| |
A new meeting place: chatting on the Internet, e-dating and sexual risk behaviour among Dutch men who have sex with men
|
| |
| |
"... among HIV-positive men, a significantly higher proportion reported unprotected anal intercourse with their last e-date compared with HIV-negative and never-tested men..... These findings call for increased efforts to target preventative interventions at MSM who visit Internet chatrooms...."
AIDS: Volume 19(10) 1 July 2005
Hospers, Harm Ja; Kok, Gerjoa; Harterink, Paulb; de Zwart, Onnob
From the aMaastricht University, Faculty of Psychology, Department of Experimental Psychology, Maastricht, the Netherlands
bMunicipal Health Service Rotterdam and Environs, Rotterdam, the Netherlands.
Abstract
Objective: To describe the process of Internet chatting, and subsequent dating and sexual (risk) behaviour among Dutch men who have sex with men (MSM), and to compare the demographic profile of the Internet sample with a traditional Dutch MSM sample.
Design: A self-selected anonymous Internet sample.
Methods: Enrolment via a prominent gay chat room. Potential respondents were asked to complete an online questionnaire about chatting and dating, and sexual (risk) behaviour with their last date via chatting (e-date). In 26 days, 5302 questionnaires were returned. Of these, 4984 were evaluated as reliable and were used in the analyses.
Results: Eighty-two per cent chatted at least once a week, 88% had ever dated through chatting, and of these 89% had had sex with one or more e-dates. With respect to the last e-date, almost 50% had had anal sex with their last e-date, and 15% reported unprotected anal sex. Especially among HIV-positive men, a high percentage of unprotected anal sex was reported (39%). After correcting for the disclosure of HIV status, this percentage remained twice as high compared with HIV-negative and never-tested men (28 versus 14%). Compared with a traditional MSM sample, the Internet sample was significantly younger, and comprised more non-Dutch and bisexual men, whereas the level of sexual risk behaviour with casual partners was comparable.
Conclusion: The Internet is a popular new meeting place for MSM, and attracts men with a different demographic profile. The level of risk behaviour warrants that opportunities for interventions on the Internet are explored, with special attention to HIV-positive chatters.
Introduction
The Internet has become increasingly popular for seeking sex partners [1], especially among men who have sex with men (MSM) [2-5]. In 1999, the Dutch MSM Monitoring Study (DMMS) showed that 17% of respondents reported chatting [6]. In 2002, this percentage had increased to 46% [7]. Results from sexually transmitted infection clinics have shown that among MSM who were diagnosed with syphilis, increasing proportions use the Internet to find sexual partners [8-10]. Studies that have investigated Internet MSM samples reported that these are usually younger and comprise more bisexual men in comparison with traditionally recruited MSM samples [2,3,11-14].
A limited number of studies have compared the sexual behaviour of MSM from Internet samples with traditional samples [1,2,11-15], using a variety of sampling strategies, and with varying sample sizes (283 ≦ N ≦ 2097). Four of the studies found that Internet MSM reported more casual partners [2,3,11,13], whereas two did not find a difference [12,15]. Two studies showed that Internet MSM were likely to report more sexually transmitted infections [2,13], whereas one study did not find a difference [3]. Finally, three studies reported more unprotected anal intercourse with casual partners among Internet MSM [2,13,14], and two other studies found no difference [11,12].
The present study investigated chatting, dating, and sex with e-dates among Dutch MSM. Second, we compared the online sample with a traditional MSM sample, i.e. the 2002 wave of the DMMS (N = 1238) [7].
RESULTS
The mean age was 33.2 years (median 32, SD 10.05, range 14-77). The majority (58%) had a higher educational level, 42% a lower educational level. Most respondents (81%) reported a Dutch cultural background, 15% a combination of a Dutch/non-Dutch background (e.g. Dutch-Moroccan), and 4% a non-Dutch background. Most respondents (80%) reported being mostly or exclusively attracted to men (hereafter gay men), whereas 20% were attracted to men and women or mostly to women (hereafter bisexual men).
