Unifying multiple radiology worklists into a single worklist can make coverage of after-hours imaging studies more efficient and reduce report turnaround times, according to a clinical practice management case study published online March 3 in the Journal of the American College of Radiology.
After implementing a quality improvement project aimed at improving report turnaround time during call shifts for certain inpatient radiography exams, a team of researchers led by Dr. Chintan Shah of the University of Pennsylvania Health System in Philadelphia found that a unified worklist led to a nearly 20% improvement in turnaround time for these studies.
"Overall, these results indicate that unified worklists can improve radiologists' efficiency by eliminating the need to switch between worklists and improving the visibility of studies that would have been deprioritized in a segmented workflow," the authors wrote.
Low priority, longer turnaround
The researchers had sought to address an issue with their plain radiography call shift, which is covered at their academic health system by a first-year radiology resident and an attending radiologist. This call shift, which is responsible for reading radiographs from two hospitals for both inpatients and emergency department (ED) patients, involved cycling through five separate worklists. Consequently, the worklist for inpatients from one of the hospitals was often given a low priority, resulting in longer report turnaround times and delayed care in these patients compared with the ED patients.
To help, the researchers created a unified evening call worklist using their institution's Radiant RIS software (Epic Systems). They then quantified the benefits of this approach by calculating report turnaround time for a 10-week baseline period and a 16-week implementation period.
| Report turnaround time for on-call radiography exams | ||
| Before unified worklist implementation | After unified worklist implementation | |
| Inpatient chest radiographs | 64.4 minutes | 52.9 minutes (-18%) |
| Emergency department chest radiographs | 36 minutes | 41.1 minutes (+12.5%) |
During the initial implementation period, the researchers also received positive feedback on the new system. Residents and attending radiologists reported that the unified worklist simplified their workflow and that they strongly preferred this method to their old system, according to the researchers. What's more, the objective improvement in turnaround time for inpatient chest radiography studies also came with subjective improvement in how the radiologists assessed their own efficiency.
Turnaround for ED radiographs
The researchers found that the higher turnaround time for ED chest radiographs was due primarily to a two-week period during which a greater number of preliminary reports were produced. After a post-hoc analysis of the proportion of radiographs read by residents was performed for each period, the researchers found a large increase in the proportion of inpatient radiographs that were read by the radiology residents in the implementation period.
"This may be related in part to improvements in efficiency of the [radiology residents] over the study period," the authors wrote. "However, the much larger increase in the proportion of [inpatient] radiographs read by [radiology residents] relative to ER radiographs also suggests that the combined worklist did indeed improve the visibility of these studies to the residents, with associated improvements in [inpatient turnaround time], a process that was initially lagging."
Even though the ED studies were also more readily picked up by residents than before the unified worklist was implemented, the slightly slower turnaround time for these exams occurred because overreading residents is inherently a longer process, according to the researchers.
"This finding highlights the limitations of [turnaround time] as an outcome metric, and we acknowledge this as a limitation of this work," they wrote.
The researchers said the unified worklist has now been fully implemented at their institution and that the project is in the sustaining phase. In addition, they have also developed suggested practices for developing worklists, including listing all patient sites and settings, determining which studies should be included or excluded for each site and setting combination.
"We also do not use time-based criteria or cutoffs, to avoid studies from other times of day not being visible to the following shift and falling through the cracks," the authors wrote. "Additional on-call unified worklists are now being developed in our department."
