文献速递中毒与环境急症消化、代谢与肾脏急症

机器学习算法预测接受阿片类处方的Medicare受益人阿片类过量风险的评价

Evaluation of Machine-Learning Algorithms for Predicting Opioid Overdose Risk Among Medicare Beneficiaries With Opioid Prescriptions.
PubMed 文献 · JAMA Netw Open · 2019-03-01
一句话结论在560057名接受阿片类处方的Medicare受益人(无癌症)中,深度神经网络(DNN)预测后续3个月阿片类过量的C统计量为0.91(95%CI 0.88-0.93),优于LASSO、随机森林和多变量逻辑回归;DNN可将患者分为低、中、高风险组,低风险组(占76.2%)过量发生率仅约万分之一,但阳性预测值很低(0.18%)。

该预后研究纳入560057名2011年1月1日至2015年12月31日期间填写至少1张阿片类处方的非癌症Medicare按服务收费受益人,随机等分为训练、测试和验证样本,利用268个预测因子(社会人口学、健康状况、阿片类使用模式、医生层面和区域层面因素)在3个月窗口内预测启动阿片类治疗后后续3个月的过量风险。在验证样本中,DNN(C统计量=0.91;95%CI 0.88-0.93)和梯度提升机(GBM)(C统计量=0.90;95%CI 0.87-0.94)优于LASSO(C统计量=0.84;95%CI 0.80-0.89)、随机森林(RF)(C统计量=0.80;95%CI 0.75-0.84)和多变量逻辑回归(MLR)(C统计量=0.75;95%CI 0.69-0.80)。在优化的敏感度和特异度下,DNN敏感度为92.3%、特异度为75.7%、需评估人数(NNE)为542、阳性预测值为0.18%、阴性预测值为99.9%。DNN将患者分为低风险(占队列76.2%[142180])、中风险(占队列18.6%[34579])和高风险(占队列5.2%[9747])亚组,低风险亚组中仅万分之一有过量事件;超过90%的过量事件发生在高风险和中风险亚组,尽管因过量结局罕见,阳性预测值较低。作者认为机器学习算法似乎在阿片类过量的风险预测和分层方面表现良好,尤其能识别出过量风险极低的低风险亚组。

要点

  • 在560057名非癌症Medicare受益人(至少1张阿片类处方)中,DNN预测后续3个月阿片类过量的C统计量为0.91(95%CI 0.88-0.93),优于LASSO(0.84;95%CI 0.80-0.89)、RF(0.80;95%CI 0.75-0.84)和MLR(0.75;95%CI 0.69-0.80)。
  • 验证样本中,DNN在优化敏感度和特异度下敏感度为92.3%、特异度为75.7%、NNE为542、阳性预测值为0.18%、阴性预测值为99.9%。
  • DNN将队列分为低风险(76.2%[142180])、中风险(18.6%[34579])和高风险(5.2%[9747])亚组,低风险亚组过量发生率仅万分之一,超过90%的过量事件发生在中、高风险亚组。
  • 尽管DNN的C统计量高且能识别低风险亚组,但因过量结局罕见,阳性预测值低(0.18%),高风险分层不等于个体高概率事件。
  • 该模型基于Medicare按服务收费、非癌症、有阿片类处方的人群,预测启动阿片类治疗后后续3个月经住院和急诊索赔识别的过量事件。

适用人群与场景

  • 非癌症Medicare按服务收费受益人(n=560057),在2011年1月1日至2015年12月31日期间填写1张或以上阿片类处方。
  • 受益人随机等分为训练(n=186686)、测试(n=186685)和验证(n=186686)样本;平均(SD)年龄68.0(14.5)岁,约63%为女性,82%为白人,35%有残疾,41%为双重资格,0.60%有至少1次过量事件。
  • 预测窗口为启动阿片类治疗后后续3个月;预测因子在3个月窗口中测量,从启动阿片类前3个月开始,直至失访或观察结束。

诊断标准与关键阈值

  • 阿片类过量事件通过住院和急诊科索赔识别。
  • 使用Youden指数确定预测评分的优化阈值,以平衡敏感度和特异度。
  • DNN模型将患者分为低风险、中风险和高风险亚组;原文未报告各亚组的具体评分界值。

