Abstract
INTRODUCTION: Intensive care unit (ICU) visiting restrictions in hospitals, implemented due to infection control and other factors, limited contact between patients and family members, affecting patients' well-being and clinical outcomes. With the evolution of voice-cloning based on artificial intelligence (AI) technology, it has become possible for health providers to communicate with critically ill patients using highly similar synthetic voices of their loved ones during ICU care. This current randomised, controlled trial will explore the effect of voice-cloning care on improving mental health outcomes for critically ill patients, with ICU-acquired anxiety as the primary indicator. METHODS AND ANALYSIS: This study will enrol 234 adult ICU patients, who are expected to require mechanical ventilation for over 24 hours and an ICU stay exceeding 72 hours. Participants will be randomly assigned to two groups. The control group will receive standardised communication as part of usual ICU care, while the intervention group will receive Speech-to-Speech Voice-Cloning Care (SVCC) in addition to standard ICU care. The SVCC interventions involve nurses using a participant's family member's cloned voice to deliver pre-set communication scripts during awakening, soothing and preparation for endotracheal tube removal. The primary outcome is expected to lead to improvements in ICU-acquired anxiety for the intervention group, measured by the Hospital Anxiety and Depression Scale. Secondary outcomes include ICU-acquired depression, the incidence of delirium, the mean duration of mechanical ventilation and the average length of ICU stay. ETHICS AND DISSEMINATION: This protocol was approved by the Ethics Committee of Peking Union Medical College Hospital (Approval Number: K-6842). The protocol version number is 1.2 and the version date is 8 October 2024. Study findings will be disseminated through peer-reviewed publications and presentations at national and international conferences. TRIAL REGISTRATION NUMBER: ClinicalTrials.gov, ID: NCT06743321.