Compare all-cause mortality following nonsurgical ICU admission for opioid users with nonusers.
Nationwide register-based cohort study.
All 43 ICUs in Denmark (7,028,668 citizens cumulatively during the study period). The Danish National Health Service provides universal healthcare, guaranteeing equal access to healthcare along with partial reimbursement for prescribed drugs.
All 118,388 nonsurgical patients admitted to an ICU from 2005 to 2014.
Patients were categorized according to timing of last redeemed opioid prescription before admission: current user (prior 0–30 d), recent user (prior 31–365 d), former user (prior 365+ d), or nonuser (no prescription since 1994).
All-cause mortality 0–30 days and 31–365 days following ICU admission was calculated using the Kaplan-Meier method. Crude and adjusted hazard ratios with 95% CIs were computed using Cox regression, comparing users with nonusers. Adjusted models included age, gender, socioeconomic factors, comedications, and comorbidity.
Fifteen percent of the patients were current opioid users, 15% recent users, 30% former users, and 40% nonusers. Zero- to 30-day mortality was 35% for current users, 29% for recent users, 24% for former users, and 21% for nonusers. After confounder adjustment, current users remained at elevated risk during the first 30 days following ICU admission (hazard ratio, 1.20; 95% CI, 1.15–1.24). No association remained for recent or former users. A similar pattern was evident for 31–365-day all-cause mortality: 24% for current users, 19% for recent users, 13% for former users, and 10% for nonusers. During 31–365 days of follow-up, both current users and recent users remained at elevated risk of mortality after adjustment (hazard ratio, 1.47; 95% CI, 1.39–1.55 and hazard ratio, 1.20; 95% CI, 1.13–1.27, respectively).
Current opioid users experience increased mortality during the first year following ICU admission.
1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark.
2Department of Health Research and Policy (Epidemiology), Stanford University, Stanford, CA.
*See also p. 1005.
The study was designed by Drs. Munch, Christiansen, and Sørensen. Drs. Pedersen and Munch carried out analysis. Dr. Munch drafted the article, which all authors critically revised and approved before submission. Dr. Munch had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ccmjournal).
Supported, in part, by grants from the Program for Clinical Research Infrastructure established by the Lundbeck Foundation and the Novo Nordisk Foundation.
The Department of Clinical Epidemiology is involved in studies with funding from various companies as research grants to (and administered by) Aarhus University. The authors have disclosed that they do not have any potential conflicts of interest.
For information regarding this article, E-mail: email@example.com