ACR Bulletin

Covering topics relevant to the practice of radiology

Collaborating on Quality

Jump to Article
The ACR Learning Network showcases standout quality improvement projects in a new ACR Bulletin series on prostate MR imaging, mammography positioning, lung screening and follow-up recommendations.

We felt like we could teach technologists and medical physicists and radiologists to work together as part of a quality team to show that if you learn how to score exams using PI-QUAL scores and then determine what quality is, then quality can be a part of everyone's job — as opposed to just physicians as the end user of the technique.

—Rajan T. Gupta, MD
August 30, 2023



This is the first article in a series the ACR Bulletin published in 2023 about the inaugural cohort of the ACR Learning Network quality improvement collaboratives. The Learning Network, launched in 2022, is now accepting applications for its fourth cohort. The program is aimed at improving diagnostic imaging care through a learning health systems approach. Each collaborative supports teams at the facilities of four to six sites working to solve similar problems. 

Read the other articles in the collection: 

Radiology teams across the country work diligently to provide the latest, safest and most effective imaging services to patients. A cornerstone of delivering top-notch care is an open exchange of information that can shape new clinical technology, streamline imaging processes and deliver patient care with better outcomes. Building on this formula, the ACR Learning Network set out in early 2022 to improve diagnostic imaging care through a learning health systems approach — defined by fostering strong leadership, maximizing the use of data in clinical settings and building a workplace culture committed to continuous learning and improvement.

The Learning Network was created to facilitate local improvements in patient care, establish a community of shared learning, create global approaches for lasting improvements and recognize participants with measurable and sustained performance. To date, the program has focused on four quality improvement (QI) collaboratives to address critical areas of cancer diagnosis: prostate MR image quality, mammography positioning, lung cancer screening and follow-up recommendations post-imaging.

Each collaborative supports teams at the facilities of four to six sites working to solve similar problems, enabling them to grow experience and fresh perspectives as they share insights and advance QI goals. Every improvement collaborative has the support of a radiologist, an administrative lead and dedicated staff — including a learning facilitator, an administrative support specialist and a data support specialist. The program is funded through a grant from the Gordon and Betty Moore Foundation.

The ACR Bulletin is publishing a four-part series delving into the work of the 22 sites that participated in the first cohort of QI collaboratives. These sites have laid important foundational work that will support current and future cohorts by defining and refining QI measures and developing enhanced internal imaging procedures and protocols. This first article in the series focuses on improving prostate MRI for clinically significant cancer detection and localization. The experience is defined through the perspectives of four members of various teams. 

Collaborative Leader: Examining What Quality Means

“When approached about the Learning Network improvement project, I saw it as an opportunity to not only gain experience from leading the effort at my own institution, but to learn from the other highly talented groups participating in the prostate MR imaging collaborative,” says Andrei S. Purysko, MD, clinical assistant professor of radiology at Cleveland Clinic Lerner College of Medicine and prostate MR imaging collaborative leader.

MRI is increasingly used in the diagnostic pathway of prostate cancer, but image quality in up to 40% of biparametric exams results in diagnostic errors. The prostate MR QI collaborative supports site teams in creating a standardized system for improving the percentage of MR prostate exams that meet quality standards according to Prostate Image Quality (PI-QUAL) criteria. PI-QUAL can evaluate image quality changes resulting from protocol improvements. With training, radiologists, technologists and physicists can perform PI-QUAL scoring with similar performance — and broadening the scope  of a quality improvement team can result in meaningful and sustainable change.

A radiologist expert leads each of the improvement collaboratives as participant sites develop and refine one or more measures within their respective healthcare setting. This supports the work of the individual site’s QI projects and the progress of the entire collaborative.

“The first challenge was to get a consensus on what good prostate MRI quality is,” Purysko says. “That is not easy to define, and we saw that people were all over the place in their definition. In the first cohort, we spent a significant amount of time with sites discussing this and making some refinements.” In addition, the collaborative brought in an expert on quality metrics who gave a presentation on what high quality should look like, he says.

