Rich Shannan N, Jun Inyoung, Bian Jiang, Boucher Christina, Cherabuddi Kartik, Morris J Glenn, Prosperi Mattia
Department of Epidemiology, College of Public Health and Health Professions and Medicine, University of Florida, PO Box 100009, Gainesville, FL, 32610, USA.
Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA.
Infect Dis Ther. 2022 Oct;11(5):1869-1882. doi: 10.1007/s40121-022-00677-x. Epub 2022 Jul 31.
Urinary tract infections (UTIs) are common infections for which initial antibiotic treatment decisions are empirically based, often without antibiotic susceptibility testing to evaluate resistance, increasing the risk of inappropriate therapy. We hypothesized that models based on electronic health records (EHR) could assist in the identification of patients at higher risk for antibiotic-resistant UTIs and help guide the selection of antimicrobials in hospital and clinic settings.
EHR from multiple centers in North-Central Florida, including patient demographics, previous diagnoses, prescriptions, and antibiotic susceptibility tests, were obtained for 9990 patients diagnosed with a UTI during 2011-2019. Decision trees, boosted logistic regression (BLR), and random forest models were developed to predict resistance to common antibiotics used for UTI management [sulfamethoxazole-trimethoprim (SXT), nitrofurantoin (NIT), ciprofloxacin (CIP)] and multidrug resistance (MDR).
There were 6307 (63.1%) individuals with a UTI caused by a resistant microorganism. Overall, the population was majority female, white, non-Hispanic, and older aged (mean = 60.7 years). The BLR models yielded the highest discriminative ability, as measured by the out-of-bag area under the receiver-operating curve (AUROC), for the resistance outcomes [AUROC = 0.58 (SXT), 0.62 (NIT), 0.64 (CIP), and 0.66 (MDR)]. Variables in the best performing model were sex, history of UTIs, catheterization, renal disease, dementia, hemiplegia/paraplegia, and hypertension.
The discriminative ability of the prediction models was moderate. Nonetheless, these models based solely on EHR demonstrate utility for the identification of patients at higher risk for resistant infections. These models, in turn, may help guide clinical decision-making on the ordering of urine cultures and decisions regarding empiric therapy for these patients.
尿路感染（UTIs）是常见的感染性疾病，其初始抗生素治疗决策通常基于经验，往往未进行抗生素敏感性测试以评估耐药性，从而增加了不恰当治疗的风险。我们假设基于电子健康记录（EHR）的模型可以帮助识别抗生素耐药性UTIs风险较高的患者，并有助于指导医院和诊所环境中抗菌药物的选择。
获取了佛罗里达州中北部多个中心的电子健康记录，包括患者人口统计学信息、既往诊断、处方和抗生素敏感性测试，这些记录来自2011年至2019年期间诊断为UTI的9990名患者。开发了决策树、增强逻辑回归（BLR）和随机森林模型，以预测对用于UTI管理的常见抗生素[磺胺甲恶唑-甲氧苄啶（SXT）、呋喃妥因（NIT）、环丙沙星（CIP）]的耐药性以及多重耐药（MDR）情况。
有6307名（63.1%）个体的UTI由耐药微生物引起。总体而言，该人群以女性、白人、非西班牙裔和老年人为主（平均年龄=60.7岁）。对于耐药结果，通过袋外受试者工作特征曲线下面积（AUROC）衡量，BLR模型具有最高的判别能力[AUROC=0.58（SXT）、0.62（NIT）、0.64（CIP）和0.66（MDR）]。表现最佳模型中的变量包括性别、UTI病史、导尿、肾脏疾病、痴呆、偏瘫/截瘫和高血压。
预测模型的判别能力中等。尽管如此，这些仅基于电子健康记录的模型证明了在识别耐药感染风险较高患者方面的实用性。反过来，这些模型可能有助于指导关于尿液培养检查的临床决策以及这些患者经验性治疗的决策。
