AMC, Academic Medical Center; aOR, adjusted odds ratio; CI, confidence interval; CINIMA, Center for Infection and Immunology Amsterdam; DAG, directed acyclic graph; HIV, human immunodeficiency virus; i.e., id est, it is, for example; IQR, interquartile range; MEC, Medical Ethics Committee; MSM, men who have sex with men; OR, odds ratio; RIVM, National Institute of Public Health and the Environment, Centre for Infectious Disease Control; STI, sexually transmitted infection; UAI, unprotected anal intercourse; UMCU, University Medical Center Utrecht
New research should stay up to date as it pertains to fast shifting dating methods as well as sero-adaptive behaviours (like viral sorting and pre exposure prophylaxis). With every new way of dating and preventive chances, the rules of engagements will be different. Our data are 8years old and internet-based dating has developed since then. Nevertheless these results are useful, as they demonstrate how internet-based partner acquisition can result in more information on the sex partner, and this might affect on the frequency of UAI.
Relationship online may offer other opportunities for communicating on HIV status than dating in physical surroundings. Localsex nearest Collingwood, Victoria. Localsex Near Me South Melbourne Victoria. Facilitating more online HIV status disclosure during partner seeking makes serosorting easier. Nonetheless, serosorting may raise the weight of other STI and will not prevent HIV infection entirely. Interventions to prevent HIV transmission should notably be directed at HIV negative and oblivious MSM and excite timely HIV testing (i.e., after risk events or when experiencing symptoms of seroconversion illness) as well as routine testing when sexually active.
Because determinations on UAI seem to be partially based on perceived HIV concordance, precise knowledge of one's own and the partner's HIV status is important. In HIV negative men and HIV status-unaware men, determinations on UAI will not only be based on perceived HIV status of the partner but also on one's own negative status. HIV serosorting is challenged by the frequency of HIV testing and also the HIV window period during which individuals can transmit HIV but cannot be diagnosed with the commonly used HIV tests. Hence serosorting can't be regarded as a very successful way of preventing HIV transmission 22 Besides interventions to stimulate the uptake of HIV and STI testing in sexually active men, interventions to caution against UAI based on sensed HIV-negative concordant status are in order, irrespective of whether this concerns online or offline dating.
For HIV-unaware guys the effect of dating place on UAI did not change by adding partner characteristics, but it increased when adding lifestyle and drug use. It is difficult to evaluate the real risk for HIV for these guys: do they behave as HIV-negative guys that are attempting to protect themselves from HIV infection, or as HIV positive men trying to safeguard their HIV negative partner from HIV infection? A study by Horvath et al. reported that 72% of guys who were never tested for HIV, profiled themselves online as being HIV-negative, which might be problematic if they are HIV-positive and engage in UAI with HIV negative partners 12 Previously Matser et al. reported that 1.7% of the unaware and perceived HIV-negative MSM were tested HIV positive. The study population included the MSM reported in this study 15
Online dating wasn't connected with UAI among HIV negative guys, a finding in agreement with some previous studies, mostly among young men 21 , but in comparison with other studies 1 - 5 This may be due to the fact that most earlier studies compared sexual behaviour of two groups of MSM rather than comparing two sexual behavior patterns within one group of guys. Nonetheless it could also reflect secular changes; perhaps in the beginning of online dating a more high risk group of men used the Internet, and over time online dating normalized and less high-risk MSM nowadays additionally make use of the Net for dating.
