Global survey finds doctors, nurses all in with healthcare AI

Almost half of clinicians around the world, 48%, believe it’s a good thing for AI to help direct care pathways. The broad buy-in is striking since, in the present year, only 11% of clinical decisions are assisted by AI tools.

The findings are from Elsevier Health’s latest “Clinician of the Future” survey and report. For 2023, the academic publishing company conducted online fieldwork this past spring and received completed surveys from more than 2,600 physicians and nurses.

Survey questions were designed to help forecast the future of healthcare delivery as projected by the clinicians who are generally closest to frontline patient care. Responses came in from 116 countries, led in headcounts by China (418 respondents), the U.S. (365), the U.K. (272) Spain (189), Germany (181) and India (158). The survey report covers several aspects of future care. Here are five select findings from a chapter dedicated to AI and related technologies.

1. Clinicians are open to the possibilities of evolving technologies and willing to embrace them to improve patient care.

Overall, clinicians see it as a positive to be experts in the use of digital health technology (73% for physicians to be experts and 71% for nurses)—a recognition of the accelerating developments in this area.

2. Most clinicians (68%) approve of making clinical decisions based, at least in part, on patient-collected health data.

In North America, some 71% find the practice desirable and only 11% undesirable. Nurses in the U.S. (74%) and doctors in China (75%) are the most likely subgroups to find this desirable, while nurses in China (48%) are significantly less enthusiastic.

3. The world is already seeing differences in how clinicians use generative AI tools like ChatGPT and Bard.

Worldwide use of these tools to help guide clinical decisions is significantly higher among nurses (16% of their decisions) than doctors (7% of their decisions).

4. Use of generative AI to guide clinical decisions differs greatly between regions.

The practice is much more common in China and Asia Pacific (19% of clinical decisions in both) than in North America (7%) and Europe (6%). In the Asian Pacific, AI is guiding some 28% of nurses’ decisions.

5. Globally, 51% of clinicians consider the use of AI desirable for training medical students, and 50% like it for training nurses.

In both cases, 28% of clinicians consider it undesirable. For medical student training, opinion is most split in Europe and North America. In both cases, more clinicians find it desirable (41% in Europe and 40% in North America) than undesirable (34% in Europe and 36% in North America).

The 55-page report is posted in full for free.