EnhanCed HandOffs (ECHO)

Implementing a machine learning-augmented sociotechnical intervention for postoperative handoffs is expected to improve interdisciplinary team communication, enhance patient safety, and create a more efficient care process.

Project Details -
Ongoing

Handoffs from the operating room (OR) to the intensive care unit (ICU) require seamless communication and coordination between surgical and critical care teams, as patients are extra vulnerable immediately after surgery. However, these handoffs result in about 35 percent of medical errors in the United States. A care bundle is a group of evidence-based interventions—typically between two and four—which result in better patient outcomes when performed together, than if performed individually. Standardization strategies, which include bundled interventions including checklists and processes, have been implemented across hospitals to mitigate failures in postoperative handoffs, but these approaches have been rigid and inflexible, failing to advance handoffs that are collaborative. This leads to communication errors and care continuity failures.

The EnhanCed HandOff (ECHO) project will allow the handoff process and communications to be tailored using a to-be-developed bundle called INTERACT (intelligent interactive care continuity), containing (1) a telemedicine-augmented handoff process and (2) a machine learning (ML)-augmented electronic handoff report. The INTERACT bundle will result in improved communication between interdisciplinary teams and increased shared understanding of risks for postoperative complications.

The specific aims of the research are as follows:

  • Validate a dynamic ML model to predict patient risks for postoperative complications.
  • Develop the INTERACT bundle using user-centered design principles and ascertain its usability and feasibility within the postoperative handoff process.
  • Determine the effect of the INTERACT bundle on postoperative handoff outcomes.

Researchers will update current ML risk-prediction models for postoperative complications and validate their accuracy and usefulness through a prospective user study with operating room clinicians. The project team will then utilize a user-centered design approach to develop and evaluate the INTERACT bundle, including handoff report designs, usability testing, and onsite simulations. The report will support personalized communication of both standardized and tailored content, based on predicted patient risks for postoperative complications. Finally, the team will evaluate the effectiveness and implementation potential of the bundle using research data, chart reviews, and clinician interviews and surveys. Chart data for adult patient handoffs from the OR to the surgical ICU or the cardio-thoracic-vascular ICU will be included in the study, and clinicians involved in OR-ICU transitions will be recruited for testing.

The findings from the ECHO project will inform a demonstration trial on a multisite implementation and evaluation of the INTERACT bundle before its dissemination at the national level. This research can also have pragmatic extensions with the intervention being implemented into rural hospitals and other low-volume surgical centers to support team communication and collaboration for reduced costs, greater patient safety, and a more efficient care process.

Improving Safety in Postoperative Handoff Communication with Telemedicine and Machine Learning

Theme:

Optimizing Care Delivery for Clinicians

Subtheme:

Using Digital Healthcare Tools to Improve Patient Safety

Implementing a postoperative handoff intervention augmented with telemedicine and machine learning technologies will promote effective and efficient communication, foster interdisciplinary collaborative team workflows, and ultimately, improve patient care and safety during care transitions.

Poor postoperative handoffs can risk patient safety

Effective communication during postoperative handoffs is crucial to patient safety. Currently, nearly one-third of medical errors and adverse events in healthcare can be tied to miscommunication and care coordination gaps during handoffs and care transitions. While postoperative handoffs have been standardized using communication checklists and process guiding protocols, such “one-size-fits-all” communication frameworks often don’t work and are not sustainable in routine care. These standardized checklists and protocols do not adequately support the collaborative and interactive features of effective communication practices, do not align well with the handoff care teams workflows and—lastly and most importantly—do not meet the needs of delivering the best, person-centered information in an effective and efficient manner.

Dr. Joanna Abraham and her team at Washington University recognize that care transition failures can lead to potentially unsafe and high-risk situations for surgical patients being transferred from the operating room (OR) to intensive care units (ICU). The team has identified the need to build a “flexible” yet standardized approach to support smart precision handoffs that enhance the care team’s resilience to communication errors. This project, EnhanCed HandOffs (ECHO), attempts to provide a handoff intervention that is both standardized and flexible, informed by a sociotechnical model that reflects the interactions of people and technologies in a complex healthcare system.

Using telemedicine and machine learning may improve postoperative handoff safety

With an extensive background in anesthesiology workflows and biomedical informatics, Dr. Abraham is implementing an integrated, sociotechnical intervention for OR-ICU handoffs called the intelligent interactive care continuity (INTERACT) bundle. The bundle includes two main components: a telemedicine-augmented handoff process and a machine learning–augmented handoff report. The telemedicine-augmented handoff process promotes a fresh-eye perspective during patient handoffs from telemedicine teams for the OR and ICU, whereas the machine learning–augmented report promotes the identification and sharing of patient risks to postoperative complications and anticipatory management plans during handoffs. Together, the INTERACT bundle components aim to personalize patient handoffs, thereby promoting intelligent and interactive communication of patient risks and risk mitigation strategies.

“We can do better when it comes to optimizing surgical patient experiences and patient safety and reducing healthcare costs associated with postoperative complications. By harnessing state of-the-art machine learning and telemedicine innovations, I am confident that our proposed cutting-edge intervention will transform the practice of postoperative handoffs from a purely information transactional process to a more patient-centered, thoughtful, interactive, and collaborative team process.” – Dr. Joanna Abraham

Reducing risks can increase safety of postoperative handoffs

Customizing postoperative handoffs allows for patient-specific insights to be shared across teams for promoting effective and efficient handoffs, while providing a safety net against handoff errors. The team will evaluate the effectiveness and feasibility of implementing the INTERACT bundle using mixed methods. This project is the first to leverage and implement these advanced technologies for increasing safety during handoffs.

This Research is Featured in

Principal Investigator

Abraham, Joanna

Organization

Washington University

Research Profile

Funded Amount

$1,996,227

Research Type

Emerging

This Research is Featured in

Related Stories