Abstract
Background
Internet-based platforms are increasingly prominent interfaces for social and sexual networking among men who have sex with men (MSM).
Methods
MSM were recruited through venue-based sampling in 2008, 2011, and 2014 in 20 US cities. We examined changes in internet use (IU) to meet men and in meeting the last partner online among MSM from 2008 to 2014 using Poisson regression with generalized estimating equations to calculate adjusted prevalence ratios (APRs). We also examined factors associated with increased frequency of IU using data from 2014. IU was categorized as never, infrequent use (<once a week), and frequent use (≥once a week).
Results
Frequent IU increased from 21% in 2008 to 44% in 2014 (APR = 1.39, 95% confidence interval: 1.36 to 1.42), and having met the last partner online increased from 19% in 2008 to 32% in 2014 (APR = 1.30, 95% confidence interval: 1.26 to 1.34). Those who never used the internet had fewer partners (median of 2 in the past 12 months, interquartile range: 1–4) compared with infrequent (4, 2–7) and frequent users (5, 3–12). HIV testing in the past 12 months also increased with increasing IU (58%, 68%, and 71%, respectively, P < 0.0001). Among HIV-positive participants, the percent HIV-positive awareness increased as IU increased (71%, 75%, and 79%, P < 0.005).
Conclusions
Both IU to meet men and meeting the last partner online increased since 2008. Although men who used the internet more frequently reported more partners in the past 12 months, they were also more likely to report testing in the past 12 months and were more likely to be HIV-positive aware.
Keywords: MSM, HIV, internet, partners, HIV testing, sexual behavior, trends, changes
BACKGROUND
Internet-based dating websites or applications (apps) are increasingly being used for social and sexual networking among gay, bisexual, and other men who have sex with men (referred to collectively as MSM). Before the widespread use of the internet, meeting friends and sex partners often occurred in person at venues such as bars, parties, and public cruising areas. As personal computers became available and social network software was developed in the 1990s, methods for MSM to meet other men broadened to include the internet. Websites rapidly developed in the early 2000s to provide services to those who wanted to meet people, including sexual partners, over the internet.1 Norms quickly shifted too, as one study of MSM in the United Kingdom found that the proportion of men under 30 who reported meeting their first sexual partner over the internet increased 20-fold from less than 3% in 1992 to 61% in 2002.2 Because internet speed and coverage has rapidly increased, what began as chat rooms tied to a computer with limited visual capabilities have developed into websites and apps that are on mobile phones and allow instant sharing of photographs and location.3 Today, the use of geospatial social networking and dating apps and websites is common among all segments of the population and around the world.3 Although dating apps and websites are used by many different subpopulations, Grov et al3 reported in 2013 that MSM accessed dating apps on average 22 times a week compared to heterosexuals who accessed dating apps on average 8 times a week. Apps link MSM around the globe—one of the most popular apps for MSM is available in 196 countries and reports over 1,000,000 users per hour.4
The first reported cases of HIV in the United States in the early 1980s were among MSM,5 a group that continues to be disproportionately affected by HIV. Despite representing approximately 2% of the US population,6,7 MSM accounted for 70% of all new HIV diagnoses (including MSM who inject drugs) in 2014.8 An estimated 15% of MSM Nationwide have HIV, with 11% estimated to be already diagnosed.9 Prevalence is even higher in the 20 cities that are part of the Centers for Disease Control and Prevention’s (CDC) National HIV Behavioral Surveillance (NHBS). In 2014, 22% of participating MSM were found to be HIV positive.10 Other sexually transmitted infections (STI), such as syphilis, have also increased among MSM in the United States.8,11
There is substantial evidence linking use of the internet to meet men to risky sexual behavior, including condomless anal sex and higher number of sexual partners.12,13 In addition, men recruited online report more risk behavior and more partners than those recruited offline.14 Although some studies have failed to show an association between use of networking apps to meet sex partners and risk of chlamydia and gonorrhea infection,15,16 others have found a higher prevalence of these infections among MSM who use networking apps compared with those who met partners exclusively through in-person methods.1 Furthermore, the internet has played a key role in several outbreaks of STIs among MSM.13 The high usage of dating apps and websites among MSM, coupled with high HIV and STI rates among users, make these apps potentially effective and efficient ways to reach a high-risk population with prevention and care messages, and serving as effective channels to recruit participants for public health programs and research. We sought to examine trends in the internet use (IU) to meet men among a large sample of MSM recruited offline in 20 large cities in the United States. We used data among MSM participating in NHBS to evaluate changes from 2008 to 2014 in using the internet to meet or socialize with men and in having met the last sex partner online. We also investigated the association between using the internet to meet men, and HIV-testing behavior and risk behavior in 2014.
METHODS
NHBS collects data among MSM every 3 years. All NHBS participants must be aged ≥18 years, live in a participating MSA or city, and be able to complete a behavioral survey in English or Spanish. The cross-sectional data reported in this analysis are from MSM recruited for interviews and HIV testing through venue-based, time-space sampling in surveys in the 20 cities that participated in NHBS (cities included in this analysis are: Atlanta, GA; Baltimore, MD; Boston, MA; Chicago, IL; Dallas, TX; Denver, CO; Detroit, MI; Houston, TX; Nassau-Suffolk, NY; New Orleans, LA; Los Angeles, CA; Miami, FL; Newark, NJ; New York, NY; Philadelphia, PA; San Diego, CA; San Francisco, CA; and San Juan, PR; Seattle, WA; and Washington, DC) in 2008, 2011, and 2014. NHBS-venue-based, time-space sampling procedures have been previously published.17,18 In each city, a team of staff members familiar with the local community conducted formative research to establish a list of venues frequented by MSM. Venues on the list were categorized as bar, dance club, fitness club or gymnasium, Gay Pride event, park or beach, large dance party (eg, rave or circuit party), cafe or restaurant, retail business, sex establishment or sex environment, social organization, street location, or another venue type, such as an event hosted by the local house ball community.
