Business Analytics Fuels Radiology
It's tough to solve a problem you don't know you have. Every radiology department is feeling the pressure to reduce costs and improve turn-around time, but how does an administrator pinpoint problem areas that prevent the department from reaching efficiency goals?
Business analytics, which leverages data and graphical visualizations to identify workflow issues and drive evidence-based modifications to organizational processes, provide an answer. Tools can be delivered by vendor-created dashboard software or homegrown applications built into IT systems. The goal is informed decision-making based on a global view of performance, rather than anecdotal evidence.
Healthcare is just discovering business analytics. Christopher D. Meenan, director of clinical informatics at the University of Maryland School of Medicine in Baltimore, says radiology departments are well-suited for analytics because so much workflow data are contained within existing systems and can be correlated with data points. He says the benefits of business analytics include:
"There are still a lot of islands of information systems in healthcare today," says Meenan. "People may look at a PACS, RIS or dictation system in isolation for information. There's an increasing trend, though, where people are trying to look at data across systems." This need for a global view led Meenan and his colleagues to build a business analytics platform that aggregates data from multiple hospitals.
The more common approach, according to Meenan, is a vendor dashboard. EMR vendors are following in imaging informatics' vendors footsteps and adding built-in analytics tools, but business analytics isn't just about the software
A team-based approach is as important as the software. "Even if you have analytics software that tries to solve a real problem, it's important to engage the right team of stakeholders," says Meenan. "Implementing software alone without a willing and engaged team from across the organization is unlikely to result in successfully improved operations."
Meenan suggests that organizations should focus on their own specific pain points, which vary from department to department. One of the first areas of improvement for the University of Maryland in 2005 was report turn-around time, which the department was able to substantially improve after a focused project supported by data, analytics tools and a team-based quality improvement project.
Baker says her organization decided to examine processes after reading a February 2011 New York Times article about the overuse of "babygrams," full body x-rays of infants. Nemours doesn't perform the full head-to-toe x-ray exams on infants, but she wanted to examine chest and abdomen procedures to identify any areas of improvement.
"We started to look at the whole process," says Baker. The analysis begins with necessity as the department tries to ensure that any studies ordered are in the best interest of the patient. Baker also works with technologists to make sure they collimate exams as much as possible.
Baker and colleagues then turned to analytics reports, which were produced by a platform layered into the EMR, and pulled data from the RIS. After making dose reduction a priority, they found that total infant chest/abdomen x-ray procedures had decreased, but less than expected. They are currently analyzing the data and breaking them down by department to determine areas for further improvement.
Another example of the use of analytics to control radiation dose at Nemours came when the department looked at data for voiding cystourethrogram (VCUG) studies commonly used to diagnose vesicoureteral reflux. Baker says they first looked at total fluoroscopy time, the amount of time the operator's foot is actually on the fluoroscopy pedal, but realized some operators were using pulsed fluoroscopy rather than constant x-ray. Since this would alter the total dose received, they then produced reports looking at dose area product and found differences between operators, so they made it a goal to reduce variation of dose area product, specifically for VCUG.
"The idea that we're looking at dose has folks thinking about it and being more conscientious, which translates to better quality care for kids," says Baker, adding that the department tied bonuses to these quality metrics and uses analytics in pay-for-performance programs.
The primary challenge for Nemours in terms of gathering data is that a person has to transcribe dose measurement data from the acquisition devices, as the business analytics software cannot yet pull data directly from scanners. Baker and colleagues aim to automate this process and begin reporting dose data to the American College of Radiology for benchmarking.
"First, I talked to my staff about how I believed this was going to be something they're really going to enjoy using and that they will benefit from," Barteet said during a November virtual conference hosted by Health Imaging (HealthImaging.com/VirtualConference2011).
At Trident, an analytics dashboard pulls and aggregates data from the RIS, PACS, voice recognition and ED systems. For technologists performing CT and x-ray exams, this translates into a large-screen display showing process information. When a procedure is ordered, it goes on the large display board and the clock starts ticking. Techs then check off when studies are completed, verified, dictated and signed.
In Barteet's office, a real-time stoplight display keeps her updated on procedures through each step of the process. The CT turn-around time goal is an hour. The display stoplight for an individual CT procedure is green for the first 15 minutes, but turns red after 30 minutes so that any potential delays can be addressed before the hour goal is passed. Delay codes are entered and tracked by the system.
"The techs are aware of what the turn-around time is, what their expectations are and what's going on with the results," says Barteet.
One process improvement came after the team realized a delay point in its CT process was the serum creatinine test administered prior to the CT scan. The analytics tools showed techs needed lab reports faster, spurring the department to place handheld blood analyzers at the bedside for point-of-care diagnostic testing.
Prior to implementing these tools, Barteet says that Trident's turn-around times were inconsistent, averaging 45 minutes per procedure one week and as many as 70 minutes the next. After turning to business analytics, the department consistently meets its one hour goal and has seen a 20 percent reduction in turn-around time.
Barteet says the biggest advantage to having the real-time dashboard is that administrators can see and fix issues on the spot.
In the 18 months since the system went live at Trident, it's become second nature to keep an eye on the business analytics display board, showing how quickly business analytics can move from a foreign concept to an everyday habit.
Business analytics, which leverages data and graphical visualizations to identify workflow issues and drive evidence-based modifications to organizational processes, provide an answer. Tools can be delivered by vendor-created dashboard software or homegrown applications built into IT systems. The goal is informed decision-making based on a global view of performance, rather than anecdotal evidence.
