Do No Harm: Lessons Learned About Contextualizing Your Findings

Background: Post marketing surveillance of adverse events following vaccination is necessary to monitor vaccine safety. We published a peer-reviewed paper on adverse events following HPV vaccination that included data on emergency department (ED) visits or hospitalizations within 42 days of vaccination. Our findings reported that of the 195,270 women that received 528,913 doses HPV of vaccine, 958 were hospitalized and 19,351 had an ED visit within 42 days of vaccination. We concluded that rates of adverse events after HPV vaccination in Alberta are low and consistent with types of events seen elsewhere. After publication all members of the research team received unusual emails and interview requests.

Methods: We describe the sequence of events, hypotheses of the cause of these events, and actions taken.

Results: The paper was published online February 26, 2016 with the first inquiry from a Canadian journalist on February 29. In April, two “independent” journalists emailed each research team member requesting an interview. Both emails used similar wording implying that ED utilization was high. By the end of April, a UK tabloid journalist requested an interview and two academic researchers emailed requesting data. The article scored in the 97^th percentile of article of the same age (60 days) and source for comments/responses: 95 tweets from nine countries (88 from the public, 5 from science community and 2 from bloggers). Most of the social media comments noted that “9.9% ended up in emergency rooms in just 42 days” and attributed this directly to vaccination.

We hypothesize that two factors this high rate of social media commentary: the title of the paper, and the lack of contextualization of the health service utilization. The unusual attention necessitated a plan and comparative data to contextualize the reported hospitalization data. University of Calgary Media Relations assisted us to identify a contact person, standard responses and a communications plan.

Conclusion:Our experience demonstrates the importance of contextualizing results to minimize the misinterpretation of study findings. Social media may be used to disseminate misinterpretations of research findings and researchers should prepare accordingly.