作者: Benjamin C. Sun , Nicoleta Lupulescu-Mann , Christina J. Charlesworth , Hyunjee Kim , Daniel M. Hartung
DOI: 10.1016/J.ANNEMERGMED.2017.10.007
关键词: Propensity score matching 、 Controlled substance 、 Emergency department 、 Confidence interval 、 Medicine 、 Medicaid 、 Ambulatory care 、 Medical prescription 、 Medical emergency 、 Cohort 、 Emergency medicine
摘要: Study objective The link between prescription opioid shopping and overdose events is poorly understood. We test the hypothesis that a history of associated with increased risk events. Methods This secondary analysis linked claims controlled substance dispense database. studied adult Medicaid beneficiaries in 2014 use 6 months before an ambulatory care or emergency department visit pain-related diagnosis. primary outcome was nonfatal event within cohort entry date. exposure interest (opioid shopping) defined as having prescriptions by different prescribers greater than equal to 1-day overlap filled at 3 more pharmacies entry. used propensity score match shoppers nonshoppers 1:1 ratio. calculated absolute difference rates nonshoppers. Results 66,328 patients, including 2,571 (3.9%). There were 290 patients (0.4%) overall who experienced overdose. In unadjusted analyses, had higher (rate 4.4 per 1,000; 95% confidence interval 0.8 7.9). After matching, there no differences 0.4 –4.7 5.5). These findings robust various definitions look-back periods. Conclusion Prescription not independently