分析驱动的工作列表对学术环境中肌肉骨骼MRI解读时间的影响
本研究前瞻性计算三位肌肉骨骼专科放射科医师2016年全部MRI解读时间,并按个体相对解读时间从快到慢生成定制工作列表试用7天。结果显示试验期全体总解读时间每日平均减少29.5分钟,仅两种检查个体解读时间与基线有显著差异,各身体部位工作量分布发生变化。
要点
- 分析驱动工作列表使试验期全体总解读时间每日平均减少29.5分钟。
- 仅2种检查个体解读时间与基线显著不同:读者1腕关节检查慢10分钟(p=0.01),读者3颈椎检查快9分钟(p<0.01)。
- 整体解读时间下降伴随各身体部位工作量分布改变。
核心结论
- 基于个体相对解读时间的分析驱动MRI工作列表可能减少整体组解读时间,而不显著改变个体解读速度,但需工作量分布改变。
研究设计与样本
- 前瞻性研究,纳入三位经过肌肉骨骼专科培训的放射科医师2016年签署的所有MRI检查。
- 根据初始阅片和报告签署时间计算基线解读时间,并按相对解读时间从快到慢排列身体部位生成定制工作列表。
- 定制工作列表试用连续7天,比较试验期与基线平均解读时间及工作量分布差异。
- 个体解读时间变化采用z分数评估,统计学显著性设定为≤0.05。
诊断标准与关键数值
- 个体解读时间变化采用z分数评估,统计学显著性设定为≤0.05。
- 读者1腕关节检查个体解读时间较基线慢10分钟(p=0.01)。
- 读者3颈椎检查个体解读时间较基线快9分钟(p<0.01)。
- 工作量分布变化范围:读者1为-3%至4%,读者2为-13%至14%,读者3为-24%至10%。
临床应用与阅片要点
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVE. The objective of this study was to determine how use of analytics-driven worklists for MRI based on relative individual interpretation time affects the overall group interpretation time in an academic musculoskeletal practice. SUBJECTS AND METHODS. In this prospective study, interpretation times for all MRI studies signed by three musculoskeletal fellowship-trained radiologists during 2016 were calculated from initial study view and report signing times. Custom worklists were made for each radiologist with body parts ordered from the fastest to the slowest based on relative interpretation time. These worklists were then used for a trial period of 7 consecutive days. The difference in mean interpretation times between the trial period and baseline and the differences in volume distribution were calculated. Changes in individual interpretation time were assessed by z-score with statistical significance set at ≤ 0.05. RESULTS. Across all readers, total interpretation time decreased by a mean of 29.5 minutes per day during the trial period. Only two types of studies were read with an individual interpretation time significantly different from baseline (wrist studies for reader 1 were 10 minutes slower [p = 0.01] and cervical spine studies for reader 3 were 9 minutes faster [p < 0.01]). Volume distributions changed across various body parts (-3% to 4% for reader 1, -13% to 14% for reader 2, and -24% to 10% for reader 3). CONCLUSION. Analytics-driven worklists for MRI may decrease overall group interpretation time without significant alteration in individual speed, though a change in volume distribution is required.
引用
- Wong TT, Kazam JK, Rasiej MJ, et al. Effect of Analytics-Driven Worklists on Musculoskeletal MRI Interpretation Times in an Academic Setting. AJR Am J Roentgenol. 2019. PMID: 30807228. doi:10.2214/AJR.18.20434
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。