell you where to get them free’. Only after a chatter privately instant messaged the educator did the educator privately IM the chatter. This requirement ensured that the educator respected the privacy and motivations of the chatters within the chat room.

After being instant messaged by a chatter, the educator explained the study to the potential participant. After engaging with a chatter using the established protocol and confirming that the participant wanted to participate in the CyBER/M4M study, the educator asked the chatter to complete the risk assessment. This assessment served as the intervention baseline data collection and the data for this current analysis. The educator usually used statements such as ‘Thanks for talking with me. Before you go, do you have 5 more minutes to take the men's health survey? It's the first step in your participation in CyBER/M4M. It's an online and anonymous survey that asks questions about you, sex, and the internet’. If the participant agreed to complete the assessment, the educator provided an online Web site and a password. The educator also asked the chatter IM him when he had completed the assessment. Each participant was given a unique password he could use only once.

The Web site hosting the brief assessment was available up to 1 h after the web address and password were provided to the participant.

Recruitment in the public chat room: passive recruitment

After entering a pre-determined chat room, the educator followed a standard protocol. The educator sent general messages to the public chat room; however, these messages included pertinent information about the assessment. Messages included ‘Hey guys—take our survey on men's health! Go to [web site address]’ and ‘We need men to complete an online and anonymous survey about men's health! Log on at: [web site address]’. No password was needed, and the site was available to all chatters within the chat room who saw the Web site address.

The Web site hosting the brief assessment was available up to 1 h after the final message was sent to the public chat room.

The educator–chatter profile

Chatter profiles are online responses to close- and open-ended standardized items that many chatters choose to complete and are accessible to all chatters within the chat room. Among other characteristics, profiles usually include sex/gender, sexual orientation, HIV status, ethnicity and level of ‘outness’ about one's sexual orientation. The educators were guided by the CBPR partnership to create titillating profiles to spark interest for the target community while accurately representing their roles in the chat room. The educators' profiles indicated their affiliation with research; chatters in either group could read about the educator, his purpose online and the CyBER/M4M intervention.

Data analysis

SPSS 11.5 (Chicago, IL, USA) was used for all data analysis. Descriptive analyses were performed to examine overall prevalence, group prevalence and distribution shape. Demographic and risk data were then analyzed in bivariate analyses using chi-square and Fisher's exact tests for categorical variables, and t-tests for continuous variables to compare men by group. Risks also were analyzed in a multivariable logistic regression model that adjusted for demographic differences as covariates. From this modeling, adjusted odds ratios and 95% confidence intervals were calculated for risks. The referent group was that which participated in CyBER/M4M (actively recruited).

Results

Participants

Of the 448 male participants, the mean age was 30.78 (±10.8, range 18–62). Half (n = 223, 50.7%) self-identified as Black or African American; 29.1% as White (n = 128); 12.7% as Native Hawaiian or Other Pacific Islander (n = 56) and the remaining 7.5% as American Indian or Alaska Native (n = 16), Asian (n = 8) or Hispanic or Latino (n = 6).

A third of participants (33.9%, n = 149) self-identified as bisexual, 63.9% as gay (n = 281), 1.4% as heterosexual or straight (n = 6) and 0.9% as undefined other (n = 4). All participants reported having sex with other men within the past 2 years.

Selected characteristics of participants by recruitment group are presented in Table I. When comparing demographics of the two recruitment groups, differences were identified among the groups in race/ethnicity, annual income and self-identified sexual orientation.

Table I.

