Simple outreach on dating websites increases HIV testing rate

A simple online outreach intervention – in which a health educator engaged in conversation and responded to questions about HIV testing on dating and hook-up websites for men who have sex with men – significantly increased the HIV testing rate, according to data from a randomised controlled trial published in Clinical Infectious Diseases.

These kinds of interventions are widely implemented, but until now there has been little data to show whether they are effective or not. This is the first randomised trial of such an intervention.

The intervention was developed in partnership with community organisations and involved a health educator having a presence in chat rooms on Adam4Adam, BlackGayChat, Craigslist and



Confounding exists if the true association between one factor (Factor A) and an outcome is obscured because there is a second factor (Factor B) which is associated with both Factor A and the outcome. Confounding is often a problem in observational studies when the characteristics of people in one group differ from the characteristics of people in another group. When confounding factors are known they can be measured and controlled for (see ‘multivariable analysis’), but some confounding factors are likely to be unknown or unmeasured. This can lead to biased results. Confounding is not usually a problem in randomised controlled trials. 

On each site, the health educator created a profile and publicised his availability to answer questions about HIV testing, including providing details of local services and of testing opportunities in bars and other community spaces. He could answer questions about transmission risks, symptoms and the testing process. Men using the sites could choose to begin a conversation by sending the health educator a private message.

“The intervention was based on empowerment education, social cognitive theory, and natural helping,” say the authors. It was more fully described in a previous publication. The health educator was skilled in building relationships with social media users, maintaining boundaries, being patient, having a sense of humour, and networking with other organisations, particularly those offering HIV testing.

The intervention was provided over a period of a year, only between 9am and 5pm, Monday to Friday. These aren’t the busiest moments for hook-up websites, but many health organisations only provide such interventions during office hours.

The study was conducted in four geographically distinct communities in the United States, in 2013 and 2014. In two of the communities, on the four websites, the intervention was provided. In the two control communities, no interventions were provided. Because each community was at least 200 miles away from the other ones and the sites are used to arrange face-to-face hook-ups, the researchers judged that people in control communities would be unlikely to be exposed to the intervention.

A cross-sectional survey was done with users on each site in each community before the intervention was delivered, with the survey repeated after the intervention period. Numbers completing each survey were quite small, with approximately 150 people taking part in each community in each survey. Overall, 1292 men took part in all surveys.

Most participants were in their thirties or forties; three-quarters were white; one third reported recent sex with a woman as well as a man.

Before the intervention was provided, there was no difference in HIV testing rates between the communities – around 36% had tested for HIV within the previous year.

After the intervention, 64% of those in intervention communities had tested in the previous year, compared with 42% of those in control communities. After adjustment for confounding factors, those in intervention communities were three times more likely to have tested (odds ratio 2.9, 95% confidence interval: 1.8-4.7).

The researchers note some features of the intervention – it is simple and cheap to deliver, requiring only a trained health educator and an internet connection; it can easily be replicated and delivered at a greater scale; it reaches men when they may already be thinking about sex; it reaches men who may not use gay bars, including bisexual men.

This randomised study examined an intervention delivered on websites. In a separate study the same researchers have also demonstrated that it is feasible and acceptable to provide a similar intervention on geosocial networking apps such as Grindr, A4A Radar, Jack’d and Scruff.