Sammer Marla B K, Stahl Andrew, Ozkan Eray, Sher Andrew C
Texas Children's Hospital, Singleton Department of Pediatric Radiology, 6107 Fannin Street, Suite 470, 77030, Houston, TX, USA.
Department of Radiology, Baylor College of Medicine, Houston, TX, USA.
J Digit Imaging. 2021 Jun;34(3):741-749. doi: 10.1007/s10278-021-00451-4. Epub 2021 Apr 9.
In our pediatric radiology department, radiographs (XR) are the shared responsibility of the body section and interpreted in addition to modality or site-specific assignments. Given an unequal contribution amongst radiologists to the XR workload, a software solution was developed to distribute radiographs and improve workload balance. Metrics to evaluate the intervention's effectiveness were compared before and after the intervention. Data was retrieved from the radiology analytics platform, scheduling software, and the peer learning database. Metrics were compared 12 months pre (March 2018-February 2019) and 6 months post (March 2019-August 2019) intervention on non-holiday weekdays, 7 am-5 pm. To evaluate the intervention's effectiveness, variance between radiologists' contributions to XR volume was assessed using Levene's and Fisher's tests. Changes in turnaround times (TATs) and error rates pre- and post-intervention were evaluated as secondary metrics. Following the intervention, the average number of XR interpreted on target rotations increased by 8.9% (p = 0.011) while the departmental volume of radiographs increased only 4.5%. The variance between radiologists' daily XR contribution was 21.3% (p < 0.0001) higher prior to the intervention. Days where target rotations read fewer than 5 XR decreased from 17.8 to 1.1% (p < 0.0001) after the intervention. Days in which more than 75% of all XR had a TAT less than 60 min improved from 26.8 to 39.7% (p = 0.017) after the intervention. There was no statistically significant difference in error frequency (error rate 2.49% pre and 2.72% post, p = 0.636). In conclusion, the software intervention improved XR workload contribution with decreased variability. Despite increased volumes, there was an improvement in turnaround times with no effect on error rates.
在我们的儿科放射科，X光片（XR）由各身体部位共同负责解读，除了按模态或特定部位分配的任务外，也会进行解读。鉴于放射科医生对X光片工作量的贡献不均衡，我们开发了一种软件解决方案来分配X光片并改善工作量平衡。在干预前后比较了评估该干预效果的指标。数据从放射学分析平台、排班软件和同行学习数据库中获取。在非节假日工作日的上午7点至下午5点，对干预前12个月（2018年3月至2019年2月）和干预后6个月（2019年3月至2019年8月）的指标进行了比较。为评估干预效果，使用Levene检验和Fisher检验评估放射科医生对X光片数量贡献的差异。干预前后周转时间（TAT）和错误率的变化作为次要指标进行评估。干预后，目标轮次上解读的X光片平均数量增加了8.9%（p = 0.011），而科室的X光片总量仅增加了4.5%。干预前，放射科医生每日X光片贡献的差异高出21.3%（p < 0.0001）。干预后，目标轮次解读少于5张X光片的天数从17.8%降至1.1%（p < 0.0001）。所有X光片中超过75%的TAT少于60分钟的天数在干预后从26.8%提高到39.7%（p = 0.017）。错误频率没有统计学上的显著差异（干预前错误率为2.49%，干预后为2.72%，p = 0.636）。总之，软件干预改善了X光片工作量的贡献，减少了变异性。尽管工作量增加，但周转时间有所改善，且对错误率没有影响。
