Department of Medicine, Division of General Internal Medicine and Geriatrics, Indiana University School of Medicine, Indianapolis, IN, USA.
Indiana University Health Physicians Inc, Indianapolis, IN, USA.
J Gen Intern Med. 2021 Aug;36(8):2244-2250. doi: 10.1007/s11606-021-06593-z. Epub 2021 Jan 27.
Predicting the risk of in-hospital mortality on admission is challenging but essential for risk stratification of patient outcomes and designing an appropriate plan-of-care, especially among transferred patients.
Develop a model that uses administrative and clinical data within 24 h of transfer to predict 30-day in-hospital mortality at an Academic Health Center (AHC).
Retrospective cohort study. We used 30 putative variables in a multiple logistic regression model in the full data set (n = 10,389) to identify 20 candidate variables obtained from the electronic medical record (EMR) within 24 h of admission that were associated with 30-day in-hospital mortality (p < 0.05). These 20 variables were tested using multiple logistic regression and area under the curve (AUC)-receiver operating characteristics (ROC) analysis to identify an optimal risk threshold score in a randomly split derivation sample (n = 5194) which was then examined in the validation sample (n = 5195).
Ten thousand three hundred eighty-nine patients greater than 18 years transferred to the Indiana University (IU)-Adult Academic Health Center (AHC) between 1/1/2016 and 12/31/2017.
Sensitivity, specificity, positive predictive value, C-statistic, and risk threshold score of the model.
The final model was strongly discriminative (C-statistic = 0.90) and had a good fit (Hosmer-Lemeshow goodness-of-fit test [X (8) =6.26, p = 0.62]). The positive predictive value for 30-day in-hospital death was 68%; AUC-ROC was 0.90 (95% confidence interval 0.89-0.92, p < 0.0001). We identified a risk threshold score of -2.19 that had a maximum sensitivity (79.87%) and specificity (85.24%) in the derivation and validation sample (sensitivity: 75.00%, specificity: 85.71%). In the validation sample, 34.40% (354/1029) of the patients above this threshold died compared to only 2.83% (118/4166) deaths below this threshold.
This model can use EMR and administrative data within 24 h of transfer to predict the risk of 30-day in-hospital mortality with reasonable accuracy among seriously ill transferred patients.
预测入院时院内死亡率的风险具有挑战性，但对于患者结局的风险分层和制定适当的治疗计划至关重要，尤其是在转院患者中。
开发一种模型，该模型使用转院后 24 小时内的管理和临床数据来预测学术医疗中心（AHC）的 30 天院内死亡率。
回顾性队列研究。我们在全数据集（n=10389）中使用了 30 个假定变量的多元逻辑回归模型，以确定 20 个候选变量，这些变量是从入院后 24 小时内的电子病历（EMR）中获得的，与 30 天院内死亡率相关（p<0.05）。这些 20 个变量使用多元逻辑回归和曲线下面积（AUC）-接受者操作特征（ROC）分析进行测试，以在随机分割的推导样本（n=5194）中确定最佳风险阈值评分，然后在验证样本（n=5195）中进行检查。
2016 年 1 月 1 日至 2017 年 12 月 31 日期间，10389 名年龄大于 18 岁的患者转至印第安纳大学（IU）成人学术医疗中心（AHC）。
模型的灵敏度、特异性、阳性预测值、C 统计量和风险阈值评分。
最终模型具有很强的判别能力（C 统计量=0.90），拟合度良好（Hosmer-Lemeshow 拟合优度检验[X（8）=6.26，p=0.62]）。30 天院内死亡的阳性预测值为 68%；AUC-ROC 为 0.90（95%置信区间 0.89-0.92，p<0.0001）。我们确定了一个风险阈值评分-2.19，在推导和验证样本中具有最高的灵敏度（79.87%）和特异性（85.24%）（灵敏度：75.00%，特异性：85.71%）。在验证样本中，34.40%（354/1029）的患者超过该阈值，而只有 2.83%（118/4166）的患者低于该阈值死亡。
该模型可以使用转院后 24 小时内的 EMR 和管理数据，在严重转院患者中以合理的准确性预测 30 天院内死亡率的风险。
