A Characterization of Internet Dating Network Structures among Nordic Men Who Have Sex with Men Antonella Villani 1 , Arnoldo Frigessi 2 , Fredrik Liljeros 3,4 , Monica K. Nordvik 3,5 , Birgitte Freiesleben de Blasio 2,6 * 1 Department of Mathematics, University of L’Aquila, L’Aquila, Italy, 2 Department of Biostatistics, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway, 3 Department of Sociology, Stockholm University, Stockholm, Sweden, 4 Institute for Future Studies, Stockholm, Sweden, 5 Department of Social Work, Mid Sweden University, O ¨ stersund, Sweden, 6 Department of Infectious Disease Epidemiology, Division of Infectious Disease Control, Norwegian Institute of Public Health, Oslo, Norway Abstract Background: The Internet has become an important venue for seeking sexual partners and may facilitate transmission of sexually transmitted infections. Methods: We examined a 64-day data log of flirt messages expressing sexual interest among MSM within the Qruiser.com community. We used logistic regression to analyze characteristics of MSM sending and receiving flirt messages and negative binomial regression to examine individual activity and popularity. The structural properties, including the core structure of the flirt network, were analyzed. Results: The MSM population consisted of approximately 40% homosexuals and 37% bisexuals, while the remaining 23% included men who identified as heterosexual but searched for sex with men and ‘‘experimental’’. MSM were more likely to send flirt messages if they were homosexual and aged 40+ years; young people aged , 30 years were more likely to receive a flirt. Possession of a webcam was strongly associated with both sending flirt messages and being a flirt target. The distributions of flirts sent (max k out = 2162) and received (max k in = 84) were highly heterogeneous. Members in central cores were more likely homosexuals, singles, and aged 31–40 years. The probability of a matched flirt (flirt returned from target) increased from 1% in the outer core to 18% in the central core (core size = 4). Discussion: The flirt network showed high degree heterogeneity similar to the structural properties of real sexual contact networks with a single central core. Further studies are needed to explore use of webcam for Internet dating. Citation: Villani A, Frigessi A, Liljeros F, Nordvik MK, de Blasio BF (2012) A Characterization of Internet Dating Network Structures among Nordic Men Who Have Sex with Men. PLoS ONE 7(7): e39717. doi:10.1371/journal.pone.0039717 Editor: Vittoria Colizza, INSERM & Universite Pierre et Marie Curie, France Received September 20, 2011; Accepted May 25, 2012; Published July 13, 2012 Copyright: ß 2012 Villani et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: Birgitte Freiesleben de Blasio was supported by the Norwegian Research Council (Projects 166056/V50). Fredrik Liljeros was supported by Riksbankens Jubileumsfond (The Bank of Sweden Tercentenary Foundation). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There are no external funding sources for this study. Competing Interests: The authors have declared that no competing interests exist. * E-mail: birgitte.deblasio@basalmed.uio.no Introduction In recent years, the internet has become a major driver for social interactions and has made it possible for people to communicate quickly and over long distances. Alongside this development, internet dating has become increasingly popular and a socially acceptable way to meet partners for dates and relationships [1]. In 2000 an outbreak of syphilis was traced linked to men who have sex with men (MSM) chat room participants [2], and this has fueled speculation that the internet may function as a risk-promoting environment for spread of sexually transmitted infections [3]. However, recent findings seem to suggest that seeking sexual partners online is a marker of high- risk sexual behavior, whereas meeting online partners does not in itself promote high-risk behaviors [4,5]. Online dating is different from traditional ways of meeting sex partners, e.g. frequenting bars, due to the larger user base, the greater anonymity and flexible self-presentation options [6], and dating people need not be online at the same time. Furthermore, the internet has the ability to transcend social and geographic barriers, which may have an impact on sexual contact networks, and influence disease transmission [7]. Studies of sexual contact networks show a large variation in numbers of contacts [8] and indicate that the structural properties of sexual contact networks are crucial to the spread of STIs as highly sexually active individuals are responsible for a dispropor- tionate number of transmission events. Highly sexually active individuals are more likely to become infected, and when infected they can spread infection more effectively. The spread of infection can be further enhanced if individuals with many contacts most frequently mix with each other (assortative mixing) [9]. Such individuals with high turnover rates of partners and a high internal contact may form one or several cores separated by less dense regions in the network which may permit persistence and spread of STIs which are not able to reproduce themselves in the general population. A study on network models for STI transmission PLoS ONE | www.plosone.org 1 July 2012 | Volume 7 | Issue 7 | e39717