In early 2021, the University of Rochester Medical Center (URMC) in upstate New York began an enterprise-wide rollout of point-of-care ultrasound (POCUS) devices in 64 departments. Surgeon Michael Rotondo, MD, executive director of the University of Rochester Medical School Group and senior vice president of URMC, recently spoke with Healthcare innovation about the project objectives and the work still to be done.
URMC partnered with a company called Butterfly to launch an enterprise-wide ultrasound program designed to improve access to images and improve clinical workflows. To date, URMC has deployed approximately 1,000 Butterfly iQ devices across 64 academic departments and clinical programs, integrated with Butterfly’s Compass workflow software. URMC plans to expand the program to 2,500 devices by 2026.
Health innovation: Could we first talk about some of the advantages of point-of-care ultrasound technology? How does it change the way procedures or diagnoses are performed in the health system?
rotund: Healthcare systems have primarily used cart-based ultrasound technology. It has an ultrasound machine that is on a cart that it carries to the patient’s bed. It is usually based on units. So if you work in the emergency department, that’s where the emergency department is. If you work in the ICU, you are in the ICU. It has been for intensive care doctors who treat acute illnesses. This has some limitations, one of which is that it is not always accessible. You could be in another location and have to go to a different unit to bring that device. Most importantly, the images that were acquired would be primarily stored on that ultrasound cart, not in the cloud, and could not be easily transported to the electronic health system. record. Additionally, primary care physicians’ offices do not have ultrasound capabilities at all.
The Butterfly iQ Plus is a wearable device. They came up with a chip-based technology, so they can generate those ultrasound waves on a small chip, and that allows the device to fit in your pocket. The imaging device is your phone, so through a Bluetooth interface, the ultrasound goes directly to your phone. They created a cloud-based solution that allows those images to be immediately uploaded to the cloud and then to the electronic medical record.
Because of its portability, because of the ability to capture those images in the cloud, because of the ability to put them in an electronic medical record and use them virtually anywhere: in the home care environment, in the nursing home environment. , in primary care offices and surgeons’ offices, even in general hospital care units, the game changes dramatically.
HCI: So almost every doctor working in the fields you just described could have one in their pocket, basically?
rotund: Exactly. Since 2022, we have deployed just under 1,000 of these devices across our company. We have 2,200 physicians who are in our multi-specialty group.
We also use it for our medical students, because we believe this is the stethoscope of the future. Everyone will have this in their pocket. As time goes on, being able to do imaging and some diagnosis at the time of the physical exam will basically become a standard of care. For a physical exam, a doctor uses his ears, eyes, and hands to visualize what anatomical structures look like and think about what the pathology might be. Now you have images in your hands that will help you move to the next level. Some artificial intelligence built into the probe can not only help you identify what that structure is, but also what is not normal about that structure, what the potential pathology might be. It’s a really incredibly interesting technology, which we’re still learning a lot about.
HCI: How widespread is this in healthcare systems across the country? Is URMC pioneering this or is it being implemented in many places?
rotund: We are at the forefront of this because of the relationship with Butterfly. His leadership approached. They were looking for academic medical centers that might be interested in working with them. It turns out we have two things: the University of Rochester Health Lab, which is a think tank, kind of an incubator/accelerator that I helped establish 10 years ago, where we look at completely new technologies. Additionally, our radiologists had been doing research in ultrasound in the Department of Orthopedics. So we decided to work with them on a large-scale implementation, but also study that implementation. What works? No? What are the challenges in trying to change medical practice? How does it benefit patients? What do you do for quality?
HCI: That was going to be my next question. Is there a way to study the impact on the quality of care?
rotund: Before, it was not really known how ultrasound was used. We now have image capture in almost 80% of the studies. We have an image; we have a report and we can look at the image and the conclusions, just as you would with any x-ray or any image, and determine if something was missed, if it’s accurate or not. Therefore, it has given us the ability to have a robust quality program that otherwise simply would not exist.
