Editor’s Note: We at Healthcare Informatics were once again ecstatic with the exceptional quality of the submissions we received from innovating patient care organizations across the U.S. In addition to the four winning teams this year (whose stories will be posted throughout this week), our editorial team also selected several runners-up. Below, please find descriptions of the initiatives of the 14 teams whom we have awarded semifinalist status in this year’s program.
Centre for Addiction and Mental Health (Ontario, Canada)
Improving patient care through achievement of HIMSS EMRAM Stage 7
In May 2014, CAMH implemented a clinical information system using a big-bang approach with an integrated team of clinicians, information technology, and other staff. But after implementation, CAMH noted a lack of clinical practice standardization. A new initiative emerged that included work to refine the inputting of clinical documentation to the EHR, the development of electronic whiteboards to display and manage assessment and risk factors, leveraging data to inform improvement initiatives, and many other requirements as defined by the HIMSS EMRAM (EMR Adoption Model) Stage 7 criteria.
CAMH is the first academic teaching hospital to achieve HIMSS Stage 7 in Canada; this achievement is a milestone in both the Canadian and international health landscape. Now, more than 99 percent of CAMH clinically-relevant documentation is completed directly within the EHR and CPOE (computerized provider order entry) rates have been over 90 percent since December 2016. What’s more, the creation of a suicide risk dashboard has led to 90 percent of patients having a suicide risk assessment completed within 24 hours of admission.
Cleveland Clinic (Cleveland, Ohio)
An enterprise imaging service
The goal of the enterprise imaging service is to provide a comprehensive longitudinal medical record through incorporation of all medical images into a single archive. Through a universal viewer, the archive is integrated with the EHR and provides a foundation for image distribution to all caregivers throughout the enterprise. The archive also serves as a foundation for image sharing. Implementation required a comprehensive assessment of all image generating equipment throughout all hospitals and outpatient centers.
The Clinic’s officials say that the establishment of an enterprise imaging program has led to the consolidation of imaging archives throughout the health system. Images which were not previously easily accessible are now readily viewable through the EHR (electronic health record) with access points both within the firewall and from home. To date, 11 different service lines and more than 440 pieces of image generating equipment outside of radiology have been integrated.
Compass Medical (Massachusetts)
Annual wellness, chronic care management and quality outcomes
By leveraging new information technology, Compass Medical has been able to follow proven population health management and care management principles, allowing patient care leaders to identify and target specific population groups, stratify and prioritize care gaps and engage and individualize care plan activities. In 2016, for example, Compass Medical was able to identify and target more than 14,000 Medicare patients that were struggling to manage their chronic health conditions and needed a more personalized and comprehensive care plan. One year later, Compass Medical developed and launched a new Chronic Care Management Program to help engage with and closely manage Medicare patients that suffer from two or more chronic health conditions. With the help of its EHR and big data platform, Compass Medical positioned itself to automate many of the workflows for care management nurses.
The Annual Wellness Visit (AWV) is another example of a preventive care service that has been positively affected by leveraging IT. In 2017, national trends suggested utilization of AWV were still hovering around the low 20-percent range with the highest performing state reaching 35 percent. Utilizing EHR-based patient engagement campaigns for increasing focused outreach, incorporating a team based care model with scribes, and creating standard work processes for reducing provider burden have helped Compass Medical reach 57 percent AWV utilization for its Medicare eligible population by the end of year 2017.
Duke University School of Medicine (Durham, N.C.)
A NICU discrete event simulation model
Duke’s neonatal clinicians care for more than 800 babies each year in the Duke Neonatal Intensive Care Unit (NICU). Although the majority do well, about 40 babies do not survive. How could they improve outcomes and save lives? Duke’s neonatal research team partnered with analytics company SAS to create an analytics-based model of Duke Children’s Hospital’s Level IV neonatal intensive care unit. The result was the creation of a discrete event simulation model that closely resembled the clinical outcomes of Duke’s training unit, which was validated using data held back from the original model, which also closely tracked actual unit outcomes.
The model uses a vast resource of clinical data to simulate the experience of patients, their conditions and staff responses in a computerized environment. It creates virtual babies experiencing care within a simulated NICU environment, including virtual beds staffed by virtual nurses. The research team attests that they cannot find any evidence of discrete event simulation modeling being used in a NICU setting, making this a first in neonatal care.
Houston Methodist (Houston, Texas)
A coordinated care/Medicare Shared Savings Program (MSSP) initiative
Houston Methodist's MSSP program, Houston Methodist Coordinated Care (HMCC), can track and report Medicare patients’ healthcare visits and medical details. The successful execution of the program is a layering of technologies with the foundation being the organization’s integrated EHR platform and a separate population management tool.
