KenSci Shines the Light on Nursing Informatics at ANIA18
Whende Carroll, MSN RN BC, Director, Nursing Informatics, KenSci
The ANIA18 annual conference was held in Orlando, Florida from May 9–12th. The theme of the American Nursing Informatics Association conference was ‘Shine the light on Nursing Informatics.’ KenSci presented a poster in partnership with Madigan Army Medical Center. We presented a poster that summarizes of a study of nurses expanding on our co-development of a 30-day Risk of Readmission risk prediction tool which identifies patients at highest risk for being readmitted to the hospital in a Military Treatment Facility.
American Nursing Informatics Association (ANIA) 101
ANIA members are a national community of nursing informaticists who are committed to the dedicated nursing specialty of Informatics Nursing. The nursing niche integrates nursing science, computer science and information science to manage and communicate data, information, knowledge, and wisdom in nursing and informatics practices. ANIA members identify informatics practice as a specialty that is essential to the delivery of high quality, and cost-effective health care, and use informatics to improve the health of populations, communities, families, and individuals by optimizing information management and communication.
Nurse informaticists work in inpatient, outpatient, HIT vendor, clinical research, education, and startup (!) settings. Their work is unique as they speak many languages, particularly tech and clinical, and they serve as are the liaisons between both worlds. Many are board certified through the American Nurses Credentialing Center (ANCC) in conjunction with the American Nurses Association (ANA) as Informatics Nurses (RN-BCs) and through the Health Information Management and Systems Society (HIMSS) as a Certified Professional in Health Information & Management Systems (CAHIMS).
ANIA 2018 — Various Sessions and Posters
SESSIONS (32)
· Nursing Informatics in Clinical Transformation — A Process, Not a Project
· Interoperability for Better Care
· Increasing Patient Engagement in Their Care Utilizing Mobile Technology
· EHR Optimization: Blueprint for Sustainable Change
· Making the Stars Align when Time Matters: Leveraging Actionable Data to Combat Sepsis
POSTERS (67)
· Clinical Decision Support — Importance of Nursing Informatics Involvement
· The Architect of a Successful Implementation: Nursing Informatics’ Vital Role
· Revising Patient Discharge Process to Reduce Readmission Rates and Focus on Patient Engagement
· Using Technology to Enhance Patient Safety
The MAMC/ KenSci Poster
The inception of our study and subsequent poster, “Nurse Satisfaction and Experience Using a 30-day Readmission Predictive Analytics Tool in a Military Treatment Facility Patient Centered Medical Home (PCMH)” focuses on how nurses leverage the risk predictions of our solutions. Expanding on our work with MAMC on the RoR risk prediction tool, I was very interested to know if the evaluation of the end-user experience with the risk of readmission tool provides evidence of its dynamic use and value in nursing practice, and if so to share that with other nurses.
After gaining feedback from nurses in the internal medical clinic (IMC) PCMH and cardiology clinic, I came to understand that during the pilot, nurses began utilizing the tool effectively to ultimately develop patient-focused population health management and chronic disease management programs in their clinical settings. These insights are what KenSci and nurses attempting to work with data need to know: it is possible to develop ML solutions that serve the Quadruple Aim — reducing cost, improving quality of care, patient engagement and experience and caregiver experience.
Lessons learned from the study include:
- New clinical workflows and interventions were developed and revised using the risk predictions, enabling nurses to
o Arrange follow-up appointments face-to face-while high-risk patients were still hospitalized
o Identify high-risk patients for members of the care teams and advocating for and coordinate care of VA patients
o Provide one-on-one nurse education high-risk patients, and facilitate group classes for patients with new onset of heart failure
- Improvements recommended by nurses for the RoR tool:
o Add filter options and identify the drivers of patients’ risk score, (i.e., diagnoses, procedures, and medications)
o Direct the nurse to patients’ charts when clicking on the risk score
o Continue to improve model accuracy and reliability
New knowledge in for nursing is gained from this study. Insights showed that population health management strategies to improve transitional care from the inpatient setting to Patient Centered Medical Home’s clinics can be realized through the use of an ML predictive analytics 30-day readmission tool for all patients. Users reported improved chronic disease management by interdisciplinary teams working in a large MTF. Also, the collaboration between an MTF and healthcare informatics company empowered nurses to suggest enhancements to the predictive RoR tool itself, ensuring improved usability and reliability in the future.
Poster Feedback and Opportunity for Nursing Impact
“This is cool!”
“We need to do this at our hospital.”
“We’re implementing this with Epic for Sepsis.”
“We’re doing this with Epic for Readmissions.”
“Hospitals aren’t doing this yet.”
To my surprise, no sessions or posters at ANIA18 focused on machine learning or multiple data source prediction for value-based care. As evidenced by the application of the RoR tool and what we learned about nurse end-user satisfaction and experience with our solution, KenSci has an immense opportunity for nurses to leverage data for prediction. There is a great need for nurses to how to learn about, embrace and adopt ML solutions in their organizations. Informaticists are our champions in hospitals and clinics and can be super users who will help nurses select and implement KenSci solutions. We hope to return to the American Nursing Informatics Association conference in 2019 and take the stage with a customer to highlight successful implementation, uses, and outcomes from our ML solutions for patient-centered care.