Purysko was surprised by how much variability he discovered throughout the process. “You think you have established processes in place — using SOPs, for example — but there is sill so much variability,” he says. “It is something I found internally and validated what was happening on a broader scale among the collaborative sites. This project has given them a chance to take control and standardize the way things should be done at their facilities.”

It is acceptable to have certain differences between sites — different vendors and scanners with varying capabilities, for instance. “We still need to follow some basic rules, however, and explore why MRI results on initial scans and follow-up imaging may not be reproducible,” Purysko says. “How do you know why things look different? Is it because a cancer has progressed versus the quality of a scanner? That is why standardization is so important. You must work with the technologists and schedulers and understand how the entire imaging process works, because it is more complex than one would think. We have a chance not only to better understand their role in the imaging process but to offer guidance and talk to them about the importance of image quality.”

From giving patients instructions before they come in for imaging services to the scan itself, there are a lot of moving parts, he says. “We have also been working with providers — the physicians and nurses who order prostate MRI — to discuss quality and to validate that the interventions they are doing have a positive impact.”

ImPowering Shared Learning

“The Learning Network project teaches all participants more about QI in general but also moves the collaborative team toward improving processes and upping the level of quality at their own institutions,” Purysko says. “It gives participating sites a way to learn about QI in a very structured way using the ACR ImPower Program. Each of the collaborative sites participates in ImPower, which encourages and facilitates shared learning as colleagues identify new ways to improve." 

ImPower focuses on structured improvement and provides QI collaborative site teams the opportunity to learn and work alongside other organizations solving the same problem at the same time and using the same processes. The program is based on the premise that when the people doing the work are taught and empowered to solve problems, there will be better solutions. Staff are more likely to embrace changes and an organization strengthens its ability to improve.

The Learning Network project teaches all participants more about QI in general but also moves the collaborative team toward improving processes and upping the level of quality at their own institutions.

—Andrei S. Purysko, MD

Throughout participation in the program, teams share what they have learned and support one another in implementing sustainable best practices at their local sites. Site teams participate in 10 learning sessions over four months, tracking and reporting data about their progress on a weekly basis. Participants have access to online tutorials and standardized templates to assist with problem solving and change management. This QI approach helps teams reach agreement on specific challenges and to develop, test and manage potential solutions.

QI Physician Leader: Targeting PI-QUAL Scores

“We jumped at the chance to be a part of the ACR Learning Network collaborative,” says Rajan T. Gupta, MD, chief of abdominal imaging at Duke University Medical Center and physician leader of the prostate MR imaging collaborative. “At Duke, we have been doing prostate MRI at a high level for over 15 years. At the same time, we are constantly looking for opportunities to improve our image quality and our patient experience.”

Gupta began his role by asking several questions around the problem they wanted to solve. How do we continue to obtain high-quality imaging of the prostate without using an endorectal coil? And more importantly, how do we quantify the improvements that we are making in a more structured way?

“Internally, we have held a lot of meetings to talk about our goal and there has been a consistent desire to advance our efforts,” Gupta says. “We focused on measurable results using PI-QUAL scores.” PI-QUAL was created to assess the diagnostic quality of a scan against a set of objective criteria as per Prostate Imaging Reporting and Data System (PI-RADS®) recommendations, together with criteria obtained from the image. The PI-QUAL score  is a 1-to-5 scale where a score of 1 indicates that all MR sequences are below the minimum standard of diagnostic quality, a score of 3 means that the scan is of sufficient diagnostic quality and a score of 5 implies that all sequences are of optimal diagnostic quality.

“We felt like we could teach technologists and medical physicists and radiologists to work together as part of a quality team to show that if you learn how to score exams using PI-QUAL scores and then determine what quality is, then quality can be a part of everyone's job — as opposed to just physicians as the end user of the technique,” Gupta says. “We set a desired measurable change in our PI-QUAL scores. Our post-intervention goal was to have 85% of our exams with a non-endorectal coil getting a score above 4 — and we were proudly able to do that.”

