Sharp Adam L, Nagaraj Ganesh, Rippberger Ellen J, Shen Ernest, Swap Clifford J, Silver Matthew A, McCormick Taylor, Vinson David R, Hoffman Jerome R
Department of Research and Evaluation, Kaiser Permanente Southern California, Pasadena, CA.
Department of Emergency Medicine, Los Angeles Medical Center, Kaiser Permanente Southern California, Los Angeles, CA.
Acad Emerg Med. 2017 Jan;24(1):22-30. doi: 10.1111/acem.13061.
Millions of head computed tomography (CT) scans are ordered annually, but the extent of avoidable imaging is poorly defined.
The objective was to determine the prevalence of likely avoidable CT imaging among adults evaluated for head injury in 14 community emergency departments (EDs) in Southern California.
We conducted an electronic health record (EHR) database and chart review of adult ED trauma encounters receiving a head CT from 2008 to 2013. The primary outcome was discordance with the Canadian CT Head Rule (CCHR) high-risk criteria; the secondary outcome was use of a neurosurgical intervention in the discordant cohort. We queried systemwide EHRs to identify CCHR discordance using criteria identifiable in discrete data fields. Explicit chart review of a subset of discordant CTs provided estimates of misclassification bias and assessed the low-risk cases who actually received an intervention.
Among 27,240 adult trauma head CTs, EHR data classified 11,432 (42.0%) discordant with CCHR recommendation. Subsequent chart review showed that the designation of discordance based on the EHR was inaccurate in 12.2% (95% confidence interval [CI] = 5.6% to 18.8%). Inter-rater reliability for attributing CCHR concordance was 95% (κ = 0.86). Thus, we estimate that 36.8% of trauma head CTs were truly likely avoidable (95% CI = 34.1% to 39.6%). Among the likely avoidable CT group identified by EHR, only 0.1% (n = 13) received a neurosurgical intervention. Chart review showed none of these were actually "missed" by the CCHR, as all 13 were misclassified.
About one-third of head CTs currently performed on adults with head injury may be avoidable by applying the CCHR. Avoidance of CT in such patients is unlikely to miss any important injuries.
每年有数百万例头部计算机断层扫描（CT）检查，但可避免的成像检查范围尚不明确。
确定南加州14家社区急诊科对成年头部受伤患者进行CT成像检查时，可能避免的检查的患病率。
我们对2008年至2013年接受头部CT检查的成年急诊科创伤患者进行了电子健康记录（EHR）数据库和病历审查。主要结果是与加拿大CT头部规则（CCHR）高危标准不一致；次要结果是在不一致队列中使用神经外科干预措施。我们查询了全系统的电子健康记录，以使用离散数据字段中可识别的标准来识别CCHR不一致情况。对一部分不一致的CT进行明确的病历审查，提供了错误分类偏差的估计，并评估了实际接受干预的低风险病例。
在27240例成年创伤性头部CT检查中，EHR数据显示11432例（42.0%）与CCHR建议不一致。随后的病历审查显示，基于EHR的不一致判定在12.2%（95%置信区间[CI]=5.6%至18.8%）的情况下不准确。判断CCHR一致性的评分者间信度为95%（κ=0.86）。因此，我们估计36.8%的创伤性头部CT检查实际上可能是可以避免的（95%CI=34.1%至39.6%）。在EHR识别出的可能避免的CT组中，只有0.1%（n=13）接受了神经外科干预。病历审查显示，这些病例中没有一例实际上被CCHR“漏诊”，因为所有13例都被错误分类。
应用CCHR，目前对成年头部受伤患者进行的头部CT检查中约有三分之一可能是可以避免的。在此类患者中避免CT检查不太可能遗漏任何重要损伤。