Interviews were conducted by trained interviewers using a standardized questionnaire covering demographics, HIV-associated behavior, and the use of HIV prevention and testing services. HIV testing was performed by collecting blood or oral specimens for either rapid testing in the field or laboratory-based testing followed by laboratory confirmation. Participants received incentives for participating in the interview and the HIV test. The incentive format (cash or gift card) and amount varied by city based on formative assessment and local policy. A typical incentive included $25 for completing the interview and $25 for providing a specimen for HIV testing. NHBS activities were approved by local institutional review boards in each participating city and as nonengaged research by the CDC.
Measures
IU to meet men was assessed with the question: “In the past 12 months, how often have you used the internet to meet or socialize with gay men either for friendship or sex? This could include social networking websites (such as Facebook or MySpace), websites directed toward gay men (such as Manhunt or Gay.com), or dating websites, or the use of mobile social apps (such as Foursquare or Grindr).” The 2008 measure of IU was slightly different and specified using the internet to look for sex partners. Response options were grouped as follows: frequent (once a week/more than once a week/daily), infrequent (less than once a month/once a month/more than once a month), and never. Place where participants met their last sex partner was assessed with the following question: “Where did you first meet this partner?” We grouped responses as having met the last sex partner on the internet versus all other responses. Discordant condomless anal sex was defined as condomless anal sex at last sex with a partner of different or unknown HIV status. During the interview, participants were asked questions to determine if they had previously tested positive for HIV. After the interview, each participant was offered HIV testing. A nonreactive rapid test was considered a definitive negative result; a reactive (preliminary positive) rapid test result was considered a definitive positive only when confirmed by supplemental laboratory testing (eg, western blot, immunofluorescence assay or nucleic acid amplification test). Participants with a confirmed positive HIV-test result who reported having previously tested positive for HIV were considered to be aware of their infection.
Analysis
Men who consented to and completed the survey and had a male sex partner in the past 12 months were included in analyses. The main outcomes for the analyses were having met the last sex partner online and frequent IU to meet or socialize with gay men. To determine if these outcomes changed over time, we used Poisson models with generalized estimating equations clustered on recruitment event to calculate adjusted prevalence ratios (APRs) and 95% confidence intervals (CIs). Year was included in the model as an ordinal variable. Individual interaction terms for each covariate by year were included in models to examine changes over time by subgroup. Each PR measures change in the outcome for a 3 year increase in interview year (ie, 2008–2011 or 2011–2014). Age, race, and income were included as covariates. Using data from 2014, we also investigated the association between HIV testing, awareness of HIV status and risk behaviors, using the Wilcoxon rank sum and chi-square tests.
RESULTS
The analysis sample was n = 8881 in 2008, n = 9253 in 2011, and n = 9636 in 2014. The race, age, education, and income composition were similar across years. A larger proportion of participants were HIV positive based on the NHBS test results in 2014 (20%) compared with 2011 (17%) and 2008 (17%) (Table 1).
TABLE 1.
Characteristics of Sexually Active Men Who Have Sex With Men, NHBS, 20 Cities, United States, 2008, 2011, and 2014*
| 2008 | 2011 | 2014 | ||||
|---|---|---|---|---|---|---|
|
|
|
|
||||
| Characteristic | n | % | n | % | n | % |
| Race/ethnicity | ||||||
| American Indian/Alaska Native | 48 | 0.5 | 75 | 0.8 | 63 | 0.7 |
| Asian/Native Hawaiian/Other Pacific Islander | 262 | 3.0 | 272 | 2.9 | 224 | 2.3 |
| Black | 2125 | 23.9 | 2485 | 26.9 | 2652 | 27.7 |
| Hispanic† | 2219 | 25.0 | 2407 | 26.1 | 2523 | 26.3 |
| White | 3762 | 42.4 | 3665 | 39.7 | 3668 | 38.3 |
| Other‡ | 459 | 5.2 | 325 | 3.5 | 455 | 4.7 |
| Age group, yrs | ||||||
