León Casey, Sung Minhee L, Reisman Joel I, Liu Weisong, Kerns Robert D, Gordon Kirsha S, Mitra Avijit, Kwon Sunjae, Yu Hong, Becker William C, Li Wenjun
Department of Public Health and Center for Health Statistics, University of Massachusetts Lowell, Lowell, MA.
Pain Research, Informatics, Multimorbidities and Education Center, Veterans Affairs Connecticut Healthcare System, West Haven, CT.
Clin J Pain. 2025 Feb 1;41(2):e1266. doi: 10.1097/AJP.0000000000001266.
Neurocognitive symptoms (NCSs) may be early indicators of opioid-related harm. We aimed to evaluate the incidence and potential attribution of opioid-related NCS among patients on long-term opioid therapy (LOT) by using natural language processing to extract data from the electronic health records within the Veterans Health Administration.
We conducted a retrospective cohort study of patients prescribed LOT in 2018. NCS were identified from clinical notes across 3 periods: 6 months before LOT initiation, during LOT, and 1-year post-LOT discontinuation. Opioid-related NCS included cognitive impairment, sedation, light-headedness, altered mental status, and intoxication. We calculated incidence rates, incidence rate ratios, and hazard ratios to evaluate the occurrence and potential opioid attribution of NCS across these periods.
Among 55,652 patients, 3.1% experienced opioid-related NCS, with the highest incidence observed during LOT. Prevalence of NCS was greater in patients who were: between 55 and 64 (3.6%) or 65 and 74 years old (3.2%), Asian (4.8%, P = 0.02), and had received treatment for substance use disorders (7.1%, P = 0.01). In adjusted proportional hazards models, identified Asian race (hazard ratio: 2.20 [95% CI: 1.09-4.44], P = 0.03), and cooccurring conditions dementia (1.50 [1.12-2.00], 0.01), depression (1.31 [1.14-1.49], <0.01), posttraumatic stress disorder (1.18 [1.02-1.37], 0.02), substance use disorder (1.62 [1.36-1.92], 0.01), cardiovascular disease (1.18 [1.01-1.37], 0.04), chronic obstructive pulmonary disease (1.16 [1.01-1.33], 0.03), cirrhosis (1.73 [1.34-2.24], 0.01), chronic kidney disease (1.41 [1.19, 1.66]; 0.01) and traumatic brain injury (1.42 [1.06-1.91], 0.02) were associated with increased likelihood of NCS. Likelihood of NCS increased with LOT dose and decreased with LOT duration.
This study suggests that opioid-related NCS are most likely to occur during LOT, indicating a potential temporal association with opioid use. These findings highlight the importance of monitoring NCS in patients on LOT as part of a broader strategy to mitigate opioid-related harms.
神经认知症状(NCSs)可能是阿片类药物相关危害的早期指标。我们旨在通过使用自然语言处理从退伍军人健康管理局的电子健康记录中提取数据,评估长期接受阿片类药物治疗(LOT)的患者中阿片类药物相关NCS的发生率和潜在归因。
我们对2018年开具LOT处方的患者进行了一项回顾性队列研究。通过3个时间段的临床记录识别NCS:LOT开始前6个月、LOT期间以及LOT停药后1年。阿片类药物相关NCS包括认知障碍、镇静、头晕、精神状态改变和中毒。我们计算了发病率、发病率比和风险比,以评估这些时间段内NCS的发生情况和潜在的阿片类药物归因。
在55652名患者中,3.1%经历了阿片类药物相关NCS,其中LOT期间的发病率最高。NCS的患病率在以下患者中更高:年龄在55至64岁(3.6%)或65至74岁(3.2%)之间、亚洲人(4.8%,P = 0.02)以及曾接受物质使用障碍治疗的患者(7.1%,P = 0.01)。在调整后的比例风险模型中,确定亚洲种族(风险比:2.20 [95%置信区间:1.09 - 4.44],P = 0.03)、同时存在痴呆症(1.50 [1.12 - 2.00],0.01)、抑郁症(1.31 [1.14 - 1.49],<0.01)、创伤后应激障碍(1.18 [1.02 - 1.37],0.02)、物质使用障碍(1.62 [1.36 - 1.92],0.01)、心血管疾病(1.18 [1.01 - 1.37],0.04)、慢性阻塞性肺疾病(1.16 [1.01 - 1.33],0.03)、肝硬化(1.73 [1.34 - 2.24],0.01)、慢性肾脏病(1.41 [1.19, 1.66];0.01)和创伤性脑损伤(1.42 [1.06 - 1.91],0.02)与NCS发生可能性增加相关。NCS的可能性随LOT剂量增加而增加,随LOT持续时间减少。
本研究表明,阿片类药物相关NCS最有可能在LOT期间发生,表明与阿片类药物使用存在潜在的时间关联。这些发现凸显了在接受LOT的患者中监测NCS的重要性,作为减轻阿片类药物相关危害的更广泛策略的一部分。