Colleen M Carey1, Anupam B Jena2, Michael L Barnett3. 1. Cornell University, Ithaca, New York (C.M.C.). 2. Harvard Medical School and Massachusetts General Hospital, Boston, Massachusetts, and National Bureau of Economic Research, Cambridge, Massachusetts (A.B.J.). 3. Harvard T.H. Chan School of Public Health and Brigham and Women's Hospital, Boston, Massachusetts (M.L.B.).
Abstract
Background: Providers are increasingly being expected to examine their patients' opioid treatment histories before writing new opioid prescriptions. However, little evidence exists on how patterns of potential opioid misuse are associated with subsequent adverse outcomes nationally. Objective: To estimate how a range of patterns of potential opioid misuse relate to adverse outcomes during the subsequent year. Design: Observational study comparing outcomes for Medicare enrollees with potential opioid misuse patterns versus those for beneficiaries with no such patterns, adjusting for patient characteristics. Setting: Medicare, 2008 to 2012. Patients: A 5% sample of beneficiaries who had an opioid prescription without a cancer diagnosis. Measurements: Several measures for opioid misuse were defined on the basis of drug quantity, overlapping prescriptions, use of multiple prescribers or pharmacies, and use of out-of-state prescribers or pharmacies. The primary outcome was a diagnosis of opioid overdose in the year after a 6-month index period. Secondary outcomes included subsequent opioid-related or overall mortality. Results: Overall, 0.6% to 8.5% of beneficiaries fulfilled a misuse measure. Subsequent opioid overdose was positively associated with successively greater numbers of prescribers or pharmacies or higher opioid quantities during the index period. For example, patients who obtained opioids from 2, 3, or 4 prescribers were increasingly more likely to have an opioid overdose (adjusted absolute risk per 1000 beneficiary-years [aAR], 3.5 [95% CI, 3.3 to 3.7]; 4.8 [CI, 4.5 to 5.2]; or 6.4 [CI, 5.8 to 6.9], respectively) than those with a single prescriber (aAR, 1.9 [CI, 1.8 to 2.0]). Subsequent overdose risk increased meaningfully with any deviation in the single prescriber-single pharmacy opioid use pattern. All misuse measures examined had a positive association with subsequent opioid overdose and death. Limitation: Risk estimates provide measures of association and may not generalize to non-Medicare populations. Conclusion: To fully assess patients' opioid overdose risk, clinicians should examine a wide range of misuse patterns. Primary Funding Source: National Institutes of Health.
Background: Providers are increasingly being expected to examine their patients' opioid treatment histories before writing new opioid prescriptions. However, little evidence exists on how patterns of potential opioid misuse are associated with subsequent adverse outcomes nationally. Objective: To estimate how a range of patterns of potential opioid misuse relate to adverse outcomes during the subsequent year. Design: Observational study comparing outcomes for Medicare enrollees with potential opioid misuse patterns versus those for beneficiaries with no such patterns, adjusting for patient characteristics. Setting: Medicare, 2008 to 2012. Patients: A 5% sample of beneficiaries who had an opioid prescription without a cancer diagnosis. Measurements: Several measures for opioid misuse were defined on the basis of drug quantity, overlapping prescriptions, use of multiple prescribers or pharmacies, and use of out-of-state prescribers or pharmacies. The primary outcome was a diagnosis of opioid overdose in the year after a 6-month index period. Secondary outcomes included subsequent opioid-related or overall mortality. Results: Overall, 0.6% to 8.5% of beneficiaries fulfilled a misuse measure. Subsequent opioid overdose was positively associated with successively greater numbers of prescribers or pharmacies or higher opioid quantities during the index period. For example, patients who obtained opioids from 2, 3, or 4 prescribers were increasingly more likely to have an opioid overdose (adjusted absolute risk per 1000 beneficiary-years [aAR], 3.5 [95% CI, 3.3 to 3.7]; 4.8 [CI, 4.5 to 5.2]; or 6.4 [CI, 5.8 to 6.9], respectively) than those with a single prescriber (aAR, 1.9 [CI, 1.8 to 2.0]). Subsequent overdose risk increased meaningfully with any deviation in the single prescriber-single pharmacy opioid use pattern. All misuse measures examined had a positive association with subsequent opioid overdose and death. Limitation: Risk estimates provide measures of association and may not generalize to non-Medicare populations. Conclusion: To fully assess patients' opioid overdose risk, clinicians should examine a wide range of misuse patterns. Primary Funding Source: National Institutes of Health.
Authors: Zachary F Meisel; Nicoleta Lupulescu-Mann; Christina J Charlesworth; Hyunjee Kim; Benjamin C Sun Journal: Ann Emerg Med Date: 2019-06-20 Impact factor: 5.721
Authors: Victor Puac-Polanco; Stanford Chihuri; David S Fink; Magdalena Cerdá; Katherine M Keyes; Guohua Li Journal: Epidemiol Rev Date: 2020-01-31 Impact factor: 6.222
Authors: Matthew L Maciejewski; Valerie A Smith; Theodore S Z Berkowitz; David E Arterburn; Katharine A Bradley; Maren K Olsen; Chuan-Fen Liu; Edward H Livingston; Luke M Funk; James E Mitchell Journal: Surg Obes Relat Dis Date: 2020-05-07 Impact factor: 4.734
Authors: Mara A G Hollander; Julie M Donohue; Bradley D Stein; Elizabeth E Krans; Marian P Jarlenski Journal: J Gen Intern Med Date: 2019-12-02 Impact factor: 5.128
Authors: Martin S Andersen; Vincent Lorenz; Anurag Pant; Jeremy W Bray; G Caleb Alexander Journal: Am J Manag Care Date: 2022-02-01 Impact factor: 2.229
Authors: Bhushan R Deshpande; Ellen P McCarthy; Yoojin Jung; Timothy S Anderson; Shoshana J Herzig Journal: J Hosp Med Date: 2021-12 Impact factor: 2.960
Authors: John Jake Magel; Gerald Cochran; Nancy West; Julie M Fritz; Mark D Bishop; Adam J Gordon Journal: Subst Abus Date: 2021-07-20 Impact factor: 3.716