In 1998,,20% of households in the United States had

Transcription

1 EPIDEMIOLOGY AND SOCIAL SCIENCE Characteristics of Recently HIV-Infected Men Who Use the Internet to Find Male Sex Partners and Sexual Practices With Those Partners Davey M. Smith, MD, MAS,* Lydia N. Drumright, PhD, MPH,* Simon D. W. Frost, DPhil,* W. Susan Cheng, MPH,* Stephen Espitia, MS, Eric S. Daar, MD, Susan J. Little, MD,* and Pamina M. Gorbach, MHS, DrPH Objective: To examine (1) characteristics of recently HIV-infected men who have sex with men (MSM) who find sex partners through the Internet and (2) differences in characteristics of and sexual behaviors practiced with Internet partners as compared to other partner types. Methods: From May 2002 to 2005, a computer-assisted selfinterview was administered to 194 recently HIV-infected MSM in southern California. MSM who used the Internet to find sex partners were compared with those who did not report Internet use, and partners found from the Internet were compared with those who were found from other venues using x 2 analyses, t tests, logistic regression, and generalized estimating equations. Results: Seventy percent of participants reported using the Internet to find partners in the prior 3 months. In multivariate analysis, Internet users as compared to non-internet users reported higher education levels (some college vs. high school: odds ratio [OR] = 5.04; P, 0.01 and college or greater vs. high school: OR = 9.61; P = 0.01), were more likely to be white (OR = 2.16; P = 0.04), reported more partners in the prior 3 months (OR = 1.05; P = 0.04), were more likely to have had sexual contact with all their last 3 partners after HIV diagnosis (OR = 3.43; P, 0.01), and were more likely to report Received for publication May 11, 2006; accepted July 27, From the *University of California, San Diego, La Jolla, CA; Veterans Affairs San Diego Healthcare System, San Diego, CA; and Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA. D. M. Smith and L. N. Drumright contributed equally to this work. Supported by the Universitywide AIDS Research Program grants ID01- SDSU-056, ID01-LA-056, IS02-SD-701, and CCTG-CC99 SD003; National Institutes of Health grants 5K23AI055276, AI27670, AI38858, AI043638, AI43752, AI29164, AI47745, AI57167, AI29164, AI27660, and M01-RR00425; University of California, San Diego Center for AIDS Research (AI36214); and Research Center for AIDS and HIV Infection of the San Diego Veterans Affairs Healthcare System ( ). The authors have reported no financial conflict with the publication of this paper. Written informed consent was obtained from all patients, and the human experimentation guidelines of the US Department of Health and Human Services and the individual institutions were followed in conducting this research. Reprints: Davey M. Smith, MD, MAS, University of California, San Diego, 9500 Gilman Drive 0679, La Jolla, CA ( davey@ucsd. edu). Copyright Ó 2006 by Lippincott Williams & Wilkins that all their last 3 partners were HIV-negative (OR = 3.35; P = 0.02), but none were main partners (OR = 2.36; P = 0.02). When compared with partners who were found in other venues, Internet partners were less likely to be main partners (OR = 0.52; P, 0.01) and were more likely to be younger (OR = 0.98; P = 0.05), to be HIV-negative (OR = 1.88; P = 0.02), and to become sex partners after HIV diagnosis (OR = 1.58; P = 0.03). Conclusions: The Internet is a popular venue for recently HIVinfected MSM to find partners, many of whom are HIV-negative. Because finding sex partners through the Internet occurs after HIV diagnosis, the Internet could be a valuable target for new HIV prevention strategies. Key Words: Internet, men who have sex with men, partnerships, primary HIV, risk (J Acquir Immune Defic Syndr 2006;43: ) In 1998,,20% of households in the United States had Internet access. By 2003, more than 65% of adults in California had Internet access. 1 Men who have sex with men (MSM) may find the Internet especially attractive for finding sex partners because it offers some anonymity, allowing them to connect to other MSM or to find specific types of partners without negative social consequences. It may also be more time-efficient to find sex partners on the Internet than in bars or dance clubs. Rejection from a potential sex partner may also be less embarrassing on-line than in other settings. These factors help to explain why the Internet has become a popular venue for MSM to find sex partners Frequent use of the Internet to find partners, 4,7 the large number of sites that cater to finding sex partners for MSM, 3 and the recent increases in HIV incidence among MSM 11,12 have led to growing concern about the Internet as a potential catalyst for HIV transmission, however. 