1Center for Clinical Effectiveness, Baylor Scott & White Health, Dallas, TX; 2Center for Applied Health Research, Baylor Scott & White Health, Temple, TX3University of California San Francisco, School of Pharmacy, San Francisco, CA; 4Health Economics and Outcomes Research Department, Mallinckrodt Pharmaceuticals, Hampton, NJ; *Affiliated with Mallinckrodt Pharmaceuticals at the time this study was conducted.
To examine the incidence of opioid-related adverse events (ORADE) in a large, integrated healthcare delivery system of multiple acute care hospitals
To evaluate the association of ORADE with patient outcomes including mortality, readmission, length of stay, discharge disposition, and costs
Over 73 million patients undergo surgical procedures in the United States each year.
Opioids are commonly used to manage moderate to severe postoperative pain.
ORADE lead to increased risk of poor patient outcomes.
The rate of postoperative ORADE and factors that increase the risk of post-surgical ORADE are not well established.
Retrospective study utilizing clinical and administrative data from 21 acute care hospitals in a single state-wide integrated healthcare delivery system.
Included patients who underwent surgery from January 1, 2013 to September 30, 2015
Qualifying surgical procedures were identified by Current Procedural Terminology (CPT) or International Classification of Disease (ICD-9) codes.
ORADE was defined as occurrence of one or more well-known side effects of opioids or use of an opioid antagonist.
ORADE types were classified as mild, moderate, or severe by study investigators based on potential for harm.
Types and amount of opioid use were also examined.
Opioid doses were converted to daily morphine equivalent doses based on a formula from the Centers for Disease Control and Prevention (CDC).
Univariate analysis was used to examine unadjusted differences in patient characteristics and outcomes between patients with and without ORADE.
Multivariable analyses was used to estimate risk-adjusted association between patient outcomes and ORADE occurrence.
Covariates included age, sex, race, Charlson Comorbidity Index score, history of drug/alcohol abuse, and payer type
Approximately 88% of included patients received opioids and 11% experienced an ORADE.
Patients with an ORADE were significantly more likely to be older, male, white, and non-Hispanic, and on Medicare, and have more comorbidities and history of alcohol/drug abuse.
ORADE associated with longer duration and higher doses of opioids.
ORADE associated with worse risk-adjusted patient outcomes.
Risk of death increased 27x
Risk of prolonged length of stay increased 80%
Risk of high cost admission increased 70%
Need for discharge to other care facilities increased 3x
LOS increased by 1.6 day
Cost increased by $6595
Readmissions increased by 20%
Endoscopic surgeries had highest volume of ORADE (3,681) followed by abdominal, cardiac, vascular procedures.
Open thoracic procedures had the highest rate (33%) of ORADEs per procedure.
The majority of ORADE were moderate (56%) or severe (37%) and associated with significantly worse outcomes
Unable to determine the timing of ORADE relative to opioid administration
Demonstrate association between ORADE and outcomes, not causality
Attributing all adverse events to opioids may overestimate ORADE
Common mild adverse events may be underestimated
Limited generalizability to other geographical locations
Conclusions and Implications for Research and Clinical Practice
ORADE are common after surgery and are associated with worse patient outcomes.
ORADE were associated with an increase in LOS and costs per patient, resulting in a potential cost increase of $95 million to the system over the study period.
Hospital-acquired harm from ORADE in the surgical population is an important opportunity for health system patient safety interventions and cost variability reduction programs.
This study was funded by Mallinckrodt Pharmaceuticals.