JMIR Publications

ABSTRACT

Background:

Large Language Model (LLM) artificial intelligence (AI) tools have the potential to streamline healthcare administration by enhancing efficiency in document drafting, resource allocation, and communication tasks. Despite this potential, the adoption of such tools among hospital administrators remains understudied, particularly at the individual level.

Objective:

To explore factors influencing the adoption and utilization of LLM AI tools among hospital administrators in China, focusing on enablers, barriers, and practical applications in daily administrative tasks.

Methods:

A multi-center, cross-sectional, descriptive qualitative design was employed. Three tertiary hospitals located in Beijing (Site 1), Shenzhen (Site 2), and Chengdu (Site 3) were selected to represent diverse geographic regions and institutional profiles. Middle-level administrators were recruited using purposive sampling. Data were collected from June 11 to August 16, 2024 through face-to-face semi-structured interviews guided by a collaboratively developed and piloted interview guide. Each interview was audio-recorded and transcribed verbatim. Colaizzi’s method was employed for thematic analysis. Data saturation was determined on a per-site basis by continuously reviewing transcripts during biweekly meetings until no new themes emerged from additional interviews.

Results:

A total of 31 participants (Site 1: 9; Site 2: 10; Site 3: 12) completed interviews lasting an average of 27.3 min (range: 21–39 min). Only 22.6% of participants reported high familiarity with LLM AI tools, and 25.8% were frequent users while 45.2% were rare users. Adoption varied by site. Site 3 had the highest proportion of high-familiarity participants who consistently used the tools more frequently. Qualitative analysis revealed that positive early experiences and prior technological expertise facilitated adoption, whereas mistrust in tool accuracy, limited prompting skills, and insufficient training were significant barriers. Participants predominantly used the tools for document drafting and strongly advocated for structured tutorials and institutional support to enhance broader utilization.

Conclusions:

Familiarity with technology, positive early experiences, and openness to innovation may facilitate adoption, while barriers such as limited knowledge, mistrust in tool accuracy, and insufficient prompting skills can hinder broader use. LLM AI tools are now primarily used for basic tasks such as document drafting, with limited application to more advanced functionalities due to a lack of training and confidence. Structured tutorials and institutional support are needed to enhance usability and integration. Targeted training programs, combined with organizational strategies to build trust and improve accessibility, could enhance adoption rates and broaden tool usage. Future quantitative investigations should validate the adoption rate and influencing factors.