10,13 Some cross-sectional studies indicate that the Internet has the potential to aid in serostatus disclosure and may promote serosorting among HIV-positive MSM, 14,15 whereas other studies indicate that MSM who use the Internet to find sex partners are more likely to report unprotected anal intercourse (UAI) with partners who are serodiscordant for HIV or of unknown serostatus. 4,16 Other risky behaviors such as drug use, increased number of sex partners, 16 sex with 582 J Acquir Immune Defic Syndr Volume 43, Number 5, December 15, 2006

2 J Acquir Immune Defic Syndr Volume 43, Number 5, December 15, 2006 Finding Sex Partners Through the Internet anonymous partners, 17 group sex, 18 and trading sex for money or drugs 2 have been reported at higher frequencies by MSM who use the Internet to find partners than by their non Internet-using peers. Using the Internet to find sex partners has also been associated with syphilis outbreaks among MSM in San Francisco 19 and Los Angeles. 17 Although these studies indicate that the Internet may be a venue that has the potential to increase risky sexual behaviors, little is known about specific behaviors with Internet partners. To investigate HIV risks associated with finding sex partners through the Internet, we compared characteristics of MSM who were recently infected with HIV who used the Internet to find partners with characteristics of those who did not and examined differences between their recent Internet and non-internet partners. By studying MSM with recent HIV infection, we focus on the part of the sexual network in which HIV transmission is most likely to occur, providing insight into risks for acquiring HIVand the subsequent transmission to sex partners. 20 METHODS Between May 2002 and August 2005, 207 of the 211 recently HIV-infected individuals who were recruited into the Southern California Acute Infection and Early Disease Research Program (AIEDRP) cohort completed a computerassisted self-interview (CASI). Eligibility of enrollment in the AIEDRP requires that all participants have laboratorydocumented evidence of HIV infection within 12 months before enrollment as previously described. 21,22 Among the 207 participants who completed the questionnaire, responses from 13 (4 women, 8 men who reported sexual contact with women only, and 1 man who reported no sex partners) were not included in these analyses because the focus of this study was Internet use among MSM. Data from all 194 male participants who reported sexual activity with men in the previous 12 months were included in these analyses. Interviews were completed, on average, 5 weeks (median = 3 weeks) after HIV diagnosis and 15 weeks (median = 13 weeks) after the estimated date of infection. Participants were asked about their sexual histories, including where they found sex partners in the 12-, 3-, and 1-month periods before their interview. Those who reported finding at least 1 sex partner through the Internet in the 3 months before program enrollment were classified as finding sex partners through the Internet. Detailed information was also collected on the respondents last 3 sex partners, including meeting location, types of sexual contact, partner type (main partner vs. other), partner s age, partner s HIV status, timing of the first and last sexual episodes with that partner, and any recreational drug use during sexual activity with that partner. All participants were deemed mentally fit at the time of enrollment, gave informed consent, and received a study code to protect confidentiality. The study was approved by the Institutional Review Boards of the University of California, San Diego; University of California, Los Angeles (UCLA), Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance; and Cedar-Sinai Medical Center, Los Angeles. Data were entered by participant response into Ci3 (Sawtooth Technologies, Northbrook, IL) and analyzed using STATA version 8.2 SE (StataCorp LP, College Station, TX). Associations between finding sex partners on the Internet in the previous 3 months and demographics and sexual histories of the participants were examined using x 2 analyses, t tests, and univariate and multivariate logistic regression. Multivariate generalized estimating equations (GEEs) were used to examine differences in