| 18–24 | 2005 | 22.6 | 2352 | 25.4 | 1960 | 20.3 |
| 25–29 | 1641 | 18.5 | 1750 | 18.9 | 2109 | 21.9 |
| 30–39 | 2475 | 27.9 | 2190 | 23.7 | 2487 | 25.8 |
| 40–49 | 1899 | 21.4 | 1873 | 20.2 | 1686 | 17.5 |
| ≥50 | 861 | 9.7 | 1088 | 11.8 | 1394 | 14.5 |
| Education | ||||||
| <High school | 578 | 6.5 | 427 | 4.7 | 348 | 3.6 |
| High school diploma or equivalent§ | 3367 | 37.9 | 2215 | 24.1 | 2081 | 21.8 |
| Some college or technical degree | 2064 | 23.2 | 3143 | 34.2 | 3089 | 32.3 |
| College degree or postgraduate education | 2871 | 32.3 | 3392 | 37.0 | 4048 | 42.3 |
| Annual household income | ||||||
| 0–$19,999 | 2624 | 30.0 | 2906 | 31.9 | 2917 | 30.7 |
| $20,000–$39,999 | 2228 | 25.5 | 2275 | 25.0 | 2285 | 24.0 |
| $40,000–$74,999 | 2162 | 24.7 | 2193 | 24.1 | 2343 | 24.6 |
| $75,000 or more | 1735 | 19.8 | 1732 | 19.0 | 1961 | 20.6 |
| Sexual identity | ||||||
| Homosexual | 7197 | 81.2 | 7573 | 82.1 | 7831 | 81.5 |
| Bisexual | 1570 | 17.7 | 1550 | 16.8 | 1649 | 17.2 |
| Heterosexual | 101 | 1.1 | 102 | 1.1 | 125 | 1.3 |
| Frequency of IU to meet men‖ | ||||||
| Never | 5471 | 61.9 | 3285 | 35.5 | 2989 | 31.0 |
| Infrequent | 1543 | 17.5 | 2102 | 22.7 | 2419 | 25.1 |
| Frequent | 1821 | 20.6 | 3858 | 41.7 | 4225 | 43.9 |
| Self-reported HIV status | ||||||
| HIV positive | 1060 | 11.9 | 1244 | 13.4 | 1586 | 16.5 |
| HIV negative | 6767 | 76.2 | 7131 | 77.1 | 7424 | 77.0 |
| HIV unknown | 1054 | 11.9 | 878 | 9.5 | 626 | 6.5 |
| HIV-test result | ||||||
| Positive | 1516 | 17.1 | 1553 | 16.8 | 1888 | 19.6 |
| Negative | 6315 | 71.1 | 6867 | 74.2 | 6846 | 71.0 |
| Not tested or invalid | 1050 | 11.8 | 833 | 9.0 | 902 | 9.4 |
| City | ||||||
| Atlanta, GA | 347 | 3.9 | 558 | 6.0 | 507 | 5.3 |
| Baltimore, MD | 502 | 5.7 | 452 | 4.9 | 497 | 5.2 |
| Boston, MA | 283 | 3.2 | 417 | 4.5 | 302 | 3.1 |
| Chicago, IL | 566 | 6.4 | 501 | 5.4 | 519 | 5.4 |
| Dallas, TX | 509 | 5.7 | 471 | 5.1 | 500 | 5.2 |
| Denver, CO | 544 | 6.1 | 547 | 5.9 | 515 | 5.3 |
| Detroit, MI | 388 | 4.4 | 460 | 5.0 | 512 | 5.3 |
| Houston, TX | 448 | 5.0 | 509 | 5.5 | 508 | 5.3 |
| Los Angeles, CA | 537 | 6.0 | 520 | 5.6 | 524 | 5.4 |
| Miami, FL | 529 | 6.0 | 504 | 5.4 | 534 | 5.5 |
| Nassau-Suffolk, NY | 281 | 3.2 | 339 | 3.7 | 338 | 3.5 |
| New Orleans, LA | 478 | 5.4 | 488 | 5.3 | 517 | 5.4 |
| New York, NY | 554 | 6.2 | 521 | 5.6 | 508 | 5.3 |
| Newark, NJ | 98 | 1.1 | 250 | 2.7 | 246 | 2.6 |
| Philadelphia, PA | 563 | 6.3 | 545 | 5.9 | 655 | 6.8 |
| San Diego, CA | 549 | 6.2 | 471 | 5.1 | 538 | 5.6 |
| San Francisco, CA | 487 | 5.5 | 465 | 5.0 | 388 | 4.0 |
| San Juan, PR | 355 | 4.0 | 363 | 3.9 | 515 | 5.3 |
| Seattle, WA | 362 | 4.1 | 371 | 4.0 | 503 | 5.2 |
| Washington, DC | 501 | 5.6 | 501 | 5.4 | 510 | 5.3 |
| Total | 8881 | 9253 | 9636 | |||
Numbers might not add to total because of missing or unknown data. Percentages might not sum to 100 because of rounding.
Includes men who consented to and completed the survey, had a male sex partner in the past 12 months and reported their HIV status.
Hispanic can be of any race.
Other race includes multiple races.
General educational development (GED) diploma.
Infrequent use of the internet is defined as <once a week, frequent use is defined as ≥once a week.
Having Met the Last Sex Partner Online
Across years, white MSM and those with an annual income of $75,000 or more had higher reports of meeting the last sex partner online. Overall, having met the last sex partner online increased from 19% in 2008 to 24% in 2011 and 32% in 2014 (APR = 1.30, 95% CI: 1.26 to 1.34). IU increased over time in all age, race, income, and HIV-status categories (Table 2).
TABLE 2.
Met the Last Sex Partner on the Internet, Men Who Have Sex With Men, NHBS, 20 Cities, United States, 2008–2014*
| 2008 | 2011 | 2014 | ||||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|
||||||
| n | % | n | % | n | % | Unadjusted PR (95% CI)† | Adjusted PR (95% CI)‡ | |
| Overall | 1424 | 19.1 | 1908 | 24.3 | 2617 | 32.1 | 1.30 (1.26 to 1.34) | 1.30 (1.26 to 1.34) |
| Age, yrs | ||||||||
| 18–24 | 338 | 18.0 | 513 | 22.9 | 583 | 31.5 | 1.33 (1.25 to 1.41) | 1.30 (1.22 to 1.38) |
| 25–29 | 311 | 21.6 | 399 | 25.3 | 678 | 35.9 | 1.30 (1.23 to 1.38) | 1.32 (1.24 to 1.39) |
| 30–39 | 414 | 20.2 | 514 | 28.0 | 733 | 35.3 | 1.32 (1.25 to 1.39) | 1.32 (1.25 to 1.38) |
| 40–49 | 261 | 17.8 | 331 | 23.1 | 383 | 29.3 | 1.28 (1.20 to 1.37) | 1.29 (1.21 to 1.38) |
