新生儿戒断综合征预测模型的开发与验证
本研究基于2009至2014年田纳西州医疗补助计划中218 020对母婴数据,开发并内部验证了两种预测新生儿戒断综合征(NAS)的临床风险模型。模型纳入人口学及产前暴露协变量,区分度良好(AUC均为0.89),对低风险婴儿校准良好,可用于风险分层。
要点
- 基于30个产前暴露协变量构建了通用和高效两种NAS预测模型。
- 通用模型纳入母亲年龄、丙肝感染、阿片类使用天数、吸烟量及特定药物使用。
- 两模型AUC均为0.89,对低风险婴儿校准良好,可助风险分层。
核心结论
- 基于人口学及妊娠最后30天产前暴露可构建预测NAS的模型,能对婴儿进行风险分层。
研究设计与样本
- 基于田纳西州医疗补助计划2009至2014年参保者数据,开发并内部验证(bootstrap交叉验证)NAS预测模型。
- 共纳入218 020对母婴,其中3208名婴儿诊断为NAS。
- 使用逻辑回归和基于AIC改善的后向选择构建模型。
主要发现
- 通用模型纳入母亲年龄、丙肝感染、按类型阿片使用天数、每日吸烟数及妊娠最后30天使用安非他酮、抗恶心药、苯二氮䓬类、抗精神病药和加巴喷丁。
- 婴儿特征包括出生体重、小于胎龄儿和婴儿性别。
- 高效模型使用更少的预测变量。
- 两模型区分度良好,AUC均为0.89,对低风险婴儿校准良好。
关键数值与推荐
- 通用模型纳入了母亲年龄、丙肝感染、按类型阿片使用天数、每日吸烟数及妊娠最后30天使用安非他酮、抗恶心药、苯二氮䓬类、抗精神病药和加巴喷丁。
- 婴儿特征包括出生体重、小于胎龄儿和婴儿性别。
- 高效模型使用更少的预测变量。
- 两模型AUC均为0.89,对低风险婴儿校准良好。
临床怎么用
- 可利用模型对婴儿发生NAS的风险进行分层,辅助临床监测与资源分配。
局限性
- 模型仅内部验证(bootstrap交叉验证),未进行外部验证。
- 数据来源于田纳西州医疗补助计划参保者,外推性需谨慎。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: To develop and validate clinical risk prediction tools for neonatal abstinence syndrome (NAS). STUDY DESIGN: We developed prediction models for NAS based on a set of 30 demographic and antenatal exposure covariates collected during pregnancy. Data (outpatient prescription, vital, and administrative records), were obtained from enrollees in the Tennessee Medicaid Program from 2009 to 2014. Models were created using logistic regression and backward selection based on improvement in the Akaike information criterion, and internally validated using bootstrap cross-validation. RESULTS: A total of 218 020 maternal and infant dyads met inclusion criteria, of whom 3208 infants were diagnosed with NAS. The general population model included age, hepatitis C virus infection, days of opioid used by type, number of cigarettes used daily, and the following medications used in the last 30 day of pregnancy: bupropion, antinausea medicines, benzodiazepines, antipsychotics, and gabapentin. Infant characteristics included birthweight, small for gestational age, and infant sex. A high-risk model used a smaller number of predictive variables. Both models discriminated well with an area under the curve of 0.89 and were well-calibrated for low-risk infants. CONCLUSIONS: We developed 2 predictive models for NAS based on demographics and antenatal exposure during the last 30 days of pregnancy that were able to risk stratify infants at risk of developing the syndrome.
引用
- Patrick SW, Slaughter JC, Harrell FE, et al. Development and Validation of a Model to Predict Neonatal Abstinence Syndrome. J Pediatr. 2020. PMID: 33080277. doi:10.1016/j.jpeds.2020.10.030
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。