Selected characteristics of chatters by recruitment strategy

Characteristics Activea (n = 210) Passivea (n = 238) P 
Age 30.9 (11.2) 30.6 (10.5) 0.20 
Race/ethnicity   <0.001 
    American Indian/Alaska Native 2 (1.0) 14 (5.9)  
    Asian 1 (0.5) 7 (2.9)  
    Black or African American 76 (37.6) 147 (61.8)  
    Hispanic or Latino 6 (3.0) .0 (0.0)  
    Native Hawaiian or other Pacific Islander .0 (0.0) 56 (23.5)  
    White 114 (56.4) 14 (5.9)  
    Multiracial/ethnic 3 (1.5) .0 (0.0)  
Some college or above 160 (79.2) 196 (82.4) 0.20 
Estimated annual income   <0.001 
    <$10 000 41 (20.3) 35 (14.7)  
    $10 000–19 999 30 (14.9) 35 (14.7)  
    $20 000–39 999 56 (27.7) 77 (32.4)  
    $40 000–59 999 45 (22.3) 63 (26.5)  
    $60 000–79 999 13 (6.4) 28 (11.8)  
    ≥$80 000 17 (8.4) .0 (0.0)  
Have medical insurance 171 (84.7) 189 (79.4) 0.08 
Self-identified sexual orientation   <0.001 
    Bisexual 51 (25.7) 98 (41.2)  
    Gay 141 (71.2) 139 (58.4)  
    Heterosexual or straight 6 (3.1) 1 (0.4)  
Characteristics Activea (n = 210) Passivea (n = 238) P 
Age 30.9 (11.2) 30.6 (10.5) 0.20 
Race/ethnicity   <0.001 
    American Indian/Alaska Native 2 (1.0) 14 (5.9)  
    Asian 1 (0.5) 7 (2.9)  
    Black or African American 76 (37.6) 147 (61.8)  
    Hispanic or Latino 6 (3.0) .0 (0.0)  
    Native Hawaiian or other Pacific Islander .0 (0.0) 56 (23.5)  
    White 114 (56.4) 14 (5.9)  
    Multiracial/ethnic 3 (1.5) .0 (0.0)  
Some college or above 160 (79.2) 196 (82.4) 0.20 
Estimated annual income   <0.001 
    <$10 000 41 (20.3) 35 (14.7)  
    $10 000–19 999 30 (14.9) 35 (14.7)  
    $20 000–39 999 56 (27.7) 77 (32.4)  
    $40 000–59 999 45 (22.3) 63 (26.5)  
    $60 000–79 999 13 (6.4) 28 (11.8)  
    ≥$80 000 17 (8.4) .0 (0.0)  
Have medical insurance 171 (84.7) 189 (79.4) 0.08 
Self-identified sexual orientation   <0.001 
    Bisexual 51 (25.7) 98 (41.2)  
    Gay 141 (71.2) 139 (58.4)  
    Heterosexual or straight 6 (3.1) 1 (0.4)  
a

Mean ± SD or n (%), as appropriate.

Table II presents risk behavior comparisons for the total sample and by recruitment group. Overall, participants reported spending >16 h in Internet chat rooms during the past 7 days, and 21.2% of participants reported having had intercourse with a female partner within the past 2 years. More than three-fourths (75.4%) reported ever chatting online with a man they did not know, meeting him in person and having oral sex with him; and 57.1% reported ever chatting with a man they did not know, meeting him in person and having anal intercourse with him. More than one-third (35.4%) of the sample also reported inconsistent condom use during anal intercourse with men met online during the past 3 months.

Table II.

Comparison of risk of participants by recruitment strategy: bivariate analysis

Risk Overalla (N = 448) Activea (n = 210) Passivea (n = 238) P 
Number of hours spent in online chat rooms in past 7 days 16.7 (22.7) 13.7 (24.3) 19.2 (21.0) 0.006 
