Assessing usability of intelligent guidance chatbots in Chinese hospitals: Cross-sectional study.

ABSTRACT

Objective: This study aimed to assessing usability of intelligent guidance chatbots (IGCs) in Chinese hospitals. Methods: A cross-sectional study based on expert survey was conducted between August to December 2023. The survey assessed the usability of chatbots in 590 Chinese hospitals. One-way ANOVA was used to analyze the impact of the number of functions, human-like characteristics, number of outpatients, and staff size on the usability of the IGCs. Results: The results indicate that there are 273 (46.27%) hospitals scoring above 45 points. In terms of function development, 581(98.47%) hospitals have set the number of functions between 1 and 5. Besides, 350 hospitals have excellent function implementation, accounting for 59.32%. In terms of the IGC's human-like characteristic, 220 hospitals have both an avatar and a nickname. Results of One-way ANOVA show that, the number of functions(F = 202.667, P < 0.001), human-like characteristics(F = 372.29, P < 0.001), staff size(F = 9.846, P < 0.001), and the number of outpatients(F = 5.709, P = 0.004) have significant impact on the usability of hospital IGCs. Conclusions: This study found that the differences in the usability of hospital IGCs at various levels of the number of functions, human-like characteristics, number of outpatients, and staff size. These findings provide insights for deploying hospital IGCs and can inform improvements in patient's experience and adoption of chatbots.