Almost half of all respondents (49%) were in a steady relationship. Of those, 73% had a relationship with a man and 27% had a relationship with a woman. The large majority (92%) reported sex with men in the 6 months preceding the assessment. These men reported a median of four sex partners (mean 8.4, SD 14.1, range 1-300). Among men who reported casual partners, 57% reported engaging in anal sex with casual sex partners, and 23% reported unprotected anal intercourse. Of all respondents, 44% had ever been tested for HIV. Among tested men, 6% reported being HIV-positive.
Chatting and dating behaviour
More than a quarter of the respondents (26%) chatted daily, 56% at least once a week, and 18% less frequently. The majority of the sample (88%) reported experience with e-dating. Men with e-dating experience reported a median of three e-dates during the previous 6 months (mean 5.3, SD 9.00, range 0-190). Among men who report e-dating, 89% had also engaged in sex with one or more e-dates.
Last e-date with sex
Respondents who reported experience with sex with e-dates (78% of total sample) were asked about their last sexual encounter with and e-date. Most men (56%) had the date immediately after the chat session, 44% later. Non-disclosure of HIV status: in 79% of encounters neither the respondent nor the sex partner disclosed their status between the chat and sex.
Of all the respondents who had had sex with e-dates (78% of total sample), approximately half had had anal sex with their last e-date, and 15% reported unprotected anal intercourse with their last e-date. This percentage was significantly higher among men younger than 24 years (18%) compared with men aged 33 years and over, among men with a lower educational level (17%), among non-Dutch men (19%), and among HIV-positive men (39%). As we asked whether respondents had disclosed their HIV status before sex with their last e-date, we were able to correct for positive-positive sex. This lowered the percentage of unsafe sex among HIV-positive men from 39 to 28%. This percentage is still significantly higher compared with HIV-negative men or never-tested men.
The Internet sample compared with the Dutch MSM Monitoring Study
Finally, we compared demographics and sexual behaviour with casual partners between the Internet sample and the 2002 DMMS. Men in the Internet sample were substantially younger compared with the DMMS. Furthermore, the Internet sample comprised considerably more men with a mixed Dutch/non-Dutch cultural background or a non-Dutch cultural background (Internet sample 19%, DMMS 5%), and more bisexual men (Internet sample 20%, DMMS 5%). Fewer men in the Internet sample had ever had an HIV test compared with men in the DMMS (44 and 53%, respectively). Differences in educational level were not significant.
A slightly higher percentage in the DMMS reported anal sex with casual partners in the 6 months preceding assessment, compared with the Internet sample (61 versus 57% respectively, ns). The level of unprotected anal sex with casual partners was comparable between the samples (Internet sample 23%, DMMS 21%). The median number of casual sex partners was significantly higher in the DMMS compared with the Internet sample (five versus four).
AUTHOR DISCUSSION
The present study succeeded in attracting over 5000 MSM through the Internet within a month, and to collect detailed information on chatting and dating behaviour. The Internet sample was substantially younger than our offline MSM sample, comprised considerably more bisexual men, and men with a mixed or non-Dutch cultural background. The Internet is obviously providing a more convenient and confidential means of meeting people and finding sexual partners, and is leading to a greater diversity in those men seeking sex with men. This may cause changes in traditional sexual networks, with possible subsequent implications for HIV transmission.
In contrast to several studies [2,3,11,13,14], and in line with other studies [12,15], we did not find differences in the level of unprotected anal intercourse with casual partners between the online and offline samples. Whether or not the Internet attracts more risk-taking MSM thus remains inconclusive. However, it should be noted that sampling differences may contribute to these discrepant results.