| ≥50 | 100 | 16.6 | 151 | 19.4 | 240 | 23.5 | 1.19 (1.08 to 1.33) | 1.22 (1.10 to 1.35) |
| Race | ||||||||
| Black | 240 | 13.3 | 410 | 19.2 | 554 | 24.3 | 1.34 (1.25 to 1.44) | 1.31 (1.23 to 1.41) |
| Hispanic§ | 336 | 17.6 | 508 | 24.3 | 702 | 32.2 | 1.35 (1.27 to 1.43) | 1.32 (1.25 to 1.40) |
| White | 700 | 22.8 | 833 | 27.4 | 1133 | 37.4 | 1.29 (1.24 to 1.35) | 1.30 (1.25 to 1.36) |
| Other‖ | 146 | 22.3 | 153 | 26.4 | 212 | 34.3 | 1.20 (1.09 to 1.32) | 1.22 (1.11 to 1.34) |
| Annual household income | ||||||||
| 0–$19,999 | 310 | 13.4 | 512 | 19.8 | 633 | 24.4 | 1.33 (1.25 to 1.42) | 1.33 (1.25 to 1.42) |
| $20,000–$39,999 | 373 | 19.4 | 484 | 24.5 | 678 | 33.9 | 1.33 (1.26 to 1.41) | 1.32 (1.25 to 1.40) |
| $40,000–$74,999 | 415 | 23.2 | 495 | 26.7 | 679 | 34.6 | 1.23 (1.16 to 1.30) | 1.24 (1.17 to 1.31) |
| $75,000 or more | 309 | 23.7 | 389 | 29.3 | 589 | 40.0 | 1.30 (1.23 to 1.38) | 1.32 (1.24 to 1.40) |
| Self-reported HIV status | ||||||||
| HIV positive | 181 | 21.1 | 245 | 25.2 | 421 | 32.8 | 1.30 (1.25 to 1.34) | 1.31 (1.27 to 1.35) |
| HIV negative | 1104 | 19.4 | 1516 | 24.8 | 2060 | 32.7 | 1.25 (1.16 to 1.35) | 1.28 (1.19 to 1.38) |
| HIV unknown | 139 | 15.3 | 147 | 19.0 | 136 | 24.1 | 1.26 (1.12 to 1.41) | 1.25 (1.12 to 1.40) |
| City | ||||||||
| Atlanta, GA | 74 | 22.9 | 129 | 27.9 | 138 | 31.5 | 1.17 (1.02 to 1.34) | 1.20 (1.04 to 1.38) |
| Baltimore, MD | 64 | 15.7 | 49 | 13.2 | 109 | 27.0 | 1.35 (1.14 to 1.61) | 1.34 (1.14 to 1.56) |
| Boston, MA | 57 | 24.2 | 111 | 31.7 | 107 | 41.2 | 1.30 (1.12 to 1.52) | 1.32 (1.13 to 1.53) |
| Chicago, IL | 101 | 21.2 | 106 | 24.8 | 167 | 38.8 | 1.37 (1.22 to 1.54) | 1.33 (1.19 to 1.49) |
| Dallas, TX | 59 | 13.6 | 65 | 16.2 | 98 | 22.6 | 1.31 (1.10 to 1.55) | 1.30 (1.10 to 1.54) |
| Denver, CO | 76 | 17.1 | 116 | 24.2 | 143 | 34.3 | 1.42 (1.27 to 1.58) | 1.40 (1.26 to 1.56) |
| Detroit, MI | 40 | 11.8 | 76 | 18.8 | 110 | 24.8 | 1.42 (1.23 to 1.66) | 1.41 (1.22 to 1.62) |
| Houston, TX | 49 | 13.3 | 91 | 20.9 | 113 | 27.0 | 1.40 (1.18 to 1.67) | 1.42 (1.21 to 1.68) |
| Los Angeles, CA | 99 | 21.7 | 131 | 28.7 | 176 | 39.1 | 1.35 (1.20 to 1.51) | 1.36 (1.22 to 1.52) |
| Miami, FL | 63 | 13.7 | 117 | 25.8 | 132 | 28.0 | 1.38 (1.19 to 1.59) | 1.35 (1.17 to 1.56) |
| Nassau-Suffolk, NY | 62 | 29.5 | 75 | 26.3 | 129 | 45.9 | 1.30 (1.10 to 1.54) | 1.30 (1.12 to 152) |
| New Orleans, LA | 46 | 12.3 | 61 | 15.3 | 97 | 22.7 | 1.37 (1.19 to 1.58) | 1.39 (1.21 to 1.60) |
| New York, NY | 85 | 18.0 | 87 | 18.9 | 170 | 39.3 | 1.53 (1.35 to 1.74) | 1.41 (1.25 to 1.58) |
| Newark, NJ | 12 | 15.2 | 58 | 26.5 | 51 | 25.6 | 1.18 (0.94 to 1.48) | 1.18 (0.93 to 1.49) |
| Philadelphia, PA | 99 | 22.5 | 115 | 27.2 | 158 | 27.3 | 1.10 (0.98 to 1.22) | 1.12 (1.01 to 1.25) |
| San Diego, CA | 115 | 25.5 | 112 | 28.5 | 172 | 40.1 | 1.26 (1.15 to 1.39) | 1.28 (1.15 to 1.40) |
| San Francisco, CA | 104 | 25.0 | 114 | 28.7 | 134 | 42.1 | 1.30 (1.15 to 1.47) | 1.28 (1.14 to 1.45) |
| San Juan, PR | 55 | 18.0 | 94 | 30.0 | 140 | 30.6 | 1.26 (1.08 to 1.46) | 1.26 (1.09 to 1.46) |
| Seattle, WA | 73 | 23.6 | 72 | 23.3 | 137 | 32.2 | 1.19 (1.02 to 1.38) | 1.23 (1.06 to 1.42) |
| Washington, DC | 91 | 20.7 | 129 | 30.1 | 136 | 31.4 | 1.21 (1.09 to 1.36) | 1.21 (1.09 to 1.35) |
Numbers might not add to total because of missing or unknown data. Percentages might not sum to 100 because of rounding.
Includes men who consented to and completed the survey, had a male sex partner in the past 12 months and reported their HIV status.
Reference is 2008; PR corresponds with the interaction term with year.
Reference is 2008; PR corresponds to the interaction term with year; model includes year, race/ethnicity, and annual household income, and their interactions with year as fixed effects.
Hispanic can be of any race.
Includes MSM reporting American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, other race, or multiple races.
Frequency of Internet Use to Meet Men
The percentage of men who reported never using the internet to meet men decreased over time from 62% in 2008 to 36% in 2011, and 31% in 2014. Across all years, white MSM were more likely to report frequent IU compared with other races. The percent of men who reported frequent IU increased from 21% in 2008 to 42% in 2011 and 44% in 2014 (APR = 1.39, 95% CI: 1.36 to 1.42) and increased in all age, race, income, and HIV-status categories. Significant increases were seen in all cities but one (Table 3).
TABLE 3.