HCI: What about the cost impact? If someone were looking at this from a payer perspective, are suddenly a lot more ultrasounds being billed?
rotund: That’s a great question, and I think there’s still no consensus on that, but I can tell you that our charge capture is now up 116%, so we’re now capturing charges for these studies and then trying to design a way to determine if this actually leads to further studies. Because after you do a portable ultrasound and get a specific impression, do you just generate more diagnostic tests or do you have enough information in that particular use case to go directly to therapy and treatment?
So there are two parts to analyzing impact: timeliness and appropriateness. Is the timeliness such that you make the diagnosis early and immediately move on to care, or is it that you were having a test that creates increased sensitivity and leads to unnecessary testing in the future or inappropriate care? I think the jury is still out on that. I will say that that’s the kind of thing we’ve been very careful about. One of the things we haven’t done is use this to bill for many more services. We’ve focused very clearly on what the key use cases are, on abdominal pain and shortness of breath, on chest pain and on a whole range of conditions that are present where it helps our primary care physicians help the patients. Help our home care nurses help patients. Helps staff nurses help patients. So I think it’s an exceptionally important question. It is the one that really requires additional study.
HCI: You mentioned that integration with the EHR is valuable. Did that require a lot of customization on Butterfly’s part or URMC’s part?
rotund: The fact that they created the cloud-based solution and had it as advanced as they did is just exceptional work on their part. I mean, it really is a turning point. This also allows images from any cart-based device to be uploaded to the cloud.
But it still required some additional work. Specifically, to ensure that the providers’ iPhones are secure and can be used as a device through which an image can then be uploaded to the registry. Both we and the company underestimated the amount of work it would entail, but we overcame it. So the answer is that we have been learning together. On the one hand, they brought the cloud-based technology part into the future. On the other hand, we both learned that there were still some barriers we had to overcome, but we were able to find our way through them.
HCI: In addition to studying the impact on clinical quality, you are also studying workflows and adoption. Any lessons learned so far?
rotund: I think the results are more or less what you would expect. We’ve had early adopters who wouldn’t practice without it now. And then there’s an evolving group that’s in the middle. And then some people who see it as something extra that they have to learn and do, and they don’t necessarily see the usefulness of it. We want to be respectful of what people can and cannot do. We only have 175 primary care doctors in our group and they belong to all age groups. We’ve skewed this more toward our younger workforces and our students, because we know they’re much more technology-oriented and will end up driving change.
HCI: Do you have a governance committee established around this project?
rotund: We have good practice governance committees generally across the medical school group governance structure. Whenever we introduce new technology into the healthcare system, it doesn’t happen simply because someone wants it. We put it through a process and make sure it is fully vetted so it is safe. We think very clearly about which use cases will be applicable and go from there. You have to have a governance policy. Then you have to monitor and make sure you do it safely and responsibly.
HCI: Is there still work to be done in this regard?
rotund: We’ve been rolling out deployments sequentially, rather than just launching 3,000 devices and seeing what happens. We are learning as we go. Therefore, we anticipate that this rollout will continue over the next year through 2026. As we do this across various use groups, we are also studying the use cases and then using that accumulated experience to benefit the new people we bring on board. program.
It will be interesting to see what happens with the home hospitalization movement across the country, with more people than ever receiving home care. Ultrasound on a chip, regardless of who the device is, although it is owned by Butterfly at this point, I think will be essential in home imaging.
HCI: When we started the conversation, you mentioned its use in the hospital setting. Was there a group you started with first to test it out?
rotund: In fact, we started with primary care. We thought it was pretty innovative to start with primary care, not jump to places where there might be resistance in the emergency department and cardiology. They are really used to using cart-based technology. We started with new non-imaging users and it’s really fascinating how, unsurprisingly, some people took to it right away and others not so much.
HCl: In the primary care setting, are there a couple of main things you would use it for?
rotund: The use case that comes to mind is congestive heart failure, where fluid can be identified around the lungs. It can identify specific changes within the lung that occur as a result of congestive heart failure. Another would be pain in the right upper quadrant looking for gallbladder problems. Thickening of the gallbladder or gallbladder stones may be seen. It can actually have you diagnosing them, rather than sending the patient for an x-ray or trying to schedule a study. In some cases, an immediate referral to a specialist may be made. However, modeling around that and what happens on the cost side is complicated and will take us longer to really understand.
