Young Adults on the Internet: Risk Behaviors for Sexually Transmitted Diseases and HIV

JOURNAL OF ADOLESCENT HEALTH 2002;31:11 16 ORIGINAL ARTICLE Young Adults on the Internet: Risk Behaviors for Sexually Transmitted Diseases and HIV MARY McFARLANE, Ph.D., SHEANA S. BULL, Ph.D., M.P.H. AND CORNELIS A. RIETMEIJER, M.D., M.P.H. From the Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta, Georgia (M.M.); AMC Cancer Research Center, Atlanta, Georgia (S.S.B.); and Denver Health and Hospitals Authority, Atlanta, Georgia (C.A.R.). Address correspondence to: Mary McFarlane, Ph.D., 1600 Clifton Road NE, Mail stop E44, Atlanta, Georgia 30333. E-mail: xzm3@cdc.gov. Manuscript accepted January 30, 2002. The full text of this article is available via JAH Online at http:// www.elsevier.com/locate/jahonline Purpose: To examine the sexual behaviors and related risk factors for sexually transmitted diseases and HIV among young adults who seek sex partners on the Internet. Methods: Study staff recruited participants in online chat rooms, bulletin boards, and other online venues. A total of 4507 participants responded to a 68-item, selfadministered, online survey of Internet sex-seeking practices. The survey solicited information on sexual behavior with partners found on the Internet; in addition, a parallel set of questions addressed sexual behaviors with partners found off the Internet. Of the respondents, 1234 (27.4%) were 18 24 years old. Of the young adults, 61% were male and 75% were white. Responses from young adults were compared to those received from older adults. In addition, responses from young adults who seek sex partners online were compared to responses from young adults who do not seek sex partners online. Analyses, including logistic regression, Chi-square tests, Student s t-tests, and analyses of variance, focused on the difference between young and older adults, as well as the differences in sexual behavior with partners located online and offline. Results: Young adults who seek sex on the Internet report substantially different sexual behavior patterns than young adults who do not seek sex on the Internet. Young adults with online partners reported sexual behaviors similar to older respondents who used the Internet to find sex partners; however, older respondents were more likely than young adults to have been tested for sexually transmitted diseases and HIV. Conclusions: Young adults who seek sex partners online may be at significantly greater risk for sexually transmitted diseases than their peers who do not seek sex partners online. These data point to an urgent need for online sexual health promotion. Society for Adolescent Medicine, 2002 KEY WORDS: Internet Sexual behavior Sexually transmitted diseases Human immunodeficiency virus Risk behavior Young adults The Internet has become an important part of daily American life, with over 45 million households online in the year 2000, and a projected increase to 90 million households (170 million users) in 2004 [1]. According to a report by Grunwald Associates [2], more than 25 million children are on the Internet, representing 40% of American young adults between the ages of 2 and 17 years. That number is predicted to climb to 44 million by the year 2005. In addition, there are more ways than ever to gain access to the Internet, including desktop computers in the home, school, workplace, and recreational facilities, and handheld or portable devices allowing Internet access from virtually any location. In previous reports, we have shown that one use of the Internet is the initiation of risky sexual contact between anonymous partners [3 5]. Individuals who Society for Adolescent Medicine, 2002 Published by Elsevier Science Inc., 655 Avenue of the Americas, New York, NY 10010 1054-139X/02/$ see front matter PII S1054-139X(02)00373-7

