1 Temple Fox Chase Cancer Center , Department of Internal Medicine, Section of Hematology Oncology, Philadelphia, Pennsylvania.
J Palliat Med. 2014 May;17(5):521-6. doi: 10.1089/jpm.2013.0454. Epub 2014 Apr 10.
Prognostication is an important element of palliative care consultations. Research has shown that estimated survivals offered by clinicians are often inaccurate; however, few of these studies have focused on the documentation and prognostic accuracy of palliative care providers.
Our aim was to determine whether palliative care clinicians document specific estimates of survival in the electronic medical record and whether these survival estimates are accurate.
We retrospectively analyzed 400 consecutive, new palliative care consults at an urban, academic medical center from October 1, 2009 to December 31, 2010. Descriptive statistics were used to summarize patient demographics, median patient survival, documented estimated survival, agreement between estimated and actual survival, and agreement differences among disease groups.
The inpatient consult note template was utilized by the clinicians in 94.2% of the patients analyzed, and 69.4% of the patients analyzed had a specific survival estimate documented. Of the patients with specific survival estimates documented, 42.6% died in the time frame estimated. Weighted kappa coefficients and Kaplan survival estimators showed fair to moderate agreement between actual survival and estimated survival offered by palliative care clinicians. Survival groups with the shortest prognosis had the most accurate estimates of prognosis. Cancer had the least agreement between estimated and actual survival among disease types. Overestimation of survival was the most common prognostic error. Use of a template resulted in significantly greater documentation of a specific estimated survival.
The prognostic accuracy of palliative care physicians in this study was similar to physician accuracy in other studies. Trends toward overestimation were also similar to those seen in previous research. Use of a template in the electronic medical record (EMR) increases documentation of estimated prognosis.
预后是姑息治疗咨询的一个重要组成部分。研究表明，临床医生提供的预计生存时间往往不准确；然而，这些研究很少关注姑息治疗提供者的记录和预后准确性。
我们的目的是确定姑息治疗临床医生是否在电子病历中记录了具体的生存估计值，以及这些生存估计值是否准确。
我们回顾性分析了 2009 年 10 月 1 日至 2010 年 12 月 31 日在一家城市学术医疗中心进行的 400 例连续新的姑息治疗咨询。使用描述性统计来总结患者人口统计学特征、中位患者生存时间、记录的估计生存时间、估计生存时间与实际生存时间的一致性以及不同疾病组之间的一致性差异。
在分析的 400 例患者中，94.2%的患者使用了住院会诊记录模板，69.4%的患者记录了具体的生存估计值。在记录了具体生存估计值的患者中，42.6%的患者在估计的时间范围内死亡。实际生存时间和姑息治疗临床医生提供的估计生存时间之间的加权 kappa 系数和 Kaplan 生存估计器显示出良好到中度的一致性。预后最短的生存组具有最准确的预后估计。癌症在疾病类型中与估计和实际生存之间的一致性最低。生存时间的高估是最常见的预后错误。在电子病历（EMR）中使用模板可显著增加估计预后的记录。
本研究中姑息治疗医生的预后准确性与其他研究中的医生准确性相似。高估的趋势也与之前的研究相似。在电子病历中使用模板可增加对估计预后的记录。
