Medical AI can’t do much without info-sharing patients—and younger generations aren’t thrilled to be asked
To feed medical AI the massive volumes of clinical data it needs to succeed, researchers must offer patients individualized options for sharing personal health information.
The demand for such tailored consent is strongest among younger generations, further pressing the need to speed the evolution of PHI policies to propel AI development.
So report researchers in Canada who surveyed 222 patients and had their findings published March 24 in BMC Medical Ethics.
Senior author Fei-Fei Liu, MD, of the University of Toronto and colleagues found 83% of all respondents having no qualms about sharing PHI with researchers at their own hospital.
However, 13% of those aged 49 and younger said they’d be uncomfortable with the prospect while only 2% of those 50 and older had similar misgivings.
Other key findings drawn from the representative sampling:
- 70% of patients expect to be notified of study results.
- 63% want to be asked for permission before being entered into a contact pool. The younger cohort want more of a say (76%).
- 61% want the ability to track use of their PHI, and the younger patients comprise the bulk of this subgroup.
- 56% of all patients are OK giving broad consent for PHI sharing with researchers, while 38% prefer specific consent and 6% would not consent at all.
Meanwhile 51% of the 222 indicated feeling uncomfortable about sharing PHI with commercial enterprises, but 27% were fine with the idea and 22% were neutral.
“We would be remiss to allow these results to reverberate in an esoteric echo chamber; further work must uncover strategies to translate these findings into effective policy changes,” Liu and co-authors comment in their discussion section. “While this seems like a vast undertaking, the very technological advancements that have prompted the need for massive volumes of PHI can no doubt be applied to the design and implementation of modern consent platforms that exemplify a respect for well-identified patient values and preferences.”
The study is available in full for free.