McFarlane M, Bull SS, Rietmeijer CA. The Internet as a Newly Emerging Risk Environment for Sexually Transmitted
Diseases. JAMA. 2000;284(4):443-446. doi:10.1001/jama.284.4.443
Author Affiliations: Division of Sexually Transmitted Disease Prevention, Centers for Disease Control and Prevention, Atlanta, Ga (Dr McFarlane); and Denver Public Health Department, Denver Health and Hospital Authority, Denver, Colo (Drs Bull and Rietmeijer).
Context Transmission of sexually transmitted diseases (STDs) such as human immunodeficiency
virus (HIV) infection is associated with unprotected sex among multiple anonymous
sex partners. The role of the Internet in risk of STDs is not known.
Objective To compare risk of STD transmission for persons who seek sex partners
on the Internet with risk for persons not seeking sex partners on the Internet.
Design Cross-sectional survey conducted September 1999 through April 2000.
Setting and Participants A total of 856 clients of the Denver Public Health HIV Counseling and
Testing Site in Colorado.
Main Outcome Measures Self-report of logging on to the Internet with the intention of finding
sex partners; having sex with partners who were originally contacted via the
Internet; number of such partners and use of condoms with them; and time since
last sexual contact with Internet partners, linked to HIV risk assessment
and test records.
Results Of the 856 clients, most were white (77.8%), men (69.2%), heterosexual
(65.3%), and aged 20 to 50 years (84.1%). Of those, 135 (15.8%) had sought
sex partners on the Internet, and 88 (65.2%) of these reported having sex
with a partner initially met via the Internet. Of those with Internet partners,
34 (38.7%) had 4 or more such partners, with 62 (71.2%) of contacts occurring
within 6 months prior to the client's HIV test. Internet sex seekers were
more likely to be men (P<.001) and homosexual
(P<.001) than those not seeking sex via the Internet.
Internet sex seekers reported more previous STDs (P
= .02); more partners (P<.001); more anal sex
(P<.001); and more sexual exposure to men (P<.001), men who have sex with men (P<.001), and partners known to be HIV positive (P<.001) than those not seeking sex via the Internet.
Conclusions Seeking sex partners via the Internet was a relatively common practice
in this sample of persons seeking HIV testing and counseling (representative
of neither Denver nor the overall US population). Clients who seek sex using
the Internet appear to be at greater risk for STDs than clients who do not
seek sex on the Internet.
Several studies have shown that transmission of sexually transmitted
diseases (STDs), such as human immunodeficiency virus (HIV) infection, often
involves persons with multiple and sometimes anonymous sex partners.1- 10
Sex with anonymous partners typically has been initiated in bars, bathhouses,
clubs, or parks.4 It has been suggested that
the Internet may be another venue for the initiation of sexual contact.11 Observations of chat rooms and other Internet sites
reveal that the Internet facilitates communication of sexual desires, as well
as in-person meetings resulting in sexual contact.12
Identifying information such as full name, address, or place of work may be
withheld from Internet-based sex partners. Due to the impossibility of observing
this behavior from initiation of conversation to completion of a sexual encounter,
it is difficult to gauge the rate at which persons engage in Internet-initiated
sexual relationships. Also, it is difficult to assess, based on Internet observations,
the risk of STD/HIV transmission resulting from these encounters. Our goal
was to determine whether the use of the Internet to solicit sex partners should
be considered a potential risk factor for STD/HIV. To ascertain whether high-risk
persons seek sex partners on the Internet, we surveyed clients of an HIV counseling
and testing site regarding Internet behaviors and risk for STD/HIV.
This study was reviewed and approved by the institutional review board
at the Centers for Disease Control and Prevention (CDC), as well as by the
Colorado Multiple Institutional Review Board. Clients aged 18 years or older
who sought HIV testing at the Denver Public Health HIV Counseling and Testing
Site (DPHHCTS) were eligible for the study. All eligible clients presenting
for counseling and testing during the study period (September 1999-April 2000)
were invited to participate. Clients completed a detailed consent form referencing
the risk-assessment questionnaire administered by a counselor. The counselor,
who had completed CDC–supported training for client-centered HIV counseling,
reviewed the consent form with the patient before obtaining consent.
The HIV testing protocol included a risk-assessment questionnaire covering
sexual orientation (heterosexual, homosexual, or bisexual), number of sex
partners in the past 12 months, and history of sexual exposure to risky partners
(ie, persons with history of STDs, HIV, or injection drug use; men who have
sex with men; or hemophiliacs). Questions about drug and alcohol use, condom
use during the last sexual encounter, and types of sex performed (ie, oral,
anal, or vaginal) were included. Clients were asked to provide a blood specimen.
Test results were entered into the record.
