The 2017 Healthcare Informatics Innovator Awards: Semifinalists

Jan. 26, 2017
Inside, please find descriptions of the initiatives of the nine teams whom we have awarded semifinalist status in this year’s program.

We at Healthcare Informatics were once again elated by the outstanding quality of the submissions we received from innovating teams from across the U.S. In addition to the four winning teams this year, our editorial team also selected several runners-up. Below, please find descriptions of the initiatives of the nine teams whom we have awarded semifinalist status in this year’s program.

Association of Ontario Health Centers

A BI reporting tool supporting 85 organizations

In 2011, the Community Health Centers (CHCs) in Ontario, Canada had a vision to unify its data asset into a single enterprise data warehouse and associated business intelligence reporting tool (BIRT). Supporting 85 independent organizations, the vision was to unify accountability reporting to funders and create a robust self-service analytic environment. This was delivered within a security infrastructure that permits containment of sensitive clinical information but a shared BI environment where all organization users can share and collaborate the ad hoc reports among peers. The solution gives members a holistic view of operations by consolidating key data and presenting it in an integrated and easy-to-analyze manner. What makes this BI solution particularly unique is that it was built based on the needs of the members it serves, officials attest.

Bridging Access to Care

Leveraging IT to integrate trauma informed care in everyday practice

Bridging Access to Care (BAC) aimed to facilitate trauma informed care by enhancing care delivery and augmenting plans of care to incorporate trauma specific objectives. The project plan included: developing a systematic approach using the electronic health record (EHR) system to inform and guide staff in delivering trauma sensitive activities; developing an electronic tool that could capture specific data and trend changes in consumer health outcomes; identifying validated tools that were simple to use, met industry standards and were recognized by payers; and creating a system for sharing information in real-time to facilitate access to treatment and retention in care. The project revealed that systematizing trauma informed care can improve consumer health outcomes. Post-implementation data revealed that 100 percent of clients were screened for trauma (compared to 0 prior to implementation); 37 percent of them were screened positive for trauma. And 92 percent of clients screened were referred to and are receiving seeking safety interventions, while 42 percent of clients that were screened were referred to and are receiving mental health services.

CHI Franciscan

Leveraging virtual care to transform healthcare delivery

CHI Franciscan Health, the Tacoma, Wash.-based health system that consists of eight hospitals and a large network of physician clinics, has a mandate to redesign its healthcare delivery network towards a value-based system serving the Pacific Northwest. The health system’s Care Transformation initiative is a massive project that is patient-focused in scope and has the dual benefit of added value to business and clinical outcomes. In the last three years, the Care Transformation team has grown from only four to a multi-disciplinary and multidimensional service line with more than 150 clinicians and IT professionals. The Care Transformation team consists of a number of providers, nurses, information technology specialists, analysts and others roles and responsibilities, all of which focus around providing care to patients where they are, and eliminating any geographic, time commitment and language barriers in accessing that care.

As a result of this team’s work, CHI Franciscan Health has successfully implemented an impressively broad range of programs and services leveraging technology, such as telemedicine. A virtual urgent care program has helped to save $1.5 million in healthcare costs and 10,511 hours of wait and travel time. A virtual intensive care unit program has helped to shorten ICU stays and increase ICU survivors. The ventilator bundle compliance has improved by 17 percentage points and is consistently staying at goal. A virtual diabetes education program was implemented to complement care services. Since its inception in 2012, 35 patients with diabetes have received high-touch interaction and education from diabetes educators and have demonstrated sustained decreases in their HgbA1C, have lost weight, stopped smoking and improved their activity levels. A regional telemetry monitoring program, the first at a U.S. health system, has enabled CHI Franciscan Health to proactively monitor patients’ heart rhythms and rates for over 2 million hours and engage 24 patients in the first six months from the launch before cardiac issues occur.

CHOC Children’s

Improving pediatric asthma management through population health

As part of Orange, Calif.-based Children’s Hospital of Orange County’s (CHOC Children’s) strategic plan for population health management, the CHOC Pediatric System of Care was formed to better manage pediatric lives. To support the management of the pediatric asthma population, hospital leadership implemented the Cerner Healthe Registries application in CHOC Children’s primary care clinics. Foundational to this project was the development and definition of asthma measures that, when completed on a routine basis, would improve the wellbeing of the patient by keeping their asthma well managed and out of emergency situations. Another important element of the initiative was developing tools to make the status measures available in real-time in the clinical workflow. To accomplish this, a team of providers developed ten defined measures that were built into the electronic health record (EHR). In the primary care physicians’ office, new workflows were rolled out to include a morning huddle with the entire care team where important information about the patients to be seen that day are reviewed including any outstanding measures for the asthma patients. Additionally, in order to track improvement, the project leaders built dashboards to provide clinicians with information on the completion of measures.

As a result of this initiative, emergency department visits for asthma related issues decreased by 18 percent and the hospital saw about $1 million in avoided emergency room costs. Additionally, children with an asthma action plan completed in the past year went from 15 percent to 27 percent. Children with an asthma control test completed in the past year went from 15 percent to 26 percent.                                                       

Geisinger Health System

Moving ahead to create a unified data architecture

The Danville, Pa.-based Geisinger Health System, already renowned across the U.S. healthcare system for its pioneering work in developing consensus- and evidence-based clinical pathways and its population health and care management innovations, has been moving ahead to facilitate further work to improve outcomes and curb costs, through its project to create a Unified Data Architecture (UDA), in order to integrate all of the organization’s analytic platforms. After concepting the initiative, Geisinger leaders decided to create a big data (Hadoop) platform, as the foundation for the UDA, and as the first phase of the project. Within a year, the team established a big-data platform, based on Hadoop and other open-source components. Within that time, Geisinger project team members developed code for a source ingestion pipeline to pull in source data, perform necessary transformations, and load data into various views; and having done so, pulled in all of the source data currently populating the data warehouse (EDW), plus additional sources not in the EDW.

