JMIR Publications

ABSTRACT

Background:

Development of artificial intelligence (AI)-based technologies in healthcare is proceeding rapidly. In Germany, widespread implementation of such technologies in care settings are still missing and pools of data derived from daily patient care are not available to research and development. Sharing and releasing real-world data are key practical issues surrounding implementation of AI solutions into existing clinical practice. However, data derived from daily patient care are necessary for initial training, continued data supply is needed for ongoing training, validation, and improvement of AI-based solutions. Data may need to be shared across multiple institutions for widespread implementation and high-quality usage of such solutions. So far, efforts have not been undertaken in Germany to meet the challenge of providing a sufficient data volume for the development of AI-based technologies. The pAItient project (Protected Artificial Intelligence Innovation Environment for Patient-Oriented Digital Health Solutions) aims to meet this challenge by creating a large data pool feeding on donation of data derived from daily patient care. Prior to building this data pool, physician perspectives regarding data donation for AI-based solutions need to be studied.

Objective:

Exploration of physician perspectives on providing and utilizing real-world care data for the development of AI solutions in healthcare in Germany.

Methods:

Within a requirements analysis preceding the pAItient project, this qualitative study explored physician’s perspectives and expectations regarding the use of data derived from daily patient care in AI-based solutions. Semi-structured, guide-based problem-centered interviews were audio-recorded, de-identified, transcribed verbatim and analyzed inductively in a thematically structured approach.

Results:

Interviews (n=8; mean duration 24 minutes) were conducted with six General Practitioners and two hospital-based physicians. Mean participant age was 54 years (range: 30-74), with an average experience as physician of 25 years (range: 1-45). Self-rated affinity towards modern information technology varied from very high to low (5-point Likert scale, mean: 3,75). All participants would support development of AI-based solutions in research contexts by donating de-identified data derived from daily patient care if subsequent data usage was made transparent to them and their patients and benefits for patient care were clear. Contributing to care optimization and efficiency were cited as motivation for potential data donation. Discussed concerns referred to workflow integration (time and effort), appropriate de-identification and to involvement of economic third-party interests. Donation of data referring to psychosomatic treatment needs was viewed critical.

Conclusions:

Our findings indicate that in research contexts, development of AI-based solutions with a clear benefit for daily patient care would be supported by physicians with the donation of their real-world care data. Joint ventures with third-party entities should focus on care optimization and patient benefits, not on financial interests.