Ethical challenges of using artificial intelligence in healthcare delivery: a thematic analysis of a systematic review of reviews

The COVID-19 pandemic boosted the digitalization and virtualization of health and social services, leading to an era of AI and AI digital platforms for healthcare. However, various studies have addressed the ethical challenges and issues of AI in healthcare. This study aimed to investigate and accumulate the ethical challenges and issues regarding using artificial intelligence in healthcare service delivery. We performed a systematic review of the existing reviews published between 2000 and 2023 on multiple databases, such as PubMed, Scopus, ProQuest, and Cochrane database of systematic reviews. After completing the screening, quality and risk of bias assessment of the articles using the CASP checklist, we selected and included the final articles. In the subsequent step, we conducted a thematic analysis of the data extracted from the final articles. Out of 1808 records that were identified, six papers met the eligibility criteria and were included in the analysis. The findings consisted of two major themes and seven sub-themes: the major themes were ethical challenges in the design and development of AI and ethical challenges of AI implementation and impact; the sub-themes were bias, transparency, privacy, accountability, equity, trust, and replacement. Also, the findings implied that ethical principles and values may be context-dependent and dynamic. Hence, it is essential to adopt a multidisciplinary and participatory approach to address the ethical issues of AI in healthcare and to align AI technologies with the goals and values of healthcare.