PDF(Sci-hub)
Abstract：
Construct: We aimed to develop an instrument to measure the quality of inpatient electronic health record- (EHR-) generated progress notes without requiring raters to review the detailed chart or know the patient.
BACKGROUND: Notes written in EHRs have generated criticism for being unnecessarily long and redundant, perpetuating inaccuracy and obscuring providers\' clinical reasoning. Available assessment tools either focus on outpatient progress notes or require chart review by raters to develop familiarity with the patient.
METHODS: We used medical literature, local expert review, and attending focus groups to develop and refine an instrument to evaluate inpatient progress notes. We measured interrater reliability and scored the selected-response elements of the checklist for a sample of 100 progress notes written by PGY-1 trainees on the general medicine service.
RESULTS: We developed an instrument with 18 selected-response items and four open-ended items to measure the quality of inpatient progress notes written in the EHR. The mean Cohen\'s kappa coefficient demonstrated good agreement at .67. The mean note score was 66.9% of maximum possible points (SD = 10.6, range = 34.4%-93.3%).
CONCLUSIONS: We present validity evidence in the domains of content, internal structure, and response process for a new checklist for rating inpatient progress notes. The scored checklist can be completed in approximately 7 minutes by a rater who is not familiar with the patient and can be done without extensive chart review. We further demonstrate that trainee notes show substantial room for improvement.
BACKGROUND: Notes written in EHRs have generated criticism for being unnecessarily long and redundant, perpetuating inaccuracy and obscuring providers\' clinical reasoning. Available assessment tools either focus on outpatient progress notes or require chart review by raters to develop familiarity with the patient.
METHODS: We used medical literature, local expert review, and attending focus groups to develop and refine an instrument to evaluate inpatient progress notes. We measured interrater reliability and scored the selected-response elements of the checklist for a sample of 100 progress notes written by PGY-1 trainees on the general medicine service.
RESULTS: We developed an instrument with 18 selected-response items and four open-ended items to measure the quality of inpatient progress notes written in the EHR. The mean Cohen\'s kappa coefficient demonstrated good agreement at .67. The mean note score was 66.9% of maximum possible points (SD = 10.6, range = 34.4%-93.3%).
CONCLUSIONS: We present validity evidence in the domains of content, internal structure, and response process for a new checklist for rating inpatient progress notes. The scored checklist can be completed in approximately 7 minutes by a rater who is not familiar with the patient and can be done without extensive chart review. We further demonstrate that trainee notes show substantial room for improvement.
摘要：
构造：我们旨在开发一种工具来测量住院电子健康记录（EHR）生成的进度记录的质量，而无需评估者查看详细的图表或了解患者。
背景：用EHR编写的笔记因不必要的冗长和冗余而受到批评，持续的不准确和模糊的提供者\'临床推理。可用的评估工具要么专注于门诊进度记录，要么需要评估者进行图表审查以了解患者。
方法：我们使用医学文献，当地专家审查，并参加焦点小组，以开发和完善评估住院进度记录的工具。我们测量了评分者之间的可靠性，并对PGY-1学员在普通医学服务中撰写的100个进度记录样本的清单中的选定响应元素进行了评分。
结果：我们开发了一种包含18个选择响应项目和4个开放式项目的仪器，以测量EHR中编写的住院进度记录的质量。科恩的平均卡帕系数在.67时表现出良好的一致性。平均音符评分为最大可能点的66.9％（SD=10.6，范围=34.4％-93.3％）。
结论：我们在内容领域提供了有效性证据，内部结构,以及用于对住院进度记录进行评级的新检查表的响应过程。评分清单可以由不熟悉患者的评估者在大约7分钟内完成,并且可以在没有广泛的图表审查的情况下完成。我们进一步证明，见习笔记显示出很大的改进空间。
背景：用EHR编写的笔记因不必要的冗长和冗余而受到批评，持续的不准确和模糊的提供者\'临床推理。可用的评估工具要么专注于门诊进度记录，要么需要评估者进行图表审查以了解患者。
方法：我们使用医学文献，当地专家审查，并参加焦点小组，以开发和完善评估住院进度记录的工具。我们测量了评分者之间的可靠性，并对PGY-1学员在普通医学服务中撰写的100个进度记录样本的清单中的选定响应元素进行了评分。
结果：我们开发了一种包含18个选择响应项目和4个开放式项目的仪器，以测量EHR中编写的住院进度记录的质量。科恩的平均卡帕系数在.67时表现出良好的一致性。平均音符评分为最大可能点的66.9％（SD=10.6，范围=34.4％-93.3％）。
结论：我们在内容领域提供了有效性证据，内部结构,以及用于对住院进度记录进行评级的新检查表的响应过程。评分清单可以由不熟悉患者的评估者在大约7分钟内完成,并且可以在没有广泛的图表审查的情况下完成。我们进一步证明，见习笔记显示出很大的改进空间。
