The First Branch, the First Affiliated Hospital of Chongqing Medical University, Chongqing, China.
Department of Clinical Epidemiology and Health Economics, School of Public Health, The University of Tokyo, Tokyo, Japan.
BMC Med Inform Decis Mak. 2024 Aug 2;24(1):219. doi: 10.1186/s12911-024-02622-z.
This study aimed to create and validate robust machine-learning-based prediction models for antipsychotic drug (risperidone) continuation in children and teenagers suffering from mania over one year and to discover potential variables for clinical treatment.
The study population was collected from the national claims database in China. A total of 4,532 patients aged 4-18 who began risperidone therapy for mania between September 2013 and October 2019 were identified. The data were randomly divided into two datasets: training (80%) and testing (20%). Five regularly used machine learning methods were employed, in addition to the SuperLearner (SL) algorithm, to develop prediction models for the continuation of atypical antipsychotic therapy. The area under the receiver operating characteristic curve (AUC) with a 95% confidence interval (CI) was utilized.
In terms of discrimination and robustness in predicting risperidone treatment continuation, the generalized linear model (GLM) performed the best (AUC: 0.823, 95% CI: 0.792-0.854, intercept near 0, slope close to 1.0). The SL model (AUC: 0.823, 95% CI: 0.791-0.853, intercept near 0, slope close to 1.0) also exhibited significant performance. Furthermore, the present findings emphasize the significance of several unique clinical and socioeconomic variables, such as the frequency of emergency room visits for nonmental health disorders.
The GLM and SL models provided accurate predictions regarding risperidone treatment continuation in children and adolescents with episodes of mania and hypomania. Consequently, applying prediction models in atypical antipsychotic medicine may aid in evidence-based decision-making.
本研究旨在建立并验证用于预测儿童和青少年躁狂发作一年以上患者继续使用抗精神病药物（利培酮）的稳健机器学习预测模型，并发现潜在的临床治疗变量。
研究人群来自中国国家索赔数据库。共纳入 2013 年 9 月至 2019 年 10 月期间开始利培酮治疗躁狂的 4532 名 4-18 岁患者。数据随机分为训练集（80%）和测试集（20%）。采用了五种常用的机器学习方法，以及 SuperLearner（SL）算法，来开发预测非典型抗精神病药物治疗延续的预测模型。采用具有 95%置信区间（CI）的接收者操作特征曲线下面积（AUC）来评估。
在预测利培酮治疗延续方面，广义线性模型（GLM）在区分度和稳健性方面表现最佳（AUC：0.823，95%CI：0.792-0.854，截距接近 0，斜率接近 1.0）。SL 模型（AUC：0.823，95%CI：0.791-0.853，截距接近 0，斜率接近 1.0）也表现出显著的性能。此外，本研究结果强调了一些独特的临床和社会经济变量的重要性，如非精神健康障碍急诊就诊的频率。
GLM 和 SL 模型对儿童和青少年躁狂和轻躁狂发作患者继续使用利培酮治疗的情况进行了准确预测。因此，在非典型抗精神病药物治疗中应用预测模型可能有助于基于证据的决策。
