Technological solutions such as artificial intelligence (AI) are increasingly seen as a potential solution to growing resource pressures in medicine, healthcare, and biomedical research. AI systems promise novel means to evaluate and improve the quality of clinical care, undertake biomedical research and investigate new therapeutics and pharmaceuticals, and expand care offerings to previously underserved populations. A key driver of innovation and adoption is the belief that AI may relieve health professionals from “certain time-consuming clerical tasks and could increase their time for caregiving practices.” Medical decision-making and care are increasingly supported by expert and robotics systems that assist in record management, diagnosis, treatment planning, and delivery of interventions. Home and social care are similarly transformed through the introduction of remote monitoring and management systems. Health can increasingly be monitored, modelled, and managed based on data representations of the patient, supplementing or replacing verbal accounts and face-to-face physical care.

Read more

A unique impact of AI and other emerging data-intensive and algorithmic technologies is their capacity to augment and support human decision-making by recommending the best action to take in a given situation, the best interpretation of data, and so on. But these systems can also be used to outright replace human decision-making, expertise, and face-to-face clinical care. Natural language processing applications such as OpenAI’s GPT-3, for example, suggest a future in which initial patient contact and even triage can be handled in part or entirely by artificial conversational agents. AI systems are already used by clinicians and hospitals for clinical and operational decision-making, seen for instance in risk prediction, discharge planning, diagnostics, and decision-support systems. Developments in deep learning likewise suggest a future in which drug discovery and biomedical research are increasingly driven by computational systems capable of intelligent behaviour. Recent advances in the pharmaceuticals to treat a rare form of brain cancer or Deepmind’s breakthrough in protein folding via AlphaFold already show the potential of the state of the art in medical AI.

While the promise of AI is clear, a significant area of uncertainty concerns its impact on the practice of healthcare, and in particular the doctor-patient relationship. Medical expertise is no longer the sole domain of trained medical professionals and researchers; rather, AI technologies create opportunities to provide care through a mix of public and private, professional and non-professional, and human and technological stakeholders.

In response to the growing recognition of these opportunities and risks of AI on the practice of medicine and clinical care by the Council of Europe, and the call by the Committee on Bioethics (DH-BIO) to work on trust, safety, and transparency in this context, this report investigates the known and potential impacts of AI systems on the doctor-patient relationship. This impact is framed by the human rights principles referred to in the European Convention on Human Rights and Biomedicine of 1997 otherwise known as the “Oviedo Convention,” and its subsequent amendments. Human rights principles regarding health may require certain standards to be met in the doctor-patient relationship which can be disrupted, displaced, or at least augmented by the usage of AI in clinical care.