AI in Patient Care: Majority of Americans Want Notification About Use

A majority of Americans want to be informed about the use of artificial intelligence (AI) in patient care, according to a recent study published in JAMA Network Open.

While transparency has long been central to patient care and research ethics, the rapid adoption of AI across health care settings has outpaced regulatory frameworks. Therefore, researchers sought to explore public attitudes toward AI use notifications.

Between June 27 and July 17, 2023, a team of researchers surveyed 2021 US adults using a validated questionnaire through the National Opinion Research Center AmeriSpeak Panel. Participants rated their agreement with the statement, “It is important that I am notified about the use of AI in my health care,” on a scale from 1 (“not at all true”) to 4 (“very true”). Weighted analyses were applied to reflect national demographics, including oversampling Black and Hispanic respondents to ensure robust subgroup analysis.

The findings revealed that 62.7% (95%, 60%-65.8%; n=1267) of respondents strongly agreed it was “very true” that they desired notification, while only 4.8% (95% CI, 3.7%-6.2%; n=99) considered it unimportant. The weighted mean score of 3.39 (95% CI, 3.33-3.44) reflects broad support for greater transparency in the use of AI across groups.

Significant demographic differences emerged. Women were more likely than men to value notification (mean score, 3.45 vs 3.32; P =.02). Age significantly influenced preferences, with older adults (60 years and older) showing the strongest inclination (3.57) compared to younger individuals aged 18 to 29 (3.14; P <.001).

Racial and ethnic disparities were also observed, as White respondents had a higher mean preference score (3.46) than Black respondents (3.21; P =.002). Education level further influenced preferences, as those with postgraduate degrees rated notification more favorably (3.51) than those with less than a high school education (3.14; P <.001)

This study is limited by its cross-sectional design and reliance on self-reported data, which may introduce bias.

“Nonetheless, our findings suggest that notification about AI will be necessary for ethical AI and should be a priority for organizations and policymakers,” the researchers wrote.

They concluded, “Our findings indicate differences in demographic groups, particularly in the ethical context of historical, structural, and systemic inequity.”

This research was supported by the National Institutes of Health, National Institute of Biomedical Imaging and Bioengineering.