IHI faculty member Jim Duncan, MD, PhD, is Professor of Radiology at the Washington University School of Medicine in St. Louis, Missouri. He is also faculty for the IHI Expedition on Eliminating Overuse in Medical Imaging. In this interview with IHI Communications Specialist Jo Ann Endo, Dr. Duncan talks about how ending overuse in medical imaging will reduce harm to patients, decrease waste, and cut unneeded costs.
Q: Why is it so important to eliminate overuse in medical imaging?
A: It causes harms that aren’t offset by benefits. There are three main types of potential harm. First, there’s harm from radiation used for x-rays and CT scans, as well as x-ray guided procedures. Second, there’s harm caused by incidental findings discovered during these exams that don’t have a benefit. These are cases that send people chasing something found on the exam that isn’t quite normal but wasn’t destined to cause a problem. You start doing more scans and more tests to evaluate the findings, thereby adding up harm rather than providing benefit. I think every health care professional has seen at least one case where there were unneeded treatments that caused complications. You end up with “cures” that are worse than the disease or you have work-ups that are worse than the fear. Finally, there is harm caused by the tremendous amount of resources used to do this imaging. These resources should be used elsewhere within health care and society, in general.
Q: It sounds as if you are not just talking about the risk of issues like false-positives.
A: False-positives are incidental findings. A classic example recently came to my attention — a young girl who had pain in her abdomen. Her care providers were concerned about possible appendicitis. They did a CT scan instead of an ultrasound and, while either study is common, ultrasound is preferred. The ultrasound would have just focused on the appendix. The CT scan found a lung nodule. The radiology report stated “unknown malignant potential” even though here in the Ohio and the Mississippi River valleys it is usually histoplasmosis.
This led to a series of CT scans every few months to see if the nodule was growing. So, pretty soon this 13-year-old girl was exposed to an estimated 25 millisieverts (mSv) of radiation. If you are a nuclear power plant worker, your yearly allotment is 20 mSv. Twenty-five mSv is serious exposure. The best estimate is that this level of exposure is associated with a 1 in 400 risk of a future cancer, over and above the baseline risk, where about 30 percent of the population will be diagnosed with cancer in their lifetime.
Q: What are the main causes of overuse in medical imaging?
A: Medical imaging is an amazing technology — it has so much utility — and this contributes, in part, to overuse. When a critically ill patient is transferred, the first thing that care providers do is look at the imaging studies because looking inside patients is an effective means of making a diagnosis. Medical imaging provides tremendous benefit to millions of patients every day. But use of imaging, like other things, can be too much of a “good thing.” Every medical breakthrough can be overused. Think about the overuse of antibiotics. People are taking antibiotics when they have a viral infection. Antibiotics can be life-saving, but they can also be misused.
Another driver of overuse is that when most of us go to emergency rooms or physician offices, we expect to have some sort of test or a scan. For example, when I was a radiology resident, my two-year-old son fell and hit his head. There was a huge knot on his head and he was screaming. I thought, “What do I do? He needs a CT scan.” During my training, I had seen numerous examples of serious head trauma. It’s scary when this is happening to your own son. I took him to our children’s hospital. They looked at him and said, “He is behaving exactly how we would expect for a child who has hit his head like he did. Here is what you watch for, take him home. As long as he fits this pattern, it is just a bump on the head he doesn’t need a CT scan.” They did the right thing, but it is easy to fall in the trap of going against our better judgment when the parents are looking at you saying, “I think he needs a head CT. Do something.” How many x-rays are done in situations where we say, “We don’t think it is broken, but let’s get an x-ray just in case”?
The point is that there is a risk, and there is a benefit. There is a true benefit in doing the scan and then saying, “I feel better now because we have checked.” But how much risk are you willing to accept down the road for feeling good that we checked? This comes up frequently — children fall and hit their heads every day and it really does depend on which emergency room you go to as to whether CT scans are used or not. The Dartmouth Atlas analyzed the differences in head CTs for children with minor head trauma in New England. There is a three-fold variation across the map and it wasn’t uncommon to see five-fold differences.