Frequent Use of the Internet to Meet Men, Men Who Have Sex With Men, NHBS, 20 Cities, United States, 2008–2014*
| 2008 | 2011 | 2014 | ||||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|
||||||
| n | % | n | % | n | % | Unadjusted PR (95% CI)† | Adjusted PR (95% CI)‡ | |
| Overall | 1821 | 20.6 | 3858 | 41.7 | 4225 | 43.9 | 1.39 (1.36 to 1.42) | 1.39 (1.36 to 1.42) |
| Age, yrs | ||||||||
| 18–24 | 333 | 16.7 | 1058 | 45.1 | 917 | 46.8 | 1.52 (1.46 to 1.59) | 1.51 (1.45 to 1.58) |
| 25–29 | 356 | 21.8 | 814 | 46.5 | 1064 | 50.5 | 1.43 (1.37 to 1.49) | 1.44 (1.38 to 1.50) |
| 30–39 | 588 | 23.9 | 1016 | 46.4 | 1177 | 47.4 | 1.36 (1.31 to 1.41) | 1.36 (1.31 to 1.41) |
| 40–49 | 380 | 20.1 | 657 | 35.1 | 671 | 39.8 | 1.38 (1.31 to 1.45) | 1.37 (1.30 to 1.44) |
| ≥50 | 164 | 19.2 | 313 | 28.8 | 396 | 28.4 | 1.18 (1.10 to 1.27) | 1.18 (1.10 to 1.27) |
| Race | ||||||||
| Black | 363 | 17.2 | 975 | 39.3 | 1051 | 39.7 | 1.41 (1.34 to 1.48) | 1.40 (1.33 to 1.47) |
| Hispanic§ | 432 | 19.5 | 1038 | 43.1 | 1112 | 44.1 | 1.41 (1.35 to 1.47) | 1.41 (1.35 to 1.47) |
| White | 863 | 23.1 | 1553 | 42.4 | 1695 | 46.2 | 1.38 (1.34 to 1.42) | 1.38 (1.34 to 1.43) |
| Other‖ | 161 | 21.1 | 286 | 42.6 | 342 | 46.1 | 1.38 (1.28 to 1.48) | 1.39 (1.29 to 1.49) |
| Annual household income | ||||||||
| 0–$19,999 | 471 | 18.1 | 1130 | 38.9 | 1172 | 40.2 | 1.40 (1.35 to 1.46) | 1.40 (1.35 to 1.46) |
| $20,000–$39,999 | 475 | 21.4 | 987 | 43.4 | 1076 | 47.1 | 1.42 (1.36 to 1.48) | 1.41 (1.36 to 1.47) |
| $40,000–$74,999 | 478 | 22.2 | 945 | 43.1 | 1055 | 45.1 | 1.37 (1.31 to 1.42) | 1.36 (1.31 to 1.42) |
| $75,000 or more | 370 | 21.4 | 736 | 42.5 | 872 | 44.5 | 1.38 (1.32 to 1.44) | 1.39 (1.33 to 1.46) |
| Self-reported HIV status | ||||||||
| HIV positive | 270 | 25.5 | 525 | 42.2 | 694 | 43.8 | 1.41 (1.37 to 1.44) | 1.41 (1.38 to 1.45) |
| HIV negative | 1359 | 20.2 | 3018 | 42.4 | 3294 | 44.4 | 1.26 (1.20 to 1.33) | 1.27 (1.21 to 1.34) |
| HIV unknown | 192 | 18.3 | 315 | 35.9 | 237 | 37.9 | 1.42 (1.31 to 1.53) | 1.42 (1.32 to 1.53) |
| City | ||||||||
| Atlanta, GA | 121 | 35.1 | 219 | 39.3 | 246 | 48.5 | 1.18 (1.08 to 1.30) | 1.20 (1.09 to 1.32) |
| Baltimore, MD | 75 | 15 | 127 | 28.4 | 158 | 31.8 | 1.42 (1.26 to 1.60) | 1.43 (1.28 to 1.61) |
| Boston, MA | 72 | 25.5 | 173 | 41.5 | 157 | 52.2 | 1.40 (1.27 to 1.55) | 1.38 (1.25 to 1.53) |
| Chicago, IL | 112 | 20 | 230 | 45.9 | 248 | 47.9 | 1.47 (1.36 to 1.58) | 1.47 (1.37 to 1.58) |
| Dallas, TX | 80 | 15.8 | 179 | 38 | 187 | 37.4 | 1.44 (1.31 to 1.59) | 1.46 (1.33 to 1.61) |
| Denver, CO | 108 | 20 | 245 | 44.9 | 236 | 45.8 | 1.44 (1.33 to 1.56) | 1.44 (1.33 to 1.56) |
| Detroit, MI | 60 | 15.5 | 197 | 42.9 | 222 | 43.4 | 1.48 (1.34 to 1.65) | 1.50 (1.34 to 1.67) |
| Houston, TX | 69 | 15.4 | 171 | 33.6 | 181 | 35.6 | 1.43 (1.27 to 1.61) | 1.42 (1.26 to 1.60) |
| Los Angeles, CA | 109 | 20.8 | 212 | 40.8 | 255 | 48.7 | 1.48 (1.37 to 1.59) | 1.49 (1.38 to 1.61) |
| Miami, FL | 123 | 23.3 | 241 | 47.8 | 243 | 45.5 | 1.33 (1.20 to 1.48) | 1.32 (1.20 to 1.46) |
| Nassau-Suffolk, NY | 86 | 30.8 | 140 | 41.3 | 152 | 45 | 1.20 (1.06 to 1.35) | 1.17 (1.04 to 1.32) |
| New Orleans, LA | 57 | 11.9 | 156 | 32 | 197 | 38.1 | 1.64 (1.47 to 1.83) | 1.64 (1.47 to 1.84) |
| New York, NY | 115 | 21 | 170 | 32.6 | 243 | 47.8 | 1.51 (1.39 to 1.63) | 1.46 (1.34 to 1.58) |
| Newark, NJ | 6 | 6.4 | 171 | 68.4 | 144 | 58.5 | 1.43 (1.24 to 1.66) | 1.46 (1.26 to 1.69) |
| Philadelphia, PA | 162 | 28.9 | 215 | 39.4 | 228 | 34.9 | 1.08 (1.00 to 1.18) | 1.10 (1.01 to 1.19) |
| San Diego, CA | 130 | 23.7 | 219 | 46.5 | 244 | 45.4 | 1.33 (1.22 to 1.46) | 1.36 (1.24 to 1.49) |
| San Francisco, CA | 119 | 24.4 | 224 | 48.2 | 192 | 49.5 | 1.38 (1.28 to 1.49) | 1.39 (1.28 to 1.50) |
| San Juan, PR | 61 | 17.3 | 158 | 43.5 | 216 | 41.9 | 1.41 (1.27 to 1.57) | 1.43 (1.28 to 1.59) |
| Seattle, WA | 73 | 20.2 | 199 | 53.6 | 241 | 47.9 | 1.39 (1.27 to 1.53) | 1.43 (1.30 to 1.57) |
| Washington, DC | 83 | 16.6 | 212 | 42.3 | 235 | 46.1 | 1.55 (1.42 to 1.68) | 1.54 (1.41 to 1.67) |
Numbers might not add to total because of missing or unknown data. Percentages might not sum to 100 because of rounding. Frequent use of the internet is defined as ≥once a week.
