JMIR Publications

ABSTRACT

Background:

Emergency departments (EDs) worldwide face increasing challenges due to rising patient volumes, crowding, prolonged waiting times, workforce shortages, and growing administrative demands. In addition to providing urgent medical care, ED staff are required to perform numerous non-clinical tasks, including patient registration, information provision, and redundant communication with patients and accompanying persons about non-medical demands. These challenges can negatively affect patient experience and staff workload. Advances in artificial intelligence (AI) and robotics offer new opportunities to support communication processes and improve operational efficiency in healthcare settings.

Objective:

This study aimed to evaluate the feasibility, acceptance, and perceived usefulness of an AI-based communication robot deployed in an ED waiting area. Particular attention was given to patients’ prior experiences with technology, contextual factors influencing robot use, social influences on interaction behavior, perceptions of the robotic system, and potential future application scenarios.

Methods:

A qualitative proof-of-concept study was conducted in the waiting area of the ED at Charité – Universitätsmedizin Berlin. The communication robot TARA (Talking Autonomous Reception Assistant) was implemented alongside a self-check-in terminal to support patient orientation, multilingual communication, and administrative processes. Data were collected during two one-week field phases through participant observations and semi-structured interviews with patients and accompanying persons. The interview guide was informed by the Godspeed Questionnaire Series, covering dimensions such as anthropomorphism, animacy, likeability, perceived intelligence, and safety, as well as by the Unified Theory of Acceptance and Use of Technology (UTAUT) framework, including performance expectancy, effort expectancy, social influence, and facilitating conditions. Data were analyzed using qualitative content analysis according to Mayring.

Results:

A total of 15 interviews and more than 14 hours of participant observation were conducted. Participants generally viewed the robot positively and highlighted its potential to facilitate orientation, provide information, support registration processes, and overcome language barriers. The robot was often perceived as friendly, approachable, and capable of reducing uncertainty in the waiting area. However, interaction with the robot was strongly influenced by the ED context. Participants prioritized rapid and intuitive registration procedures, while stress, pain, anxiety, and uncertainty regarding the robot’s role frequently limited engagement. Privacy concerns in the open waiting area and a preference for human interaction represented additional barriers. Participants highlighted the need for further development, particularly regarding clearer communication of the robot’s role, more visible benefits for patients, and stronger integration into existing ED workflows, which were considered important prerequisites for future acceptance and use.

Conclusions:

AI-based communication robots may support patient guidance, multilingual communication, and administrative processes in ED waiting areas while potentially reducing staff workload. Successful implementation depends not only on technical performance but also on intuitive usability, workflow integration, privacy considerations, and the preservation of opportunities for human interaction. Further research is needed to evaluate long-term acceptance and clinical impact in routine emergency care. Clinical Trial: The study is registered with the German Clinical Trials Registry (DRKS-ID DRKS00038333) under https://drks.de/search/en/trial/DRKS00038333.