12 McFARLANE ET AL. JOURNAL OF ADOLESCENT HEALTH Vol. 31, No. 1 log onto the Internet with the intention of finding sex partners (referred to as online seekers ) are at greater risk for sexually transmitted diseases (STDs), including human immunodeficiency virus (HIV) than those who do not seek sex partners online [3]. Internet partner-seeking may hold special appeal for individuals who prefer to keep their sexual activities hidden, because of the perceived anonymity, the availability of partners, and the obstacles to social interaction in bars, clubs, or other social venues. Furthermore, it may be easier for those interested in homosexual activity to explore these interests in an online environment, where stigma is lower, social groups are larger, and where anonymity provides a measure of perceived safety. Our previous research shows that these online environments are well-used and are not predisposed to discussions of STD/HIV status or condom use [4]. Issues related to increased risk for STD/HIV are particularly salient for young adults. Considering the large number of youth who have access to and use the Internet frequently, it is important to understand online partner-seeking habits and related STD/HIV risk of this young population. Here we examine data from an online survey of sex behavior ( SexQuiz ), designed to assess the nature and prevalence of online partner-seeking. Methods As a result of our previous research [4], we developed a 68-item, self-administered, anonymous survey designed to document behaviors related to seeking sex partners online, as well as HIV/STD related risk behaviors with Internet and non-internet partners. The study was approved for persons aged 18 years and older by the Colorado Multiple Institutional Review Board (COMIRB) and by Institutional Review Boards at the Centers for Disease Control and Prevention and the AMC Cancer Research Center. Persons logging onto www.sexquiz.org during the study period (April 3 through August 3, 2000) saw a home page of information about the study, with links to other web sites offering HIV/STD prevention information, local testing information, and frequently asked questions (FAQs) about Sex- Quiz. Those interested in completing the survey clicked on a button to continue to the informed consent page. After reading the consent, those who wished to continue with the study clicked on a button saying I agree and began to self-administer the 68-item survey. The survey included questions on demographics (i.e., age, gender, employment, insurance, and education). We included general health questions on smoking, exercise, and diet. Participants were asked to indicate age of onset of sexual intercourse and lifetime number of sex partners. We gathered information on sexual risk behavior separately for Internet partners (those with whom first contact occurred via e-mail, chat rooms, bulletin boards, or other electronic media) and non-internet partners. Risk behavior information included number of partners in the past year; gender of partners (all male partners, all female partners, or both male and female partners); types of sex (anal, oral, vaginal); drug and alcohol use concomitant with sex; and condom use during last sexual encounter; and discussion of HIV/ AIDS or STD with partners. The survey was posted online at the web site www.sexquiz.org from April 3 through August 3, 2000. In an earlier phase of the study, staff had identified chat rooms and bulletin board sites on which people exhibited behaviors suggesting partner solicitation [4]. Staff returned to these locations to recruit persons for the online survey. They posted information about, and links to, the survey in chat rooms, on bulletin boards, and through multiple HIV/STD prevention list-serves and web sites. Staff made every effort to identify potential participants from rooms where ethnic/minority participation was likely (e.g., rooms titled Ethnic Ebony Love, Ebony and Ivory, Latin Lover, BlkM4M). Once in a chat room, staff would send out private messages (invisible to other chatters) to potential participants to encourage their enrollment in SexQuiz. In addition to staff recruitment, media attention about the survey resulted in strong recruitment. We received 64% of all responses (3503/5474) in one week coinciding with national media attention paid to the research. Persons enrolled were offered a risk score after the survey as a small incentive for completion. This was a crude assessment developed algorithmically from the risk assessment that weighted various risk behaviors (e.g. sex without a condom, number of partners, being high or drunk during sex more than 50% of the time) and placed people into limited, moderate, and higher risk categories, with explanations of what each risk behavior was, why it was risky, and how risk could be reduced. Persons were redirected to the page containing links to other HIV/STD related web sites, including testing resources, for further information. By August 3, 2000, 5474 surveys were completed.

