AI Tools Deployed to Reduce Nurse Documentation
Hackensack Meridian Health has deployed Ekam AI integrated with Google Cloud to accelerate clinical note summaries and combat nurse burnout. This follows recent discussions highlighting a 2025 Providence Health trial where ambient AI reportedly cut after-hours EHR charting by 2.5 hours per week, demonstrating a growing trend of using AI to address workflow inefficiencies.
- The underlying technology at Hackensack Meridian Health is a semantic data layer named Ekam, built on Google Cloud's BigQuery and Looker, designed to create a single source of truth from disparate data sources. This platform is now powering several AI agents, including the clinical note summarization tool built on Gemini, which is used by over 7,000 clinicians across 18 hospitals. - Frontline nurses often criticize Epic EHR for being cumbersome, requiring excessive time for data entry which detracts from direct patient care, and for having a counter-intuitive design. Specific ICU-related complaints include difficulties in getting a quick overview of a patient's medical history, charting complexities for titratable medications, and AI-driven sepsis alerts that are often inaccurate. - To transition into nursing informatics, the American Nurses Credentialing Center (ANCC) offers the Informatics Nursing Certification (NI-BC™). Eligibility typically requires a BSN, two years of RN experience, and either 2,000 hours of informatics practice in the last three years or a combination of 1,000 hours and relevant graduate coursework. - The 21st Century Cures Act, implemented through ONC and CMS rules, mandates the use of application programming interfaces (APIs) to promote data exchange between providers, payers, and patients. This has spurred the adoption of the HL7 FHIR (Fast Healthcare Interoperability Resources) standard, which uses modern web technologies to allow different EHR systems and third-party apps to share data more easily. - A key EHR optimization strategy is to directly involve nurses in reviewing and simplifying documentation templates and flowsheets to better reflect actual clinical workflows. One successful multi-year Epic optimization project at UCHealth reduced documentation time for acute care nurses by 18 minutes per 12-hour shift by eliminating unnecessary data fields and redesigning flowsheets. - In critical care, AI-powered clinical decision support (CDS) tools analyze EHR data in real-time to predict patient deterioration, sepsis, and cardiac events, often outperforming traditional scoring systems like MEWS and SOFA. These tools can improve diagnostic accuracy and reduce time-to-decision by synthesizing vast amounts of data, allowing clinicians to focus on high-priority tasks. - While ambient AI has shown promise in reducing documentation, some nurses express concerns about its reliability, citing instances where automated handoffs missed critical medication information that was only caught by manually reviewing the chart. This highlights the ongoing need for informaticists to bridge the gap between technological capabilities and safe clinical practice. - Recent lawsuits allege that Epic has engaged in information blocking by restricting third-party access to health data, potentially violating the 21st Century Cures Act. These legal challenges underscore the ongoing tension between the market dominance of major EHR vendors and the broader goals of system-wide interoperability.