behavior with and characteristics of partners who were met on the Internet as compared to those who were met elsewhere. GEEs were selected so as to correct variance estimates for repeated measures on the same individual. For all multivariate analyses, models were constructed by including variables that were considered theoretically significant from a review of the literature, were potential confounders, or were significant by univariate analysis; however, some variables, such as male only versus male and female sex partners, were removed from these models because data were too sparse to assess associations appropriately and inclusion reduced the efficiency of the model. RESULTS Of the 194 participants, 70% reported using the Internet to find sex partners in the previous 12 months. The median age of the participants was 35 years, most reported white ethnicity (69.6%), and most (87.6%) had at least some college education (Table 1). Most were employed (70.6%) and had lived in their home city for more than 12 months (85.6%). The average reported gross annual income was $50,000 US. Most (93.8%) reported having sexual contact with men only, as compared to men and women, in the previous 12 months. The average number of male partners reported in the previous 3 months was 9.2, with a median of 4. More than one quarter (27.8%) reported that sexual contact with all the last 3 partners occurred after their HIV diagnosis, and approximately half (54.6%) reported that none of the last 3 partners was a main partner. Most participants (84.0%) reported UAI with at least 1 of their last 3 partners; however, only 17.5% reported that all these partners were HIV-negative. In univariate (Table 1) and multivariate (Table 2) analyses, participants who used the Internet to recruit sex partners in the past 3 months were significantly more likely than those who did not use the Internet to be white (74.6% vs. 60.3%, adjusted odds ratio [AOR] = 2.16; P = 0.04), to be more highly educated (56.4% vs. 32.4% for college or higher vs. high school, AOR = 9.61; P, 0.01), to report more male sex partners during the prior 3 months (11.1 vs. 5.7; P, 0.01 and AOR = 1.05; P = 0.04), to report no main partners among their last 3 (62.7% vs. 39.7%; P, 0.01 and AOR = 2.36; P = 0.02), and to have had sexual contact with all their last 3 partners after their HIV diagnosis (34.9% vs. 14.7%, AOR = 3.43; P, 0.01). Additionally, in multivariate but not univariate models, Internet users were more likely than nonusers to report that all their last 3 partners were HIVnegative (AOR = 3.35; P = 0.02). For the index participant, employment status, duration of time spent in the current city, and UAI with any of the last 3 partners were not associated with recruiting sex partners on the Internet (Tables 1, 2). q 2006 Lippincott Williams & Wilkins 583

3 Smith et al J Acquir Immune Defic Syndr Volume 43, Number 5, December 15, 2006 TABLE 1. Univariate Associations Between Recently Infected MSM Who Used the Internet to Find Partners in the Past 3 Months and Participant Demographics and Sexual Histories (n ¼ 194) Characteristic Used Internet (n = 126) No Internet Use (n = 68) OR* (95% CI) P* Total (n = 194) Age, (y) 35.2 (35) 34.8 (35) 1.01 (0.97 to 1.04) (35) Ethnicity, white 74.6 (94) 60.3 (41) 1.93 (1.03 to 3.63) (135) Respondent education High school 6.4 (8) 23.5 (16) Ref 12.4 (24) Some college 37.3 (47) 44.1 (30) 3.13 (1.20 to 8.22) (77) Completed college or higher 56.4 (71) 32.4 (22) 6.46 (2.44 to 17.1), (93) Employed 73.0 (92) 66.2 (45) 1.38 (0.73 to 2.62) (137) Has resided in current city #12 months 14.3 (18) 14.7 (10) 0.97 (0.42 to 2.23) (28) Income 52k (40k) 48k (20k) 1.00 (0.99 to 1.01) k (34k) Sexual contact with men only 97.6 (123) 86.8 (59) 6.25 (1.63 to 24.0), (182) No. male partners in past 3 months 11.1 (7) 5.7 (3) 1.06 (1.02 to 1.11), (4) All sexual contact with the last 3 partners occurred after HIV diagnosis 34.9 (44) 14.7 (10) 3.11 (1.45 to 6.69), (54) UAI with any of the last 3 partners 83.3 (105) 85.3 (58) 0.86 (0.38 to 1.95) (163) None of the last 3 partners was a main partner 62.7 (79) 39.7 (27) 2.55 (1.39 to 4.68), (106) All the last 3 partners were HIV-negative 18.3 (23) 16.2 (11) 1.16 (0.53 to 2.55) (34) *ORs, confidence intervals, and probability values from univariate logistic regression models. CI indicates confidence interval; Ref, reference. Of the 572 partners, respondents met 2.1% (12 partners) in the park; 3.0% (17 partners) at work or school, 12.6% (72 partners) in a bathhouse; 13.1% (75 