Preventing All Intensive Care Unit Harms - What Would It Take

2016 - 1 February - Technological Game Changers
Adam Sapirstein, MD; Alan Ravitz, PhD; Peter J. Pronovost, MD, PhD, FCCM
Three experts discuss purpose and early success of Project Emerge in aiding effective ICU communication.


Ventilator-associated harms are among the most common and dangerous threats faced by patients in intensive care units (ICUs).

Yet when you look at the steps we must follow to prevent these harms, you wouldn’t know that stopping them—really stopping them—is a high priority for policy makers, technology companies, hospital managers and clinicians. In fact, it would seem that typical ICUs are designed to make it much harder than necessary to do so. New technologies, such as electronic medical records (EMRs) and medical devices, are not well integrated either with each other or into caregivers’ work flows. These failures of integration and work flow design cause inefficiencies and practice failures that can lead to harm and unintended outcomes.

Is the patient’s bed at an angle of 30 degrees or greater? We check the notoriously imprecise angle-finder on the bedrail, if one exists, or use a makeshift paper protractor.

Has the ventilator been set to the optimal tidal volume? We must make calculations, often by hand or our heads.

Have spontaneous awakening and breathing trials been performed? We review the narrative in physician notes or different flow sheets for the answers.

It goes on like this for several more evidence-based practices, taking clinicians on an inefficient scavenger hunt to different corners of the EMR, to medical devices or to checking the patient visually.

Ventilator-associated harms are a microcosm of the challenge of designing a safer ICU. Devices, monitors and health information technology do not communicate with each other, putting the onus on the clinician to piece together the full patient story. Could we introduce a new checklist, engage staff in preventing one or two harms, and achieve lower rates of these complications? We probably could. National quality improvement projects have achieved improvements in this area. For instance, today, central-line-associated bloodstream infections (CLABSIs) are more than 80% lower across the United States in all types of ICUs compared with 1999 rates, before the Institute of Medicine’s publication of To Err Is Human.(1)

But there are many other common harms, such as delirium and blood clots, that ICU patients may face. Each comes with a similar checklist. If we were to follow the evidence, ICU patients may require up to 200 preventive therapies a day. In today’s under-designed ICU, it is impractical to think that overburdened staff can follow these steps all the time.

It will take new ways of thinking and designing to create ICUs in which we can target all harms at the same time, rather than just one at a time. At Johns Hopkins School of Medicine, we wished to prove that systems engineering techniques can be used to implement such a system of care in the ICU—the ICU of the future. We were fortunate to partner with the university’s Applied Physics Laboratory (APL) and the University of California at San Francisco (UCSF), and to have the support of the Gordon and Betty Moore Foundation. Our project goal seemed simple—create a new system, Project Emerge, to eliminate seven preventable harms in the ICU. Creating Project Emerge required both a systems approach and the tools and best practices of systems engineering. The Project Emerge team is a diverse group that includes 18 disciplines and was guided by systems engineers from the APL.

At the heart of Project Emerge is an application that pulls data from the EMR, devices and equipment, and displays it on an easy-to-understand harms monitor. Project Emerge is typically accessed through a bedside tablet that allows any clinician to look at a list of all ICU patients. If there is a red check mark next to a patient’s name, the clinician can drill down to see exactly for which harms the patient is at risk, which therapies the patient should receive to prevent the harms, and whether the patient has received those therapies. The tool was introduced to the surgical ICU at the Johns Hopkins Hospital in July 2014; early results are promising. Five preventable harms that used to take about 20 minutes to check are now checked with greater accuracy in just a few minutes with Project Emerge.

The novel visual display of Project Emerge is designed to be intuitively clear to enhance the ICU team’s situational awareness. This awareness is similar for each of the targeted harms. For instance, we no longer need to dig through the clinical documentation to see how long a central line has been in place. The app clearly displays that data, prompting regular discussions about whether the line should be removed to decrease infection risk.

Now, instead of a clinician going to the patient’s bedside to check the bed angle, a sensor takes the measurement and communicates with the app. The clinician can easily see when the head of the bed dips below 30 degrees. Rather than forcing clinicians to calculate tidal volume with pencil and paper, the system makes the simple calculation by pulling in data from the EMR. Spontaneous awakening and breathing trials, no longer hidden in clinical narratives, are among other pieces of data pulled into the app and displayed where they can be easily found. These examples are among dozens or hundreds for which multiple processes and systems must integrate to reliably ensure positive outcomes. And, as in the bed example, technology is a tool that can be used to reduce risk and elevate reliability to prevent many harms.

We hope to not only eliminate complications in ICUs, but also to free up clinicians’ time to allow them to devote more of it to direct patient care. This may include more in-depth discussions with patients and their loved ones about their care goals.

Project Emerge also targets harms that stem from lack of patient and family engagement. The system is deliberately designed to enhance respect for the patient. Patients and their loved ones have their own tablets, giving them opportunities to share information about themselves, as well as their care goals. Project Emerge coordinates communication between the clinician and patient/family tablets to inform and prompt discussions.

Today, a project that reduces a single patient harm is a victory in flawed healthcare systems. While this lifesaving work is important and sorely needed, the future potential for the ICU rests in creating a system that helps clinicians reduce harms, rather than hindering them from doing so.

The early success of Project Emerge at Johns Hopkins Hospital is exciting. In November 2015, UCSF launched their version of Project Emerge to gather evidence of wider applicability. The goal for Project Emerge is to become a self-sustaining ICU care system. This requires a path to commercialization. The Project Emerge team has been able to rely on the support of the Gordon and Betty Moore Foundation to give proof of concept and is now engaged with Microsoft Corporation to create a commercial version of Project Emerge. It is our hope that this next generation of Project Emerge will soon be available for any ICU.

​1. Pronovost PJ, Cleeman JI, Wright D, Srinivasan A. Fifteen years after To Err is Human: a success story to learn from. BMJ Qual Saf. 2015 Dec 15. pii: bmjqs-2015-004720. doi: 10.1136/bmjqs-2015-004720. [Epub ahead of print].