Are Social Networks an Effective Channel for Disseminating Medical Information?

By November 30, 2012

While social networks might be seen as a potentially viable channel for passing on medical information, one should not count on them as a reliable means of targeted dissemination as there are many factors impacting the spread of information through networks.

If you want to study how a disease spreads through a population, onlinenetworks such as Twitter provide a fairly good picture of how this viral effect works. So could social media now be used not just for studying how diseases propagate, but also for sending out advice on disease prevention or an alert to the population? A group of researchers* has recently published a study** which maps observed dissemination of preventive medical advice on Campylobacter, a bacteria which causes food-borne disease, on to two types of human networkmodels: the  Erdös and Rényi random model, which assumes that everyone has the same random probability of contaminating another person, independently of any diseases they have contracted before, and the Small World model(SW) which starts from the hypothesis that human society is a network in which everyone is connected to everyone else by a series of short path-lengths. The study found that using social networks – in the broad sense of the word– is too unreliable a method for disseminating vital information.

A propitious environment for spreading information

The researchers found that effective transmission of information depends on both quantity – i.e. the number of network connections; and quality –  how frequent, close and intense a given connection is, and also on having a sufficient proportion of heterogeneous people in the network. The team identify two major factors which have an impact on whether this type of medical advice percolates through. The first is endogenous; a person who is directly (him/herself) or indirectly (through someone close) affected by the disease in question will of course have a greater propensity to transmit information and advice to his/her network or the person affected, and to be receptive to such information. The second factor is exogenous, i.e. it has to do with the way in which people are connected to each other. The researchers also list a number of aspects of the social ties which affect the process.  

Word of mouth remains a powerful channel

So, the way in which information percolates through existing links plus the number of connections have an impact – a logical finding since people who have a greater number of social ties will have more opportunities to transmit or receive information. But the qualitative aspect of these ties is equally important in its impact on the spread of information.  In terms of the dynamics of information flows, there are two key factors. If too few people receive the information in the initial dissemination phase, dispersion can stop dead before the viral effect kicks in and it reaches the entire population.  Alternatively, if a network becomes saturated, leaving too little heterogeneity to stimulate further spread, this will also dampen transmission, the researchers point out. This seems to be the case in the Small World model, under which the total maximum number of recipients of the information was put at around 20% of the population. However, the researchers conclude that this 2-step model is ‘flawed’ as it fails to take account of all the steps that contribute to widespread dissemination.

 * Charanpal DhanjalSandrine BlanchemancheStéphan ClémençonAkos Rona-Tas, and Fabrice Rossi

** Dissemination of Health Information within Social Networks

Legal mentions © L’Atelier BNP Paribas