partners) through friends; 14.3% (82 partners) at a bar or club; 37.2% (213 partners) on the Internet; and 17.7% (101 partners) in various other locations, including during travel. On average, partners were 33.5 years old (median = 33 years), and most (61.3%) were reported to be of white ethnicity (Table 3). Participants reported that they did not know the HIV status of 47.7% of partners and believed that 40.9% were HIV-negative and that only 11.4% were HIV-positive; however, UAI was reported with 54.9% of partners. Few partners (18.0%) were described TABLE 2. Multivariate Logistic Regression Model Demonstrating Associations Between Using the Internet to Look for Sex Partners Within the Previous 3 Months and Participant Demographics and Sexual Histories (n ¼ 194) Characteristic AOR 95% CI P Age, (y) to Ethnicity, white to Respondent education High school Ref Some college to 16.0,0.01 Completed college or higher to 31.7,0.01 No. male partners in past 3 months to All sexual contact with the last 3 partners occurred after HIV diagnosis to 8.22,0.01 All the last 3 partners were HIV-negative to UAI with any of the last 3 partners to None of the last 3 partners was a main partner to CI indicates confidence interval; Ref, reference. as main partners, substance use during sexual activity was reported with 43.3% of partners, and sexual contact occurred after participant HIV diagnosis with 46.7% of partners. In univariate GEE analyses, those partners met on the Internet were significantly less likely to be main partners (odds ratio [OR] = 0.66; P = 0.04) and were more likely to be younger (OR = 0.8; P = 0.04) than partners met in other venues (Table 3). In multivariate GEE models containing all 194 participants (accounting for 572 partnerships), partners that MSM met through the Internet were less likely to be main partners (OR = 0.52; P = 0.03), were more likely to be younger (OR = 0.98; P = for every year of age), were more likely to be HIV-negative (OR = 1.88; P = 0.02), and were more likely to be partners after the participant was diagnosed with HIV (OR = 1.58; P = 0.03) as compared with partners met in all other venues (Table 4). Furthermore, higher educational attainment was associated with meeting one s partner on the Internet (OR = 3.03; P = 0.02 for some college vs. high school and OR = 3.28; P = 0.01 for college or more vs. high school). There were no significant interactions between UAI and partner HIV status with respect to meeting partners on the Internet and no significant association between UAI and meeting a partner on the Internet when examining data on HIV-negative or unknown status partners only (data not shown). To explore changes in behavior before and after HIV diagnosis, the sample was split into 305 partnerships from 140 of the 194 participants in which sexual activity occurred before HIV diagnosis and 267 partnerships from 135 of the 194 participants in which sexual activity occurred after HIV diagnosis. Before HIV diagnosis, 33% of sex partners were recruited through the Internet (n = 102). Slightly more partners (n = 111 [42%]) were recruited through the Internet after HIV diagnosis. Among those partners in which sexual contact 584 q 2006 Lippincott Williams & Wilkins

4 J Acquir Immune Defic Syndr Volume 43, Number 5, December 15, 2006 Finding Sex Partners Through the Internet TABLE 3. Univariate Associations Between Meeting Locations for Partners and Partner Characteristics and Sexual Activities Practiced With Partners Partner Meeting Location Characteristic/Sexual Activity Internet (n = 213 ) Elsewhere (n = 369) OR* (95% CI) P* Total (n = 572) Partner s age, (y) 32.6 (33) 33.9 (33) 0.98 (0.96 to 0.99) (33) Partner s ethnicity, white 61.0 (130) 61.5 (227) 0.92 (0.65 to 1.29) (357) Partner was a main partner 14.1 (30) 20.3 (75) 0.66 (0.44 to 0.98) (105) Partner HIV status HIV-positive 8.9 (19) 12.8 (46) Ref 11.4 (65) HIV-negative 45.5 (97) 38.2 (137) 1.65 (0.95 to 2.85) (234) HIV status unknown 45.5 (97) 49.0 (176) 1.55 (0.91 to 2.65) (273) UAI with partner 55.4 (118) 54.6 (196) 0.98 (0.72 to 1.34) (314) Substance use during sex 39.4 (84) 45.5 (168) 0.92 (0.64 to 1.32) (252) Sex occurring after diagnosis 52.1 (111) 43.5 (156) 1.24 (0.88 to 1.75) (267) *ORs, confidence intervals, and probability values calculated using univariate GEEs. CI indicates confidence interval; Ref, reference. occurred after HIV diagnosis, Internet partners were significantly less likely to be main partners (OR = 0.47; P = 0.03) and more likely to be younger (OR = 0.96; P = 0.01 for