Nurses can spend up to 40% of their workdays on documentation, sometimes more in acute care settings. In 2023, the American Academy of Nursing noted that the constant addition of electronic health record-related (EHR-related) tasks is a significant contributor to increasing documentation. This can limit the time available for direct patient care, increase staff burnout and turnover, and increase the risk of medical errors.
In 2025, UW Health’s Practitioner Ambient Listening clinical trial demonstrated positive results from the implementation of Abridge’s ambient AI listening tool among physicians in outpatient clinics. Use of the tool, which captures clinical encounters and generates draft documentation for the EHR, resulted in a significant reduction in practitioner work exhaustion, time spent on clinical documentation notes, and work outside of normal working hours, as well as an increase in interpersonal engagement. Encouraged by the study’s results, UW Health leaders decided to explore whether the technology could benefit nurses.
Led by Becky Kohler, UW Health chief nursing information officer, Dr. Ann Wieben, assistant professor of nursing, and Jann Pfaff, nurse scientist with UW Health, the 26-week clinical trial was launched to nursing staff across three inpatient/surgical units at East Madison Hospital. While ambient listening has been studied in physician settings, evidence of its use in nursing remains limited. This study is among the first efforts to evaluate the technology in inpatient nursing workflows.
Thoughtful implementation through partnership
To make sure the AI tool is implemented safely and effectively, UW Health partnered with the UW Institute for Clinical and Translational Research (ICTR) Learning Health System (LHS) Program to integrate rigorous research methods into their operational decision-making and to evaluate the technology’s impact on nursing staff and patients.
Before rollout, the project team used readiness assessment strategies to anticipate potential challenges during implementation, including staffing constraints, leadership turnover, fatigue from recent technology implementations, and skepticism about AI. Many of the challenges are amplified by the inpatient nursing environment, in which nursing workflows are very complex, notes Wieben.
The AI tool maps conversations between nursing team members and patients in real time, with human verification before documentation in the EHR. Researchers and unit managers observe and track nursing workflows to identify operational roadblocks and opportunities to integrate the technology into existing care processes.
Felice Resnik, PhD, an implementation scientist with ICTR’s Dissemination and Implementation Launchpad, notes that successful adoption often depends as much on local context as on the intervention itself. “Whether it is the available technology, clinical workflows, organizational processes, or how staff schedules are organized, these contextual factors can make it easier or harder to successfully put a new tool or practice into everyday use.” Understanding the clinical environment and having strong support from leadership are important contributors to a successful intervention.
Advancing research and innovation
The rapid growth of AI in the last year has raised concerns about its costs, the resources needed to support it, and those that will be depleted by it, diminishing an appreciation for the opportunities it presents in healthcare.
Kohler understands the anxiety surrounding AI in healthcare settings and believes that UW’s scientific research infrastructure and rigorous safeguards, combined with UW Health’s high standards and advanced security measures, are leading the way for the study and use of AI tools in the design and delivery of safe patient care. She points out that consumer expectations for high-quality care will help drive the use of AI in healthcare when patients see the difference such tools can make.
Healthcare organizations increasingly emphasize the need for a rigorous scientific approach when adopting AI. This year, the Health Information and Management Systems Society (HIMSS) committed to developing an AI assessment framework, similar to that used by the LHS program, that helps health systems measure outcomes and demonstrate the value of AI in healthcare.
The abundance of AI tools that are becoming available to health systems may also present an opportunity for more research in clinical settings that include nursing services. Wieben hopes that the structured, evidence-based innovations UW Health and LHS are currently implementing will increase interest in nursing research and help build the capacity of UW Health and other academic health centers to engage in nursing research.
After the clinical trial ends, researchers, nurses, and health system leaders from across the country will be watching, eager to learn from UW’s pioneering study. Project leaders say the study will provide valuable insights into how health systems can evaluate emerging technologies and effectively manage change while keeping patient safety, workforce well-being, and quality of care at the center of innovation. Early signs suggest the tool has the potential to make a positive impact. “Ambient Listening has significantly shortened my charting time, allowing me to focus more on patient care,” reports Susan Colemer, a nurse at East Madison Hospital. “I actively encourage my peers to try it, sharing effective strategies I’ve learned to maximize its utility in our practice.”
For additional information about UW Health’s Ambient Listening for Nursing LHS project, download the most recent LHS Learning Loop podcast episode, featuring an interview with Wieben and Kohler. You can also view and download the playbook for this project using GitLab, an open‑source collaboration platform.