The team did not stop there, he says. “We want to continue to build — 85% is not our final goal. That was our first goal, but we want to keep going and are engaging additional radiologists and technologists. Our participation in future Learning Network projects will probably be more of a mentorship role — sharing what we have been doing and why we think it has been successful.” 

The team has been writing about the process in publications, including an article  in the European Journal of Radiology, Gupta says. It was exciting to include as many participants as possible on that paper as coauthors given the incredible work they did, he adds.

One of the biggest things that impressed Gupta during the QI collaborative was how much of a team effort it fostered. “We had an incredible team of people involved in the project,” he says. They included Scott Robertson, PhD, a medical physicist and Duke’s QI champion, as well as Waleska (Waly) Pabon-Ramos, MD, Duke’s vice chair of quality and safety, and a host of experienced and dedicated technologists. “The technologists were huge in terms of helping us see where our image quality improvement opportunities were — and then helping us to quantify the impact those imaging changes or protocol changes have on image quality,” Gupta says.

“Some of our team members have said the collaborative is the best professional initiative they’ve ever been a part of,” Gupta says. “That is great to hear in an era where burnout is becoming an epidemic. Finding something that brings you joy at work is what keeps you there, and I think this opportunity for participating and staying engaged has increased job satisfaction.” 

That satisfaction begins with team members articulating what they believe about their role in the process — and what they think it should be. “There can be discrepancies there, and listening to what people need and offering team support helps us move things forward,” Gupta says.

Collaborative Team Member: Pushing for Better Image Quality

“We do a lot of prostate imaging and are constantly trying to improve prostate image quality,” says Nicholas (Nick) Marshall, BS, RT (R)(CT)(MR)(ARRT), MRI coordinator at St. George Regional Hospital, part of Intermountain Healthcare in St. George, Utah. “We had a bad, and rather embarrassing, experience at our facility about five years ago where a patient had a prostate MRI with no cancer detected. The same patient had the same scan done at another imaging facility a few months later and they detected cancer.” 

The patient revisited the Intermountain facility and asked to speak with a radiologist who could explain the different findings. The radiologist reviewed the images from both scans and found nothing evident to suggest cancer on the first MRI but did see it on the MRI from the other facility. “From that experience, we learned that we were not providing the highest-quality prostate imaging,” Marshall says.

“When our section chief  here at the hospital found out about the Learning Network collaborative, she asked if anyone from my team would be willing to join,” Marshall says. “My team and I started spending a lot of time working on how to improve prostate image quality. We knew we could do better and began working with GE and their device reps with the goal of providing the best-quality diagnostic scans for all patients.” 

At one point, Marshall says, a new corporate protocol was suggested for prostate imaging. “I didn’t agree with it because I knew what we had in place was better. We had been working hard to improve our protocol and I felt the new direction from corporate did not meet ACR Appropriateness Criteria. As we went through the process — collecting data, meeting each week and then using the data — we started seeing big improvements in image quality. We were making strides on our own, but our efforts have been bolstered since participating in the collaborative. We have moved from PI-QUAL 3 to PI-QUAL 5 on a lot of cases, for example.” 

Some of the equipment has been updated, Marshall says, “but really the improvements have come from protocol modifications that fall in line with ACR standards.” 

Proven success can be contagious. Intermountain as a whole has seen huge improvements in imaging at its other sites as well. “They have really stepped up their game,” Marshall says. “Even though other corporate sites did not have representation in the collaborative, they have improved prostate MRI at those facilities.”

QI Coach: Urging Teamwork and Process Refinement

“When invited to participate in the Learning Network program, we knew the initiative would offer the radiology department new ways to learn from the ACR,” says Andrea (Griffin) Rego, MHA, BSN, RN, a quality and patient safety nurse specialist at Boston Medical Center and quality improvement coach for the Prostate MR Imaging Quality Improvement Collaborative. “It was also exciting to see what we could gain from the process for our own professional development.

