A Systematic Review of Collaborative Robots for Nurses: Where are we now, and where is the evidence? Provisionally Accepted

Grace T. Babalola^{1, 2*} Jenna-Marie Gaston¹ Joseph Trombetta^{1, 2} Stephanie Tulk Jesso^{1, 2*}

• ¹Department of Systems Science and Industrial Engineering, Thomas J. Watson College of Engineering and Applied Science, Binghamton University, United States
• ²Human-Centered Mindful Technologies Lab, SUNY Binghamton, United States

Collaborative robots (cobots) have been proposed as a solution to assist nurses and enhance healthcare delivery. However, there is a lack of evidence on how cobots are being developed and implemented to support nursing practice. This systematic review investigates the current state of cobotic technologies designed for hospital settings, their intended applications, and impacts on nurses and patient care. A comprehensive database search identified 28 relevant peer-reviewed articles published since 2018. Results show cobots are predominantly patient-centered rather than nursecentered, focused on tasks like medication delivery, vital monitoring, and social interaction. Few cobots directly target reducing nursing workload through administrative or logistical assistance. Most applications emerged from India, with limited evidence from the United States despite commercial availability of nurse-assistive cobots such as TUG and Moxi. Robots ranged from proof-of-concept (technology readiness level 4-5) to commercially deployed systems (level 9). This review highlights opportunities to apply human-centered design engaging frontline nurses to develop innovative cobots that seamlessly integrate into clinical workflows. Rather than comprehensive automation, collaborative robots allowing nurses to focus on high-value caregiving show promise for improving satisfaction and outcomes. Committed research-practice partnerships are needed to guide the technical development of nurse-centered cobotic solutions. 1 **Because our final dataset only included robot platforms which are intended for direct human interaction and assistance to humans in some way, we consider all included applications as