Sex with a female partner during past 2 years 90 (21.2) 34 (16.4) 56 (23.5) 0.04 
Oral sex with multiple partners, past 3 months 362 (80.0) 159 (75.7) 203 (85.3) 0.007 
Receptive anal sex with multiple partners, past 3 months 243 (54.2) 89 (42.4) 154 (64.7) <0.000 
Insertive anal sex with multiple partners, past 3 months 255 (56.9) 101 (48.1) 238 (53.1) <0.000 
Chatted with a man did not already know and had oral sex with him 338 (75.4) 156 (74.3) 182 (76.5) <0.33 
Chatted with a man did not already know and had anal sex with him 256 (57.1) 109 (51.9) 147 (61.8) 0.02 
Inconsistent condom use during anal sex with men met online, past 3 months 156 (35.4) 58 (27.6) 98 (41.2) 0.02 
Ever had an STD 148 (33.6) 57 (28.2) 91 (38.2) <0.001 
Positive HIV serostatus 60 (13.6) 18 (8.9) 42 (17.6) <0.001 
Never been tested for HIV 52 (11.8) 31 (15.3) 21 (8.8) 0.09 
Illicit drug use, past 30 days 122 (27.7) 52 (25.7) 70 (29.4) 0.28 
Methamphetamine use, past 30 days 9 (2.0) 2 (1.0) 7 (2.9) 0.13 
Cocaine use, past 30 days 49 (11.1) 21 (10.4) 28 (11.8) 0.38 
Crack use, past 30 days 11 (2.5) 4 (2.0) 7 (2.9) 0.37 
Heroin use, past 30 days .0 (0.0) .0 (0.0) .0 (0.0)  
Ecstasy use, past 30 days 22 (5.0) 8 (4.0) 14 (5.9) 0.24 
Poppers use, past 30 days 40 (9.1) 33 (16.3) 7 (2.9) <0.001 
Viagra, Cialis or Levitra use, past 30 days 68 (15.5) 26 (12.9) 42 (17.6) 0.11 
Risk Overalla (N = 448) Activea (n = 210) Passivea (n = 238) P 
Number of hours spent in online chat rooms in past 7 days 16.7 (22.7) 13.7 (24.3) 19.2 (21.0) 0.006 
Sex with a female partner during past 2 years 90 (21.2) 34 (16.4) 56 (23.5) 0.04 
Oral sex with multiple partners, past 3 months 362 (80.0) 159 (75.7) 203 (85.3) 0.007 
Receptive anal sex with multiple partners, past 3 months 243 (54.2) 89 (42.4) 154 (64.7) <0.000 
Insertive anal sex with multiple partners, past 3 months 255 (56.9) 101 (48.1) 238 (53.1) <0.000 
Chatted with a man did not already know and had oral sex with him 338 (75.4) 156 (74.3) 182 (76.5) <0.33 
Chatted with a man did not already know and had anal sex with him 256 (57.1) 109 (51.9) 147 (61.8) 0.02 
Inconsistent condom use during anal sex with men met online, past 3 months 156 (35.4) 58 (27.6) 98 (41.2) 0.02 
Ever had an STD 148 (33.6) 57 (28.2) 91 (38.2) <0.001 
Positive HIV serostatus 60 (13.6) 18 (8.9) 42 (17.6) <0.001 
Never been tested for HIV 52 (11.8) 31 (15.3) 21 (8.8) 0.09 
Illicit drug use, past 30 days 122 (27.7) 52 (25.7) 70 (29.4) 0.28 
Methamphetamine use, past 30 days 9 (2.0) 2 (1.0) 7 (2.9) 0.13 
Cocaine use, past 30 days 49 (11.1) 21 (10.4) 28 (11.8) 0.38 
Crack use, past 30 days 11 (2.5) 4 (2.0) 7 (2.9) 0.37 
Heroin use, past 30 days .0 (0.0) .0 (0.0) .0 (0.0)  
Ecstasy use, past 30 days 22 (5.0) 8 (4.0) 14 (5.9) 0.24 
Poppers use, past 30 days 40 (9.1) 33 (16.3) 7 (2.9) <0.001 
Viagra, Cialis or Levitra use, past 30 days 68 (15.5) 26 (12.9) 42 (17.6) 0.11 
a