A vital strength of the study was that it explored the relationship between online dating and UAI among MSM who had recent sexual contact with both online and also offline casual partners. This avoided prejudice caused by potential differences between guys just dating online and those simply dating offline, a weakness of several previous studies. By recruiting participants at the largest STI outpatient clinic in the Netherlands we could include a high number of MSM, and prevent potential differences in guys sampled through Internet or face-to-face interviewing, weaknesses in a few previous studies 3 , 11
Among HIV positive guys, in univariate analysis UAI was reported significantly more frequently with online associates than with offline associates. When correcting for partner characteristics, the effect of online/offline dating on UAI among HIV positive MSM became somewhat smaller and became non-significant; this indicates that differences in partnership factors between online and offline partnerships are responsible for the increased UAI in online established partnerships. This could be because of a mediating effect of more information on associates, (including perceived HIV status) on UAI, or to other variables. Among HIV negative guys no effect of online dating on UAI was detected, either in univariate or in the multivariate models. Localsex near Collingwood Victoria. Among HIV-unaware guys, online dating was correlated with UAI but only important when adding partner and venture variants to the model.
In this large study among MSM attending the STI clinic in Amsterdam, we found no evidence that online dating was independently related to a higher risk of UAI than offline dating. Localsex closest to Collingwood Victoria. For HIV negative men this lack of assocation was clear (aOR = 0.94 95 % CI 0.59-1.48); among HIV positive guys there was a nonsignificant association between online dating and UAI (aOR = 1.62 95 % CI 0.96-2.72). Only among guys who indicated they weren't aware of their HIV status (a little group in this study), UAI was more common with on-line than offline partners.
The number of sex partners in the preceding 6months of the index was likewise associated with UAI (OR = 6.79 95 % CI 2.86-16.13 for those with 50 or more recent sex partners compared to those with fewer than 5 recent sex partners). UAI was significantly more likely if more sex acts had occurred in the venture (OR = 16.29 95 % CI 7.07-37.52 for >10 sex acts within the venture compared to only one sex act). Other factors significantly associated with UAI were group sex within the partnership, and sex-connected multiple drug use within partnership.
In multivariate model 3 (Tables 4 and 5 ), also including variants concerning sexual behaviour in the partnership (sex-associated multiple drug use, sex frequency and partner type), the independent effect of online dating location on UAI became somewhat more powerful (though not essential) for the HIV positive men (aOR = 1.62 95 % CI; 0.96-2.72), but remained similar for HIV-negative men (aOR = 0.94 95 % CI 0.59-1.48). The effect of online dating on UAI became more powerful (and critical) for HIV-oblivious men (aOR = 2.55 95 % CI 1.11-5.86) (Table 5 ).
In univariate analysis, UAI was significantly more inclined to happen in online than in offline ventures (OR = 1.36 95 % CI 1.03-1.81) (Table 4 ). The self-perceived HIV status of the participant was strongly correlated with UAI (OR = 11.70 95 % CI 7.40-18.45). The result of dating place on UAI differed by HIV status, as can be seen best in Table 5 Table 5 shows the organization of online dating using three different reference groups, one for each HIV status. Among HIV positive guys, UAI was more common in online compared to offline ventures (OR = 1.61 95 % CI 1.03-2.50). Among HIV-negative men no association was apparent between UAI and internet ventures (OR = 1.07 95 % CI 0.71-1.62). Among HIV-unaware men, UAI was more common in online when compared with offline partnerships, though not statistically significant (OR = 1.65 95 % CI 0.79-3.44).
Features of online and offline partners and partnerships are revealed in Table 2 The median age of the partners was 34years (IQR 28-40). Compared to offline partners, more on-line partners were Dutch (61.3% vs. 54.0%; P 0.001) and were defined as a known partner (77.7% vs. 54.4%; P 0.001). The HIV status of on-line partners was more frequently reported as known (61.4% vs. 49.4%; P 0.001), and in on-line partnerships, perceived HIV concordance was higher (49.0% vs. 39.8%; P 0.001). Participants reported that their on-line partners more often knew the HIV status of the participant than offline partners (38.8% vs. 27.2%; P 0.001). Participants more frequently reported multiple sexual contacts with internet partners (50.9% vs. 41.3%; P 0.001). Sex-associated material use, alcohol use, and group sex were less frequently reported with internet partners.