The graph I often use shows the sources of our biggest exposure to radiation in the US. There are several natural sources — radon in homes and caves, solar radiation. It never gets to zero. But now medical imaging is the dominant source. We have effectively doubled our dose and we won’t know for probably a generation or more what effect that is having.
Q: What are some benefits of eliminating overuse in medical imaging?
A: It is a classic “win-win” every time you eliminate waste. It’s the Triple Aim: better care, better health for the population, and reducing costs. My trip to the emergency room with my son is a good example of better care in that somebody examined him, made good decisions, and got us on our way. If we eliminated overuse, we would have plenty of resources available to tackle underuse. There is tremendous potential benefit to redirecting resources in health care.
Q: How will the IHI Expedition help hospitals meet the newly revised Joint Commission requirements?
A: In the Expedition, we will explore not only what you have to do, but why you should be doing it. It’s not just about checking a box to check a box. The revised Joint Commission requirements include tracking CT dose metrics and reporting the results in the medical record. We want to help the Expedition participants understand why the Joint Commission believes these are important requirements. The point we want to make is that the way we are going to improve is by measuring. It’s going to take some work, but there is considerable value in creating the infrastructure that allows participants to drive improvements in medical imaging. We have to create an infrastructure, a measurement system, a safety culture — in other words, an improvement culture that includes medical imaging.
Q: Would you talk about “The Five Rights of Imaging”?
A: Imaging's role within the larger health care system is nicely summarized as the “The Five Rights of Imaging.” The five rights are: right study, right order, right way, right results, and right action. You start by asking the question, “Do we need an imaging study?” In some children with minor head trauma you don’t. The right study is to observe them, not image them. If you decide to image, then the right type of scans should be ordered. The right way is using the proper method, protocol, or equipment. For example, for a child, you adjust the settings on the machine so that it matches the child’s size. It is pretty amazing how much a dose can vary in these images. The right result is reading and interpreting the study, making the right recommendations and report, and handing that information to the next person. The right action is that the results of a report lead to the right next steps for the patient. It is an extraordinarily complex operation. If all these steps don’t lead to the right action, it’s all for naught.
Q: It sounds as if there has been a lot of progress in reducing overuse in imaging in pediatric care that needs to extend to adult-focused care.
A: My hat goes off to the pediatric centers and the Image Gently campaign. They really saw the issue, and they are making corrections. But they account for only a small fraction of the places where children are imaged. Most imaging for children occurs in adult-focused hospitals. Imagine that you have 20 adults lined up for imaging and then a child comes in, and then you have another 20 adults in queue. What do you think your machine is set at? It is hard to make these adjustments. Everywhere we look, the pediatric hospitals are doing a fantastic job, but it hasn’t yet crossed over to the adult population.
Q: Do you have any tips for teams that are ready to address these issues?
A: Becoming a learning culture starts with measurement. You won’t know where you are until you start measuring. The Joint Commission is now requiring a record of some measure of the CT exposure, the radiation exposure from a CT scan, after every CT. It won’t be enough to measure, but that is a good place to start. Compare ourselves to the teams that we think are our peers and start figuring out what we can do to get better. It is pretty amazing how quickly you can start and there is a lot of room for improvement. For adult hospitals, two-fold improvement is as easy as “falling off a log” and I believe ten-fold improvements are achievable.
Q: At IHI, we are always eager to engage a variety of disciplines in improving patient safety. As a radiologist who is also an Improvement Advisor, why is this work so important to you?
A: It is the quest for knowledge. I see improvement as trying to learn. It always comes down to two fundamental questions: What changes do we predict will lead to improvement? How do we prove that those ideas do improve the lives of the patients we serve? For me, it’s extraordinarily gratifying to work with the teams that I work with. Their mission in life is to serve the members of their community.