Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Patterns of Opioid Use and Risk of Opioid Overdose Death Among Medicaid Patients

Garg, Renu K., PhD*; Fulton-Kehoe, Deborah, PhD; Franklin, Gary M., MD, MPH†,‡,§,∥

doi: 10.1097/MLR.0000000000000738
Original Articles

Background: The Centers for Disease Control and Prevention recognizes Medicaid as a high-risk population for fatal opioid overdose. Further research is needed to identify factors that put Medicaid patients at increased risk.

Objective: To determine whether patterns of opioid use are associated with risk of opioid-related mortality among opioid users.

Design: This is a retrospective cohort study.

Patients: In total, 150,821 noncancer pain patients aged 18–64 years with ≥1 opioid prescription, April 2006 to December 2010, Washington Medicaid.

Measures: Average daily dose (morphine equivalents), opioid schedule/duration of action, sedative-hypnotic use.

Results: Compared with patients at 1–19 mg/d, risk of opioid overdose death significantly increased at 50–89 mg/d [adjusted hazard ratio (aHR), 2.3; 95% confidence interval (CI), 1.4–4.1], 90–119 mg/d (aHR, 4.0; 95% CI, 2.2–7.3), 120–199 mg/d (aHR, 3.8; 95% CI, 2.1–6.9), and ≥200 mg/d (aHR, 4.9; 95% CI, 2.9–8.1). Patients using long-acting plus short-acting Schedule II opioids had 4.7 times the risk of opioid overdose death than non-Schedule II opioids alone (aHR, 4.7; 95% CI, 3.3–6.9). Sedative-hypnotic use compared with nonuse was associated with 6.4 times the risk of opioid overdose death (aHR, 6.4; 95% CI, 5.0–8.4). Risk was particularly high for opioids combined with benzodiazepines and skeletal muscle relaxants (aHR, 12.6; 95% CI, 8.9–17.9). Even at opioid doses 1–19 mg/d, patients using sedative-hypnotics concurrently had 5.6 times the risk than patients without sedative-hypnotics (aHR, 5.6; 95% CI, 1.6–19.3).

Conclusions: Our findings support Federal guideline-recommended dosing thresholds in opioid prescribing. Concurrent sedative-hypnotic use even at low opioid doses poses substantially greater risk of opioid overdose.

Departments of *Epidemiology

Environmental and Occupational Health Sciences

Health Services

Neurology, University of Washington, Seattle

§Washington State Department of Labor and Industries, Olympia, WA

Supported by the Centers for Disease Control and Prevention (grant number 5R21CE001850-01).

The authors declare no conflict of interest.

Reprints: Renu K. Garg, PhD, Department of Epidemiology, University of Washington, 130 Nickerson Street, Suite 212, Seattle, WA 98109. E-mail:

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.