The project was centered around six core elements: 1) becoming the first ACO (accountable care organization) in Texas to acquire real-time admission, discharge and transfer (ADT) notification capability that links all health providers; 2) chronic heart failure home monitoring; 3) real-time notification when HMCC patients came into the ED; 4) risk assessments for emergency room visits, hospital readmissions and the need for complex care; 5) same-day appointment facilitation; and 6) care team alerts. In sum, there were 17,000 Houston Methodist patients in HMCC in 2017, year-to-date, with 105 participating physicians. Total healthcare cost savings year-to-date are more than $1.3 million, according to officials.
Indiana University Health (Indianapolis, Ind.)
FHIR HIEdrant: making big data actionable at the point of care
One of the difficult challenges for many HIEs (health information exchanges) is the time and effort that it takes to reach out to a second system to search for needed data at the point of care. As such, the goal at IU Health was to develop an application within the clinical workflow that will, at the click of a single button, bring back data to that workflow relating to the patient’s chief complaint from the HIE.
The first phase of this project was building the framework and the mechanisms to make this a possibility and apply it to a single context: an emergency department patient with chest pain. Leaders at IU Health are utilizing the Fast Healthcare Interoperability Resources (FHIR) standard to communicate out from the IU Health Cerner EHR to the HIE to retrieve five specific data elements that are germane to caring for a chest pain patient in the emergency department and understanding their risk. Within the workflow, the clinician is being presented the most recent: ECG, cardiology note, discharge summary, catheterization report, and more. According to IU Health officials, this is the first FHIR-based application that directly accesses an HIE and delivers context-specific data about a patient directly to the clinical workflow.
Johns Hopkins Health System (Baltimore, Md.)
inHealth precision medicine initiative
The precision medicine initiative at Johns Hopkins Medicine and University–inHealth–seeks to improve individual and population health outcomes through scientific advances at the intersection of biomedical research and data science. Through a collaboration of The Johns Hopkins Applied Physics Laboratory (APL), and Johns Hopkins Medicine (JHM), inHealth is building a big-data precision medicine platform with the goal of accelerating the translation of insight into care delivery.
The first result of this broad, multidisciplinary effort was the successful creation of two Precision Medicine Centers of Excellence (PMCoE) focused on multiple sclerosis and prostate active surveillance. The organization’s Technology Innovation Center has developed applications to garner new data and learnings from clinical practice and feedback into discovery. Physicians have begun using the discovery platform to facilitate conversations with their patients about their treatment options and risks. The experiences of these centers will lead the next wave of PMCoEs, expanding the utility of the platform.
Lakeland Health (St. Joseph, Mich.)
Something wicked this way comes
Leaders at Lakeland Health set three core cybersecurity goals: (a) put risk management and cybersecurity near the top of health system leadership agenda; (b) use innovative strategies and tools to execute the cybersecurity program; and (c) shift focus from fear to clinical integrity. The cybersecurity program covered the hospitals, clinics, home care, hospice and all the different legal entities which comprised the health system. In order to ensure strategic direction and alignment, a steering committee was set up which met every two weeks.
The cybersecurity program execution was focused on three work-streams—process, technology and team members. In the process work stream, execution covered implementation and audit of policies and procedures, risk assessment and HIPAA (Health Insurance Portability and Accountability Act) compliance, and a monthly information security executive dashboard which was reviewed by the steering committee. Despite this continuing threat, the cybersecurity program delivered strong results in different areas, including: more than 100 business associate agreements (BAA) were signed; annual HIPAA risk assessment and remediation plans were put in place; the initial internal phishing campaign eventually lowered the click rate to 10 percent; there was a five-fold increase in the suspicious emails forwarded to the security team; and more than 1,000 laptops were encrypted.
Lexington Clinic (Lexington, Ky.)
Development of a direct-to-employer network
Costs of certain services often vary dramatically between providers, so by selectively designing benefits to increase cost-sharing at providers who provide more expensive care, enrollees are incentivized to see the more efficient providers who provide care at a lower cost, reducing average overall expenditure. Savings can then be passed on to the employer. In this project, implementation was examined with an organization with a self-funded insurance model. Steering beneficiaries toward a tighter network of providers resulted in significant overall reductions in expenditure while improving the health of the overall employee population. Rather than limiting their employee health plan to a lower percentage of area providers like most similar plan designs, the employer entered into a direct-to-employer program with a local, multispecialty physician group: Lexington Clinic.
A key component of a direct-to-employer plan is population health. Lexington Clinic was able to utilize analytics software to deliver value to the employer by implementing high cost/high utilization analysis, undetected chronic disease engagement, and ancillary modality management. Lexington Clinic also determined that there were specific interventions that could be made at critical junctures in the care continuum of the employee population. These interventions would be designed to prevent health issues before they arise, reducing future expenditures and worsened health outcomes. Via the Lexington Clinic premier network, the employer demonstrated a clear reduction in aggregate expenditure from the 2015 to 2016 time period of more than 4 percent.