研究设计与关键结果

  • 研究为预后研究,于2017年9月1日至2018年12月31日进行;纳入560057名非癌症Medicare按服务收费受益人,随机等分为训练(n=186686)、测试(n=186685)和验证(n=186686)样本。
  • 在验证样本中,DNN(C统计量=0.91;95%CI 0.88-0.93)和GBM(C统计量=0.90;95%CI 0.87-0.94)预测阿片类过量的性能优于LASSO(C统计量=0.84;95%CI 0.80-0.89)、RF(C统计量=0.80;95%CI 0.75-0.84)和MLR(C统计量=0.75;95%CI 0.69-0.80)。
  • 在优化的敏感度和特异度下,DNN敏感度为92.3%、特异度为75.7%、NNE为542、阳性预测值为0.18%、阴性预测值为99.9%。
  • DNN将患者分为低风险(76.2%[142180])、中风险(18.6%[34579])和高风险(5.2%[9747])亚组,低风险亚组中仅万分之一有过量事件;超过90%的过量事件发生在高风险和中风险亚组,但阳性预测值低,因过量结局罕见。

证据类型、等级与局限

  • 证据类型:预后研究(机器学习算法开发与验证)。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为机器学习算法似乎在阿片类过量的风险预测和分层方面表现良好,尤其能识别出过量风险极低的低风险亚组。
  • 原文未据此给出针对急诊或ICU的具体推荐。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开

IMPORTANCE: Current approaches to identifying individuals at high risk for opioid overdose target many patients who are not truly at high risk.

OBJECTIVE: To develop and validate a machine-learning algorithm to predict opioid overdose risk among Medicare beneficiaries with at least 1 opioid prescription.

DESIGN, SETTING, AND PARTICIPANTS: A prognostic study was conducted between September 1, 2017, and December 31, 2018. Participants (n = 560 057) included fee-for-service Medicare beneficiaries without cancer who filled 1 or more opioid prescriptions from January 1, 2011, to December 31, 2015. Beneficiaries were randomly and equally divided into training, testing, and validation samples.

EXPOSURES: Potential predictors (n = 268), including sociodemographics, health status, patterns of opioid use, and practitioner-level and regional-level factors, were measured in 3-month windows, starting 3 months before initiating opioids until loss of follow-up or the end of observation.

MAIN OUTCOMES AND MEASURES: Opioid overdose episodes from inpatient and emergency department claims were identified. Multivariate logistic regression (MLR), least absolute shrinkage and selection operator-type regression (LASSO), random forest (RF), gradient boosting machine (GBM), and deep neural network (DNN) were applied to predict overdose risk in the subsequent 3 months after initiation of treatment with prescription opioids. Prediction performance was assessed using the C statistic and other metrics (eg, sensitivity, specificity, and number needed to evaluate [NNE] to identify one overdose). The Youden index was used to identify the optimized threshold of predicted score that balanced sensitivity and specificity.

RESULTS: Beneficiaries in the training (n = 186 686), testing (n = 186 685), and validation (n = 186 686) samples had similar characteristics (mean [SD] age of 68.0 [14.5] years, and approximately 63% were female, 82% were white, 35% had disabilities, 41% were dual eligible, and 0.60% had at least 1 overdose episode). In the validation sample, the DNN (C statistic = 0.91; 95% CI, 0.88-0.93) and GBM (C statistic = 0.90; 95% CI, 0.87-0.94) algorithms outperformed the LASSO (C statistic = 0.84; 95% CI, 0.80-0.89), RF (C statistic = 0.80; 95% CI, 0.75-0.84), and MLR (C statistic = 0.75; 95% CI, 0.69-0.80) methods for predicting opioid overdose. At the optimized sensitivity and specificity, DNN had a sensitivity of 92.3%, specificity of 75.7%, NNE of 542, positive predictive value of 0.18%, and negative predictive value of 99.9%. The DNN classified patients into low-risk (76.2% [142 180] of the cohort), medium-risk (18.6% [34 579] of the cohort), and high-risk (5.2% [9747] of the cohort) subgroups, with only 1 in 10 000 in the low-risk subgroup having an overdose episode. More than 90% of overdose episodes occurred in the high-risk and medium-risk subgroups, although positive predictive values were low, given the rare overdose outcome.

CONCLUSIONS AND RELEVANCE: Machine-learning algorithms appear to perform well for risk prediction and stratification of opioid overdose, especially in identifying low-risk subgroups that have minimal risk of overdose.

来源:JAMA Netw Open,查看原文。版权归原作者所有。

参考来源

  1. Lo-Ciganic WH, Huang JL, Zhang HH, et al. Evaluation of Machine-Learning Algorithms for Predicting Opioid Overdose Risk Among Medicare Beneficiaries With Opioid Prescriptions. JAMA Netw Open. 2019. PMID: 30901048. doi:10.1001/jamanetworkopen.2019.0968

供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。