Includes men who consented to and completed the survey, had a male sex partner in the past 12 months and reported their HIV status.
Reference is 2008; PR corresponds with the interaction term with year.
Reference is 2008; PR corresponds to the interaction term with year; model includes year, race/ethnicity, and annual household income, and their interactions with year as fixed effects.
Hispanic can be of any race.
Includes MSM reporting American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, other race, or multiple races.
Characteristics of MSM by Frequency of Internet Use
Median number of male sex partners in the past 12 months increased with increasing IU. In 2014, those who never used the internet had a median of 2 partners [interquartile range (IQR): 1–4] compared with a median of 4 among infrequent internet users (IQR: 2–7) and 5 among frequent users (IQR: 3–12) (P < 0.0001). The percentage reporting any condomless sex in the past 12 months increased with increasing IU (55%, 67%, and 70%, P < 0.001). HIV testing in the past 12 months also increased with increasing IU (58%, 68%, and 71%, respectively, P < 0.0001). HIV prevalence decreased with increasing IU (25%, 20%, and 20%, P < 0.0001). Similarly, the percent of HIV-positive MSM who were aware of their status was higher among nonfrequent and frequent internet users (71%, 75%, and 79%, P < 0.005) (Table 4).
TABLE 4.
Characteristics of Men by Frequency of Internet Use, Men Who Have Sex With Men, NHBS, 20 Cities, United States, 2014*
| Internet Usage | |||||||
|---|---|---|---|---|---|---|---|
|
|
|||||||
| Never | Infrequent | Frequent | |||||
|
|
|
|
|||||
| n | % | n | % | n | % | P | |
| Used noninjection drugs, past 12 mo | |||||||
| Yes | 1510 | 50.5 | 1438 | 59.5 | 2503 | 59.3 | <0.0001 |
| No | 1479 | 49.5 | 981 | 40.6 | 1721 | 40.7 | |
| STD, past 12 mo | |||||||
| Yes | 234 | 7.8 | 363 | 15.0 | 750 | 17.8 | <0.0001 |
| No | 2749 | 92.2 | 2053 | 85.0 | 3470 | 82.2 | |
| Anal sex with a male without a condom, past 12 mo | |||||||
| Yes | 1636 | 54.8 | 1609 | 66.7 | 2955 | 70 | <0.0001 |
| No | 1349 | 45.2 | 803 | 33.3 | 1267 | 30 | |
| HIV status of last male sex partner | |||||||
| Discordant | 260 | 8.7 | 187 | 7.7 | 353 | 8.4 | 0.08 |
| Concordant | 1525 | 51.1 | 1333 | 55.2 | 2255 | 53.5 | |
| Unknown | 1200 | 40.2 | 896 | 37.1 | 1611 | 38.2 | |
| Anal sex without a condom with last male sex partner by partners status | |||||||
| Discordant | 76 | 2.6 | 80 | 3.3 | 162 | 3.8 | 0.52 |
| Concordant | 2532 | 84.9 | 2032 | 84.4 | 3532 | 83.8 | |
| Unknown | 374 | 12.5 | 297 | 12.3 | 523 | 12.4 | |
| Type of male sex partners, past 12 mo | |||||||
| Main only | 1024 | 34.3 | 395 | 16.3 | 506 | 12.0 | <0.0001 |
| Casual only | 1118 | 37.4 | 835 | 34.5 | 1514 | 35.8 | |
| Main and casual | 845 | 28.3 | 1188 | 49.1 | 2205 | 52.2 | |
| Exchanged sex, past 12 mo | |||||||
| Yes | 388 | 13.0 | 229 | 9.5 | 438 | 10.4 | <0.0001 |
| No | 2598 | 87.0 | 2188 | 90.5 | 3786 | 89.6 | |
| Tested for HIV, past 12 mo | |||||||
| Yes | 1716 | 57.6 | 1631 | 67.6 | 3011 | 71.4 | <0.0001 |
| No | 1264 | 42.4 | 782 | 32.4 | 1209 | 28.7 | |
| HIV test result† | |||||||
| Positive | 673 | 24.9 | 437 | 20.0 | 781 | 20.1 | <0.0001 |
| Negative | 2025 | 75.1 | 1745 | 80.0 | 3097 | 79.9 | |
|
| |||||||
| Mean/Median | Mean/Median | Mean/Median | |||||
|
| |||||||
| No. of male sex partners, past 12 mo | |||||||
| Mean (range) | 6.1 (1–1000) | 7.5 (1–500) | 14.2 (1–1000) | <0.0001 | |||
| Median (IQR) | 2 (1–4) | 4 (2–7) | 5 (3–12) | <0.0001 | |||
| No. of casual male anal sex partners with inconsistent condom use, past 12 mo | |||||||
| Mean (range) | 1.1 (0–500) | 1.8 (0–500) | 3.5 (0–320) | <0.0001 | |||
| Median (IQR) | 0 (0–0) | 0 (0–1) | 0 (0–2) | <0.0001 | |||
Numbers might not add to total because of missing or unknown data. Percentages might not sum to 100 because of rounding.