July 2002 YOUNG ADULTS ON THE INTERNET 13 There were 231 persons (4% of those completing) who visited the web site but did not agree to participate in the study. Most common reasons offered for refusal to participate included lack of relevancy (persons who never used the Internet to find a sex partner, although this was not a criterion for eligibility) 38%, invasion of privacy (19%), and no time (11%). Of completed surveys, 64 surveys were deleted owing to age ineligibility or obvious false age (i.e. age less than 18 years or age 99 years or older). An additional 561 persons living outside North America and 342 persons with irreconcilable inconsistencies in survey responses were removed from data analysis. The final sample includes 4507 persons. Of those, 1234 or 27.4% were aged 18 24 years. We refer to respondents to the SexQuiz who reported having had sex with an Internet partner as SIPs. Respondents with no Internet-identified partners are called NIPs. SIPs numbered 2219 or 49.2% of the sample, and 495 (22.3%) of these were aged 18 24 years. Results Comparing Young SIPs to Young NIPs Table 1 presents comparisons of the risk behaviors and background data provided by young SIPs and young NIPs. Young SIPs are demographically similar to young NIPs who responded to the SexQuiz, with one notable exception: Young males are more likely to report having located sex partners online than young females. Both groups are largely white, with approximately 6% self-identifying as Hispanic. The groups are educationally similar, and are equally likely to be employed or working for pay (85%). Young SIPs report an earlier onset of sexual intercourse than those who did not have sex partners who they met online. Approximately 55% of young SIPs had been tested for HIV, while only 35% of their young NIP counterparts had been tested. Young SIPs were more likely to report having had an STD (10%) than young adults who did not have sex partners who they met online, and 14 young adults reported having been infected with an STD by an Internet partner. Over 40% of young adults who had met sex partners online reported having traveled 100 miles or more to meet sex partners. Young SIPs clearly can only be compared to their NIP counterparts by examining sexual behaviors with non-internet partners (i.e., partners who were not met over the Internet). Young SIPs had more lifetime and 12-month non-internet partners than Table 1. Young Persons (Age 18 24 Years) Reporting Sex With Internet Partners (SIPs; N 495) and Young Persons With No Internet Partners (NIPs; N 739) to SexQuiz Measure SIPs N 495 NIPs N 739 p-value Demographics Male 66.7 56.8 0.001 White 76.3 74.4 0.18 Black 5.1 5.3 Hispanic 6.1 6.4 0.88 Age at first sex (yrs) 16.6 (2.8) 17.1 (2.8) 0.003 Tested for HIV 55.3 35.2 0.001 Tested for STD 48.5 36.3 0.001 Had an STD 10.4 6.2 0.02 Non-Internet partners Lifetime partners 14.6 (47.1) 5.8 (11.1) 0.001 12-month partners 4.3 (16.2) 1.7 (1.5) 0.001 Partner sex 0.001 Same sex 30.8 5.7 Opposite sex 53.3 86.7 Both sexes 15.9 7.6 Oral sex 12 months 87.6 91.1 0.001 Anal sex 12 months 45.9 24.1 0.001 Vaginal sex 12 months 60.0 87.8 0.001 Condom/barrier last 3.3 4.2 0.32 oral sex Condom last anal/ 46.6 38.1 0.001 vaginal sex Drunk/high during sex 57.6 60.5 0.07 Discuss HIV with partners 60.5 60.3 0.93 Discuss STD with partners 62.3 59.7 0.66 Prevention chat room 31.5 26.1 0.04 Prevention e-mail 49.2 43.6 0.15 Meet partners in bar 14.2 8.6 0.009 Meet partners at work/ 22.3 30.9 0.006 school Meet partners in public environment 8.0 5.8 0.02 STD sexually transmitted disease. young NIPs. Young SIPs were far more likely than young NIPs to have same-sex partners. In general, SIPs and NIPs met their non-internet partners in similar venues; however, SIPs were more likely to meet their partners in bars or public environments, whereas NIPs were more likely to meet partners at school or work. Young SIPs were less likely than young NIPs to report oral sex with non-internet partners, but were equally unlikely to use a condom or barrier during oral sex with the most recent non-internet partner. SIPs reported far less vaginal sex with non-internet partners than NIPs. Interestingly, young SIPs were slightly more likely than their young NIP counter-

