A Yardstick To Improvement
June 25, 2019
Robert A. Rosenbaum, M.D., FACEP; Mike Thompson; and Mike Taigman
There are few things in our world more gratifying than meeting someone you’ve helped resuscitate from cardiac arrest. Increasing the percentage of people who survive is obviously good for the people who survive, and it’s also good for the bystanders, dispatchers, EMTs, paramedics, nurses, and physicians who are part of the system of care.
A few EMS systems have made such dramatic improvements in their resuscitation rates that folks have shifted from expecting that most cardiac arrest patients will die to believing that most of them will survive. The path to saving more lives involves dozens of small improvements that collectively result in better outcomes.
For the dispatch part of the process, key leverage points for improving resuscitation rates are increasing the percentage of arrests recognized by the calltaker, shortening the time to getting someone’s hands on chest, and overcoming barriers to bystanders starting CPR. The team in New Castle County (Delaware, USA) is using the Institute for Healthcare Improvement’s Model for Improvement as the framework to improve their performance.
Delaware has a population of just under a million people with nearly 600,000 living in New Castle County. New Castle County Dispatch Center EMDs provide all pre-arrival and post-dispatch instructions in New Castle County and is the busiest call center in the state. It’s an IAED™ Accredited Center of Excellence. Staff members use ProQA® to handle about 75,000 medical incidents per year, of which 650 are workable cardiac arrests. They have seen an increase in the number of cardiac arrests each year, primarily because of opioid overdoses.
Early 2018, the emergency communications leadership team and EMS medical director in New Castle County chartered a project to improve bystander CPR. The project aims were increasing the overall rate of bystander CPR from 30 percent to 45 percent, increasing witnessed bystander CPR from 40 percent to 55 percent, and shortening the time for telecommunicators to recognize cardiac arrest. They hoped to complete the project in 12 months.
The outcome measure for this project is the percentage of cardiac arrest patients that are discharged from the hospital.
Process measures are those things that if done well will cause improvement in the outcome measures. For this project they include:
- Time from the first ring of the 911 phone to recognition that the patient is in cardiac arrest. This is captured in ProQA and by doing an audio review of the call.
- The percentage of patients in cardiac arrest who receive bystander CPR.
- The percentage of patients with a witnessed (not by EMS) cardiac arrest that receive bystander CPR.
Ideas for change
Too often systems have one favorite idea that they think will make things better. They implement that idea without testing, and if it fails to produce the results they hope for, they have nothing else to try. New Castle County had several ideas or theories about what to do to inspire change.
- Implementation of a 100 percent call review with feedback of all calls where the Telecommunicator CPR bundle was used will result in better performance.
- Implementation of the two question “No-No-GO!” model would shorten time to recognition of cardiac arrest, shorten the time to dispatch of resources, and shorten time to initiation of CPR instructions and first compressions.
- Providing rebuttal coaching to callers who were reluctant to start CPR would result in a higher percentage of people who received bystander CPR.
- Having supervisors monitor calls in real time would allow them to provide ideas in the moment that would improve performance.
- Having the system medical director provide dispatch continuing education on shift in the dispatch center would increase the percentage of new ideas that were implemented.
- Direct education from the medical director to EMDs on the importance of pushing the caller to start CPR and the reassurance that applying this pressure was OK. The medical director even offered to buy lunch for the first calltaker who had a family member complain that they were too aggressive with pushing the family member to do CPR
The medical director began by recruiting the communications chief and other leadership to participate in the project team. They asked each person for a full, “Are you in?” commitment before they began. Even though there were lots of distractions during the time this project was implemented, including changing to a new CAD, they stayed focused on making improvements.
This team extended the participation to all shift supervisors to learn how to review a call using the CARES T-CPR metrics and data dictionary. The medical director was invited to meet with each platoon to review the aim and potential value.
The ideas for change were tested using Plan, Do, Study, Act (PDSA) cycles, and the results of each test were shared with the entire team. Communications were written carefully to avoid blame or criticism. The team focused all their feedback and communications squarely on the patients that they care for.
Initially, the calltakers had lots of concerns about the “No-No-GO!” model of questioning and how that was going to work. They also had lots of concerns about dealing assertively with people who refused to do CPR. Rather than pushing back and trying to overcome the resistance, supervisors, the performance improvement officer, and medical director listened to their team and collaborated on small tests so that they could learn together what would work and what would not. The medical director and performance improvement officer showed up for every crew on every shift to explore their concerns and gather their ideas for how to test their change ideas.
The team members openly shared all of their performance data with everyone. They regularly updated their charts and graphs throughout the project such as this Pareto analysis of the barriers to doing CPR:
They “benchmarked” the information from the Resuscitation Academy for best practice ideas to implement in their system.
While the project is ongoing, they have made significant improvements in the time from first ring of the phone in their dispatch center to recognition that a patient is in cardiac arrest. They cut that time in half and have sustained their improvement for nearly a year.
Using the Institute for Healthcare Improvement’s Model for Improvement with solid measurements and feedback to your team, it’s possible to make measurable improvements in the things that matter.
The authors of this article would like to thank the Project Team of New Castle County Emergency Communications Division Chief Jeff Miller, Assistant Chief Donald Holden, PI Officer Robert Snyder, and IT Officer Jay Richwine. Additional thanks for case review and data collection by Supervisors Jamie Rosseel, Christopher Williams, Robert Kerr, and Scott Dunkelberger and to all personnel of NCC Emergency Communications for their commitment to improving patient care.
This project was mentored by Ernesto Rodriguez, EMS Chief, Austin, Texas, as part of NAEMSP’s yearlong Quality Improvement and Safety Course!