every 1 year of age; see Table 4). Participants who were more highly educated were more likely to find partners from the Internet after HIV diagnosis (OR = 3.49; P = 0.04 for some college vs. high school and OR = 3.29; P = 0.06 for college or more vs. high school). Additionally, increased likelihood of UAI was marginally associated with meeting a sex partner on the Internet after HIV diagnosis (OR = 1.61; P = 0.08). Among partners with whom sexual activity occurred before HIV diagnosis, Internet partners were marginally less likely to be main partners (OR = 0.47; P = 0.06). No other partner or partnership characteristic was associated with finding partners from the Internet among partners with whom sexual contact occurred before HIV diagnosis. To develop a further understanding of potential HIV transmission events, several stratified analyses were conducted using GEEs to examine associations between UAI and HIV status (Fig. 1). Before stratifying by timing of sexual activity with respect to HIV diagnosis, UAI was significantly more likely with HIV-positive partners in general (P = 0.02) and with those met at venues other than the Internet (P = 0.01) but not with those met on the Internet (P = 0.52). Conversely, there were no associations between UAI and HIV status with partners before HIV diagnosis (Fig. 1). After HIV diagnosis, TABLE 4. Multivariate GEEs: Demographic and Behavioral Characteristics of Internet Partners Versus Partners Met at Any Other Location (n ¼ 194 Individuals, 572 Partners) Sexual Contact Before HIV Diagnosis (n = 140) (n = 305 of 572 partners) Sexual Contact Only After HIV Diagnosis (n = 135) (n = 267 of 572 partners) Complete Sample (N = 194) (N = 572 of 572 partners) AOR (95% CI) P AOR (95% CI) P AOR (95% CI) P Partner was a main partner 0.47 (0.22 to 1.03) (0.24 to 0.92) (0.32 to 0.83),0.01 Respondent age, (y) 1.01 (0.98 to 1.05) (0.97 to 1.05) (0.98 to 1.04) 0.47 Partner s age, (y) 0.99 (0.96 to 1.03) (0.93 to 0.99) (0.96 to 1.00) 0.05 Respondent education High school Ref Ref Ref Some college 2.17 (0.70 to 6.70) (1.04 to 11.7) (1.22 to 7.56) 0.02 Completed college or higher 2.55 (0.82 to 7.93) (0.97 to 11.2) (1.31 to 8.22) 0.01 UAI 1.00 (0.61 to 1.62) (0.95 to 2.73) (0.81 to 1.61) 0.45 Any drug use with partner 0.73 (0.41 to 1.28) (0.42 to 1.31) (0.56 to 1.22) 0.34 Respondent ethnicity, white 1.59 (0.79 to 3.18) (0.63 to 2.71) (0.79 to 2.31) 0.28 Partner HIV status HIV-positive Ref Ref Ref HIV-negative to 5.29) (0.82 to 4.02) (1.05 to 3.37) 0.02 HIV status unknown 1.27 (0.50 to 3.23) (0.65 to 2.93) (0.77 to 2.41) 0.29 Had sexual contact only after diagnosis NA NA 1.58 (1.08 to 2.32) 0.03 CI indicates confidence interval; NA, not applicable because of stratification on this variable; Ref, reference. q 2006 Lippincott Williams & Wilkins 585

5 Smith et al J Acquir Immune Defic Syndr Volume 43, Number 5, December 15, 2006 FIGURE 1. Percentage of partners in which participants reported UAI by HIV status stratified by meeting location and timing of sexual activity with regard to HIV diagnosis (percentages from contingency tables, P values calculated using GEEs). however, UAI was again more common with HIV-infected partners in the overall sample (P = 0.01) and among those partners who were met at venues other than the Internet (P = 0.02) but not with partners met on the Internet (P = 0.33). DISCUSSION Our results support previously published studies that the Internet is a common venue for MSM to meet sex partners. 4,5,7,8,23 Similarly, we found that recently HIV-infected MSM who use the Internet to find sex partners were more likely to be white and educated. 3,7 Our current investigation extends those previous observations with findings important to the understanding of how MSM with recent HIV infection use the Internet to find sex partners. We found that recently HIVinfected MSM who use the Internet to find sex partners were more often white and more educated and reported more male partners than non-internet users. Additionally, sex partners found through the Internet were less likely to be main partners, more likely to be younger, more likely to be become partners after HIV diagnosis, and more likely to be considered HIVnegative than partners from other venues. These results suggest that after HIV diagnosis, younger and more educated MSM may be more likely to use the Internet to find younger nonmain partners. Although the Internet seemed to bridge HIV-positive MSM from our sample with negative-msm, UAI with Internet partners was no more likely than with partners found from other venues, regardless of partner HIV status. Some studies have shown that after HIV diagnosis, individuals reported less risky HIV transmission behavior, 24 which is consistent with the findings of our own previous reports from this cohort. 