Penn Medicine (Philadelphia, PA) has been selected as the winner of ECRI Institute's 13th Health Devices Achievement Award for its development of a computerized dashboard and clinician alert system that uses electronic health record (EHR) data to promote more rapid weaning of patients from mechanical ventilation and sedation. The ABC application, as the computer program is called, leverages the EHR to provide a meaningful, real-time display of sedation and ventilation protocol data.

The dashboard allows clinicians to view the barriers to ventilator weaning, the sedation depth and medication doses, and automated screening results. The alert system prompts clinicians when action is needed. The ABC app has helped reduce the duration of mechanical ventilation by more than 24 hours compared with baseline data, and it has led to a corresponding reduction in ICU and hospital lengths of stay.

The Health Devices Achievement Award recognizes outstanding initiatives undertaken by member healthcare institutions to improve patient safety, reduce costs, or otherwise facilitate better strategic management of health technology. ECRI Institute announced Penn Medicine as the winner of the 13th award in May 2019. For details about the other submissions that achieved recognition, see The Health Devices Achievement Award: Recognizing Exceptional Health Technology Management.

ECRI Institute congratulates the project team members: Barry Fuchs, MD; Mike Restuccia; David Do, MD; Mike Draugelis; Venkat Panchanadam; Corey Chivers; Ann Huffenberger, DBA; Steve Gudowski; Kevin Pollock; Katherine Choi, MD; Yevgeniy Gitelman, MD; Shivan Mehta, MD; Margie Pierce; William D. Schweickert, MD; Brian Anderson, MD; Andrea Gabrielli, MD; Julianne Jablonski.
 

The Challenge

To shorten the time patients stay on mechanical ventilation, both to improve patient care—for example, by reducing the risk of complications—and to reduce the length of stay in the ICU.
 

The Context

Mechanical ventilators are a lifesaving technology, but their use is not without risks for the patient. Complications can include ventilator-associated pneumonia, ventilator-induced lung injury, deep venous thrombosis, pneumothorax, gastrointestinal bleeding, and sepsis. To minimize the risk of complications, clinicians seek to wean patients from mechanical ventilation as soon as possible.

Additionally, the longer a patient remains on a ventilator, the longer that patient needs to stay in the ICU. Consequences of longer ICU stays include:

1. Increased costs—Care in the ICU is expensive to provide.

2. Reduced bed access for other patients requiring ICU care—A lack of available ICU beds can affect other patients within the hospital, and it can prevent the hospital from accepting patients from referring institutions.

Several years ago, Penn Medicine found that its patients remained on mechanical ventilators for 12 to 24 hours longer than expected, compared with external benchmarks. When investigating this finding, the health system discovered several contributing factors:

1. Delays by nurses in reducing sedative medications

2. Delays by respiratory therapists (RTs) in reducing ventilatory support

3. An inefficient and complex screening process for RTs to assess patient readiness for a ventilator weaning trial

a) This was a manual process performed only once a day.

b) Thus, if a patient wasn't screened or failed to pass the screening trial, the patient would not be screened again until 24 hours later.

In addition, the reports that clinicians received about these care processes included information that was outdated and not actionable.
 

The Process

To address these challenges, clinicians teamed up with Penn Medicine's Department of Data Science, along with its Center for Health Care Innovation and its Information Services department, to develop a computer program (app) that would:

1. Enable continuous screening of patient eligibility for a reduction in sedative medications and for weaning from the ventilator

2. Prompt staff to take the actions needed to promote faster weaning of patients from sedative and ventilatory supports

3. Provide compliance data to the ICU team and local quality managers

The app developed by the team leveraged information stored in the EHR to provide clinicians with real-time access to meaningful ventilation and sedation protocol data. Named the ABC app—for awakening and breathing coordination, which are core components of an ICU bundle of care used at Penn—the app includes two key components:

1. A dashboard that reflects the live status of patients at any time—based on current vital signs, ventilator settings, sedative depth, and medications—and that highlights the screening results

2. An alert system that sends simultaneous text alerts to the mobile devices carried by bedside staff (RTs, nurses), who may not be looking at the dashboard when opportunities are displayed. The alert system relies on current data to provide relevant and actionable alerts—specifically, to notify respiratory and nursing staff when interventions are required to move patients through the weaning process. For instance:

a) The RT is alerted when a patient meets criteria to undergo a ventilator weaning trial.

b) If that same patient is oversedated (based on the documentation of sedative infusions and their sedation score), the nurse and provider are alerted with advice to stop the sedation.

Core to the functioning of the app are real-time data streams from the EHR. The EHR vendor (Epic) enabled real-time streaming of HL7 clinical data, which was stored in a time series format and aggregated and normalized. This database was leveraged for the in-house development of the web dashboard and notification system.

Similarly, the app relies on clinical decision support (CDS) rules—based on Penn Medicine's complex ICU treatment protocols and input from clinical experts—to guide the data analysis, display, and transmittal. The "awakening and breathing coordination" component of the ICU bundle of care, on which the app is based, is supported by evidence from a multicenter ABC clinical trial. Penn Medicine participated in that trial, which showed that coordinating sedation and ventilation protocols improves patient outcomes.

App developers tested the CDS algorithms and the alerts for "proof of concept" by sending the alerts to nurses on Penn Medicine's telemedicine team. These nurses, located off-site in the Penn E-lert center, independently validated that the alert was appropriate. Penn Medicine then implemented the system in the medical intensive care unit (MICU) at the University of Pennsylvania, which cares for one of the health system's most complex and critically ill patient populations. 
 

The Results

The Penn Medicine team described several measures of success associated with implementing the ABC app:

1. The health system observed a statistically significant reduction in the duration of mechanical ventilation. The duration was reduced by more than 24 hours, compared with the baseline period.

2. A corresponding significant reduction in ICU length of stay was also observed, as was a trend toward a significant reduction in hospital length of stay.

3. The dashboard and alerts were well received by ICU staff.

4. Nurses stopped sedative medications faster during the intervention period and RTs performed ventilator liberation trials sooner, both of which contributed to more rapid extubation and ICU transfer.
 

Key Takeaways

1. The ABC app helped Penn Medicine address what the team members consider to be one of the biggest challenges in healthcare today: translating important evidence-based care processes to the bedside. That is, putting the evidence into practice. The Penn Medicine team notes:

a) The medical literature includes strong evidence to support widespread use of ventilator and sedation weaning protocols, showing that such protocols favorably impact patient outcomes. However, many institutions have not implemented these protocols effectively.

b) The ABC app facilitates staff compliance with the weaning protocols, as well as documentation of these care processes.

c) Additionally, the app provides an infrastructure with which to implement other evidence-based clinical protocols across the health system to achieve better patient outcomes.

2. The collaboration of a multifaceted production team, involving relevant leaders and key departments across the health system, was paramount to the program's success. For example:

a) Clinical experts conceived the CDS rules and alerts.

b) The Department of Data Science programmed the rules engine.

c) Penn Medicine's Center for Health Care Innovation and its Information Services department designed and built the electronic dashboard.

d) Frontline caregivers in the telemedicine program helped test and validate the system.

3. The app's ability to provide immediate feedback about the documentation of, and adherence to, ICU protocols was judged to be a key feature in improving the outcomes of ICU patients.

a) The health system found tremendous power in having meaningful, real-time data and CDS available to bedside clinicians, quality administrators, and the entire team.

b) Whereas most EHRs function as data repositories for documentation of the care process, the ABC app makes use of that technology—real, meaningful use—to facilitate the delivery of care.