It has been five years since Martin Doerfler, MD, came to North Shore-Long Island Jewish Health System as senior vice president for clinical strategy and development and associate chief medical officer. When he started, the 18-hospital system based in Great Neck, NY, had a sepsis rate that was above the national average. Healthgrades noted the system was “underperforming” in the area. Before he started, sepsis was the largest single contributor to mortality in the health system. They created a task force to try to deal with it, he says.
As a critical care physician, sepsis was something that Doerfler had seen plenty of, too much of. While he was not brought in specifically to address that problem, he was as keen as anyone else at North Shore to try to get a handle on it, and soon was deeply involved, he says. His work, and that of the rest of the team involved in reducing sepsis in the North Shore system, paid off. This year, the health system was given the National Quality Forum/Joint Commission John M. Eisenberg Award for Patient Safety and Quality (other winners included Mark Graber and the American College of Surgeons, as reported in the April issue of Hospital Peer Review).
“Most of the research on sepsis has been focused on septic shock,” Doerfler explains. “That is when you can see it.” Mortality is as high as 50% when sepsis moves to severe sepsis and septic shock.
When he started at the system in 2009, the work to reduce sepsis had begun, but had not had much traction. “There was a lot of controversy about the best way to care for people when they progress to septic shock,” he says. “Our critical care folks were in a debate with the emergency department. If the ED sees septic shock, they go right to the intensive care unit. But the guidelines that were out then referred to sepsis, even though the items that were in the guidelines were things you mostly saw once sepsis progressed to septic shock.” The emergency physicians rightly noted that sepsis, severe sepsis, and septic shock are all different things.
“A 22-year-old college student with a severe sore throat and a high heart rate has all the elements required to meet sepsis according to those old guidelines from Surviving Sepsis,” he says. “That kid, who looks fine and is being given ibuprofen and sent home, could have been sent to the intensive care unit under those guidelines. You can’t apply those rules to everyone.”
Doerfler’s boss at the time, the chief quality officer, asked him to get involved in the project, and when he joined the team they started to modify those guidelines. “That was the key to us moving forward,” he says.
Time zero
One of the first things that had to change was the definition of “time zero”. “The ED would tell you if a patient comes into the department in shock, you start the clock from when the patient showed up,” he explains. But let’s say there is a 55-year-old with pneumonia, a fever, and a cough. The nurse checks his vitals and they are fine and he goes into a cube. If you have an hour to get something done, when is the zero time? Under the National Quality Forum and Surviving Sepsis rules, time zero is when the nurse first saw the patient, or triage time, if the patient is very sick. For other patients, it is after certain vitals are taken by the nurse and the doctor orders a serum lactate test. That indicates a physician is considering a diagnosis of sepsis.
“There are about a dozen things that you have to do in the first six hours of a sepsis patient coming in,” he says. “Some have to be done in 12. But there are four things you need to do in the first three hours. We decided to ignore everything but those first four items, as they were the most important.”
The four items were getting blood cultures before ordering antibiotics, starting antibiotics within those three hours, sending serum lactate tests, and for patients that meet severe sepsis/septic shock criteria, ensuring they get appropriate fluids.
“Then we started looking at how to ensure those things were getting done, barriers that existed, and how to break through them,” he says.
The lactate test was one of the simpler problems to solve. “We do not just want to send the test out, we want to get it back, so we need the lab to understand that this is a priority and that they should call us with results fast,” says Doerfler.
The antibiotic choice can’t wait several days for a culture, so other factors have to come into play for a physician to make a decision on what bug he or she is treating. There might be X-rays to see if there is evidence of pneumonia, which can take 90 minutes, he says. “So we have to look at what are the things that might slow that process down.”
Strategic partnership
During this process, the North Shore system created a strategic partnership with the Institute for Healthcare Improvement to focus on a couple issues related to the project, and used the IHI’s method for process improvement during the remainder of the project. Doerfler says they also decided to focus efforts on the emergency department, as most cases came in the front door, and did not originate within the hospital.
“We worked on a lot of little pieces,” he says. “How long does it take you to get a patient from coming in the door to getting a central temperature? How can we work to cut a few minutes off of that? If we cut a few minutes off of every step, it adds up. And we continue to do that work.” As an example, they looked at what antibiotics were commonly kept in the ED and made some changes so that the most likely antibiotics a physician would need were available on demand, and there were fewer instances of waiting for someone from the pharmacy to bring a needed drug.
Key to those efforts was asking frontline teams to participate in the project and all its pieces. “They are the people who know where the slowdowns are and where there are opportunities for improvement,” he says.
The results were impressive. In 2009, mortality for sepsis was 32% in the system. Currently, it is 16.1%, although in March it was as low as 13%, Doerfler says. “We see a thousand patients a month that meet the criteria for sepsis, so that is pretty good,” he says. “The goal is to get it to single digits. Intermountain Healthcare in Utah is in the high single digits. They are smaller than we are, but we think it is possible and that we will reach that goal in the next few years.”
Doerfler says they are also pretty reliable on three of the four items that need to be done in those three golden hours. Scores are in the 80-90% range for those metrics for patients that come through the emergency department. They are starting to work on the inpatient sepsis population next, which accounts for about 15% of the total. And they also only hit about the 50% in mark in making sure that the right patients get the appropriate fluids before that 180 minutes are up. “This is the year of the fluids for us,” he says. “We want it to meet the level of reliability the other metrics do.”
Sepsis is a problem at every hospital, whether it is seen and counted and considered or not, he says. “Not everyone is focused on this problem. But just because you do not see it, does not mean you do not have it.”
He says that any hospital can focus on getting the four items of the three-hour bundle done for sepsis patients. “And if you want to focus on just one part, it is the antibiotics. The evidence is unequivocal that this is what will decrease mortality.”
Then work on making all the moving parts run more efficiently. “Empower your nurses, pharmacists, and labs to do this work. They are the true experts. They are the ones who can solve the problems around this at your facility.”
For more information on this topic, contact Martin Doerfler, MD, Senior Vice President, Clinical Strategy and Development, Associate Chief Medical Officer, North Shore-Long Island Jewish Medical System, Great Neck, NY. Email: [email protected]