14 McFARLANE ET AL. JOURNAL OF ADOLESCENT HEALTH Vol. 31, No. 1 parts to use a condom during anal or vaginal sex with the last non-internet partner. However, young SIPs also reported more anal sex than NIP young adults. Young SIPs were just as likely as their young NIP counterparts to discuss HIV or STD with their non- Internet partners. In addition, both groups were equally likely to report being drunk or high with the last non-internet partner. Young SIPs were at least as willing as NIPs to visit prevention-oriented chat rooms or to open prevention-oriented electronic mail. Comparing Younger vs. Older SIPs In-depth comparisons of older (25 years or older) vs. younger (age 18 24 years) SIP respondents appear in Table 2. Young SIPs were less likely to be white than older SIPs, but were more likely to be Hispanic, more likely to be female, and equally likely to use the Internet each day or almost each day. Approximately 73% of respondents in each group logged on to the Internet from their homes. [Education and work-status differences are not reported here owing to obvious confounds with the grouping variable (age).] As expected, young SIP respondents had significantly fewer lifetime partners than their older counterparts; however, young and older respondents reported similar numbers of Internet partners (both lifetime and in the past 12 months), as well as similar numbers of non-internet partners in the past year. Younger SIP respondents report younger age at first sex than older respondents, but were less likely ever to have been tested for HIV or STD. Younger SIPs were less likely than older SIPs to have had sex with partners not met over the Internet, suggesting that for some young adults, the Internet is the only means of initiating sexual contact. Younger SIPs were more likely to have opposite-sex partners than older SIPs: this is true for both Internet and non- Internet partners. Condom use with Internet partners did not differ across age groups. Younger SIPs report less anal and oral sex with Internet partners than their older counterparts, but equal amounts of vaginal sex with Internet partners. With non-internet partners, young adults were more likely than older respondents to have used a condom during their last episode of anal or vaginal sex, but were equally likely to have used a condom or barrier during their last episode of oral sex. Young adults were slightly less likely than older respondents to have anal sex with non-internet partners, Table 2. Comparing Younger SIPs (18 24 Years) (N 495) With Older (25 Years or More) SIPs (N 1724) Measure Young SIPs N 495 Older SIPs N 1724 p-value Demographics Male 66.7 76.6 0.001 White 76.3 85.6 0.001 Black 5.1 3.4 Hispanic 6.1 3.8 0.03 Age at first sex (yrs) 16.6 (2.8) 17.1 (4.2) 0.01 Tested for HIV 55.3 75.8 0.001 Tested for STD 48.5 67.3 0.001 Had an STD 10.4 25.3 0.001 Non-Internet partners Lifetime partners 14.6 (47.1) 42.8 (93.5) 0.0001 12-month partners 4.3 (16.2) 4.7 (21.2) 0.72 Partner Sex 0.009 Same sex 30.8 38.7 Opposite sex 53.3 48.6 Both sexes 15.9 12.7 Oral sex 12 months 87.6 88.2 0.87 Anal sex 12 months 45.9 52.2 0.04 Vaginal sex 12 months 60.0 51.9 0.02 Condom/barrier last 3.3 5.3 0.12 oral sex Condom last anal/ 46.6 39.8 0.03 vaginal sex Drunk/high during sex 57.6 65.6 0.17 Discuss HIV with partners 60.5 60.3 0.93 Discuss STD with partners 59.7 59.8 0.47 Meet partners in bar 14.2 17.6 0.001 Meet partners at work/ 22.3 19.1 0.14 school Meet partners in public environment 8.0 9.5 0.33 Internet partners Lifetime partners 9.5 (56.2) 14.4 (78.2) 0.19 12-month partners 6.7 (55.7) 6.0 (23.5) 0.69 Partner sex 0.01 Same sex 38.3 45.6 Opposite sex 53.1 46.1 Both sexes 8.6 8.3 Oral sex 12 months 87.2 92.6 0.001 Anal sex 12 months 46.9 54.5 0.04 Vaginal sex 12 months 51.7 52.3 0.97 Condom/barrier last 5.5 7.3 0.02 oral sex Condom last anal/ 39.9 41.5 0.26 vaginal sex Drunk/High during sex 39.7 36.7 0.26 Discuss HIV with partners 64.5 75.6 0.001 Discuss STD with partners 63.6 67.8 0.004 Would read or visit.... Prevention chat room 31.5 34.4 0.23 Prevention e-mail 49.2 47.6 0.55 SIPs sex with Internet partners; NIPs no Internet partners; STD sexually transmitted disease.

