Association of Cannabis Use Disorder With Hospitalizations for Pulmonary Embolism and Subsequent in-Hospital Mortality in Young Adults: A Contemporary Nationwide Analysis.

Abstract

With the increase in popularity of cannabis and its use and the lack of large-scale data on cannabis use and venous thromboembolism and pulmonary embolism (PE), we used a nationally representative cohort of young adults (aged 18-44 years) to compare the odds of admissions and in-hospital mortality of PE with and without cannabis use disorder (CUD). Identified patients with PE using the National Inpatient Sample (2018) were compared for baseline, comorbidities, and outcomes. Multivariable regression analysis, adjusted for covariates, was used to compare the odds of PE in young patients with CUD (CUD+) versus those without (CUD-) and those with prior venous thromboembolism. Propensity score-matched analysis (1:6) was also performed to assess in-hospital outcomes. A total of 61 965 (0.7%) of 8 438 858 young adult admissions in 2018 were PE related, of which 1705 (0.6%) had CUD+. On both unadjusted (odds ratio, 0.80 [95% CI, 0.71-0.90]; P<0.001) and adjusted regression analyses, the CUD+ cohort had a lower risk of PE admission. The CUD+ cohort had fewer routine discharges (58.3% versus 68.3%) and higher transfers to short-term (7.9% versus 4.8%) and nursing/intermediate care (12.6% versus 9.5%) (P<0.001). The PE-CUD+ cohort of in-hospital mortality did not differ from the CUD- cohort. Propensity score-matched (1:6) analysis revealed comparable mortality odds with higher median hospital stay and cost in the CUD+ cohort. Young adults with CUD demonstrated lower odds of PE hospitalizations without any association with subsequent in-hospital mortality. The median hospital stay of the CUD+ cohort was longer, they were often transferred to other facilities, and they had a higher cost.