Includes men who consented to and completed the survey, had a male sex partner in the past 12 months and reported their HIV status.
Limited to those with a valid HIV test result in NHBS.
DISCUSSION
In a sample of men recruited offline in gay venues in large cities in the United States, IU to meet men was very common; in 2014, two-third of men had used the internet to meet other men and one-third had met their last sex partner online. Both the percentage of MSM using the internet frequently to meet men and the percentage of men who met the last sex partner online have increased since 2008, illustrating how common IU for social and sexual connection has become among MSM. Data from 2014 showed that men who used the internet more frequently had higher number of partners and higher reports of condomless sex. However, frequent internet users also had higher testing rates and were less likely to be HIV positive but unaware of their status.
As others have done before, we documented sexual risk among MSM using the internet to meet men. A recent meta-analysis found that meeting sex partners online was associated with increased risk of condomless anal sex, group sex, serosorting, and strategic positioning.13 We have reported previously on an association between frequent IU and a higher occurrence of recent and acute HIV infection among MSM.19 Case control studies in the United Kingdom found that men who use the internet to meet sex partners had twice the odds of seroconverting to being HIV positive compared with those who did not use the internet.20 Furthermore, several studies have shown increased prevalence of chlamydia and gonorrhea among men who use the internet to meet sex partners.1 However, it is important to note that it is not clear if the association between IU and high-risk sex is because of the internet facilitating riskier sex or to riskier men using the internet to meet sex partners. The internet environment may accentuate risk taking and make it easier to meet partners for condomless sex by providing an anonymous interface to disclose sexual preferences.12 However, the association could also be because of high-risk men being more likely to use the internet to meet sex partners, as suggested by research showing higher number of partners21 and STIs1 among those who meet partners online versus those meeting partners offline only.
Our data and data from others also suggest that MSM who use the internet more frequently are taking steps to protect their health. We found a lower HIV prevalence among those using the internet, and HIV-positive MSM who used the internet to meet men were more likely to be aware of their status. Furthermore, we found higher HIV testing among those using the internet more frequently. The formation of online partnerships may include intensive communications including risk negotiations before an offline, in-person contact is made.12 In contrast, typical offline venues such as bars and clubs may not be conducive to such negotiations. However, risk negotiations do not invariably result in higher levels of protected sex; the accuracy of self-reported status among prospective sex partners cannot be taken for granted, as it depends on when the last HIV test occurred and on the veracity of the reported status.22 A recent survey of website/app owners, users, and public health professionals found agreement about several ways that dating apps and websites could be used to promote public health, including: automated HIV and STI testing reminders, local STI testing directories, access to sexual-health experts, profile options to include safer sex preference, and other strategies.23 There is also evidence that many MSM across age and racial/ethnic groups are using the internet generally, and dating apps and websites specifically, to receive and obtain sexual-health information.24,25 Encouraging frequent HIV testing among HIV-negative MSM using the internet to meet men and disclosure of HIV status among HIV-positive MSM may reduce the likelihood of discordant condomless sex and the potential for HIV transmission. A recent meta-analyses of HIV testing among MSM recruited online suggests that HIV-testing frequency is suboptimal and there is room for improvement.26 The ability to target messages to specific populations, on specific apps or websites, or in specific geographic areas gives great power to public health practitioners to tailor messages for maximum impact.
The findings in this report are subject to limitations. NHBS data are from MSM who were recruited at venues in cities with high HIV burden. Thus, results may not be generalizable to all MSM. Recruiting men at venues may lead to an underestimate of IU, however, if venues such as bars are popular locations to meet partners found online, it could lead to an overestimate of IU. Furthermore, analyses were based on self-reported data and may be subject to social desirability bias. In addition, all data are cross-sectional, thus we are not able to causally link use of the internet to meet men with risk or protective behaviors. The question to assess frequency of IU changed from 2008 to 2011, when most of the change was seen for increases in frequent use. However, these changes were accompanied by increases in having met the last sex partner online, a measure that remains unchanged. This provides evidence that the observed trends are real and not a measurement artifact. NHBS does not collect data on specific websites or apps used. There are likely important differences between using the internet for socializing and specifically to find sex partners that our study cannot investigate. Finally, data are not weighted to account for the complex sampling methodology used to recruit MSM. Point estimates may therefore be biased by over- or under-represented subgroups of the population. However, multivariate analysis of differences across years should not be affected by a lack of weighting, especially given the consistency of data distribution across years.
Gay, bisexual, and other MSM represent two-thirds of new HIV diagnoses in the United States and are increasingly using the internet to meet men. It is encouraging to see more testing and high awareness of status among those using the internet frequently, but our data underscore the need for innovative and effective strategies to best HIV-prevention needs of MSM who use dating apps and websites. The implications of these trends for HIV or STI transmission risk are complex and suggest the need to develop online prevention strategies to reach MSM who use apps, who engaged in more condomless sex, and prevention strategies for men not using apps, who were less likely to be aware of their HIV status.
Supplementary Material
Acknowledgments
Supported by the Centers for Disease Control and Prevention.
We thank the NHBS participants.
For the full list of NHBS Study Group participants, please see Supplemental Digital Content, http://links.lww.com/QAI/B38.
Footnotes
Presented in part at the International AIDS Conference; July 18–22, 2016; Durban, South Africa.
The authors have no conflicts of interest to disclose.
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.jaids.com).
