Social Media for HIV prevention: More research needed

The potential for Social Media (#SoMe) to contribute to HIV prevention is worth exploring and researching but studies need to be conducted with strict criteria to discern the evidence from the hype in such a fuzzy area.

A study published in the Annals of Internal Medicine is gathering some momentum (read here and here) for allegedly demonstrating the power of the Internet and Social Media like Facebook as potential or “serious” HIV prevention medium, depending of the reporting outlet. Though the results of this study are somewhat encouraging, it is worth looking at it more closely.

The research

The team of researchers led by Thomas Coates from UCLA recruited 112 MSM based in Los Angeles, more than 85% of whom were African American or Latino and randomly allocated them to an intervention or control study groups. In each group the intervention was led  by specially trained peer leaders who communicated with participants on Facebook by sending messages, chats, and wall posts.

In the intervention group peer leaders communicated with participants about HIV prevention and testing. In the control group they communicated about the importance of exercising, healthy eating, and maintaining a low-stress lifestyle. Every 4 weeks, participants were told that they could ask for a free, home-based testing kit.

The study lasted 12 weeks during which participants could request one free home-based HIV testing kit. A survey was conducted at baseline and 12 weeks later.

Limitations and caveats

Before going into the limitations and caveats of this study, it is worth noting that it draws heavily on work also worth reading previous published by the same group earlier this year.

The researchers identified a number of limitations in their study such as possible bias in the recruitment, the sole use of Facebook as Social Media (hence the impossibility to generalise the results to all Social Media or platform) but also the absence of known best practice in the field; it many have been useful starting here (See this). The intervention was also limited to 12 weeks, and there are evidence that to be sustainable this type of interventions required to be sustained in time.

The study was initially powered with an 80% chance to detect a modest between-group difference in HIV testing of 16 percentage points or more, which was not particularly ambitious.  For that purpose it needed 7 clusters of 25 participants. However “Fiscal constraints required [the researchers] to scale back the number of clusters to 2”, making the study statistically worthless.

The main observation reported by the researchers was that more participants in the HIV intervention group requested an HIV testing kit than in the control participants (25 of 57 [44%] vs. 11 of 55 [20%]; mean difference, 24 percentage points [95% CI, 8 to 41 percentage points]).

The choice of  general health as discussion topic in the control group is open to question. If the purpose of the study was to show that social media can be used to deliver health messages, whether they are about HIV or something else, it is adding little to a large body of research. If it was to show that Social Media are able to do better or more or something different than traditional interventions, this study provides limited information.

Indeed, imagine voluntering to enrol in a cooking class not knowing if you will attend either a general cooking class or a Japanese cooking class. Now, every 4 weeks, you are offered a selection of cooking books, including a Wagamama Cooking book. What do you think would happen? What will be the uptake of the Japanese cooking book in each cooking class? Is a randomised clinical trial needed?

There are a number of other gaps that this study did not fill. For example did the intervention allow for reaching people who otherwise would not have been reached? This is critical in an epidemic who is driven by people who do not know their HIV status and may be hard to reach. What were the results of the HIV tests? Did the intervention allow for detecting more infections than a classic intervention such as community based would have? Isn’t it what it should have been aiming at testing? And what did the participants enrolled in the control group learned?

Talking about it!

There was no other major differences between the two groups with one exception that the researchers did not really explore: there was a greater frequency of chatting and sending of personal messages in the intervention group than in the control group.

This is an important piece of observation that warrant further research but also confirms what many activists working with real people know: People want to know and talk about HIV and they are prepared to engage with you if you are prepared to discuss the issues. What most people don’t want is a one-way conversation and being told about HIV. This is something I feel strongly about and not only because it is a reflexion based on personal experience.

In search of deeper answers

This approach to research illustrates a trend in HIV prevention research to conduct studies for the sake of conducting studies without consideration for the bigger picture and for asking the questions whose answers really matter, the first two being “why are we doing this research?” and “how to do it best?”

Social Media are not everywhere and everything as advertisers, corporations and even the non-profit sector tell you to believe. It is undeniable that Social Media’s power and influence, even if artificial, can’t be ignored and this is why ethical, meaningful, well designed (i.e. representation and sample size!) and properly funded studies are worth conducting.

What is the potential of Social Networks for HIV prevention? What can prevention through Social Networks achieve that other interventions cannot? It is possible to do better and how to do it ? Time and again I go back to an old motto: First we made a t-shirt, then a CD, followed by a DVD, now it is a YouTube video and a Facebook page. But does any of that make a difference and more importantly a sustainable and noticeable difference?

Some NGOs have understood that it does not. And that is why more research is needed so as not squander meagre prevention funds into the latest fads (think about the gamification of HIV prevention). Because let’s face it, if MySpace was still what it used to be, that research would have been done there. But what happened to MySpace, and what could happen to Facebook and Social Media in general?

What do the virtual reality of Social Networks means and represents? What projection of ourselves are we at risk to confound with the real world? These are the deeper questions for which we need some elements of answer before jumping on a fast moving and uncertain bandwagon.


Switch to our mobile site