Twitter is rich with adverse drug reaction data, but it’s hard to parse

By: Jonah Comstock | Apr 30, 2014        

Tags: | | | | |  |

twitter graphA team of researchers has found that, for 23 commonly used drugs, well over three times as many adverse drug reactions were reported on Twitter in a given time period than were reported to the FDA. Of course, the study also notes that the tweeted reports, referred to as “Proto AEs” (adverse events) weren’t necessarily as specific or useful as FDA adverse event reports.

Over a six-month period, researchers collected 6.9 million tweets by searching public Twitter for keywords related to the drugs, then randomly chose 61,000 of those tweets to manually analyze. Out of those, 4,400 were identified as adverse events. In the same time period, there were just 1,400 events related to those drugs reported to the FDA through normal channels. Interestingly, researchers said some Twitter behavior suggested that users meant to report this data, not just to discuss it with others socially.

“Further, there was evidence that patients intend to passively report AEs in social media, as evidenced by hashtags and mentions such as #accutaneprobz and @EssureProblems,” the study authors write. “Even within 140 characters, some tweets demonstrate an understanding of basic concepts of causation in drug safety, such as alleviation of the AE after discontinuation of the drug: ‘I found that Savella made my blood sugar [sic] high, once I got off, my blood sugar returned to normal.'”

The researchers, including scientists from Boston University, Boston Children’s Hospital, Harvard Medical School, and Georgetown University, stressed that the main limitation for collecting adverse event data on Twitter is that the tweets were too varied and complex to be accurately assessed by a computer. Some tweets generated false positives such as ““I got the flu shot yet somehow I’ve gotten the stomach flu twice in this month”, which inaccurately equates gastroenteritis (“stomach flu”) with influenza.

“While it may be feasible to review all social media posts with Proto-AEs for lower volume drugs, there is likely to be a point at which the volume of posts, say for a widely used and established medicine, may overwhelm the capacity for human review, requiring further automated analysis,” the researchers wrote. “We also stress that, at this time, we do not recommend the wholesale import of individual social media posts into post-marketing safety databases. Rather, in parallel with other post-marketing sources, these data should be considered for idea generation, and reasonable hypotheses followed up with formal epidemiologic studies.”

Pharmaceutical companies have been called out lately for their lack of engagement with social media, notably in a recent study from IMS Health. That report noted that until FDA guidance is made clearer, social media engagement is a challenge for pharma companies partly because of adverse event reporting. Even by having a Twitter account, for instance, the companies might be responsible for reporting any adverse reaction a user tweets at them.

“For an industry that is used to having a clear framework from regulators, this is a disconcertingly vague environment and has made social media strategies challenging for the pharmaceutical industry,” the IMS report said. “There is the concern that investment taken now in the area could be wasted should the current state of affairs change, or worse, that companies may find themselves liable for damages should new legislation be applied retroactively.”