Improving “Door to Balloon” Times and Patient Care – in Some Hospitals, Why Not Every One?
Recently, the Wall Street Journal put some focus on the issue of so-called “door to balloon” times in our hospitals. “Door to balloon” is a measure of time from a patient with a serious type of heart attack (“STEMI“) arriving at the hospital (door) to the moment when the arterial blockage that affects the heart is cleared via angioplasty (balloon).
As they say at ThedaCare and other organizations that have used Lean methods to dramatically reduce this time, “time is muscle,” as a faster DTB time means less damage to the heart, faster recoveries, and better patient outcomes (and, undoubtedly, lower costs).
Franciscan St. Francis Health System (where my Healthcare Kaizen co-author Joe Swartz works) has reduced times from 113 minutes to 75 through the development of a new process (“EHART”) and “dozens” of small Kaizen improvements (see page 50). This has led to a decrease in the average length of stay of two days.
One example of an improvement was:
…Nathan Lowder, an ER nurse, developed a Kaizen that saved more than 20 minutes during the typical patient transfer from his ER to the cath lab by having all necessary medications given to the patient while being transported in the ambulance.
As John Toussaint, MD wrote about in On the Mend, ThedaCare has reduced its average DTB time from about 90 minutes (considered the current standard of best care) to just 45 minutes. They are also working to reduce variation in these times. There were a number of improvements made over a number of years to bring this time down – it wasn't one silver bullet solution. One key in this or other “value stream” improvements is to reduce delays between the “value adding” steps of diagnosis and treatment.
ThedaCare changed the sequence of events to improve the overall flow. Changes were made to roles and responsibilities (“standardized work”) to prevent delays – for example, emergency medicine physicians making a determination about the need for the “balloon” instead of having to wait for a cardiologist. They've minimized the delays in getting the patient to the “cath lab” where the procedure is done.
ThedaCare says just about the only thing that could reduce that time even more would be to have the electrocardiogram done in the ambulance that transports the patient, rather than doing it in the hospital. The ECG can be transmitted and read before the patient even arrives. I've seen this with my own eyes when visiting Sweden a few years back. The ECG was read before the patient arrived and they could be sent IMMEDIATELY to the cath lab, bypassing the emergency department.
From the WSJ:
More than half of emergency medical systems in an American Heart Association survey reported having the devices in their vehicles, with 35% of those able to wirelessly transmit the results ahead, enabling hospitals to mobilize the cardiac catheterization team before a patient arrives.
How are hospitals doing against this 90 minute goal?
Researchers who studied Medicare data from more than 300,000 patients at 900 hospitals found so-called door-to-balloon times fell from a median of 96 minutes in 2005 to 64 minutes in 2010. The best-performing hospitals regularly achieved times under 60 minutes, which “may become the new standard,” the study, published last year in the journal Circulation, concluded.
The WSJ suggests, correctly as we see from ThedaCare, that it's possible to do better than 60 minutes. This is a great example of needing to not just meet the minimum of an accepted target – whether it's DTB times or infection rates. We need to strive for the best performance possible, never being satisfied (a key Lean attribute that ThedaCare has in spades).
A Yale cardiologist says:
“Rather than set a new, lower threshold, we are now saying the time to treatment should be as short as possible, so we treat everyone as quickly and safely as we can.” At Yale, door-to-balloon times have been as low as 14 minutes, he says.
It's great that Yale had “as low as 14 minutes” — that shows what's theoretically possible when everything flows perfectly. The challenge is repeating that consistently. Hospital labs can often get test results done in just minutes – when the lab order is “STAT” – but normal tests might take an hour or more. “Lean labs” are able to improve flow so that every specimen flows like a STAT, improving the turnaround time for all.
The WSJ highlights some of the DTB delays:
Rather than wait for a cardiologist to review results, which can take a half-hour, ER doctors are making the diagnosis and activating pager systems that scramble cardiology doctors, nurses and technicians with a single call. They are expected to drop what they are doing if in the hospital. During off hours, on-call teams are expected to rush to the hospital, usually within a half-hour, to prepare the cardiac catheterization lab where angioplastyâ€”also known as percutaneous coronary interventionâ€”is done, often followed by placement of stents to keep the artery open.
The article discusses the improvement at University Hospitals Case Medical Center in Cleveland, which has gotten times down to a median of 47 minutes (meeting the 90 minute goal 100% of the time). They spent $50,000 in training and equipment for ambulances to have portable ECG machines.
The hospital set minute-by-minute goals for individual team members and continues to provide feedback on how well they do for each patient.
That's an important characteristic of a Lean process (although we don't know that was the method used there) — measuring results and giving (constructive) feedback. When we don't meet our target times for a step, we should analyze the process rather than blaming the people involved, asking why times were so long so we can improve the process and future performance.
In the name of DTB improvement, the key, again, is reducing delays, not doing the work faster. There are concerns about “false alarms,” which waste time and add cost to the system. ThedaCare has “false alarm” rates below the national average (which is about 20%, I think), as Toussaint has told me.
From the WSJ:
It's possible a patient rushed to the lab would turn out to have something other than a serious heart attack. At the University of Michigan Health System, a program to improve door-to-balloon performance cut the median time from 67 minutes in 2007 to 55 minutes in 2011. The false-alarm rate increased to 40% of all cases from 15%, according to a study by cardiologist Geoffrey Barnes. Such false alarms can be a drain on staff and a poor use of resources, he says.
This has led the U of M to make further adjustments to their process.
It's great to see issues like this outside of medical journalist and healthcare press. These types of DTB improvements are something that EVERY hospital can achieve. It doesn't require more highly skilled physicians or more expensive equipment to get the times down from 90 to 45 minutes.