References
- 1.Beymer MR, Weiss RE, Bolan RK, et al. Sex on demand: geosocial networking phone apps and risk of sexually transmitted infections among a cross-sectional sample of men who have sex with men in Los Angeles County. Sex Transm Infect. 2014;90:567–572. doi: 10.1136/sextrans-2013-051494. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bolding G, Davis M, Hart G, et al. Where young MSM meet their first sexual partner: the role of the internet. AIDS Behav. 2007;11:522–526. doi: 10.1007/s10461-007-9224-9. [DOI] [PubMed] [Google Scholar]
- 3.Grov C, Breslow AS, Newcomb ME, et al. Gay and bisexual men’s use of the internet: research from the 1990s through 2013. J Sex Res. 2014;51:390–409. doi: 10.1080/00224499.2013.871626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Woo J. Meet Grindr, How One App Changed the Way We Connect. Canada: Jaime Woo; 2013. [Google Scholar]
- 5.CDC. Pneumocystis pneumonia—Los Angeles. MMWR Morb Mortal Wkly Rep. 1981;30:250–252. [PubMed] [Google Scholar]
- 6.Grey JA, Bernstein KT, Sullivan PS, et al. Estimating the population sizes of men who have sex with men in US states and counties using data from the American Community Survey. JMIR Public Health Surveill. 2016;2:e14. doi: 10.2196/publichealth.5365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Purcell DW, Johnson CH, Lansky A, et al. Estimating the population size of men who have sex with men in the United States to obtain HIV and syphilis rates. Open AIDS J. 2012;6:98–107. doi: 10.2174/1874613601206010098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.CDC. Diagnoses of HIV Infection in the United States and Dependent Areas, 2014. Atlanta, GA: CDC; 2015. [Google Scholar]
- 9.Rosenberg ES, Grey JA, Sanchez TH, et al. Rates of prevalent HIV infection, prevalent diagnoses, and new diagnoses among men who have sex with men in US states, metropolitan statistical areas, and counties, 2012–2013. JMIR Public Health Surveill. 2016;2:e22. doi: 10.2196/publichealth.5684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wejnert C, Hess KL, Rose CE, et al. Age-specific race and ethnicity disparities in HIV infection and awareness among men who have sex with men—20 US cities, 2008–2014. J Infect Dis. 2016;213:776–783. doi: 10.1093/infdis/jiv500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Abara WE, Hess KL, Neblett Fanfair R, et al. Syphilis trends among men who have sex with men in the United States and Western Europe: a systematic review of trend studies published between 2004 and 2015. PLoS One. 2016;11:e0159309. doi: 10.1371/journal.pone.0159309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lewnard JA, Berrang-Ford L. Internet-based partner selection and risk for unprotected anal intercourse in sexual encounters among men who have sex with men: a meta-analysis of observational studies. Sex Transm Infect. 2014;90:290–296. doi: 10.1136/sextrans-2013-051332. [DOI] [PubMed] [Google Scholar]
- 13.Liau A, Millett G, Marks G. Meta-analytic examination of online sex-seeking and sexual risk behavior among men who have sex with men. Sex Transm Dis. 2006;33:576–584. doi: 10.1097/01.olq.0000204710.35332.c5. [DOI] [PubMed] [Google Scholar]
- 14.Yang Z, Zhang S, Dong Z, et al. Prevalence of unprotected anal intercourse in men who have sex with men recruited online versus offline: a meta-analysis. BMC Public Health. 2014;14:508. doi: 10.1186/1471-2458-14-508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Al-Tayyib AA, McFarlane M, Kachur R, et al. Finding sex partners on the internet: what is the risk for sexually transmitted infections? Sex Transm Infect. 2009;85:216–220. doi: 10.1136/sti.2008.032631. [DOI] [PubMed] [Google Scholar]
- 16.Buhi ER, Klinkenberger N, McFarlane M, et al. Evaluating the internet as a sexually transmitted disease risk environment for teens: findings from the communication, health, and teens study. Sex Transm Dis. 2013;40:528–533. doi: 10.1097/OLQ.0b013e31829413f7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.MacKellar DA, Gallagher KM, Finlayson T, et al. Surveillance of HIV risk and prevention behaviors of men who have sex with men—a national application of venue-based, time-space sampling. Public Health Rep. 2007;122(suppl 1):39–47. doi: 10.1177/00333549071220S107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Finlayson TJ, Le B, Smith A, et al. HIV risk, prevention, and testing behaviors among men who have sex with men—National HIV Behavioral Surveillance System, 21 U.S. cities, United States, 2008. MMWR Surveill Summ. 2011;60:1–34. [PubMed] [Google Scholar]
- 19.Paz-Bailey G, Smith A, Masciotra S, et al. Early HIV infections among men who have sex with men in five cities in the United States. AIDS Behav. 2015;19:2304–2310. doi: 10.1007/s10461-015-1011-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Macdonald N, Elam G, Hickson F, et al. Factors associated with HIV seroconversion in gay men in England at the start of the 21st century. Sex Transm Infect. 2008;84:8–13. doi: 10.1136/sti.2007.027946. [DOI] [PubMed] [Google Scholar]
- 21.Kim AA, Kent C, McFarland W, et al. Cruising on the internet highway. J Acquir Immune Defic Syndr. 2001;28:89–93. doi: 10.1097/00042560-200109010-00013. [DOI] [PubMed] [Google Scholar]
- 22.Aral SO, Fenton KA, Lipshutz J. The New Public Health and STD/HIV Prevention: Personal, Public and Health Systems Approaches. New York, NY: Springer; 2013. [Google Scholar]
- 23.Wohlfeiler D, Hecht J, Volk J, et al. How can we improve online HIV and STD prevention for men who have sex with men? Perspectives of hook-up website owners, website users, and HIV/STD directors. AIDS Behav. 2013;17:3024–3033. doi: 10.1007/s10461-012-0375-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Badal H, Stryker J, DeLuca N, et al. Swipe left: dating/hookup website and app use and men who have sex with men; Presented at: National HIV Prevention Conference; December 6–9, 2015; Atlanta, GA. [Google Scholar]
- 25.DeLuca N, Badal H, Stryker J, et al. Sex sells: utilizing effective digital channels to reach men who have sex with men with HIV testing and prevention messages; Presented at: National HIV Prevention Conference; December 6–9, 2015; Atlanta, GA. [Google Scholar]
- 26.Noble M, Jones AM, Bowles K, et al. HIV testing among internet-using MSM in the United States: systematic review. AIDS Behav. 2017;21:561–575. doi: 10.1007/s10461-016-1506-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.