“We typically try to identify trends in our incident reports and respond to a particular department that brings an issue or an opportunity to us,” Rego says. “We facilitate in different ways. For example, sometimes we take on more of a project management role and other times a QI role. Both positions help our people identify where they want to start and the problem they are trying to solve through the collaborative process.”

Rego was relatively new to the QI team when Boston Medical started working with the prostate MR collaborative. “I had worked on QI initiatives before, but not on such a large scale,” she says. 

There are so many people involved in the process that may not seem obvious as integral but turn out to be, she says. “You end up learning about your scheduling department for prostate MRI and interpreter services, for example,” she says. “We got to talk to them before even starting any interventions.”

The Boston group in the collaborative had a core team of Rego, a prostate MR technologist and the physician who reads the prostate MRI. “We pulled several others into the project, including two staff from scheduling and other MR technologists to collect prescreening forms from patients and ask them about their prep instructions,” Rego says. 

The group changed the patient prep process with improved instructions. “We looked at how we were giving instructions to the patient, calling in advance and providing more updated, detailed instructions about what to eat or drink prior to the scan — and translating from English when necessary,” she says.

“We have been tapping into the knowledge and resources of the ACR," Rego says. "Personally, I think I have learned a lot through the prostate MR imaging project, especially about our current check-in process. I now know more about what patients take away from the experience. I have been channeling what I’ve learned to ensure that we are being very clear with our patients about checking in, the timing and potential delays. There has also been a lot of behind-the-scenes work around our MRI machine. The technologists changed various settings to see what might improve image quality that way. We couldn’t necessarily tell if it was one particular setting versus another that had an impact, but we were encouraged by leadership to try many things — because you won’t always immediately know what might be making the difference.”

The Learning Network in general encourages participants to test new approaches and share findings with the collective group. “One thing that stood out for me during the work of the prostate collaborative was the benefit of having the ACR behind us,” Rego says. The structure of regular meetings makes it easier to get people in the same room, and there is much more consistency when everyone is present to move a project along, she says. “This was evident when we reached our PI-QUAL goal and were acknowledged by the ACR for sustained quality improvement.”

Participating Is Worth It

“When you hear about something like this learning collaborative, it can feel imposing to have such a structured, rigorous approach,” Duke’s Gupta says. “Your team may wonder if you have the infrastructure to be a part of something like this. But I would tell people that they absolutely should take the opportunity to participate in the ACR’s Learning Network. It gave us a chance to empower all members of our team to truly engage in the quality of our exams.”

Cleveland Clinic’s Purysko says the program also made a difference for him personally. “The program taught me how to be a better section head because the QI collaborative emphasizes understanding how processes work by asking questions — to find out why one approach may be better than another,” he says. “This has really changed my mindset, and the tools I have gained from the program have enabled me to be a better leader. You see how the impact is not just local but can be felt at a national level. To see how processes across institutions are changing for the better has been extremely rewarding.”

Participation in the Learning Network’s collaboratives is helping raise the level of QI across sites and throughout the radiology ecosphere, Purysko says. “This is true for radiologists, technologists, nurses and administrative staff alike. I have seen how rewarding it is for the team because we are giving them a voice and encouraging them — essentially putting them in the driver’s seat and empowering them to take responsibility for sustaining a QI process. The engagement that has come from this program has been a privilege to witness.”


Benefits of Participation in the ACR Learning Network

  • Receive rigorous training in proven quality improvement (QI) strategies and processes through the ImPower Program and quality coach training that a site can use for future QI work.
  • Obtain guidance from national leaders to get the most out of participation and achieve sustained diagnostic excellence.
  • Become a regional and/or national leader in one of the collaboratives’ focus areas and continue participation in the Learning Network community.
  • Obtain continuing education credits for ImPower program participation and complete a quality improvement initiative. 

Learn how to apply at the ACR Learning Network web page.

Author Chad E. Hudnall  senior writer, ACR Press