n (%).

One-third (33.6%) of the sample reported ever having had an STD, and 13.6% reported testing seropositive for HIV. More than one-quarter (27.7%) reported some type of illicit drug use during the past 30 days. Use of one of the common drugs to enhance sexual satisfaction was reported by 15.5% of the participants.

When examining differences by recruitment group, participants who were recruited passively through general messages to public chat rooms (and therefore not participating in the intervention) were significantly more likely to report: higher mean number of hours spent in online chat rooms during the past 7 days (μ = 19.2, range 1–56 h); sex with a female partner during the past 2 years (23.5%); oral sex (85.3%) and receptive anal and insertive anal intercourse with multiple partners during the past 3 months (64.7 and 53.1%, respectively); chatting with a man they did not know, meeting him in person and having anal intercourse with him (61.8%); inconsistent condom use during anal intercourse with a man met online during the past 3 months (41.2%); a history of STDs (38.2%) and positive HIV serostatus (17.6%).

Chatters who participated in the HIV prevention outreach intervention were significantly more likely to report using poppers during the past 30 days (16.3%).

After adjusting for the demographic characteristics found to be significant in Table I, including race/ethnicity, income and self-identified sexual orientation, differences between groups remained, as presented in Table III. Passively recruited participants had an increased odds of reporting: higher mean number of hours spent in online chat rooms during the past 7 days; oral sex and receptive anal intercourse with multiple partners in the past 3 months; inconsistent condom use during anal intercourse with a man met online during the past 3 months; ever having had an STD; being HIV seropositive; illicit drug use during the past 30 days and use of drugs to enhance sexual satisfaction during the past 30 days.

Table III.

Comparison of risk of participants by recruitment strategy: multivariable logistic regression analysis

