Baseline characteristics of the participants.

Background
The impact of marijuana on the general population is largely unknown. The present study aimed to assess the association between marijuana use and liver steatosis and fibrosis in the general United States population utilizing data from the National Health and Nutrition Examination Survey (NHANES).
Methods
This cross-sectional study was performed with data from the 2017–2018 cycle of NHANES. The target population comprised adults in the NHANES database with reliable vibration controlled transient elastography (VCTE) results. The median values of the controlled attenuation parameter (CAP) and liver stiffness measurement (LSM) were used to evaluate liver steatosis and fibrosis, respectively. After adjusting for relevant confounders, a logistic regression analysis was used to assess the association between marijuana use and liver steatosis and fibrosis.
Results
A total of 2622 participants were included in this study. The proportions of never marijuana users, past users, and current users were 45.9%, 35.0%, and 19.1%, respectively. Compared to never marijuana users, past and current users had a lower prevalence of liver steatosis (P = 0.184 and P = 0.048, respectively). In the alcohol intake-adjusted model, current marijuana use was an independent predictor of a low prevalence of liver steatosis in people with non-heavy alcohol intake. The association between marijuana use and liver fibrosis was not significant in univariate and multivariate regression.
Conclusion
In this nationally representative sample, current marijuana use is inversely associated with steatosis. The pathophysiology is unclear and needs further study. No significant association was established between marijuana use and liver fibrosis, irrespective of past or current use.