Utilizing Large Language Models to Augment Family Conversations in the Intensive Care Unit

What this Trial Is About

Researchers are evaluating how artificial intelligence (AI), specifically large language models like GPT-4, can help doctors in intensive care units (ICUs) save time and improve communication with families of adult patients. This observational study aims to see if AI-generated summaries of family conversations can reduce the time doctors spend documenting notes while making the information clearer and easier for families to understand. The study involves ICU doctors and adult family members of patients, all providing informed consent, and focuses on improving interactions in the ICU setting. During the study, family conversations occurring in Dutch will be observed and analyzed. Both the ICU healthcare professionals and participating family members must provide consent. The study does not involve experimental treatments but observes the use of AI technology in documenting these conversations. There are no specified intervention groups or treatments being administered. Participants will be involved in family conversations documented by ICU staff, with researchers assessing the quality of documentation within 30 days after each conversation. They will also measure the time ICU clinicians spend on documentation within one day following the conversation and evaluate family satisfaction within 30 days. The study focuses on adult patients admitted to the ICU and their families, with data collection centered on communication and documentation practices rather than clinical treatment.

Evaluating the Use of Artificial Intelligence to Improve Family Conversations for Intensive Care Patients and Their Families