Risk Active versus passiveab P 
Number of hours spent in online chat rooms in past 7 days 1.3 (1.01–1.9) 0.03 
Sex with a female partner during past 2 years 0.9 (0.6–1.7) 0.9 
Oral sex with multiple partners, past 3 months 2.5 (1.4–4.5) 0.002 
Receptive anal sex with multiple partners, past 3 months 2.2 (1.4–3.5) <0.001 
Insertive anal sex with multiple partners, past 3 months 1.1 (0.8–1.8) 0.5 
Chatted with a man did not already know and had oral sex with him 1.2 (0.8–2.0) 0.4 
Chatted with a man did not already know and had anal sex with him 1.1 (0.7–1.6) 0.8 
Inconsistent condom use during anal sex with men met online, past 3 months 1.8 (1.2–3.0) <0.001 
Ever had an STD 1.3 (1.1–1.6) 0.004 
Positive HIV serostatus 1.5 (1.2–2.1) 0.003 
Never been tested for HIV 1.2 (0.9–1.4) 0.2 
Illicit drug use, past 30 days 2.5 (1.5–4.4) <0.001 
Methamphetamine use, past 30 days 6.7 (1.7–26.1) 0.02 
Cocaine use, past 30 days 1.3 (0.7–2.5) 0.5 
Crack use, past 30 days 1.4 (0.3–6.1) 0.7 
Ecstasy use, past 30 days 2.8 (1.1–7.1) 0.03 
Poppers use, past 30 days 0.4 (0.2–1.1) 0.07 
Viagra, Cialis or Levitra use, past 30 days 3.5 (1.8–6.8) <0.001 
Risk Active versus passiveab P 
Number of hours spent in online chat rooms in past 7 days 1.3 (1.01–1.9) 0.03 
Sex with a female partner during past 2 years 0.9 (0.6–1.7) 0.9 
Oral sex with multiple partners, past 3 months 2.5 (1.4–4.5) 0.002 
Receptive anal sex with multiple partners, past 3 months 2.2 (1.4–3.5) <0.001 
Insertive anal sex with multiple partners, past 3 months 1.1 (0.8–1.8) 0.5 
Chatted with a man did not already know and had oral sex with him 1.2 (0.8–2.0) 0.4 
Chatted with a man did not already know and had anal sex with him 1.1 (0.7–1.6) 0.8 
Inconsistent condom use during anal sex with men met online, past 3 months 1.8 (1.2–3.0) <0.001 
Ever had an STD 1.3 (1.1–1.6) 0.004 
Positive HIV serostatus 1.5 (1.2–2.1) 0.003 
Never been tested for HIV 1.2 (0.9–1.4) 0.2 
Illicit drug use, past 30 days 2.5 (1.5–4.4) <0.001 
Methamphetamine use, past 30 days 6.7 (1.7–26.1) 0.02 
Cocaine use, past 30 days 1.3 (0.7–2.5) 0.5 
Crack use, past 30 days 1.4 (0.3–6.1) 0.7 
Ecstasy use, past 30 days 2.8 (1.1–7.1) 0.03 
Poppers use, past 30 days 0.4 (0.2–1.1) 0.07 
Viagra, Cialis or Levitra use, past 30 days 3.5 (1.8–6.8) <0.001 
a

Odds ratio and (95% confidence interval) adjusted for race/ethnicity, income and sexual orientation.

b

Actively recruited chatters serve as the referent group.

Discussion

Several findings from this analysis deserve highlighting. First, the overall analysis included a majority of non-White participants. This is one of a few studies of online MSM that was able to recruit a majority of racial/ethnic minorities. Most often, participants in online research targeting MSM tend to be White [4, 10, 13, 25–28]. One study of HIV-risk behaviors reported recruiting than 1 500 Latino MSM through the use of banner advertisements [29]. Although this study differed in recruitment strategy from our own study reported here, it provides further evidence of the growing potential to recruit racial/ethnic minorities online. Moreover, aggregate trends of Internet access and use as defined by the ‘digital divide’, which suggests that within the United States younger, more educated, higher income, White men have greater access to the Internet [30–32], may be changing as the number of individuals online increases. Furthermore, populations most at risk for HIV appear to be early adopters of the Internet for both sexual and non-sexual communication [21]; thus, use among MSM may not reflect use suggested by aggregate trends. Our recruitment of a majority of Black and African American MSM runs counter to US population trends; however, White chatters were more likely to participate in the online assessment if they were actively recruited, rising questions about race and bias in access to HIV prevention interventions.

Despite this racial/ethnic diversity, however, White chatters were more likely to participate in the assessment if they were actively recruited. Because active recruitment included engaging in a one-on-one chat with an educator, racial/ethnic minority chatters may not have wanted to engage in dialogue with the educator or participate in an intervention. They may not have trusted the anonymity of the assessment when actively recruited. Trust has been identified as a barrier to participation in research for some racial and ethnic minorities [33–36], including racial/ethnic minority gay men [37]. An important consideration is that chat room-based interventions may further reinforce inequalities in health outcomes if they reach men who are more likely to participate.

Sexual orientation also was associated with recruitment strategy. Participants who self-identified as bisexual were less likely to participate in the assessment when actively recruited. Bisexual chatters may not have trusted the anonymity of the assessment when actively recruited; they may have worried about female partners discovering their same-sex behavior. Furthermore, because research has identified stigma attached to bisexuality [38–40], these men may have had a level of perceived discrimination from the gay community, either online or in person, which in turn discouraged their participation in active recruitment.

