Posted 09 January 2018
By Michael Mezher
A new study looking at the US Food and Drug Administration's (FDA) use of social media to communicate about drug safety finds that the agency could improve its impact by developing social media strategies and taking a more active role on web platforms.
The study, conducted by researchers at Harvard Medical School, Northeastern University, Boston Children's Hospital and The Ohio State University, and funded by FDA's Center for Drug Evaluation and Research, looks specifically at FDA's use of social media following two drug safety communications (DSCs) for drugs containing zolpidem, the active ingredient in Ambien.
In 2013, FDA issued two DSCs related to zolpidem, the first in January warning of the risk of next-day impairment after taking insomnia drugs and requiring lower recommended doses for drugs containing zolpidem, and the second notifying the public that the agency had approved those labeling changes.
In addition to publishing the DSCs to its website, FDA made posts related to the DSCs on its various social media accounts, including its Facebook page and three of its Twitter accounts (@US_FDA, @FDA_Drug_Info and @FDAMedWatch).
For the first DSC, FDA posted once to its Facebook account, getting 61 shares, and tweeted six times across its three accounts, for a total of 111 retweets.
To measure the impact of FDA's posts, the authors identified some 174,000 tweets and 59,000 Facebook posts referencing zolpidem between 1 October 2012 and 31 August 2013. Of those, roughly 9% of the tweets were tagged as adverse events (AEs), 74% were tagged as mentions and 17% were considered junk. For the Facebook posts, 5% were identified as AEs, 69% as mentions and 26% as junk.
However, for the second DSC, FDA did not post to Facebook and only tweeted about the DSC for its @FDA_Drug_Info account (3 tweets, 37 retweets), though the agency also tweeted generally about a group of recent prescribing changes, including for zolpidem, from its @FDAMedWatch account.
The authors observed a greater social media response from both Twitter and Facebook users following the first DSC, but did not notice an increase in social activity related to zolpidem following the second communication.
The authors say that the difference in social media response to the two safety communications can be explained by the difference in how FDA issued the communications. In addition to posting about the first DSC more frequently and to more of its accounts, the first DSC was also issued with an accompanying press release.
To make better use of its social media presence in the future, the authors recommend that FDA develop strategies for how and how often to post different types of messages.
"Since there can be a lot of questionable information shared on social media, the FDA should be able to leverage its independent expertise and position as a widely trusted source of information to help promote accurate and informative messages. In the case of the zolpidem safety alerts, the FDA took some steps to do that, but could create more outreach and disseminate its materials more frequently via varied platforms," Michael Sinha, a postdoctoral fellow at Harvard and lead author on the paper told Focus.
Sinha also emphasized that the social media landscape is substantially different now than in 2013 when these safety communications were issued.
"Social media like Twitter was seen as more of a recreational tool in 2013, but now many professionals, academic institutions and companies have vibrant social media presences. Increased health professional engagement on social media increases the likelihood that FDA content on these platforms would be more widely disseminated and have a greater impact," he said.
The paper also recommends that FDA develop a greater understanding of social media platform features and user preferences to develop its approach to posting.
"FDA should try to stay on top of the ever-evolving strategies being developed on social media to amplify messages—such as use of handles or hashtags on Twitter—that can bring attention to important drug safety content disseminated by the agency," Sinha said.
The authors also suggest that FDA should take matters into its own hands when it comes to updating information about drugs on Wikipedia.
Despite multiple edits to the Wikipedia page corresponding to both communications, the pages contained incomplete information and did not cite the DSCs themselves.
"Given that informational sites like Wikipedia are commonly accessed by the lay public for information on drugs and that anyone can edit the content, the FDA should consider a plan to formally update the pages for appropriate content at the time a DSC is released and to ensure continued accuracy of the information over time," the authors write.
Editor's note: this article was updated with comment from one of the paper's authors.