The relationship between erythema nodosum (EN) and ulcerative colitis (UC) prognosis has not been established. Thus, we used the National Inpatient Sample database to evaluate the relationship between EN, anatomical bowel involvement, and inpatient outcomes of UC. Using validated ICD-9 codes, we analyzed adults 2003-2015 with a primary diagnosis of UC with and without EN. Multivariable logistic and linear regression models were used to determine the relationship between EN and UC characteristics after adjusting for patient demographics and hospital characteristics. An estimated 415,247 discharges had a primary diagnosis of UC, of which 778 had a secondary diagnosis of EN. The EN group was less likely to have chronic proctitis (OR¼0.30, 95% CI 0.16, 0.58) or universal colitis (OR¼0.54, 95% CI 0.42, 0.68), and more likely to have left-sided colitis (OR¼1.63, 95% CI 1.23, 2.17). There was no difference in chronic enterocolitis, ileocolitis, proctosigmoiditis, pseudopolyposis, and other or unspecified forms of UC between groups (p-values>0.05). A diagnosis of EN was associated with lower odds of GI hemorrhage (OR¼0.74, 95% CI 0.56, 0.98), but higher rates of post-hemorrhagic anemia (OR¼1.46, 95% CI 1.17, 1.86) and other anemias or deficiencies (OR¼2.22, 95% CI 1.89, 2.60). Healthcare utilization analysis of the EN group revealed a lower rate of colorectal resection (OR¼0.36; 95% CI 0.26, 0.49), shorter hospital stay (0.76 days shorter), and lower hospital charges ($671 less) (p-values<0.003). There was no difference in in-hospital mortality between groups (p¼0.992). Our study shows that EN as a physical exam finding in patients with UC may be used to identify patients at lower risk of pancolitis, as well as to predict moderate disease severity and better inpatient outcomes.