Besides the demographic differences identified by recruitment strategy, risk also differed by recruitment strategy. Of 18 risks examined, passively recruited participants were more likely than the actively recruited participants to report 10 of them. Actively recruited participants were more likely to report none of the 18 risks.

Most notably, passively recruited participants had over double the odds of reporting oral sex and receptive anal sex with multiple partners during the past 3 months, and nearly double the odds of inconsistent condom use during anal sex with men met online, during the past 3 months. Coupling these increased risks with the increased odds of reporting having ever had an STD and being HIV seropositive, these are particularly important findings given that these participants were not reached by the CyBER/M4M intervention.

In addition to sexual risk and history of STD and HIV infection, drug use differed between the two recruitment groups. Passively recruited participants were at increased odds of reporting illicit drug use, methamphetamine use and ecstasy use during the past 30 days. These participants also were at increased odds of reporting use of drugs to enhance sexual satisfaction. These findings are important given the links of use of these substances, including methamphetamines, ecstasy and drugs to enhance sexual satisfaction, with sexual risk behavior [41–46].

Implications of this analysis

The implications of this analysis fall into three categories: (i) the sexual risks of MSM in Internet chat rooms, (ii) the findings in terms of chat room-based data collection and (iii) the findings in terms of chat room-based HIV prevention outreach interventions.

The sexual risks of MSM in Internet chat rooms

As described in this analysis (and in the literature on the epidemiology and behaviors of MSM who use Internet chat rooms for social and sexual networking), chatters were at risk for HIV and other STDs. Further research is needed to understand and document social and sexual networks within online communities; to explore ways to reach the various online subgroups, including those that may be harder to reach for outreach and intervention (e.g. those identified within this analysis), and to identify potential leverage points (i.e. changeable key mediators and moderators) to reduce the risk of HIV exposure, infection and reinfection. How men negotiate risk with partners met online was not explored in this study; however, risk-reduction techniques such as engaging in oral as opposed to anal sex may occur between men as they negotiate sex online or after meeting in person.

Implications for chat room-based data collection

This analysis suggested that active recruitment led to a sample of online chatters who were at risk but were not at the greatest risk within the chat room. For chat room-based data collection efforts, using general messages (as opposed to targeted messages) to the public chat room may be more effective in collecting data on the higher risk chatters. Further research is necessary to document and understand the demographic, psychological and sociocultural characteristics of participants using different recruitment strategies.

It also remains unclear how participants recruited using active versus passive recruitment in chat rooms might differ from participants recruited using other online modalities such as banner advertisements. Furthermore, we have no information how chatters who did not participate in either recruitment strategies differed. Although research has begun to compare online to traditional samples of MSM (this research is nascent and may evolve as Internet technology changes), comparisons of online samples by recruitment strategies are rare.

Implications for chat room-based HIV prevention outreach interventions

This analysis also suggested that although CyBER/M4M was reaching those at risk for HIV exposure, infection and reinfection, it did not reach chatters at higher risk. Higher risk chatters were less likely to engage with the educator and thus less likely to enroll in CyBER/M4M. Further research is needed to understand more about these chatters and to determine what types of HIV prevention outreach interventions would reach them, particularly racial and ethnic minority MSM who tend to carry a disproportionate infection burden and have worse health outcomes.

Limitations

This study is not without limitations. First, we were unable to calculate a response rate for each recruitment strategy because of the unknown number of chatters (potential participants) who read the general public chat room messages but chose not to reply or participate. Comparing the number of chatters in the chat room to the number who completed the assessment would not be accurate because chatters engaged in private chats may not attend to messages in the public room; thus, the denominator may be artificially high.

Second, the observed associations are based on cross-sectional data. Additional studies using a prospective cohort design will be necessary to evaluate the significance and stability of these findings over time. Furthermore, although the Internet provides a data collection mode that may minimize response bias among other limitations [26], these results remain based on self-reported data and their potential limitations, including halo effects. However, techniques similar to those found to increase validity of self-reported behavior when applied to paper-and-pencil assessments [10, 26, 47–49] were applied in this study.

Third, the generalizability of the findings to other populations of MSM is unclear. MSM in the United States frequently display behaviors that differ from their counterparts in other countries. However, these data provide important information about the epidemic in a region of the United States and among a population disproportionately affected by HIV infection rates; little is known about the risks of chatters in general and in the Southeastern United States in particular. Moreover, the ways in which different online recruitment strategies reach different potential participants and how these participants differ are predominately unexplored in the current research literature.

Fourth, the measurement of race/ethnicity was imprecise. Combining race and ethnicity into one item rather than as two separate items blurs distinctions; however, it does require participants to commit to a primary characterization and may inform research and prevention strategies by identifying primary groups of self-identification. It also illustrates a compromise made to create the brief risk assessment.

