P045 Development and evaluation of machine learning algorithms for the prediction of opioid-related deaths among UK patients with non-cancer pain
P045 Development and evaluation of machine learning algorithms for the prediction of opioid-related deaths among UK patients with non-cancer pain
复制标题
P045 用于预测英国非癌性疼痛患者阿片类药物相关死亡的机器学习算法的开发和评估
DOI：
10.1093/rheumatology/kead104.086
复制
发表时间：
2023
期刊：
影响因子：
5.5
通讯作者：
Benitez-Aurioles J
中科院分区：
文献类型：
--
作者：
Benitez-Aurioles J
背景/目标全球范围内使用阿片类药物治疗非癌症疼痛的人数急剧增加。尽管人们对不良反应的认识有所提高，但它们在英国仍然很常见。临床预测模型提供了评估给定结果的个体风险的可能性，从而可以更好地将资源分配给面临风险的人。机器学2017 年 12 月 31 日已在临床实践研究数据链 (CPRD) 中确定。仅包括新的阿片类药物使用者。索引日期是首次处方的日期，在退出 CPRD 时或两年没有阿片类药物处方后进行审查。从每位患者的记录中提取基线数据，包括人口统计信息、合并症、合并用药以及所开处方的阿片类药物类型，收集了 49 个候选预测因子。这些用于训练三个竞争风险模型：具有 LASSO 正则化的 Fine&Gray 回归模型、生存随机森林 (RF) 和神经网络 (DeepHit)。结果是阿片类药物相关死亡率和其他原因死亡率是竞争事件，使用精心策划的 ICD-10 代码表进行定义。使用 5 倍交叉验证计算模型的预测性能，例如受试者特征算子曲线下面积 (AUC-ROC)。 结果我们总共纳入了 1,029,681 名患者，其中 1,240 名患者经历了阿片类药物相关死亡，52,833 名患者经历了竞争性死亡。 Fine&Gray、RF 和 DeepHit 模型的平均 AUC-ROC 值为 0.83（95% CI：分别为0.81-0.85)、0.78(0.77-0.79)和0.81(0.80-0.82)。在最佳风险切点，根据 Youden 指数，模型的灵敏度为 0.82（0.78-0.85）、0.75（0.67-0.82）和 0.80（0.78-0.83），特异性为 0.78（0.73-0.82）、0.75（0.68-0.83）和预测 12 个月风险时分别为 0.78（0.75-0.80）。 在 Fine&Gray 模型中，与风险增加相关的因素是物质使用障碍史（风险比 [HR]：3.40，95% CI：3.12-3.69）和酗酒（HR：3.07，95% CI：2.93-3.22）。慢性阻塞性肺病（HR：1.53，95％CI：1.48-1.58）和中度肝病（HR：1.31，95％CI：0.99-1.63）是与最高风险相关的合并症。起始时使用吗啡（HR：2.39，95％CI：2.08-2.69）和羟考酮（HR：1.10，95％CI：1.00-1.20）以及同时使用加巴喷丁类药物（HR：1.99，95％CI：1.80-2.18）和苯二氮卓类药物（HR：1.30，95％） CI：1.24-1.36）与风险增加相关。风湿性疾病的 HR 为 1.08 (95% CI:1.01-1.14)。 结论 Fine&Gray 和 DeepHit 模型表现出相当的判别性能。药物滥用、肺部和肝脏合并症、加巴喷丁类药物和苯二氮卓类药物开始使用吗啡或羟考酮以及联合处方加巴喷丁类药物和苯二氮卓类药物是与阿片类药物相关死亡率较高风险相关的一些因素。贝尼特斯-奥里奥尔莱斯：无。D.詹金斯：没有。黄：没有。拉米雷斯·梅迪纳：无。窥视：无。M.贾尼：没有。
Background/AimsThere has been a sharp rise in the use of opioids for non-cancer pain globally. Despite increased awareness of adverse effects, they remain commonly prescribed in the UK. Clinical prediction models offer the possibility of assessing individual risk for a given outcome allowing better allocation of resources towards those at risk. Machine learning (ML) approaches can address nonlinear relationships and complex interactions between variables and are increasingly used to develop these models.Our objective is to develop, validate, and compare the performance of three clinical prediction models based on regression and ML, which leverage primary care data to estimate the risk of opioid-related death in patients prescribed opioids for non-cancer pain.MethodsPatients ≥18 years old without prior cancer who were prescribed any opioid between 01/01/2006 and 31/12/2017 were identified in the Clinical Practice Research Datalink (CPRD). Only new opioid users were included. Index date was date of first prescription, with censoring at withdrawal from the CPRD or after not having an opioid prescription for two years. Baseline data were extracted from each patient’s records, including demographic information, comorbidities, concomitant medications, and the opioid type being prescribed, collecting 49 candidate predictors. These were used to train three competing risk models: a Fine&Gray regression model with LASSO regularisation, a survival random forest (RF), and a neural network (DeepHit). The outcome was opioid-related mortality and other cause mortality the competing event, defined using a curated ICD-10 codelist. Predictive performance of the models, such as the area under the receiver characteristic operator curve (AUC-ROC), were calculated using 5-fold cross validation.ResultsWe included a total of 1,029,681 patients, of which 1,240 experienced an opioid-related death, and 52,833 experienced a competing death.The Fine&Gray, RF and DeepHit models achieved average AUC-ROC values of 0.83(95% CI: 0.81-0.85), 0.78(0.77-0.79) and 0.81(0.80-0.82) respectively. At the optimum risk cut point, as per Youden’s index, the models achieved sensitivities of 0.82(0.78-0.85), 0.75(0.67-0.82) and 0.80(0.78-0.83), and specificities of 0.78(0.73-0.82), 0.75(0.68-0.83) and 0.78(0.75-0.80) when predicting 12-month risk, respectively.In the Fine&Gray model, factors associated with an increased risk were history of substance use disorder (hazards ratio [HR]: 3.40, 95% CI:3.12-3.69) and alcohol abuse (HR:3.07, 95% CI:2.93-3.22). COPD (HR:1.53, 95% CI:1.48-1.58) and moderate liver disease (HR:1.31, 95% CI:0.99-1.63) were the comorbidities associated with highest risk. Morphine (HR:2.39, 95% CI:2.08-2.69) and oxycodone (HR:1.10, 95% CI:1.00-1.20) at initiation and concomitant gabapentinoids (HR:1.99, 95% CI:1.80-2.18) and benzodiazepines (HR:1.30, 95% CI:1.24-1.36) were associated with an increased risk. HR for rheumatologic diseases was 1.08 (95% CI:1.01-1.14).ConclusionThe Fine&Gray and DeepHit models exhibited comparable discriminative performance. Substance abuse, lung and liver comorbidities, morphine or oxycodone at initiation and co-prescription of gabapentinoids and benzodiazepines, were some of the factors associated with a higher risk of opioid-related mortality.DisclosureJ. Benitez-Aurioles:None.D. Jenkins:None.Y. Huang:None.C. Ramirez Medina:None.N. Peek:None.M. Jani:None.