Association of chronic opioid therapy with emergency department visits and hospitalizations among super-utilizers

Objectives Patients with complex medical needs and high inpatient utilization frequently use opioids for non-malignant chronic pain. We examine the association between chronic opioid therapy and healthcare utilization among high-need, high-cost patients insured by Medicare. Methods This study is a retrospective cohort study of Medicare beneficiaries in a medically underserved metropolitan area. Patients were included in the study if they had >= 2 hospitalizations or >= 1 hospitalization and >= 2 emergency department (ED) visits in a 6-month-period preceding an index hospitalization between July 2011 and June 2014. Exclusion criteria included substance abuse, psychosis and malignancy. Multivariable negative binomial regression models assessed associations between baseline opioid medication use and subsequent ED and hospital visits for the study population and two subgroups: (1) those with chronic pain and (2) those with both chronic pain and a primary care provider. Key findings The majority of high-need, high-cost patients filled opioid prescriptions of >= 7-day supply (51%). Yet only 2.7% of patients with chronic pain received an opioid prescription of >= 7-day supply from an ED provider in the baseline period. Overall (n = 677), receipt of a >= 7-day opioid supply was positively associated with subsequent ED utilization but not subsequent hospitalization. Among those with chronic pain and >= 1 primary care practice visit (N = 481), opioid use was associated with inpatient utilization but not ED utilization. Receipt of adjuvant pain prescriptions was not protective of hospitalizations or ED visits. Conclusions Chronic pain and opioid therapy are common among high-need, high-cost patients and they are independently associated with higher ED and hospital utilization. Novel patient-centred outpatient pain management strategies have potential to reduce inpatient care in this population.