Higher rates of unprotected anal sex were observed among younger men, men with a lower educational background, and men with a mixed cultural or a non-Dutch background. These demographic characteristics are usually related to sexual risk-taking behaviour [16]. Common explanations include that these men have been less exposed to preventative interventions [16], and that the available information may not be tailored to their needs or cognitive capacities [17]. Therefore, special attention for these risk groups continues to be warranted. Finally, among HIV-positive men, a significantly higher proportion reported unprotected anal intercourse with their last e-date compared with HIV-negative and never-tested men. Furthermore, after controlling for positive-positive disclosure, the proportion of HIV-positive men who had had unprotected anal sex with their last e-date was still twice as high.
These findings call for increased efforts to target preventative interventions at MSM who visit Internet chatrooms. There is limited experience with Internet-based interventions. So far, one randomized controlled trial targeting MSM has been reported [18], which resulted in safer sexual behaviour in the experimental condition compared with the control group. A focus on HIV-positive MSM seems sensible given the higher levels of unprotected sex.
The present study had several limitations. First, the response was self-selected and we were not able to assess the differences between responders and non-responders. Moreover, the exact response rate could not be computed because the number of multiple visits to our survey site was unknown. On the other hand, traditional surveys are usually similarly hampered. Second, the data were obtained from chatters, a subgroup of MSM, and the results can therefore not be generalized to the total MSM population. Finally, we had to rely on self-reports regarding HIV status and behavioural variables.
In conclusion, we believe that this study contributes to the insight into an important new meeting place for MSM. As we have shown, Internet chatting and subsequent dating are increasingly popular, and substantial levels of risk-taking sex were observed. Consequently, the opportunities for online HIV-preventative interventions for this target group should be considered.
Methods
Participants were enrolled through the major gay chat room in the Netherlands
(www.chatboy.nl ), and were asked to complete an online questionnaire.
Questionnaire
The interactive online questionnaire contained the following measures: Demographics: age, educational level, cultural background, steady partner status, sexual preference, HIV testing, HIV status; Chatting and dating behaviour: frequency of chatting, number of dates and sex with e-dates in preceding 6 months; Sexual behaviour, general and with e-dating: number of casual sexual partners in the preceding 6 months, anal intercourse, condom use; Sexual behaviour with last e-date: anal intercourse, condom use, disclosure of HIV status before having sex.
In addition, social-cognitive variables were measured that will be reported elsewhere.
Procedure
The questionnaire was uploaded to a secure website, only accessible to visitors of Chatboy. The link between Chatboy and the survey site consisted of a banner on the homepage, and the round-the-clock presence of a chat room visitor with the nickname 'E-Dating Survey 2002'. Clicking this 'visitor' showed a graphics banner, plus an invitation to participate in the survey by clicking the banner.
Enrolment
The entry page of the survey site was visited approximately 12 000 times during the period 28 March to 23 April 2002. This number includes multiple visits, and is therefore an overestimation of the number of unique visitors. From the introduction page, visitors could choose to participate by clicking an 'Enter' button.
Of those visitors arriving at the survey site 8960 (75%) continued to the questionnaire by clicking the 'Enter'-button. This is again an overestimation because of multiple visits. The first questionnaire page contained answering instructions and a button to go to the first question. Nearly 90% of visitors (8050) continued to this first question. Approximately two-thirds of them (5302) completed and submitted the questionnaire.
The response rate was thus approximately 45% of all visitors of the survey website, and 59% of those who continued on to the questionnaire. As the total number of visitors to Chatboy in the data collection period was unknown, the response rate compared with that denominator cannot be provided.
Data quality TOP
The 5302 cases were subjected to a rigorous data cleaning process. The checks included unrealistically short completion times relative to the number of questions that had to be answered, multiple questionnaires from the same IP address [every computer connected to the Internet is assigned a unique number known as an Internet Protocol (IP) address], and inconsistent answering patterns.
This cleaning process led to 318 cases being removed, resulting in 4984 cases that were used in the analyses below.
Statistical analyses
Because of the large sample size, differences were tested using a P value of 0.01.
|
|
| |
| |
|
|
|
