Cannabis Use Disorder Rising Among Urologic Cancer Surgery Patients, With Longer Hospital Stays

CUD incidence rose from 51 to 383 per 100,000 admissions between 2003-2014. On multivariable analysis, CUD was not an independent predictor of major complications (p=0.6) but was independently associated with high length of stay (OR=1.31) and high inpatient cost (OR=1.33). CUD patients were significantly younger (median 37 vs 51 years) and had higher rates of opioid abuse.

Analysis of the National Inpatient Sample (2003-2014) including 1,612,743 patients undergoing prostatectomy, nephrectomy, or cystectomy. CUD identified via ICD-9 codes. Inverse probability treatment weighting balanced groups on demographics and comorbidities.

The 7-fold increase in CUD prevalence among surgical oncology patients reflects broader trends in cannabis use. The finding that CUD increases hospital stays and costs without increasing complications suggests logistical or behavioral factors (not surgical risk) are driving the difference.

The disconnect between higher unadjusted complication rates and non-significant adjusted rates suggests that age and comorbidity differences explain the complications, while the independent LOS and cost effects may reflect post-operative management challenges in this population.

ICD-9 codes for CUD likely undercount cannabis users. Administrative data cannot capture cannabis use that does not reach disorder-level diagnosis. Cannot determine if cannabis was used perioperatively. The dramatic age difference suggests fundamentally different patient populations.