Enhancing STAT radiology exam order efficiency to reduce turnaround time
There is a chronic issue at many hospitals where STAT priority radiology reads are over used, causing actual emergency exams to get lost in the long list and delaying critical patient care. Tampa General Hospital has developed an initiative to lower non-emergency STAT requests by using a combination of additional questions and orders and staff education. The effort has significantly improved turnaround times.
"We saw a real problem, because STAT imaging means medical emergency, but it has been misused and abused because a lot of people order STAT exams thinking that maybe radiology can get it done quicker, or maybe I want to get exam done before rounds, or it just becomes a habit. However, this creates a real problem for those patients who really need STAT imaging, they get mixed up with the pseudo emergency STAT imaging and that delays their treatment and increases the turnaround time," explained Raj Kedar, MD, MBBS, FACR, vice chair of radiology at the University of South Florida and chief of radiology at Tampa General Hospital. He detailed the initiative during a session at the Radiological Society of North America (RSNA) 2024 meeting and spoke with Radiology Business in the above video.
The hospital saw almost 65% of its total exams were ordered STAT. That includes emergency department as well, and for inpatient it was 45%, and it was almost 68% of inpatient exams.
To address the problem, Tampa General Hospital launched a multidisciplinary quality improvement project involving radiologists, technologists, clinicians, administrators, and data scientists. The team emphasized education, policy changes and the strategic use of technology. Among their tactics:
• Mandatory attestation: Clinicians ordering STAT exams had to affirm that the imaging was truly emergent, based on criteria like life- or limb-threatening conditions.
• Clinical decision support (CDS): Implementing and strengthening CDS tools helped guide providers toward appropriate imaging choices.
• Workflow adjustments: The hospital restructured patient prioritization, creating distinct workflows for STAT, ASAP (semi-urgent), routine, and timed imaging requests.
• AI implementation: Artificial intelligence was used to assess patient readiness, helping to manage scheduling more efficiently and avoid delays due to unprepared patients.
The results were significant. Prior to the intervention, the average turnaround time from order to final read for a STAT exam was about two hours and 35 minutes. After the changes, this was reduced by nearly 48 minutes. Moreover, STAT CT exam orders dropped by 40% over the course of a year, alleviating pressure across the imaging department.
"We did a multidisciplinary approach. We had a team of radiologists, radiology administrators, technologists, as well as clinicians who order most of those exams. We also had a team of data scientists and administrators. We had open communication, constant feedback, and we did some changes in our management," Kedar said.
He also noted that their approach went beyond radiology, targeting the education of referring physicians who often misused STAT orders either out of habit or to accelerate patient management.
The program was able to reduce the number of STAT CT over the first year by about 6,000, or about 40%, Kedar explained.
"This not only helped the STAT patients, but they also helped the semi-urgent ASAP patients where they need urgent treatment, but it's not a question of life and death because.
When so many orders were labeled as STAT, Kedar said radiologists and technologists saw them as normal exams because they knew a large percentage of these exams were not urgent, but did not have anyway of vetting the exams. He said to many imaging staff, the STAT label was not taken 100% seriously so it did not make any difference as to which of these exams were read first.
Large number of inappropriate exams also needed to be addressed
Upon analysis of ordering habits by referring physicians at the hospital, Kedar said it was also clear education was needed for resource utilization.
"They will order sometimes an ultrasound, CT and a MRI, everything at same time, and hope that whichever exam is done first will get results that they can base treatment. But that's not the appropriate approach. They need to talk to radiologist to see which is the appropriate exam, and they need to order it properly, not marking all their exams STAT," Kedar explained.
They used artificial intelligence (AI) to look at the ordering data so they could identify which referring physicians were the worst offenders for appropriate exams.
"An important thing for clinical decision support is to take action, because if somebody is ordering inappropriate exams constantly, you need to educate those providers. So with this project, we created a dashboard," Kedar said.
He said clinical decision support systems play a critical role in helping physicians understand the most appropriate imaging modality based on patient history and symptoms. But the real key is acting on that data and holding providers accountable, and offering education to correct patterns of overuse.
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: [email protected]