July 2002 YOUNG ADULTS ON THE INTERNET 15 Table 3. Comparing Younger SIPs (N 495) to Older SIPs (N 1724) With Respect to Meeting Internet Partners Face-to-Face Measure Young SIPs N 495 Older SIPs N 1724 p-value Internet venues for meeting Chat room 69.2 63.4 0.02 Bulletin board 24.6 35.6 0.001 Browsing profiles 53.2 55.6 0.34 Dating/matchmaking service 0.8 3.2 0.02 Before meeting in person Read profile 78.9 78.3 0.82 Cybersex before meeting 26.1 24.1 0.67 Instant messaging before 80.4 79.0 0.62 meeting Exchange photo 76.3 72.5 0.10 Exchange phone numbers 81.7 79.1 0.22 Exchange addresses 45.4 37.1 0.004 Travel 100 miles to meet 43.3 50.5 0.02 Meeting in person took place at... Partner s home 58.6 53.7 0.05 Respondent s home 43.6 40.3 0.19 Bar 20.5 28.9 0.001 Restaurant/coffee House 56.0 60.3 0.09 Park/outside 34.0 26.4 0.001 Restroom 6.5 3.9 0.02 Condoms available at site (last sex) 12.6 19.9 0.001 Internet partners lied about... Age 68.9 65.4 0.31 Marital status 37.4 42.2 0.18 False photo 34.0 34.3 0.95 Gender 7.1 4.7 0.14 SIP sex with Internet partner; NIP no Internet partner. but more likely to report vaginal sex with these partners. In addition, young SIPs were less likely than older SIPs to be drunk or high during sex with non-internet partners. Table 3 presents details of the methods used by respondents to seek partners on the Internet. Also included are specific data about the arrangement and location of face-to-face meetings. Younger SIP respondents were slightly more likely to use chat rooms for meeting sex partners, while older SIP respondents were more likely to use less interactive methods such as bulletin boards or dating/matchmaker services. Across age groups, over 60% of respondents claim that partners lied about their age online. Young adults were more likely than older SIP respondents to have exchanged addresses with Internet partners. Younger and older SIP respondents were equally likely to have exchanged instant messages to arrange meetings, and were equally likely to have engaged in cybersex (analogous to phone sex) before meeting. Older SIP respondents were more likely to report meeting their sex partners face-toface at hotels or bars, while younger SIPs were more likely to meet in residences (their partners or their own) or parks. Older respondents were more likely to have traveled 100 miles or more to meet an Internet partner in person, and were more likely to have had condoms available during their last meeting with an Internet partner. Younger SIP respondents were less likely than their older counterparts to discuss HIV or STD with Internet partners, but were equally likely to discuss HIV or STD with their non-internet partners. Younger and older respondents were equally willing to participate in prevention-oriented chat rooms; similarly, respondents in the two age groups were equally willing to open prevention-oriented electronic mail. Discussion Potential Limitations of Internet Surveys Anonymous online surveys have several obvious limitations. For example, our survey may have promoted false reporting of age, as anyone under the age of 18 years was excluded from participation. In addition, anonymous reporting may have facilitated the purposeful false reporting feared by many online survey researchers. However, data from other studies by these authors [3 5] point to very similar results to those obtained on this Internet survey. The similarity between risk information reported online and risk information reported in person is encouraging evidence for the validity of these data. The sample reported here is obviously biased, in that it is a sample of Internet users. Most Internet users are Caucasian, educated (high school degree or above), male, and employed. Our goal, then was to obtain a reasonably representative sample of Internet users. We attempted to ensure representation of multiple groups by performing targeted recruitment in chat rooms and on bulletin boards; however, the majority of our data were obtained after national media attention was devoted to the study. Thus, the