A key element in all this is that Geisinger’s data and IT leaders “plan to deviate from industry standard and the common opinion that big data should augment the EDW, not replace it. We are on our way to proving that we can replace the EDW,” they state. “By running analytics from our Hadoop infrastructure, we have all of the benefits of distributed computing, plus the additional benefits of late binding and the ability to deal with non-discrete data, such as we find in clinic notes.”

Lakeland Health

Developing a hemorrhage risk assessment to improve maternal health

St. Joseph, Michigan-based Lakeland Health has applied electronic health record (EHR) technologies within its OB department at all three of its hospitals to support care providers in identifying warning symptoms in a timely manner to avoid costly complications. As the top two preventable causes of maternal morbidity and mortality are hemorrhage and preeclampsia, Lakeland Health set its sights on designing custom EHR tools to assist in the assessment and prevention of these conditions. Lakeland Health automated and implemented the following four innovative technology initiatives to manage obstetrical hemorrhage—a hemorrhage risk assessment; standardizing Pitocin administration during the third stage of labor; a quantitative blood loss calculator; and an OB emergency narrator hemorrhage risk assessment.

As a result of these projects, Lakeland Health has improved the amount of blood needed by new mothers. Before implementation in January 2015, the OB department used 37 units per 1,000 patients. In May 2016, the department used 24.8 units per 1,000 patients. Additionally, the project leaders have reported improvements in caring for obstetric patients with regard to preeclampsia. As of May 2016, 82 percent of obstetric patients had a completed hemorrhage risk assessment and with 98 percent of patients had blood pressure taken within 15 minutes of arrival and, of those with elevated blood pressure, care providers were notified 100 percent of the time upon the patient’s arrival. As part of this clinical informatics initiative, project leaders have found that process improvement reports and dashboards have been key elements for process improvement.

Texas Children’s Hospital

Working to improve patient alarm management

Looking for a way to reduce alarm noise and fatigue—one of the National Patient Safety Goals of the Joint Commission—leaders at Texas Children’s Hospital (Houston) formed an alarm management team, and reached out to Medical Informatics Corp. (MIC) to perform a baseline analysis to determine their current alarm situation and to come up with a plan. Using MIC’s dashboards, the alarm management team went beyond managing alarms, to improving care and processes. Over time, the scope of the project expanded, with the goal of making all alarms actionable to improve team communication, patient experience, and outcomes.

With participation from key stakeholder groups, including nurses, physicians, IT, biomedical engineering, and quality, and led by the hospital’s CMIO, Eric Williams, M.D., the team used baseline analysis and PDSA (Plan, Do, Study Act) methodology-based processes to reduce alarms and to modify alarm settings across one service area. As part of that process, nurse champions did daily audits of alarms at the unit and patient levels; and because the dashboards used included additional analytics, the nurse champions were able to make process improvements such as ensuring that one particular nurse didn’t end up being assigned the highest-alarming patients. The organization was able to meet NPSG goals in less than six months, and continues to see a reduction in alarm noise, with primary alarms dropping 9 percent over the course of 2016, and secondary alarms dropping 100 percent, and with specific alarms on individual patients dropping between 63 and 88 percent in frequency. The work continues, as PDSA-driven analyses are finding additional opportunities for improving alarm management.

University of California Davis Health System

Integrating patient-generated health data to improve health

The technology aim of UC Davis Health System’s diabetes and better blood pressure initiatives was to leverage patient-connected devices and integrate the patient-generated data points into the EHR, thus enabling clinicians to utilize coaching for behavior change when necessary. The recruited patients were provided an iHealth wireless blood pressure and on-boarded on how to “connect” their device to Apple HealthKit to the patient portal, which would feed into the clinicians view of the patients’ EHR. For the diabetes initiative, more than 1.4 million patient-generated health data points have been integrated into the EHR for clinical review and patient health management to-date. For the better blood pressure initiative, more than 2,700 data points have been made available to clinicians for review and patient health management. Patients participating in the initiatives have provided clinicians with improved visibility into patient health via ongoing data collection; clinicians now have the ability to collaborate with patients using real-time personalized data points.

UF Health Jacksonville

Addressing sepsis with smarter alerts

Clinician leaders at UF Health Jacksonville came together to address a combined set of problems that bedevil patient care organizations across the healthcare system: alert fatigue, and the need to manage the early signs of sepsis. Early detection of sepsis leads to early management which has been shown to decrease mortality; at the same time, as the clinical literature has noted, physicians and nurses are bombarded every day with a tsunami of alerts, which can lead to “alert fatigue,” and can ironically decrease awareness of impending dangers.

Leveraging a tool from Azrael, UF Health Jacksonville clinicians have created a program in which the standard alert interval is one alert per patient per 24-hour period. And if the program sees a worsening trend within a four-hour period along any of the parameters it is tracking, the tool will send another targeted alert. What’s more, the location-based logic embedded in the system allows each area within the hospital to customize the alert-firing threshold. The initiative has had excellent results, with the hospital’s sepsis mortality index (observed/expected ratio) falling from 0.83 in 2014 to 0.84 in 2015, and with its sepsis length-of-stay index decreasing from 1.25 in 2014 to 1.11 in 2015, while the actual sepsis mortality rate also continues to decrease.