Author Affiliations: The Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada (Ms Gomes, Drs Mamdani, Dhalla, and Juurlink, and Mr Paterson); Faculty of Medicine, King Saud University, Riyadh, Saudi Arabia (Dr Mamdani); Department of Medicine, St Michael's Hospital, Toronto (Drs Mamdani and Dhalla); Leslie Dan Faculty of Pharmacy (Ms Gomes and Dr Mamdani) and Departments of Medicine (Drs Mamdani, Dhalla, and Juurlink), Health Policy, Management, and Evaluation (Drs Mamdani, Dhalla, and Juurlink and Mr Paterson), and Pediatrics (Dr Juurlink), University of Toronto, Toronto; Keenan Research Centre of the Li Ka Shing Knowledge Institute, St Michael's Hospital, Toronto (Drs Mamdani and Dhalla); Department of Family Medicine, McMaster University, Hamilton, Ontario (Mr Paterson); Centre for Evaluation of Medicines, St Joseph's Healthcare, Toronto (Mr Paterson); and The Sunnybrook Research Institute, Toronto (Dr Juurlink).
We believe Dr Williamson has misinterpreted some aspects of the methodology of our study and its findings. Dr Williamson states that an unadjusted analysis reveals no association between higher opioid dose categories and opioid-related mortality. This is incorrect. Although not reported in our published article,1 the association between dose and opioid-related mortality was consistent in our unadjusted conditional logistic regression model for the primary outcome (odds ratio, 2.32, 2.97, and 3.69 among those prescribed 50-99 mg of morphine equivalents per day [ME/d], 100-199 mg of ME/d, and ≥200 mg of ME/d, respectively, compared with the 1-19 mg of ME/d group; P < .001 for all comparisons). Although the 95% confidence intervals overlap between high and moderate dose categories, a clear and significant association between higher doses of opioids and opioid-related mortality is apparent in both unadjusted and adjusted models.
Gomes T, Mamdani MM, Dhalla IA, Paterson JM, Juurlink DN. Opioids and Dose-Related Deaths—Association or Causation?—Reply. Arch Intern Med. 2011;171(18):1687-1689. doi:10.1001/archinternmed.2011.453