Artificial intelligence for secondary prevention of myocardial infarction: A qualitative study of patient and health professional perspectives. (May 2023)

• Record Type: Journal Article
• Authors: Pelly, Melissa; Fatehi, Farhad; Liew, Danny; Verdejo-Garcia, Antonio
• Is Part Of: International journal of medical informatics. Volume 173(2023)
• Publication Date: 2023-05
• Subjects: Artificial intelligence -- Secondary prevention -- Myocardial infarction -- Co-design -- Perspectives -- Qualitative study; PHMI People with a history of myocardial infarction; Medical informatics -- Periodicals; Information science -- Periodicals; Computers -- Periodicals; Medical technology -- Periodicals; Medical Informatics -- Periodicals; Technology, Medical -- Periodicals; Computers; Information science; Medical informatics; Medical technology; Electronic journals; Periodicals; Electronic journals; 610.285
• DOI: 10.1016/j.ijmedinf.2023.105041
• Languages: English
• ISSNs: 1386-5056

Abstract

Highlights: Trust in AI hinges on credibility, safety, personalisation and transparency. Expectations of AI: personalised advice and feedback, and a virtual assistant. AI concerns: complexity, negative experiences, confidentiality and AI literacy. Adoption hinges on preference, usability features, interest/familiarity with AI. AI benefits: an alternative service, complete privacy and usefulness. Abstract: Background: Artificial intelligence (AI) has potential to improve self-management of several chronic conditions. However, the perspective of patients and healthcare professionals regarding AI-enabled health management programs, which are key to successful implementation, remains poorly understood. Purpose: To explore the opinions of people with a history of myocardial infarction (PHMI) and health professionals on the use of AI for secondary prevention of MI. Procedure: Three rounds of focus groups were conducted via videoconferencing with 38 participants: 22 PHMI and 16 health professionals. Findings: We identified 21 concepts stemming from participants' views, which we classified into five categories: Trust; Expected Functions; Adoption; Concerns; and Perceived Benefits. Trust covered the credibility of information and safety to believe health advice. Expected Functions covered tailored feedback and personalised advice. Adoption included usability features and overall interest in AI. Concerns originated from previous negative experience with AI. Perceived Benefits included the usefulness of AI to provide advice when regular contact with healthcare services is not feasible. Health professionals were more optimistic than PHMI about the usefulness of AI for improving health behaviour. Conclusions: Altogether, our findings provide key insights from end-users to improve the likelihood of successful implementation and adoption of AI-enabled systems in the context of MI, as an exemplar of broader applications in chronic disease management.