Opioid prescribing and hospital outcomes after trauma in patients with substance use disorder: An analysis of the michigan trauma quality improvement project

BACKGROUND: Trauma patients frequently have comorbid substance use disorders (SUD), which may complicate pain management and increase the risk of opioid-related adverse outcomes after discharge. We hypothesized that trauma patients with SUD would be prescribed more opioids at discharge than those without SUD. METHODS: We retrospectively analyzed a cohort of adult trauma patients across the state from 2022 to 2025 using a statewide collaborative database. Primary outcomes were opioid prescription at discharge (yes/no) and amount prescribed standardized to oxycodone 5 mg pills. A multivariable approach assessed risk-adjusted differences in outcomes among patients with SUD versus without SUD. RESULTS: Among 29,805 total patients, 8,672 (29.1%) had a diagnosis of SUD. 71.2% of patients with SUD received a discharge opioid prescription, compared with 59.6% of patients without SUD (p<0.001). After adjustment, the predicted probability of receiving an opioid prescription at discharge was 64.2% (95% CI, 63.1%-65.2%) for patients with SUD and 62.5% (95% CI, 62.0%-63.1%) for those without. Among patients that received a prescription, those with SUD were prescribed 21.2 pills of 5 mg oxycodone, corresponding to an adjusted difference of 1.6 additional pills (95% CI, 0.89-2.31; p<0.001) compared with those without. Prescribing patterns also differed by race and ethnicity: Hispanic or Latino patients were 5.4-percentage points more likely to receive an opioid prescription than non-Hispanic or Latino patients, while, among prescription recipients, Black patients received 3.56 fewer pills than White patients. CONCLUSIONS: Our findings demonstrate that patients with SUD received opioids more frequently and in greater amounts at discharge, yet absolute differences were small in magnitude and may not represent clinically meaningful differences. Differences in prescribing by race and ethnicity were larger in magnitude and warrant further investigation. LEVEL OF EVIDENCE: Prognostic and epidemiological; Level III.