Pfizer: How Facebook can ‘unblind’ a clinical trial
Craig Lipset said internet-literate patients are discussing their treatment on blogs and in internet forums. The results can be a “false spike” in reports of safety events once participants hear others’ described side effects, he said.
Some trial volunteers use internet forums to try to determine which arm of a trial they have been assigned to, compromising the blindness of the research. In other cases, patients have “coached” each other online on how to appear eligible for a trial, said Pfizer’s expert.
‘The genie cannot be put back in the bottle’
The phenomenon affects data integrity, as well as potentially compromising safety. But there is no easy fix to the problem, Lipset told Outsourcing-Pharma.com. A patient who is active online before a trial begins will probably continue to gather information via the internet during a trial; “the genie cannot be put back in the bottle.”
While trial researchers must counsel patients at the start of trials about social media use, the burden for maintaining study integrity falls on sponsors, investigators and regulators, he told us.
Pharmaceutical companies must work with regulators to find ways to monitor social media use by participants, Lipset said. However, he warned against a blanket ban which would expel patients who contravene an internet policy. Excluding participants “would likely create anxiety for patients, be challenging to enforce, and brings the potential to influence or bias the study population.”
Researchers: not immune
Just as trials can be affected by what patients read on the internet, researchers also risk violating their objectivity when they venture online, Lipset told us.
Researchers who read participants’ online conversations may pick up on talk of an adverse effect, but there are currently no guidelines on how they should deal with the information.
“We do not currently have ways to link the blinded randomized patient to a social media posting,” explained Lipset. “As such we have no way to confirm if the patient is really in the trial, which treatment assignment there were given, or if any safety observation is one already captured by the research investigator in the study database.
“Without the ability to properly determine these details, monitoring social media introduces risk to an asset as the research sponsor and the regulator may need to take a conservative view and assign all social media findings as attributable to the study drug.”
Facebook with benefits
Social media can also have a positive role in research outcomes as Lipset pointed out monitoring online discussions can lead to insights on study complexity, recruitment, or considerations about specific sites.
In December 2012 the US Food and Drug Administration approved an Investigational New Drug Application (IND) for Lisinopril, based on a crowdsourced protocol which used telemonitoring and other remote methods to collect patient data.
At a moment when all parties in pharmaceutical research are just beginning to learn about these challenges, “this is an ideal time for collaboration between sponsor and regulator where safety events are still tracked and reported, but can be done in an unconfirmed manner for learning and without necessarily bringing risk to the medicine in development,” Lipset told us.
Equally, investigators must ensure participants are “properly informed of the potential implications of these conversations on the integrity of the study which they are supporting.” Trial managers are exploring educational videos, “consent language” and other means of keeping patients informed, he said.
Looking ahead, Lipset’s biggest recommendation to CROs (contract research organisations) concerned about social media use was to “be very creative in exploring new channels” to engage patients, while remaining mindful when they “open the door” that “any patients in that community who opt to participate may feel it is entirely appropriate to continue to share.”
Lipset’s original paper in Nature Medicine (DOI: 10.1038/nm0314-231) is available here.
