DataSheet1_Understanding Fatal and Non-Fatal Drug Overdose Risk Factors: Overdose Risk Questionnaire Pilot Study—Validation.docx

Background: Drug overdoses (fatal and non-fatal) are among the leading causes of death in population with substance use disorders. The aim of the current study was to identify risk factors for fatal and non-fatal drug overdose for predominantly opioid-dependent treatment–seeking population.Methods: Data were collected from 640 adult patients using a self-reported 25-item Overdose Risk (OdRi) questionnaire pertaining to drug use and identified related domains. The exploratory factor analysis (EFA) was primarily used to improve the interpretability of this questionnaire. Two sets of EFA were conducted; in the first set of analysis, all items were included, while in the second set, items related to the experience of overdose were removed. Logistic regression was used for the assessment of latent factors’ association with both fatal and non-fatal overdoses.Results: EFA suggested a three-factor solution accounting for 75 and 97% of the variance for items treated in the first and second sets of analysis, respectively. Factor 1 was common for both sets of EFA analysis, containing six items (Cronbach’s α = 0.70) focusing around “illicit drug use and lack of treatment.” In the first set of analysis, Factors 2 (Cronbach’s α = 0.60) and 3 (Cronbach’s α = 0.34) were focusing around “mental health and emotional trauma” and “chronic drug use and frequent overdose” domains, respectively. The increase of Factor 2 was found to be a risk factor for fatal drug overdose (adjusted coefficient = 1.94, p = 0.038). In the second set of analysis, Factors 2 (Cronbach’s α = 0.65) and 3 (Cronbach’s α = 0.59) as well as Factor 1 were found to be risk factors for non-fatal drug overdose ever occurring. Only Factors 1 and 3 were positively associated with non-fatal overdose (one in a past year).Conclusion: The OdRi tool developed here could be helpful for clinical studies for the overdose risk assessment. However, integrating validated tools for mental health can probably help refining the accuracy of latent variables and the questionnaire’s consistency. Mental health and life stress appear as important predictors of both fatal and non-fatal overdoses.

背景：药物过量（致死性及非致死性）是物质使用障碍人群中主要的死亡原因之一。本研究旨在识别致死性及非致死性药物过量的风险因素，研究对象主要为寻求治疗的阿片类药物依赖人群。方法：从640名成人患者中收集数据，使用自填式25项药物过量风险（OdRi）问卷，该问卷涉及药物使用及相关领域。主要采用探索性因素分析（EFA）来提高问卷的可解释性。进行了两套EFA分析；在第一套分析中，包含所有项目，而在第二套分析中，去除了与过量经历相关的项目。使用逻辑回归对潜在因素与致死性及非致死性过量的关联进行评估。结果：EFA结果表明，第一和第二套分析中的项目分别解释了75%和97%的方差。第一因素在两套EFA分析中均存在，包含六个项目（Cronbach’s α = 0.70），聚焦于“非法药物使用及缺乏治疗。”在第一套分析中，因素2（Cronbach’s α = 0.60）和因素3（Cronbach’s α = 0.34）分别聚焦于“心理健康及情感创伤”和“慢性药物使用及频繁过量”领域。因素2的增加被发现是致死性药物过量的风险因素（调整系数 = 1.94，p = 0.038）。在第二套分析中，因素2（Cronbach’s α = 0.65）、因素3（Cronbach’s α = 0.59）以及因素1均被认定为非致死性药物过量的风险因素。仅因素1和因素3与过去一年内的非致死性过量呈正相关。结论：本研究开发的OdRi工具可能有助于临床研究中药物过量风险的评估。然而，整合经过验证的心理健康工具可能有助于提高潜在变量和问卷一致性的准确性。心理健康和生活压力表现为致死性及非致死性过量的重要预测因素。