Infections caused by Staphylococcus aureus with reduced vancomycin susceptibility (SA-RVS; minimum inhibitory concentration [MIC], >or=4 microg/mL), including vancomycin-intermediate S. aureus (VISA; MIC, 8 microg/mL), are a new clinical and public health dilemma. Prospective surveillance and a nested case-control study of patients in the United States infected with SA-RVS was conduced from March 1999 through December 2000. Control patients were persons infected with oxacillin-resistant S. aureus (MIC of vancomycin, <or=2 microg/mL). Among 19 case patients, 4 infections were due to VISA and 15 were due to non-VISA SA-RVS. Case patients with and those without VISA infection had similar clinical presentations and outcomes; the overall attributable mortality rate was 63%. Isolates recovered from case patients had heterogeneous pulsed-field gel electrophoresis banding patterns, regardless of the MIC of vancomycin. Neither dialysis nor chronic renal failure were predictive of case status compared with control status. Independent risk factors for being a case patient included antecedent vancomycin use and prior oxacillin-resistant S. aureus infection 2 or 3 months before the current infection.