OBJECTIVES: To describe an outbreak of hospital-acquired MRSA in a NICU and to identify the risk factors for, outcomes of, and interventions that eliminated it. SETTING: An 18-bed, level III-IV NICU in a community hospital. METHODS: Interventions to control MRSA included active surveillance, aggressive contact isolation, and cohorting and decolonization of infants and HCWs with MRSA. A case-control study was performed to compare infants with and without MRSA. RESULTS: A cluster of 6 cases of MRSA infection between September and October 2001 represented an increased attack rate of 21.2% compared with 5.3% in the previous months. Active surveillance identified unsuspected MRSA colonization in 6 (21.4%) of 28 patients and 6 (5.5%) of 110 HCWs screened. They were all successfully decolonized. There was an increased risk of MRSA colonization and infection among infants with low birth weight or younger gestational age. Multiple gestation was associated with an increased risk of colonization (OR, 37.5; CI95, 3.9-363.1) and infection (OR, 5.36; CI95, 1.37-20.96). Gavage feeding (OR, 10.33; CI95, 1.28-83.37) and intubation (OR, 5.97; CI95, 1.22-29.31) were associated with increased risk of infection. Infants with MRSA infection had a significantly longer hospital stay than infants without MRSA (51.83 vs 21.46 days; P = .003). Rep-PCR with mec typing and PVL analysis confirmed the presence of a single common strain of hospital-acquired MRSA. CONCLUSION: Active surveillance, aggressive implementation of contact isolation, cohorting, and decolonization effectively eradicated MRSA from the NICU for 2 1/2 years following the outbreak.