In order to analyze the potential mediating effect of more info on partners (including perceived HIV status) on UAI, we developed three variant models. In version 1, we adapted the association between online/offline dating location and UAI for features of the participant: age, ethnicity, number of sex partners in the preceding 6months, and self-perceived HIV status. In model 2 we added the venture characteristics (age difference, ethnic concordance, lifestyle concordance, and HIV concordance). In model 3, we adjusted additionally for venture sexual risk behavior (i.e., sex-associated drug use and sex frequency) and venture type (i.e., casual or anonymous). As we assumed a differential effect of dating place for HIV-positive, HIV-negative and HIV status unknown MSM, an interaction between HIV status of the participant and dating location was included in all three models by making a fresh six-category variable. For clarity, the effects of online/offline dating on UAI are also presented separately for HIV-negative, HIV positive, and HIV-unaware men. We performed a sensitivity analysis restricted to partnerships in which just one sexual contact occurred. Statistical significance was defined as P 0.05. No adjustments for multiple comparisons were made, in order not to lose potentially significant organizations. As a rather big number of statistical tests were done and reported, this strategy does lead to a higher danger of one or more false positive associations. Evaluations were done using the statistical programme STATA, version 13 (STATA Intercooled, College Station, TX, USA).
Prior to the evaluations we developed a directed acyclic graph (DAG) representing a causal model of UAI. In this model some variants were putative causes (self-reported HIV status; on-line partner acquisition), others were considered as confounders (participants' age, participants' ethnicity, and no. Collingwood Localsex. of male sex partners in preceding 6months), and some were supposed to be on the causal pathway between the principal exposure of interest and results (age difference between participant and partner; ethnic concordance; concordance in life styles; HIV concordance; venture type; sex frequency within partnership; group sex with partner; sex-related substance use in venture).
We compared characteristics of participants by self-reported HIV status (using 2-tests for dichotomous and categorical variables and using rank sum test for continuous variables). Localsex Near Me Glen Huntly Victoria. We compared characteristics of participants, partners, and venture sexual conduct by online or offline venture, and computed P values based on logistic regression with robust standard errors, accounting for linked data. Continuous variables (i.e., age, number of sex partners) are reported as medians with an interquartile range (IQR), and were categorised for inclusion in multivariate models. Random effects logistic regression models were used to analyze the association between dating place (online versus offline) and UAI. Likelihood ratio tests were used to gauge the importance of a variable in a model.
As a way to investigate possible disclosure of HIV status we additionally asked the participant whether the casual sex partner understood the HIV status of the participant, with the reply options: (1) no, (2) possibly, (3) yes. Sexual conduct with each partner was dichotomised as: (1) no anal intercourse or merely protected anal intercourse, and (2) unprotected anal intercourse. To determine the subculture, we asked whether the participant characterised himself or his partners as belonging to one or more of the subsequent subcultures/lifestyles: casual, formal, substitute, drag, leather, military, sports, trendy, punk/skinhead, rubber/lycra, gothic, bear, jeans, skater, or, if not one of these characteristics were applicable, other. Concordant lifestyle was categorised as: (1) concordant; (2) discordant. Chance partner sort was categorised by the participants into (1) known traceable and (2) anonymous partners.
HIV status of the participant was got by asking the question 'Do you know whether you're HIV infected?', with five response options: (1) I 'm definitely not HIV-infected; (2) I think that I'm not HIV-infected; (3) I do not know; (4) I believe I may be HIV-contaminated; (5) I know for sure that I am HIV-infected. We categorised this into HIV-negative (1,2), unknown (3), and HIV-positive (4,5) status. The survey enquired about the HIV status of every sex partner with the question: 'Do you know whether this partner is HIV-contaminated?' with similar answer options as above. Perceived concordance in HIV status within partnerships was categorised as; (1) concordant; (2) discordant; (3) unknown. Localsex in Collingwood, VIC. The last category represents all partnerships where the participant did not understand his own status, or the status of his partner, or both. In this study the HIV status of the participant is self-reported and self-perceived. The HIV status of the sexual partner is as perceived by the participant.