Lutheran Medical Center (Wheat Ridge, Colo.)
An app for staff engagement
At Lutheran Medical Center, it became a priority to redesign the way in which the staff was engaged. The organization started to use an anonymous crowdsourcing platform in 2016 with the goal to create recipes for success that would help leaders in the organization ask the right questions through an anonymous tool to enhance engagement. Using the tool has established a venue for staff to engage in problem solving and design ideas on their own terms in an anonymous way where all can follow along in the conversation in real time. The application/website started its use in the pharmacy department as a means to understand low engagement scores. This tool allowed for all staff to be involved while not taking them away from their daily duties.
Lutheran Medical Center was in the 11th percentile when it came to staff satisfaction only two years ago, but now ranks in the 43rd percentile compared to the national average. Having the ability to get staff buy-in before a change happens has been critical in impacting staff satisfaction. Before, only those invited to certain meetings had the opportunity to voice their opinions; now everyone can be reached with a single email or app use. It has been used for solving several clinical problems as well, such as how to design a “cord-free” patient room, and how to transport oxygen tanks around the hospital.
Mercy (St. Louis, Mo.)
Using NLP for heart failure EHR documentation
The goal of this project was to use NLP to extract key cardiology measures from physician and other clinical notes and incorporate the results into a dataset with discrete data fields. This dataset would then be used to obtain actionable information and contribute to the evaluation of outcomes of medical devices in heart failure patients.
Three key measures that are commonly stored in clinical notes and not available in discrete fields include ejection fraction measurement; patient symptoms including dyspnea, fatigue, dizziness, palpitations, edema and pulmonary congestion; and the New York Heart Association (NYHA) heart failure classification. Mercy patients had 35.5 million clinical notes from both inpatient and outpatient encounters that were extracted, processed and then loaded onto an NLP server. NLP queries were developed by a team of Mercy data scientists to search for relevant linguistic patterns and then evaluated for both precision and recall. The use of NLP in this project facilitated the extraction of vital patient information that is not available in any discrete field in the EHR. Without the ability to track the changes in these three essential measures, it becomes much harder to identify the point of disease progression which is a crucial factor for the evaluation of current treatments and could inform future interventions, according to Mercy officials.
Mosaic Life Care (St. Joseph, Mo.)
Revenue management analytics dashboard
Mosaic Life Care provides healthcare and life care services in and around St. Joseph, Missouri and the Kansas City Northland area. The organization’s finance and revenue cycle teams faced challenges with data silos that required caregivers to manually obtain information from disparate systems and manually collate information, subjecting the process to human error, inconsistent processes and concerns about data accuracy.
With the goal of developing a flexible “source of truth” dashboard, the enterprise data warehouse team developed an integrated revenue management analytics solution with a front-end dashboard by leveraging the core EDW solution and architecture platform to extract data from the best of breed systems. Through the new dashboard, financial analysts and management teams can perform analysis and predict future trends. As a result, the dashboard enables real-time, data-driven business decisions inclusive of multi-disparate systems within a single unified platform.
NYU Langone Health (New York City, N.Y.)
Value-based medicine to improve clinical care
The goal of the project was to leverage health IT tools and related workflows to improve the value of inpatient care. Finance collaborated with the project’s physician champions to identify variations in care both internally and compared to benchmarked external institutions. The project’s physician champions collaborated with IT physician informaticists and IT project teams to design interventions to both reduce cost and improve clinical care.
The suite of interventions included: electronic clinical pathways; blood protocols; intravenous (IV) to oral (PO) medication changes; and lab ordering enhancements. Electronic pathways were created for heart failure, colon surgery, and pneumonia, and blood ordering clinical decision support and analytics were built. These projects realized significant two-year savings, including: electronic clinical pathways: $12.9 million; lab modifications: $3 million; blood utilization: $2.9 million; and IV to PO: $2.2 million.
Penn Medicine (Philadelphia, Pa.)
Standard clinical iPhone effectively enhances patient care
In January 2016, Penn Medicine met with Apple engineers to develop an economic and efficient full configuration Standard Clinical iPhone (SCiP) to work with Penn’s mobile device management tool while leveraging Apple’s Device Enrollment Program (DEP) and Volume Purchasing Program (VPP). Using this method saved the organization 975 man hours in its initial deployment using DEP streamlined setup (15 minutes versus one hour for each iPhone). Pushing additional apps to devices without needing an Apple ID and password for download or manual touch also made the implementation efficient. By implementing this project, it placed a vital tool into caregivers’ hands, officials say.
Just by using the secure texting, clinicians were able to coordinate patient flow across care settings with multiple providers and mended gaps in communication. One example had the cardiac surgery team on a thread with the patient’s current caregivers. The nurse took a picture of the surgical site and sent it securely with a description, concerned about swelling around the surgical site. The surgical team was able to provide immediate feedback and resolve the issue remotely.