Children are increasingly recognized as being at risk for C. difficile infection (CDI), even without prior exposure to antibiotics or the healthcare environment. We aimed to distinguish risk factors, clinical course, and outcomes between healthcare facility-associated (HA) and community-associated (CA) CDI. This was a retrospective, observational cohort study conducted at the Johns Hopkins Children's Center, Baltimore, Maryland. All inpatients, aged ≥1 year, hospitalized from July 2003 to July 2012 and diagnosed with CDI based on clinical characteristics and confirmatory laboratory testing were included. The main outcome was CDI, categorized as HA-CDI, CA-CDI, and "indeterminate" (classified as disease onset in the community, 4-12 weeks from hospital discharge). Two hundred two pediatric inpatients were diagnosed with CDI, of whom 38 had CA-CDI, 144 had HA-CDI, and 20 had indeterminate CDI. Children with indeterminate CDI had baseline characteristics similar to those identified for HA-CDI. Children hospitalized with CA-CDI were less likely to have comorbidities (odds ratio [OR], 0.14; 95% confidence interval [CI], .03-.65; P = .013), to have been exposed to antibiotics (OR, 0.17; 95% CI, .07-.44; P < .001), or prior surgeries (OR, 0.03; 95% CI, .00-.24; P = .001), compared to children with HA-CDI. Compared with HA-CDI, children with CA-CDI had a trend toward more episodes of septic shock (P = .07), toxic megacolon (P = .04), and recurrences (P = .04). In a hospitalized cohort, CA-CDI is more often seen in previously healthy children without antibiotic exposure or comorbid conditions and has more frequent complications and recurrences compared to HA-CDI. For surveillance purposes, "indeterminate" CDI should be allocated to HA-CDI rather than CA-CDI.