We also asked clients to complete a 6-item questionnaire, administered
via face-to-face interview, about their use of the Internet for solicitation
of sex partners. Of the 911 patients eligible for the survey, 55 did not participate,
yielding a 94% participation rate. All questions were administered by the
counselor who obtained the general risk information; counselors administering
these questions received survey-administration training. A client could refuse
to answer any question. The first question was whether the client had ever
logged on to the Internet with the intention of seeking a sex partner. Clients
answering in the negative were not asked further questions. Those answering
affirmatively were asked whether they had successfully solicited a sex partner
via the Internet, the number of partners they had ever solicited in this manner,
time since the last Internet-initiated contact, whether a condom was used
during the last Internet-initiated contact, and from what location they most
often logged on to the Internet (eg, home, work, library).
Responses to all Internet-related questions were recorded on a sheet
separate from the HIV testing record and entered into a separate database.
We linked Internet responses to the HIV test record via the test record number
alone; this method offered no linkage to information that could be used to
identify clients specifically. Each comparison among columns (in the tables
presented herein) was considered a separate family, and we used a familywise
error rate (α) of .05 to adjust for multiple comparisons.
Data were gathered from 856 clients of the DPHHCTS. Because clients
could skip any question, the number of respondents answering each question
varied. The study sample was largely white (77.8%), men (69.2%), and heterosexual
(65.3%) (Table 1). The analysis
of risk factors, presented in Table 2,
indicated that a substantial proportion of clients had a past history of STDs
(21.8% [185/850]). Overall, the sample was sexually active; 176 (23.5%) of
the 750 clients reporting had 7 or more partners in the last 12 months. The
sexual practice indicated by the most respondents was oral sex (88.7% [749/844]),
though vaginal sex (73.9% [621/840]) and anal sex (41.2% [345/838]) were also
reported. Of the 832 who responded, 335 (40.3%) reported having used a condom
during their last sexual encounter, while 13.1% (109/834) reported being under
the influence of drugs or alcohol during their last sexual encounter, and
16.8% (142/847) reported being sexually exposed to a person known to have
HIV infection; 53% of those reported using a condom during their last sexual
encounter (M.M., unpublished data, 2000).
Responses to Internet-usage questions appear in Table 3, with 135 (15.8%) clients reporting that they have logged
on to the Internet to seek sex partners, and 88 (65.2%) of these having successfully
initiated sexual contact; 34 (38.7%) of those who had sex with Internet partners
had met more than 3 partners using the Internet. Most contacts occurred in
the 6 months prior to the client's visit to the DPHHCTS, and condom use during
the last encounter with an Internet partner was reported by only 37 (44.0%)
of 84 clients. Of the 88 clients who reported initiating sexual contact via
the Internet, of 86 reporting, 78 (90.7%) used home computers and 5 clients
(5.8%) did so at work.
To compare the risk for STD/HIV in Internet sex-seekers with that of
clients who did not seek sex on the Internet, we divided the sample by responses
to the first 2 questions on our Internet-usage survey. Clients who had never
logged on to the Internet with the intention of seeking a sex partner were
called the offline group (n = 721), indicating that sex partners were not
sought using the Internet. If clients responded that they had logged on to
the Internet with the intention of seeking a sex partner, they were classified
as online seekers (n = 135). If online seekers reported having sex with partners
whom they had originally met on the Internet, then they became online partners
(n = 88). Online seekers who had not initiated sex contacts via the Internet
constituted the online-no-partner group (n = 47). The online partner and the
online-no-partner groups are combined to form the online seekers.
As shown in Table 4, there
were some similarities in age and in race between groups. However, online
seekers were more likely to be men than offline clients: 13 women reported
seeking sex partners on the Internet. Online seekers were more likely to be
homosexual than offline clients, and online partners were more likely to be
homosexual than the online-no-partner group.
As shown in Table 5, some
risk factors were more frequently reported by the online than the offline
group. Online seekers were more likely to have had a previous STD than the
offline clients, thus increasing their risk of acquiring future STDs such
as HIV.13- 15 Online
seekers had greater numbers of partners than offline clients but were more
likely to have used a condom during their last sex act. Oral and anal sex
were more frequently reported by online seekers, with vaginal sex less frequently
reported. Online seekers reported more sexual exposure to men and to men who
have sex with men than offline clients, and online partners were still more
likely to be exposed to men and to men who have sex with men than those in
the online-no-partner group. Only 14.5% of the offline clients reported sexual
exposure to a person known to be HIV-positive, while 28.9% of online seekers
reported this exposure. Of online partners, 35.2% had been sexually exposed
to a person known to be HIV positive. Though the sample size of HIV-positive
persons was small (n = 7), online seekers (2.22%) were about as likely as
offline clients (0.55%) to be HIV positive.
The sample of DPHHCTS clients is representative of neither Denver nor
the US population as a whole. However, among this sample, the Internet clearly
has had a role in the solicitation of risky sex partners. Clients who reported
seeking sex on the Internet were more likely to have concomitant risk factors
for STD/HIV than clients who did not seek sex on the Internet. Thus, seeking
sex on the Internet may be a potential risk factor for STD/HIV. These data
underscore the need for development of strategies to promote STD/HIV prevention
among online sex seekers.