X

The Health Sciences Blog covers the latest trends and advances in life sciences and healthcare.

Social Media Data Mining for Clinical Trial Safety Insight - Part 1 in a Series

Sameer Thapar
Director of Global Pharmacovigilance

Today we have defined Big Data as encompassing the parameters of social media, that deluge of tweets, Facebook messages, forum posts, videos/vlogs, blogs, with the ever fine tuning of modalities, Pinterest, Snapchat, and Instagram posts.  However, while we now know where to get the information, we still do not have a clear, defined set of rules on what to do with the information in a pharmacovigilance, and broader, life sciences context.

Early Social Media for Life Sciences

In 2007, I was eager to present at a newly christened, “Social Media Summit” and several similar conferences. We were all at the infancy of this new media. I had begun to champion this new media, as it was labeled, as another repository for life sciences outreach and understanding of trends, if not meaningful data. The digital life sciences community was still skeptical of this information, citing lack of guidance on how to deal with it, and a general disappointment in the statistical strength of insights to be gleaned.  Some of the presentations I led with at the time carried shiny, attention-grabbing titles such as, “Energizing the Drug Safety Practice with the Use of New Media” and “Mitigating the Current Drug Safety Spotlight via New Media/Social Media Initiatives” or the even the later pharmacovigilance branded presentations entitled, “Casting the Net on Social Media: Pharmacovigilance Mining of the Web”.

Social Media's Role in Today's Clinical Research

Much has changed in this past decade dealing with social media. First, we know now that we cannot ignore it. Doing so just procrastinates the need for a good internal policy to incorporate its strengths into our common lexical resources. Second, we now realize that it is not simply a matter of putting resources at the forefront. No, it is much more involved.  It requires assistance from Bots, Natural Language Processing (NLP) toolkits, and general machine learning approaches to gather the relevant articles of importance from the growing amount  of social media data.  Mining social media is the topic du jour, and there are plenty of approaches on how to conduct this activity.

Cast a search across the Internet on social media mining in life sciences or the narrower, pharmacovigilance use of social media, and you will obtain plenty of results. All these social media results are based on something that should be part of a pharmaceutical organizational game plan – a way to interact with its constituent, the general consumer. The past decade has provided many examples of “good customer engagement” advice that was ignored, or inadvertently marginalized, and the reputation of the company took a serious media hit.

So, short of ignoring this medium, what actions should a growing pharmacovigilance department take with regard to social media mining, and how much importance (aka resources) should be placed in this activity? The answer is not an absolute. It is very interdependent on which of the following areas means the most to the organization: regulatory obligations/compliance, adverse event/misinformation engagement, and market awareness/brand strength.

 

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.