Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, United States of America.
Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA, United States of America.
PLoS One. 2020 Oct 20;15(10):e0241083. doi: 10.1371/journal.pone.0241083. eCollection 2020.
With increasing rates of opioid overdoses in the US, a surveillance tool to identify high-risk patients may help facilitate early intervention.
To develop an algorithm to predict overdose using routinely-collected healthcare databases.
Within a US commercial claims database (2011-2015), patients with ≥1 opioid prescription were identified. Patients were randomly allocated into the training (50%), validation (25%), or test set (25%). For each month of follow-up, pooled logistic regression was used to predict the odds of incident overdose in the next month based on patient history from the preceding 3-6 months (time-updated), using elastic net for variable selection. As secondary analyses, we explored whether using simpler models (few predictors, baseline only) or different analytic methods (random forest, traditional regression) influenced performance.
We identified 5,293,880 individuals prescribed opioids; 2,682 patients (0.05%) had an overdose during follow-up (mean: 17.1 months). On average, patients who overdosed were younger and had more diagnoses and prescriptions. The elastic net model achieved good performance (c-statistic 0.887, 95% CI 0.872-0.902; sensitivity 80.2, specificity 80.1, PPV 0.21, NPV 99.9 at optimal cutpoint). It outperformed simpler models based on few predictors (c-statistic 0.825, 95% CI 0.808-0.843) and baseline predictors only (c-statistic 0.806, 95% CI 0.787-0.26). Different analytic techniques did not substantially influence performance. In the final algorithm based on elastic net, the strongest predictors were age 18-25 years (OR: 2.21), prior suicide attempt (OR: 3.68), opioid dependence (OR: 3.14).
We demonstrate that sophisticated algorithms using healthcare databases can be predictive of overdose, creating opportunities for active monitoring and early intervention.
随着美国阿片类药物过量率的上升，一种用于识别高危患者的监测工具可能有助于促进早期干预。
开发一种使用常规收集的医疗保健数据库预测药物过量的算法。
在一个美国商业索赔数据库（2011-2015 年）中，确定了至少有 1 份阿片类药物处方的患者。患者被随机分配到训练集（50%）、验证集（25%）或测试集（25%）。对于每个随访月，使用基于前 3-6 个月（时间更新）患者病史的汇总逻辑回归，根据弹性网络进行变量选择，预测下一个月药物过量的可能性。作为次要分析，我们探讨了使用更简单的模型（少量预测因子，仅基线）或不同的分析方法（随机森林，传统回归）是否会影响性能。
我们确定了 5293880 名开处阿片类药物的患者；在随访期间，有 2682 名患者（0.05%）发生了药物过量（平均：17.1 个月）。平均而言，药物过量的患者更年轻，有更多的诊断和处方。弹性网络模型的性能良好（C 统计量为 0.887，95%置信区间为 0.872-0.902；灵敏度为 80.2%，特异性为 80.1%，阳性预测值为 0.21，阴性预测值为 99.9%，最佳切点）。它优于基于少量预测因子（C 统计量为 0.825，95%置信区间为 0.808-0.843）和仅基线预测因子（C 统计量为 0.806，95%置信区间为 0.787-0.26）的简单模型。不同的分析技术并没有显著影响性能。在最终基于弹性网络的算法中，最强的预测因子是年龄 18-25 岁（OR：2.21）、先前自杀企图（OR：3.68）和阿片类药物依赖（OR：3.14）。
我们证明了使用医疗保健数据库的复杂算法可以预测药物过量，为主动监测和早期干预创造了机会。
