Category: Federal Forum Posters
Purpose: Clinicians face a challenging task in achieving safe and therapeutic postoperative analgesia. This is met with a competing demand to avoid incident cases of chronic opioid therapy. Due to the risk of prolonged courses of therapy, guidelines caution against use of opioids in the presence of mental health diagnoses and substance use disorders. This guidance is especially relevant in a Veteran population, where presence of such risk factors is prevalent. In this analysis, we measured probability of opioid use at 3 critical postoperative time points based on presence of risk factors for prolonged opioid use.
Methods: This cohort analysis is comprised of Veterans undergoing orthopedic surgery at VA Salt Lake Healthcare System and agreeing to participate in the Transitional Pain Service (TPS) – a perioperative care coordination and pain management program. This analysis includes Veterans discharged from surgery between 1/1/2018 and 3/2/2019. The primary outcome is probability of opioid use at post-discharge day (PDD) 30, 60 and 90. In this analysis, we assigned Veterans with no evidence of risk factors for prolonged postoperative opioid use, i.e. mental health diagnoses, history of substance use disorder and preoperative chronic opioid use, as the “unexposed” group. In contrast, we identified comparator “exposure” groups based on presence of more than one of (1) mental health diagnoses, (2) history of substance use disorder or (3) preoperative chronic opioid use. Relative risk ratio (RR) was calculated to determine risk factor “exposure” effects on opioid use at each postoperative time point.
Results: Inclusion criteria was met in 154 Veterans; 46 with no risk factors for prolonged opioid therapy, and 108 with at least 1 risk factor present. At PDD-30, we observed the following probabilities for opioid use: (1) unexposed: 6/46=0.13 (2) preoperative chronic opioid use: 25/35=0.71, crude RR: 5.48, 95% confidence interval (CI): 2.52-11.89, p< 0.0001 (3) substance use disorder: 19/52=0.37, crude RR: 2.80, 95% CI: 1.22-6.41, p=0.0147 and (4) mental health diagnoses: 25/83=0.30, crude RR: 2.31, 95% CI: 1.02-5.22, p= 0.04. At PDD-60, we observed the following probabilities for opioid use: (1) unexposed: 2/46=0.04 (2) preoperative chronic opioid use: 21/35=0.60, crude RR: 13.80 95% CI: 3.46-54.97, p=0.0002 (3) substance use disorder: 13/52=0.25, crude RR: 5.75, 95% CI: 1.37-24.15, p=0.017 and (4) mental health diagnoses: 19/83=0.23, crude RR: 5.27, 95% CI: 1.28-21.6, p=0.02. At PDD-90, we observed the following probabilities for opioid use: (1) unexposed: 1/46=0.02 (2) preoperative chronic opioid use: 20/35=0.57, crude RR: 26.29, 95% CI: 3.70-186.55, p=0.001 (3) substance use disorder: 12/52=0.23, crude RR: 10.62, 95% CI: 1.44-78.52, p=0.02 and (4) mental health diagnoses: 19/83=0.23, crude RR: 10.53, 95% CI: 1.46-76.14, p=0.02. Adjusted analyses also demonstrated sizable risk effect based on presence of risk factors.
Conclusion: Our analysis corroborates findings from previous studies with respect to increased probability to remain on opioids postoperatively based on presence of certain preoperative risk factors. Large effect sizes observed at PDD-60 and PDD-90 highlight the complexity to safely and effectively discontinue postoperative opioids in these populations. Results from our analysis support a call to action to prospectively identify Veterans at risk for prolonged opioid use and develop a multidisciplinary care plan for postoperative pain management.