Predicting Hospitalization and Outpatient Corticosteroid Use in Inflammatory Bowel Disease Patients Using Machine Learning.
Predicting Hospitalization and Outpatient Corticosteroid Use in Inflammatory Bowel Disease Patients Using Machine Learning.
复制标题
DOI：
10.1093/ibd/izx007
复制
发表时间：
2017-12-19
影响因子：
4.9
通讯作者：
Higgins PDR
中科院分区：
文献类型：
--
作者：
Waljee AK;Lipson R;Wiitala WL;Zhang Y;Liu B;Zhu J;Wallace B;Govani SM;Stidham RW;Hayward R;Higgins PDR
炎症性肠病（IBD）是一种慢性疾病，其特征是不可预测的发作和缓解期。准确预测病程的工具将极大地帮助制定治疗决策。本研究旨在建立一个模型，准确预测门诊皮质类固醇使用和住院治疗的联合终点，作为IBD发作的替代指标。评估的预测因素包括年龄、性别、种族、使用皮质类固醇免疫抑制药物（免疫调节剂和/或抗肿瘤坏死因子）、纵向实验室数据、既往ibd相关住院和门诊皮质类固醇处方的数量。我们使用逻辑回归和机器学习方法（随机森林[RF]）构建模型来预测6个月内IBD住院和/或皮质类固醇使用的联合终点。我们在2002年至2009年间确定了20,368名退伍军人健康管理局首次（指数）诊断为IBD的患者。基线logistic回归模型的受试者工作特征曲线下面积（AuROC）为0.68（95%可信区间[CI]， 0.67-0.68）。RF纵向模型的AuROC为0.85 （95% CI, 0.84-0.85）。既往住院或使用类固醇的RF纵向模型的AuROC为0.87 （95% CI, 0.87 - 0.88）。未来住院或使用类固醇的5个主要独立危险因素是年龄、平均血清白蛋白、免疫抑制药物使用、平均和最高血小板计数。当纳入特定模型时，既往住院和皮质类固醇使用具有高度预测性。一种新的机器学习模型大大提高了我们预测ibd相关住院治疗和门诊类固醇使用的能力。该模型可在护理点用于区分疾病爆发的高风险和低风险患者，从而实现个性化的治疗管理。
Inflammatory bowel disease (IBD) is a chronic disease characterized by unpredictable episodes of flares and periods of remission. Tools that accurately predict disease course would substantially aid therapeutic decision-making. This study aims to construct a model that accurately predicts the combined end point of outpatient corticosteroid use and hospitalizations as a surrogate for IBD flare. Predictors evaluated included age, sex, race, use of corticosteroid-sparing immunosuppressive medications (immunomodulators and/or anti-TNF), longitudinal laboratory data, and number of previous IBD-related hospitalizations and outpatient corticosteroid prescriptions. We constructed models using logistic regression and machine learning methods (random forest [RF]) to predict the combined end point of hospitalization and/or corticosteroid use for IBD within 6 months. We identified 20,368 Veterans Health Administration patients with the first (index) IBD diagnosis between 2002 and 2009. Area under the receiver operating characteristic curve (AuROC) for the baseline logistic regression model was 0.68 (95% confidence interval [CI], 0.67–0.68). AuROC for the RF longitudinal model was 0.85 (95% CI, 0.84–0.85). AuROC for the RF longitudinal model using previous hospitalization or steroid use was 0.87 (95% CI, 0.87–0.88). The 5 leading independent risk factors for future hospitalization or steroid use were age, mean serum albumin, immunosuppressive medication use, and mean and highest platelet counts. Previous hospitalization and corticosteroid use were highly predictive when included in specified models. A novel machine learning model substantially improved our ability to predict IBD-related hospitalization and outpatient steroid use. This model could be used at point of care to distinguish patients at high and low risk for disease flare, allowing individualized therapeutic management.
登录
查看更多内容
影响因子：
29.4
作者：
Kappelman, Michael D.;Rifas-Shiman, Sheryl L.;Porter, Carol Q.;Ollendorf, Daniel A.;Sandler, Robert S.;Galanko, Joseph A.;Finkelstein, Jonathan A.
通讯作者：
Finkelstein, Jonathan A.
影响因子：
3.7
作者：
Waljee AK;Wiitala WL;Govani S;Stidham R;Saini S;Hou J;Feagins LA;Khan N;Good CB;Vijan S;Higgins PD
通讯作者：
Higgins PD
DOI：
10.1197/jamia.m1935
发表时间：
2006-05-01
影响因子：
6.4
作者：
Khan, Agha N.;Griffith, Stanley P.;Bertolli, Jeanne
通讯作者：
Bertolli, Jeanne
影响因子：
9.3
作者：
McDonald, CJ;Huff, SM;Maloney, P
通讯作者：
Maloney, P
影响因子：
8
作者：
Waljee, Akbar K.;Sauder, Kay;Higgins, Peter D. R.
通讯作者：
Higgins, Peter D. R.