25 By offering a level of anonymity, the Internet could be a powerful tool to aid in serostatus disclosure and negotiation of harm reduction sexual practices. 14,15 Although our investigations did not specifically address these practices, our data suggest that HIV-positive MSM often select HIV-negative partners when using the Internet to meet partners. We did not find that they were more likely to report UAI with their HIV-negative Internet partners, however. Additionally, HIV status of the partner was the perceived status as reported by the respondent and may not reflect the actual HIV status of the partner, because HIV status is often poorly reported through surrogate sources. 26 Although finding partners on the Internet was not associated with an increased risk of UAI or increased UAI with HIV-negative partners, it is concerning that UAI with Internet partners was not related to these partners HIV status. After diagnosis, among partners met in locations other than the Internet, UAI was most likely if the partner was HIV-positive. Such serosorting has been reported previously 15 and would be expected if MSM with HIV infection were trying to protect their HIV-negative partners. Such behavior was not observed with Internet partners, however, suggesting that although UAI with HIV-negative partners may not increase after HIV diagnosis, something other than the HIV status of the partner is driving the decision to have UAI with Internet partners. We estimate that our sample represents only approximately 5% of MSM who are recently infected with HIV in our 2 recruitment counties (San Diego and Los Angeles), given the total number of HIV diagnoses in these 2 areas. 27,28 Therefore, we cannot claim that our sample is generalizable to the larger population of newly infected MSM, nor can we claim that they are generalizable to all MSM, because the sexual behavior of MSM at a population level has been noted to be different 586 q 2006 Lippincott Williams & Wilkins

6 J Acquir Immune Defic Syndr Volume 43, Number 5, December 15, 2006 Finding Sex Partners Through the Internet between US cities. 15 Although the number of participants enrolled into this study is smaller than samples of MSM with chronic HIV infection, this cohort focuses on individuals who are at the greatest risk for transmitting their infection, 29,30 which is important in understanding between-partner transmission dynamics of HIV. Additionally, our sample may be biased toward newly infected MSM with a higher education and socioeconomic status, because individuals with these characteristics are more likely to be aware of their HIV risks and are more likely to participate in research studies. 31 Furthermore, all the data, including sexual behavior data, were self-reported, which is the greatest limitation of most sexual behavior research. 31 Truthful reporting of sensitive information was most likely enhanced, however, because the information was collected via CASI. 32 The Internet is a relatively new venue for MSM to meet sex partners as compared to bars that cater to MSM or bathhouses, but it has rapidly become popular; one third of all sex partners reported in our study were found through the Internet. The Internet has also become a convenient tool for rapid dissemination of information. 1 By developing a better understanding of the characteristics of MSM who use the Internet to find partners and the sexual behaviors they engage in with these partners, successful Internet-based interventions may be developed for HIV prevention. In recent studies, most MSM agreed that they would be willing to read prevention messages posted on the Internet, 33 which is an encouraging sign that suggests great promise for Internet-based HIV risk reduction interventions. ACKNOWLEDGMENTS The authors are grateful to the study participants for their unwavering generosity in participating in this research project. They also express gratitude to Heidi Aiem, Tari Gilbert, Jacqui Pitt, Paula Potter, and Joanne Santangelo for their dedication and commitment to the patients and their assistance in data collection; to Laureen Copfer for editorial assistance with this manuscript; and to Douglas Richman for intellectual discussions and scientific insight. REFERENCES 1. Spooner T. Internet use by region in the United States [report]. Pew Internet & American Life Project. Available at: pdfs/pip_regional_report_aug_2003.pdf. Accessed May 1, Kim AA, Kent C, McFarland W, et al. Cruising on the Internet highway. J Acquir Immune Defic Syndr. 