"...it saves so much time": A qualitative exploration of the use of Generative Artificial Intelligence by the health workforce

Abstract

• Though Gen AI is increasingly being explored in various fields, including healthcare, there is little to no research into health professional perspectives. • Generally, there is a support for the use of Gen AI in healthcare for both clinical use and administrative offloading, however, there are still concerns with regards to data privacy and technical errors. • Organisational support and the provision of guidelines and guardrails remain the crucial priority for health professional use of Gen AI technologies in the workplace. Generative Artificial Intelligence (Gen AI) has become an increasingly prevalent conversation in healthcare over the past few years. Though there have been research projects and articles exploring the administrative and clinical uses of such technologies, there has been little exploration of health professional perspectives, hopes and concerns. This study sought to explore perspectives and examine the barriers and enablers of Gen AI in healthcare. Australian health professionals participated in a mixed-methods study. A survey (n=31) explored the Six Dimensions of Healthcare Quality Framework, capturing quantitative (Likert-scale responses) and qualitative (free-text) data. Semi-structured interviews (n=10) explored participant perceptions of Gen AI. Quantitative data was analysed using descriptive statistics. Qualitative data was thematically analysed. Most survey respondents (74.14%) reported having used Gen AI to support their work, but only a few (25.81%) reported organisational supports for use of these technologies. Analysis of the qualitative data aligned with the survey responses. Five themes were generated through thematic analysis, aligning with health professional’s perceived use of Gen AI chatbots, benefits, risks, as well as drivers of safe use and opportunities for the future. Health professionals see potential for using Gen AI to support their work, with enthusiasm about the potential of Gen AI to reduce workloads, particularly in offloading administrative tasks. There is also awareness that Gen AI chatbots pose risks both at the individual level such as limited capability in using these technologies and at the organisational level such as lack of training to support in upskilling, and systemic concerns around policy gaps. Generative Artificial Intelligence (Gen AI) is increasingly topical in all aspects of life, and the health sector is no exception. Though there have research projects focusing on Gen AI in healthcare, there has been little exploration of health professional views and concerns. This study spoke to health professionals and found that though there is a lot of interest in potential applications of Gen AI in healthcare, particularly in administrative offloading and clinical support, the benefits don’t yet outweigh the risks. Software developers must work alongside health professionals in developing a substantially beneficial program to support the safe use of Gen AI in healthcare as well as be well supported on an organisational level. There is also opportunities to develop education to build health professionals capacity to use GenAI safely and effectively, and for health service organisations to develop guidance and policies to clearly articulate what safe use looks like.

Ähnliche Arbeiten