Context: Community-acquired, methicillin-resistant (CA-MRSA) infections in children are increasing in frequency for unknown reasons.

Objectives: To compare the presence of risk factors for methicillin resistance between patients with CA-MRSA and community-acquired methicillin-susceptible (CA-MSSA) infection and to compare the presence of risk factors among household contacts of the patients from both groups. To compare the demographic and clinical characteristics between children with CA-MRSA and CA-MSSA infection.

Design: Prospective observational study conducted between February 2, 2000 and November 14, 2000, excluding the month of May and the period between September 2 and October 15.

Setting and patients: Texas Children's Hospital, Houston, TX; inpatients and outpatients with community-acquired infection.

Main outcome measures: Proportion of MRSA among all community-acquired infections. The presence of risk factors associated with methicillin resistance among patients, and their household contacts, with CA-MRSA and CA-MSSA.

Results: The monthly rates of methicillin resistance of varied between 35 and 51%. CA-MSSA isolates were associated with deep-seated infections significantly more often (30%) than CA-MRSA isolates (11%; P= 0.01). CA-MRSA isolates were generally susceptible to clindamycin and trimethoprim-sulfamethoxazole and resistant to erythromycin. There were no significant differences in the exposure to risk factors between children with CA-MRSA and CA-MSSA infection. No significant risk factors for CA-MRSA were identified among household contacts.

Conclusions: MRSA is an established, community-acquired pathogen in our area. This necessitates a change in empiric therapy of infections suspected to be caused by.