Predicting drug-resistant epilepsy - A machine learning approach based on administrative claims data.
Predicting drug-resistant epilepsy - A machine learning approach based on administrative claims data.
复制标题
DOI：
10.1016/j.yebeh.2018.10.013
复制
发表时间：
2018-12
期刊：
影响因子：
--
通讯作者：
Sun J
中科院分区：
文献类型：
--
作者：
An S;Malhotra K;Dilley C;Han-Burgess E;Valdez JN;Robertson J;Clark C;Westover MB;Sun J
确定医疗和药房索赔数据中的信息是否可以预测，在开出第一种抗癫痫药物 (AED) 时，哪些癫痫患者会对 AED 产生耐药性。我们分析了 2006 年至 2015 年 1,376,756 名癫痫患者的纵向理赔数据。其中，582,258 名患者满足所有纳入标准； 49,916 名患者最终对 AED 产生抗药性，操作上定义为提交了至少 4 种不同 AED 索赔的患者。我们构建了 1,270 个候选预测因子（“特征”），反映了人口统计、合并症、药物、手术、癫痫状况和付款人状况来描述该队列。在训练数据集（528,640 名患者）上，我们进行了方差分析 F 值测试来选择预测特征，并训练了多种预测算法，包括逻辑回归、支持向量机 (SVM) 和随机森林。仅包含年龄和性别的模型被用作基准模型。在保留测试集（53,618 名患者）上，最佳模型的受试者工作特征 (ROC) 曲线下面积 (AUC) [95% CI] 为 0.753 [0.747, 0.759]，而基准模型为 0.664 [0.658, 0.671]。此外，预测的耐药概率与观察到的频率非常匹配。与等待 2 次 AED 失败相比，我们的模型平均提前 2.25 年预测耐药性。使用机器学习方法根据大量理赔数据创建的预测模型可以准确预测哪些癫痫患者在开出第一种 AED 时会产生耐药性。预测难治性的能力可能有助于患者在癫痫病程早期考虑替代疗法。
To determine whether information in medical and pharmacy claims data can predict, at the time of prescribing the first antiepileptic drug (AED), which patients with epilepsy will become resistant to AEDs. We analyzed longitudinal claims data from 1,376,756 patients with epilepsy from 2006 to 2015. Of these, 582,258 satisfied all inclusion criteria; 49,916 were ultimately AED resistant, operationally defined as a patient with claims filed for at least 4 distinct AEDs. We constructed 1,270 candidate predictors (“features”) reflecting demographics, comorbidities, medications, procedures, epilepsy status, and payer status to characterize the cohort. On the training dataset (528,640 patients) we performed ANOVA F-value tests to select predictive features and trained several prediction algorithms, including logistic regression, support vector machines (SVM), and random forests. A model with only age and gender was used as a benchmark model. On a held-out test set (53,618 patients), the best model achieves an area under the receiver operating characteristic (ROC) curve (AUC) [95% CI] of 0.753 [0.747, 0.759], compared to 0.664 [0.658, 0.671] for the benchmark model. Moreover, predicted probabilities for drug resistance closely match the observed frequencies. Compared to waiting for 2 AED failures, our model predicts drug resistance on average 2.25 years earlier. Predictive models created from large claims data using machine learning methods can accurately predict which patients with epilepsy will prove drug resistant at the time of prescribing the first AED. The ability to predict refractoriness may help patients consider alternative therapies earlier in the course of their epilepsy.
登录
查看更多内容
统计评论14：逻辑回归。
影响因子：
15.1
作者：
Bewick, V;Cheek, L;Ball, J
通讯作者：
Ball, J
影响因子：
9.9
作者：
Fountain, N. B.;Van Ness, P. C.;Bever, C. T., Jr.
通讯作者：
Bever, C. T., Jr.
影响因子：
1.7
作者：
CHAKRAVARTI, N
通讯作者：
CHAKRAVARTI, N
糖尿病研究中的机器学习和数据挖掘方法。
影响因子：
6
作者：
Kavakiotis I;Tsave O;Salifoglou A;Maglaveras N;Vlahavas I;Chouvarda I
通讯作者：
Chouvarda I
影响因子：
5.6
作者：
Kwan, Patrick;Arzimanoglou, Alexis;French, Jacqueline
通讯作者：
French, Jacqueline