Finally, the low frequency of some drug-use behaviors made identifying potential differences difficult.

Conclusions

Well into the HIV epidemic's third decade, the Southeastern United States is experiencing a disproportionate HIV/AIDS burden in comparison to other regions of the United States [50–52]. It also has been suggested that the Internet plays a growing role in facilitating social and sexual networking among MSM in rural areas [53], for whom other options for social and sexual networking are limited. It should be recognized that the Internet serves as an important tool in the development of men's social and sexual identity [3, 15, 54]. Understanding the differences in Internet use, HIV-related risks and willingness to participate in both online data collection and prevention outreach interventions is crucial to documenting and characterizing risk and to creating targeted and tailored HIV prevention outreach interventions.

In this analysis, profound differences between participants by recruitment strategy (i.e. active versus passive) were identified that may inform future data collection and prevention approaches. Subsequent research should include (i) comparisons of the multitude of online participant recruitment modalities to one another and to traditional and emerging modalities; (ii) more comprehensive qualitative and quantitative data collection to identify and understand both risk trends among online MSM and factors associated with participation in online data collection and health-promoting interventions; (iii) exploration of sociocultural factors associated with online sexual networking and risk behavior; (iv) documentation of the influence of biphobia on risk and participation in research and (v) the development, implementation and evaluation of creative and scientifically sound online interventions to affect the diversity of MSM, to prevent a variety of health risk behaviors and promote men's health. The first step is understanding and characterizing the subgroups of online MSM.

Conflict of interest statement

None declared.

This study was funded by a grant (to S.D.R.) from University of North Carolina Center for AIDS Research, a National Institutes of Health (NIH)-funded program—P30 A150410. L.J.Y. was supported by the NIH Roadmap Multidisciplinary Clinical Research Career Development Award Grant (8K12 RR023267) from the NIH. We also thank Emily Knipper, a medical student at Wake Forest University School of Medicine, and Sara Pula, a graduate student in the George Washington University doctoral program in Counseling, for their careful and helpful reviews and critiques of the manuscript.

References

1.
Hospers
HJ
Harterink
P
Van Den Hoek
K
, et al.  . 
Chatters on the Internet: a special target group for HIV prevention
AIDS Care
 , 
2002
, vol. 
14
 (pg. 
539
-
44
)
2.
Anderton
JP
Valdiserri
RO
Combating syphilis and HIV among users of internet chatrooms
J Health Commun
 , 
2005
, vol. 
10
 (pg. 
665
-
71
)
3.
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
, vol. 
16
 (pg. 
315
-
27
)
4.
Bull
SS
McFarlane
M
Soliciting sex on the Internet: what are the risks for sexually transmitted diseases and HIV?
Sex Transm Dis
 , 
2000
, vol. 
27
 (pg. 
545
-
50
)
5.
Elford
J
Bolding
G
Sherr
L
Seeking sex on the Internet and sexual risk behaviour among gay men using London gyms
AIDS
 , 
2001
, vol. 
15
 (pg. 
1409
-
15
)
6.
Elford
J
Bolding
G
Davis
M
, et al.  . 
Web-based behavioral surveillance among men who have sex with men: a comparison of online and offline samples in London, UK
J Acquir Immune Defic Syndr
 , 
2004
, vol. 
35
 (pg. 
421
-
6
)
7.
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
, vol. 
16
 (pg. 
953
-
63
)
8.
Hospers
HJ
Kok
G
Harterink
P
, et al.  . 
A new meeting place: chatting on the Internet, e-dating and sexual risk behaviour among Dutch men who have sex with men
AIDS
 , 
2005
, vol. 
19
 (pg. 
1097
-
101
)
9.
Kim
AA
Kent
C
McFarland
W
, et al.  . 
Cruising on the Internet highway
J Acquir Immune Defic Syndr
 , 
2001
, vol. 
28
 (pg. 
89
-
93
)
10.
Rhodes
SD
DiClemente
RJ
Cecil
H
, et al.  . 
Risk among men who have sex with men in the United States: a comparison of an Internet sample and a conventional outreach sample
AIDS Educ Prev
 , 
2002
, vol. 
14
 (pg. 
41
-
50
)
11.
Ross
MW
Tikkanen
R
Mansson
SA
Differences between Internet samples and conventional samples of men who have sex with men: implications for research and HIV interventions
Soc Sci Med
 , 
2000
, vol. 
51
 (pg. 
749
-
58
)
12.
Tashima
KT
Alt
EN
Harwell
JI
, et al.  . 
Internet sex-seeking leads to acute HIV infection: a report of two cases
Int J STD AIDS
 , 
2003
, vol. 
14
 (pg. 
285
-
6
)
13.
Taylor
M
Aynalem
G
Smith
L
, et al.  . 
Correlates of Internet use to meet sex partners among men who have sex with men diagnosed with early syphilis in Los Angeles County
Sex Transm Dis
 , 
2004
, vol. 
31
 (pg. 
552
-
6
)
14.
Tikkanen
R
Ross
MW
Technological tearoom trade: characteristics of Swedish men visiting gay Internet chat rooms
AIDS Educ Prev
 , 
2003
, vol. 
15
 (pg. 
122
-
32
)
15.
Bolding
G
Davis
M
Sherr
L
, et al.  . 
Use of gay Internet sites and views about online health promotion among men who have sex with men
AIDS Care
 , 
2004
, vol. 
16
 (pg. 
993
-
1001
)
16.
MacMaster
SA
Aquino
R
Vail
KA
Providing HIV education and outreach via Internet chat rooms to men who have sex with men
J Hum Behav Soc Environ
 , 
2003
, vol. 
8
 (pg. 
145
-
51
)
17.
McFarlane
M
Kachur
R
Klausner
JD
, et al.  . 
Internet-based health promotion and disease control in the 8 cities: successes, barriers, and future plans
Sex Transm Dis
 , 
2005
, vol. 
32
 (pg. 
S60
-
4
)
18.
Chiasson
MA
Parsons
JT
Tesoriero
JM
, et al.  . 
HIV behavioral research online
J Urban Health
 , 
2006
, vol. 
83
 (pg. 
73
-
85
)
19.
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
, vol. 
33
 (pg. 
576
-
84
)
20.
Pequegnat
W
Rosser
BR
Bowen
AM
, et al.  . 
Conducting internet-based HIV/STD prevention survey research: considerations in design and evaluation
AIDS Behav.
 , 
2006
October
20
Advance Access published
 