2001;28: Bolding G, Davis M, Hart G, et al. Gay men who look for sex on the Internet: is there more HIV/STI risk with online partners? AIDS. 2005;19: Elford J, Bolding G, Sherr L. Seeking sex on the Internet and sexual risk behaviour among gay men using London gyms. AIDS. 2001;15: Bull SS, McFarlane M, Rietmeijer C. HIV and sexually transmitted infection risk behaviors among men seeking sex with men on-line. Am J Public Health. 2001;91: Bull SS, McFarlane M. Soliciting sex on the Internet what are the risks for sexually transmitted diseases and HIV? Sex Transm Dis. 2000;27: McFarlane M, Bull SS, Rietmeijer CA. The Internet as a newly emerging risk environment for sexually transmitted diseases. JAMA. 2000;284: Rietmeijer CA, Bull SS, McFarlane M. Sex and the Internet. AIDS. 2001; 15: Ross MW, Rosser BRS, Stanton J. Beliefs about cybersex and Internetmediated sex of Latino men who have Internet sex with men: relationships with sexual practices in cybersex and in real life. AIDS Care. 2004;16: Liau A, Millett G, Marks G. Meta-analytic examination of online sexseeking and sexual risk behavior among men who have sex with men. Sex Transm Dis. 2006;33: Hall HI, Song R, McKenna MT. Increases in HIV diagnoses 29 states, [reprinted from MMWR Morbidity and Mortality Weekly Report. 2003;52: ]. JAMA. 2004;291: Hurtado I, Alastrue I, Ferreros I, et al. Trends in HIV testing, serial HIV prevalence and HIV incidence among persons attending a center for AIDS prevention from 1988 to 2003; increases in HIV incidence in men who have sex with men in recent years? Sex Transm Infect [Epub ahead of print]. 13. Bull SS, McFarlane M, Lloyd L, et al. The process of seeking sex partners online and implications for STD/HIV prevention. AIDS Care. 2004;16: Elford J. Changing patterns of sexual behaviour in the era of highly active antiretroviral therapy. Curr Opin Infect Dis. 2006;19: Parsons JT, Schrimshaw EW, Wolitski RJ, et al. Sexual harm reduction practices of HIV-seropositive gay and bisexual men: serosorting, strategic positioning, and withdrawal before ejaculation. AIDS. 2005;19(Suppl 1): S13 S 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: 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;31: Mettey A, Crosby R, DiClemente RJ, et al. Associations between Internet sex seeking and STI associated risk behaviours among men who have sex with men. Sex Transm Infect. 2003;79: Klausner JD, Wolf W, Fischer-Ponce L, et al. Tracing a syphilis outbreak through cyberspace. JAMA. 2000;284: Pilcher CD, Eron JJ, Vemazza PL, et al. Sexual transmission during the incubation period of primary HIV infection. JAMA. 2001;286: Little SJ, Holte S, Routy JP, et al. Antiretroviral-drug resistance among patients recently infected with HIV. N Engl J Med. 2002;347: Acute Infection and Early Disease Research Program (AIEDRP). Available at: Accessed May 1, 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;35: Marks G, Crepaz N, Senterfitt JW, et al. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States implications for HIV prevention programs. J Acquir Immune Defic Syndr. 2005;39: Gorbach PM, Drumright LN, Daar ES, et al. Transmission behaviors of recently HIV-infected men who have sex with men. J Acquir Immune Defic Syndr. 2006;42: Niccolai LM, Farley TA, Ayoub MA, et al. HIV-infected persons knowledge of their sexual partners HIV status. AIDS Educ Prev. 2002;14: County of San Diego, Health and Human Services Agency, Public Health Services. HIV/AIDS epidemiology reports. Available at: www2.sdcounty.ca.gov/hhsa/servicedetails.asp. Accessed May 1, Los Angeles County Department of Health Services, Office of Health Assessment and Epidemiology/Public Health, HIV Epidemiology Program. Epidemiology reports HIV. Available at: hiv. Accessed May 1, Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med. 2000;342: Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis. 2005;191: Fenton KA, Johnson AM, McManus S, et al. Measuring sexual behaviour: methodological challenges in survey research. Sex Transm Infect. 2001; 77: Newman JC, Des Jarlais DC, Turner CF, et al. The differential effects of faceto-face and computer interview modes. Am J Public Health. 2002;92: 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;16: q 2006 Lippincott Williams & Wilkins 587