Enhancing healthcare worker mental health via artificial intelligence-driven work process improvements: a scoping review

Abstract

Background: Healthcare workers (HCWs) are exposed to higher rates of mental health issues, such as burnout, anxiety, cognitive overload, and stress, compared to the general population. These may be exacerbated by administrative activities like extensive paperwork and disintegrated work processes. The implementation of artificial intelligence (AI) in healthcare holds the potential to combat these challenges by streamlining workflow processes, lowering administrative load, and increasing efficiency. The role of AI in supporting HCWs' mental health is yet to be fully explored. This scoping review mapped the current evidence on how AI can enhance HCWs' mental health through workflow optimisation.

Methods: This scoping review was informed by best practice in the conduct and reporting of scoping reviews. A comprehensive search of academic and grey literature was performed without date restrictions. A two-stage dual screening process was employed using Covidence. A customised data extraction tool was developed to systematically extract data, which was then summarised descriptively.

Results: Twenty articles were included in the review, most of which were published between 2020 and 2024. These comprised empirical studies, literature reviews, position papers, as well as selected grey literature. The studies explored various AI applications such as Natural Language Processing (NLP), AI-integrated Electronic Health Records (EHR), Machine Learning (ML), Clinical Decision Support Systems (CDSS), and Generative AI-driven tools such as ChatGPT. Burnout was the most frequently addressed mental health issue, followed by stress and cognitive load. Clinical documentation emerged as the most frequently addressed workflow, followed by clinical decision-making and diagnostics. Literature indicated that AI was capable of streamlining workflows, reducing administrative burden, and improving job satisfaction among HCWs. However, challenges such as data integration, algorithmic bias, and increased oversight demands were noted as potential barriers to effective implementation.

Conclusion: AI holds significant potential to improve HCWs' mental health and well-being by addressing workflow inefficiencies and reducing administrative burden. While available evidence highlights its benefits in enhancing job satisfaction and mitigating burnout, challenges such as data standardisation and user trust must be addressed for successful adoption. Future research should focus on evaluating the long-term impacts of AI on HCWs' mental well-being and developing strategies to mitigate unintended consequences.

Keywords: Artificial intelligence (AI); Burnout; Clinical documentation; Healthcare workers (HCWs); Mental health; Workflow optimisation.