doi: 10.1007/s10461-006-9172-9
21.
Rhodes
SD
Hergenrather
KC
Duncan
J
, et al.  . 
Using community-based participatory research to develop a chat room-based HIV prevention intervention for gay men
Prog Commun Health Partnerships Res Educ Action
  
in press
22.
Arnstein
SR
A ladder of citizen participation
J Am Inst Plann
 , 
1969
, vol. 
35
 (pg. 
216
-
24
)
23.
Rhodes
SD
Yee
LJ
Hergenrather
KC
A community-based rapid assessment of HIV behavioural risk disparities within a large sample of gay men in the Southeastern USA: a comparison of African American, Latino, and white men
AIDS Care
 , 
2006
, vol. 
18
 (pg. 
1018
-
24
)
24.
Lelchuk
I.
Net syphilis issue spurs hate mail
San Francisco Examiner
 , 
1999
August
26
pg. 
A1
 
25.
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
, vol. 
31
 (pg. 
177
-
83
)
26.
Rhodes
SD
Bowie
DA
Hergenrather
KC
Collecting behavioural data using the world wide web: considerations for researchers
J Epidemiol Community Health
 , 
2003
, vol. 
57
 (pg. 
68
-
73
)
27.
Rhodes
SD
DiClemente
RJ
Yee
LJ
, et al.  . 
Correlates of hepatitis B vaccination in a high-risk population: an Internet sample
Am J Med
 , 
2001
, vol. 
110
 (pg. 
628
-
32
)
28.
Rhodes
SD
DiClemente
RJ
Yee
LJ
, et al.  . 
Factors associated with testing for hepatitis C in an internet-recruited sample of men who have sex with men
Sex Transm Dis
 , 
2001
, vol. 
28
 (pg. 
515
-
20
)
29.
Ross
MW
Rosser
BR
Stanton
J
, et al.  . 
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
, vol. 
16
 (pg. 
526
-
37
)
30.
Bernhardt
JM
Health education and the digital divide: building bridges and filling chasms
Health Educ Res
 , 
2000
, vol. 
15
 (pg. 
527
-
31
)
31.
Lorence
DP
Park
H
Fox
S
Racial disparities in health information access: resilience of the digital divide
J Med Syst
 , 
2006
, vol. 
30
 (pg. 
241
-
9
)
32.
Whaley
KC
America's digital divide: 2000–2003 trends
J Med Syst
 , 
2004
, vol. 
28
 (pg. 
183
-
95
)
33.
Cargill
VA
Stone
VE
HIV/AIDS: a minority health issue
Med Clin North Am
 , 
2005
, vol. 
89
 (pg. 
895
-
912
)
34.
Corbie-Smith
G
Ammerman
AS
Katz
ML
, et al.  . 
Trust, benefit, satisfaction, and burden: a randomized controlled trial to reduce cancer risk through African-American churches
J Gen Intern Med
 , 
2003
, vol. 
18
 (pg. 
531
-
41
)
35.
Freimuth
VS
Quinn
SC
Thomas
SB
, et al.  . 
African Americans' views on research and the Tuskegee Syphilis Study
Soc Sci Med
 , 
2001
, vol. 
52
 (pg. 
797
-
808
)
36.
Sengupta
S
Strauss
RP
DeVellis
R
, et al.  . 
Factors affecting African-American participation in AIDS research
J Acquir Immune Defic Syndr
 , 
2000
, vol. 
24
 (pg. 
275
-
84
)
37.
Rhodes
SD
Hergenrather
KC
Exploring hepatitis B vaccination acceptance among young men who have sex with men: facilitators and barriers
Prev Med
 , 
2002
, vol. 
35
 (pg. 
128
-
34
)
38.
Elaison
M
Bi-negativity: the stigma facing bisexual men
J Bisexuality
 , 
2001
, vol. 
1
 (pg. 
137
-
54
)
39.
Steinman
E
Interpreting the invisibility of male bisexuality theories, interactions, politics
J Bisexuality
 , 
2001
, vol. 
1
 (pg. 
15
-
45
)
40.
Worth
H
The myth of the bisexual infector? HIV risk and men who have sex with men and women
J Bisexuality
 , 
2003
, vol. 
3
 (pg. 
71
-
88
)
41.
Fernandez
MI
Perrino
T
Collazo
JB
, et al.  . 
Surfing new territory: club-drug use and risky sex among Hispanic men who have sex with men recruited on the Internet
J Urban Health
 , 
2005
, vol. 
82
 (pg. 
Si79
-
88
)
42.
Parsons
JT
Bimbi
DS
Intentional unprotected anal intercourse among men who have sex with men: barebacking-from behavior to identity
AIDS Behav
 , 
2007
, vol. 
11
 (pg. 
277
-
87
)
43.
Semple
SJ
Zians
J
Grant
I
, et al.  . 
Sexual compulsivity in a sample of HIV-positive methamphetamine-using gay and bisexual men
AIDS Behav
 , 
2006
, vol. 
10
 (pg. 
587
-
98
)
44.
Mansergh
G
Shouse
RL
Marks
G
, et al.  . 
Methamphetamine and sildenafil (Viagra) use are linked to unprotected receptive and insertive anal sex, respectively, in a sample of men who have sex with men
Sex Transm Infect
 , 
2006
, vol. 
82
 (pg. 
131
-
34
)
45.
Mansergh
G
Purcell
DW
Stall
R
, et al.  . 
CDC consultation on methamphetamine use and sexual risk behavior for HIV/STD infection: summary and suggestions
Public Health Rep
 , 
2006
, vol. 
121
 (pg. 
127
-
32
)
46.
Theall
KP
Elifson
KW
Sterk
CE
Sex, touch, and HIV risk among ecstasy users
AIDS Behav
 , 
2006
, vol. 
10
 (pg. 
169
-
78
)
47.
Fishbein
M
Pequegnat
W
Evaluating AIDS prevention interventions using behavioral and biological outcome measures
Sex Transm Dis
 , 
2000
, vol. 
27
 (pg. 
101
-
10
)
48.
Pequegnat
W
Fishbein
M
Celentano
D
, et al.  . 
NIMH/APPC workgroup on behavioral and biological outcomes in HIV/STD prevention studies: a position statement
Sex Transm Dis
 , 
2000
, vol. 
27
 (pg. 
127
-
32
)
49.
Rhodes
SD
Grimley
DM
Hergenrather
KC
Integrating behavioral theory to understand hepatitis B vaccination among men who have sex with men
Am J Health Behav
 , 
2003
, vol. 
27
 (pg. 
291
-
300
)
50.
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
Morb Mortal Wkly Rep
 , 
2005
, vol. 
52
 (pg. 
597
-
601
)
51.
Centers for Disease Control and Prevention
HIV/AIDS Surveillance Report, 2003
 , 
2005
, vol. 
volume 15
 
Atlanta, GA
US Department of Health and Human Services
(pg. 
1
-
46
)
52.
Southern State Directors Work Group
Southern States Manifesto: HIV/AIDS and STD's in the South
2003
Washington, DC
National Alliance of State and Territorial AIDS Directors
53.
Horvath
KJ
Bowen
AM
Williams
ML
Virtual and physical venues as contexts for HIV risk among rural men who have sex with men
Health Psychol
 , 
2006
, vol. 
25
 (pg. 
237
-
42
)
54.
Macias
W
Lewis
LS
Smith
TL
Health-related message boards/chat rooms on the Web: discussion content and implications for pharmaceutical sponsorships
J Health Commun
 , 
2005
, vol. 
10
 (pg. 
209
-
23
)

Comments

0 Comments