Having a young child scheduled for a magnetic resonance imaging (MRI) or computed tomography (CT) scan to diagnose a potentially serious condition can be worrying for parents.
Adding to the anxiety is the fact that these scans often are done under general anesthetic to ensure that young children remain motionless and calm.
Under these circumstances, the last thing a nervous parent needs is a delay in the scheduled start time of the scan. Apart from the increased anxiety caused by waiting, there also is the practical consideration of a young child having to go without food or drink, a pre-anesthesia requirement, for longer than necessary.
At UI Hospitals and Clinics, efforts aimed at improving scheduling for these anesthetic procedures uncovered unexpected and fundamental problems in predicting case durations. UI researchers have now figured out the cause of this troublesome patient scheduling problem, and have determined and implemented a solution that has improved delivery of health care for pediatric patients. The results of the study were published in the May issue of Anesthesia and Analgesia.
Despite a perception that it should be fairly easy to accurately predict how long these diagnostic scans will take, the UI operations research team led by Franklin Dexter, M.D., Ph.D., UI professor of anesthesia in the UI Roy J. and Lucille A. Carver College of Medicine, found that estimates for the duration of an MRI or CT with anesthesia were strikingly inaccurate - more inaccurate than estimates of notoriously unpredictable surgery times.
"This was very unexpected because while we know that it is very hard to accurately predict how long a surgery will take, we assumed that it would be easy to predict how long a CT or MRI under anesthesia would take. However, we found that was not the case," Dexter said.
Having discovered this surprising problem, Dexter and Jack Yue, Ph.D., an ecological statistician at National Chengchi University in Taipei, Taiwan, determined that the inaccuracy resulted from the wide diversity of billing codes that were used to calculate how long diagnostic scans should take.
"Typically, we used historical anesthesia times classified by billing codes to determine how long a procedure should take," Dexter said. "The problem, it turns out, is there are so many billing codes for the CTs and MRIs and the codes reflect the organs imaged, not how long it takes to do the scanning procedures."
An example of how billing codes can produce problems is the fact that MRI or CT scans of separate pieces of the spine each have a separate billing code, which makes sense from a medical point of view. However, there is no billing code for a scan of the total spine, and summing all the billing codes for each part is not an accurate way to determine the time it takes to perform a scan of the spine. Also, there is no billing code to differentiate between different scanners, which can take different amounts of time to do the same scan.
Just as the root cause of the problem turned out to be quite simple, so was the solution. Dexter asked the experts -- the technicians running the CT and MRI instruments -- to estimate how long each procedure should take. Dexter and Yue analyzed the experts' responses and determined that using these estimates significantly improved the accuracy of predicting how long each case would take.
A third member of Dexter's team, Angella Dow, helped to implement the solution into a simple Web-based system that allows UI schedulers to more accurately schedule diagnostic scans with anesthesia and minimize the necessary pre-anesthesia fasting times for very young patients.
"Doing a better job predicting start times not only means less inconvenience, it also means that children are required to fast for only short amounts of time," Dexter said.
The UI Web site, which has been in effect for several months, allows schedulers to check boxes that define the scanning procedure -- the type of scan, the body part or parts involved and the instrument used, for example. The computer program, using the MRI/CT experts' knowledge, then calculates how long the scan will take. Using the patient's age and the scheduled start time for the procedure, the program also can calculate the exact times at which a child can safely have her last food and drink.
While this new procedure works and has benefited patients at UI Hospitals and Clinics, Dexter also wanted to know if the science behind this study could be universally useful in other facilities. Further analysis proved that the methodology could be applied to a wide variety of hospital settings.
"We are not only meeting the health care needs of our own patients, but we are doing the research that will improve health care worldwide, which is also part of our mission," Dexter said.