Background: Incidence of community-associated (CA) and healthcare-associated, community-onset (HACO) USA300 methicillin-resistant Staphylococcus aureus (MRSA) bloodstream infections has remained unchanged in recent years. Traditionally considered a CA strain, USA300 is increasingly associated with healthcare settings. We examined whether antimicrobial nonsusceptibility among USA300 strains could distinguish epidemiologic class (community vs hospital), and whether divergences in susceptibility were occurring over time. Methods: We used data on invasive MRSA infections from active, population, and laboratory-based surveillance during 2005–2016 from 11 counties in 3 states. Invasive cases were defined as MRSA isolation from a normally sterile site in a surveillance area resident. Cases were considered hospital-onset (HO) if the culture was obtained >3 days after hospitalization and HACO if ≥1 of the following risk factors was present: hospitalization, surgery, dialysis, or residence in a long-term care facility in the past year; or central vascular catheter ≤2 days before culture. Otherwise, cases were considered CA. Sites submitted a convenience sample of clinical MRSA isolates for molecular typing and antimicrobial susceptibility testing. Molecular typing was performed by pulsed-field gel electrophoresis until 2008, when typing was inferred using a validated algorithm based on molecular characteristics. Reference broth microdilution was performed for 8 antimicrobials and interpreted based on CLSI interpretive criteria. We compared USA300 nonsusceptibility for HO and CA isolates. For antimicrobials with >5% nonsusceptibility and for which HO isolates had greater nonsusceptibility than CA isolates, we compared nonsusceptibility for HACO and CA and analyzed annual trends in nonsusceptibility within each epidemiologic class (ie, CA, HACO, and HO) using linear regression. Results: Of 17,947 MRSA cases during 2005–2016, isolates were available for 6,685 (37%), and 2,120 were USA300 (34% CA, 52% HACO, 14% HO). HO isolates had more nonsusceptibility than CA isolates to gentamicin (2.2% vs 0.6%; P = .03), levofloxacin (47.8% vs 39.7%; P = .02), rifampin (3.7 vs 1.1%; P = .01), and trimethoprim-sulfamethoxazole (3.4% vs 0.6%; P = .04). HACO isolates also had more nonsusceptibility than CA isolates to levofloxacin (50.9% vs 39.7%; P < .01). Levofloxacin nonsusceptibility increased during 2005–2016 for HACO and CA isolates (P < .01), but not among HO isolates (P = .36) (Fig. 1). Conclusions: Overall, nonsusceptibility across drugs cannot distinguish USA300 isolates causing HO versus CA disease. Although HO isolates had higher levofloxacin nonsusceptibility than CA and HACO isolates early on, USA300 MRSA HACO isolates now have levofloxacin nonsusceptibility most similar to that of HO isolates. Further study could help to explore whether increases in fluoroquinolone nonsusceptibility among CA and HACO cases may be contributing to the persistence of USA300 strains.