Healthcare is just discovering business analytics. Christopher D. Meenan, director of clinical informatics at the University of Maryland School of Medicine in Baltimore, says radiology departments are well-suited for analytics because so much workflow data are contained within existing systems and can be correlated with data points. He says the benefits of business analytics include:
- Productivity and turn-around time: Reports allow departments to see delays in organizational processes such as patient transport, report turnaround and delivery.
- Quality assurance: Metrics to track image quality can help coordination between radiologists and technologists to produce optimal images.
- Equipment utilization: Modality utilization tools help allocate departmental resources more efficiently and provide a global view of patient flow.
"There are still a lot of islands of information systems in healthcare today," says Meenan. "People may look at a PACS, RIS or dictation system in isolation for information. There's an increasing trend, though, where people are trying to look at data across systems." This need for a global view led Meenan and his colleagues to build a business analytics platform that aggregates data from multiple hospitals.
The more common approach, according to Meenan, is a vendor dashboard. EMR vendors are following in imaging informatics' vendors footsteps and adding built-in analytics tools, but business analytics isn't just about the software
A team-based approach is as important as the software. "Even if you have analytics software that tries to solve a real problem, it's important to engage the right team of stakeholders," says Meenan. "Implementing software alone without a willing and engaged team from across the organization is unlikely to result in successfully improved operations."
Meenan suggests that organizations should focus on their own specific pain points, which vary from department to department. One of the first areas of improvement for the University of Maryland in 2005 was report turn-around time, which the department was able to substantially improve after a focused project supported by data, analytics tools and a team-based quality improvement project.
Beyond turn-around time
Besides reducing turn-around time, business analytics tools provide clinical applications to improve quality of care, says Annie Baker, administrative director of medical imaging at Nemours, a children's health system with locations in Florida, Delaware, New Jersey and Pennsylvania.Baker says her organization decided to examine processes after reading a February 2011 New York Times article about the overuse of "babygrams," full body x-rays of infants. Nemours doesn't perform the full head-to-toe x-ray exams on infants, but she wanted to examine chest and abdomen procedures to identify any areas of improvement.
"We started to look at the whole process," says Baker. The analysis begins with necessity as the department tries to ensure that any studies ordered are in the best interest of the patient. Baker also works with technologists to make sure they collimate exams as much as possible.
Baker and colleagues then turned to analytics reports, which were produced by a platform layered into the EMR, and pulled data from the RIS. After making dose reduction a priority, they found that total infant chest/abdomen x-ray procedures had decreased, but less than expected. They are currently analyzing the data and breaking them down by department to determine areas for further improvement.
Another example of the use of analytics to control radiation dose at Nemours came when the department looked at data for voiding cystourethrogram (VCUG) studies commonly used to diagnose vesicoureteral reflux. Baker says they first looked at total fluoroscopy time, the amount of time the operator's foot is actually on the fluoroscopy pedal, but realized some operators were using pulsed fluoroscopy rather than constant x-ray. Since this would alter the total dose received, they then produced reports looking at dose area product and found differences between operators, so they made it a goal to reduce variation of dose area product, specifically for VCUG.
"The idea that we're looking at dose has folks thinking about it and being more conscientious, which translates to better quality care for kids," says Baker, adding that the department tied bonuses to these quality metrics and uses analytics in pay-for-performance programs.
The primary challenge for Nemours in terms of gathering data is that a person has to transcribe dose measurement data from the acquisition devices, as the business analytics software cannot yet pull data directly from scanners. Baker and colleagues aim to automate this process and begin reporting dose data to the American College of Radiology for benchmarking.
A win for management and staff
Business analytics isn't just for meetings between managers; it also can make life easier for front line staff and technologists. Jodi Barteet, imaging director at Trident Medical Center in Charleston, S.C., says that analytics helps directors manage staff and provides tools for staff to deliver better care."First, I talked to my staff about how I believed this was going to be something they're really going to enjoy using and that they will benefit from," Barteet said during a November virtual conference hosted by Health Imaging (HealthImaging.com/VirtualConference2011).
At Trident, an analytics dashboard pulls and aggregates data from the RIS, PACS, voice recognition and ED systems. For technologists performing CT and x-ray exams, this translates into a large-screen display showing process information. When a procedure is ordered, it goes on the large display board and the clock starts ticking. Techs then check off when studies are completed, verified, dictated and signed.
In Barteet's office, a real-time stoplight display keeps her updated on procedures through each step of the process. The CT turn-around time goal is an hour. The display stoplight for an individual CT procedure is green for the first 15 minutes, but turns red after 30 minutes so that any potential delays can be addressed before the hour goal is passed. Delay codes are entered and tracked by the system.
"The techs are aware of what the turn-around time is, what their expectations are and what's going on with the results," says Barteet.
One process improvement came after the team realized a delay point in its CT process was the serum creatinine test administered prior to the CT scan. The analytics tools showed techs needed lab reports faster, spurring the department to place handheld blood analyzers at the bedside for point-of-care diagnostic testing.
Prior to implementing these tools, Barteet says that Trident's turn-around times were inconsistent, averaging 45 minutes per procedure one week and as many as 70 minutes the next. After turning to business analytics, the department consistently meets its one hour goal and has seen a 20 percent reduction in turn-around time.
Barteet says the biggest advantage to having the real-time dashboard is that administrators can see and fix issues on the spot.
In the 18 months since the system went live at Trident, it's become second nature to keep an eye on the business analytics display board, showing how quickly business analytics can move from a foreign concept to an everyday habit.
Evan Godt, Writer
Evan joined TriMed in 2011, writing primarily for Health Imaging. Prior to diving into medical journalism, Evan worked for the Nine Network of Public Media in St. Louis. He also has worked in public relations and education. Evan studied journalism at the University of Missouri, with an emphasis on broadcast media.