How medical imposters are ruining health studies

Medical researchers are witnessing a rise in “imposter participants” taking part in surveys and trials, fuelling concerns for patient safety and the reliability of studies.

Imposters vary from automated bots to people lying about previous health issues or deliberately misleading researchers over their conditions and diagnoses.

A review published this year by the British Medical Journal looked at 23 studies and found that 18 contained compromised data sources, with infiltration rates ranging from 3% to as high as 94% of participants. This growing trend could “threaten the integrity of health research and, by extension, the policies and clinical decisions built on it”.

The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

‘Seismic transformation’ from online practices

Academics are in the dark over the exact motivations of the imposters. When there isn’t a financial interest, feelings of “boredom” and “curiosity” are major players. In more serious cases, “an ideological intent to disrupt research” may be a factor.

Online recruitment methods have triggered a “seismic transformation” across the health landscape, said Bioengineer. The attraction is clear: the “expanded accessibility” has made it easier than ever to enrol in studies, but this has opened the door to individuals and artificial sources to “falsify data” or “mimic” authentic responses.

Most trials are secure but the move to online information gathering could exacerbate the latent problems in the industry, said The Times. “Survey-style studies”, which are one of the most common methods, and experience-related data gathered on health apps were “more at risk” of attracting “lying” individuals.

Other ways to detect whether a participant is credible is to submit them to “Turing tests”, which can differentiate between human and machine responses. If answers are recorded “between midnight and 4am”, or “filled out suspiciously quickly”, then a bot is likely to have been involved.

Researchers already have to walk a tightrope to gain credible and fair results, said Alan Martino and Arielle Perrotta on The Conversation. They must maintain a balance between providing “thorough vetting”, while simultaneously preserving the “ethical commitment to respect the autonomy of participants”.

Certain studies are more susceptible than others. For quantitative research, data collection usually allows “anonymous” entry points, with “minimal direct interaction”, such as filling in a form online. Interference can be greatly minimised by employing eligibility pre-screenings or “automated verification methods”.

This approach “doesn’t always translate well” to qualitative studies, which include more “personal” steps, such as one-on-one interviews, or extended observation sessions with the subject. Routinely requesting information can feel “intrusive or even alienating”, especially for “marginalised groups like those with intellectual disabilities”.

‘Systemic threat’

To combat the “systemic threat” from imposters, further action is urgently required, said the BMJ. Journals should “encourage consistent and transparent reporting” of safeguards and “acknowledge limitations”. Investors need to “keep pace” as the industry and tactics evolve, and policy makers should be “cautious” of results if imposter participant measures are not expressly addressed in the findings.

There is a fine line between ensuring the “validity” of data, but not “creating barriers for legitimate participants”, said Martino and Perrotta. If measures are too stringent, then credible concerns could be ignored. Even common “red flags” such as an unwillingness to be seen on camera, not being able to answer basic questions, or a preference for unverified payment practices can be explained away with “valid reasons” of “privacy concerns or technical limitations”.

“The key to success is balancing caution with empathy”, where patients’ experiences are weighed up, analysing the risks of including or excluding them. If recruitment processes are “clear and specific”, there is caution around compensation and interview practices are revised, there can be increased credibility among researchers.