16 McFARLANE ET AL. JOURNAL OF ADOLESCENT HEALTH Vol. 31, No. 1 representative nature of the sample must be called into question. Discussion of Results The use of the Internet to initiate risky sexual contact is not a new phenomenon, but our data suggest that it may be a growing trend. Of particular interest is the demographic makeup of young adults who have sex with Internet partners (young SIPs). In the past, the digital divide has limited access to the Internet on the part of minorities and women. Our data suggest that this is changing, as young adults who have found Internet partners are more likely to be female and Hispanic than their older counterparts. The trend toward a more demographically representative Internet bears watching, particularly from the perspective of epidemiologists studying STDs. Young SIPs reported having more Internet partners in the past 12 months than non-internet partners, suggesting that the Internet is the venue where the majority of these individuals sexual partnerships are initiated. In fact, some young SIPs reported having no non-internet partners at all in the past 12 months. For people who use the Internet to initiate sexual contacts, this venue may be of great importance to their sex lives. The fact that the average number of 12-month Internet partners is close to the average number of lifetime Internet partners suggests that the Internet may be growing in its importance to young adults sex lives. Young adults appear to be more comfortable than older SIPs with finding partners in chat rooms. In addition, they are more ready to exchange identifying or locating information, such as addresses, than are older SIPs. It is possible that the interactive, real-time chat environments allow young adults a (potentially false) sense of security and intimate knowledge of potential sex partners. Their readiness to exchange locating information with potential sex partners brings up safety concerns independent of risk for STDs and HIV. Our data suggest that people who use the Internet to initiate sexual contact may be having sex with partners who have different characteristics than non- Internet partners. That is, partners found on the Internet were more likely to be same-sex partners than partners found offline. This discrepancy between Internet and non-internet partner genders suggests that the Internet may be a tool for exploration of sexual activities that diverge from activities intiated with non-internet partners. The relative anonymity of the Internet may facilitate sexual experimentation that is considered impossible in traditional venues. This may be especially salient for young adults, many of whom still may be exploring their sexual orientation. However, experimentation with sexual orientations is only one function of the Internet in the sexual lives of young adults, the most important of which is that the Internet apparently increases the efficiency of partner-seeking. These data point to an urgent need for online STD/HIV prevention interventions targeting young adults. These interventions must be sensitive to issues of privacy and confidentiality, while emphasizing the health benefits of condom use, regular STD screening, and HIV testing. In addition, young adults who seek partners online should be encouraged to discuss STD/HIV status with potential partners, and to negotiate condom use before a face-to-face encounter. By intervening on these behaviors in the same arena in which risky behavior is initiated, we may be able to effect important changes in risk for STDs and HIV. References 1. Internet service providers: Financial benchmarks and market potential, 2000. Washington, DC: The Strategis Group, 2000. 2. Children, families, and the Internet. Burlingame, CA: Grunwald Associates, 2000. 3. McFarlane M, Bull SS, Rietmeijer CA. The Internet as a newly emerging risk environment for sexually transmitted diseases. JAMA 2000;284:443 6. 4. Bull SS, McFarlane M. Soliciting sex on the Internet. What are the risks for sexually transmitted diseases and HIV? Sex Transm Dis 2000;27:545 550. 5. Bull SS, McFarlane M, Rietmeijer CA. HIV/STD risk behaviors among men seeking sex with men online. Am J Public Health 2001;91:988 9.