BACKGROUND:: Mycoplasma pneumoniae is a common cause of community-acquired pneumonia in childhood. Few studies have addressed the association of antimicrobial treatment and outcomes. OBJECTIVE:: To determine whether macrolide therapy is associated with improved outcomes among children hospitalized with M. pneumoniae pneumonia. DESIGN:: Multicenter retrospective cohort study. SETTING:: Thirty-six childrenʼs hospitals which contribute data to the Pediatric Health Information System. PATIENTS:: Children 6-18 years of age discharged with a diagnosis of M. pneumoniae pneumonia. MAIN EXPOSURE:: Initial macrolide therapy. MAIN OUTCOME MEASURES:: Length of stay (LOS), all-cause readmissions, and asthma-related hospitalizations. RESULTS:: Empiric macrolide therapy was administered to 405 (58.7%) of 690 patients. The median LOS was 3 days (interquartile range, 2-6 days). Eight (1.2 %) patients were readmitted within 28 days, and 160 (23.2%) were readmitted within 15 months of index discharge. Ninety-five (13.7%) patients were hospitalized for asthma within 15 months of index discharge. Empiric macrolide therapy was associated with a 32% shorter overall LOS (adjusted beta-coefficient, −0.38; 95% confidence interval [CI]: −0.59 to −0.17). Macrolide therapy was not associated with all-cause readmission at 28 days (adjusted odds ratio, 1.12; 95% CI: 0.22-5.78) or 15 months (adjusted odds ratio, 1.00; 95% CI: 0.59-1.70) or with asthma-related hospitalizations at 15 months (adjusted odds ratio, 0.85; 95% CI: 0.36-1.97). CONCLUSION:: In this large multicenter study of children hospitalized with M. pneumoniae pneumonia, empiric macrolide therapy was associated with a shorter hospital LOS. Macrolide therapy was not associated with 28-day or 15-month hospital readmission. Journal of Hospital Medicine 2012; © 2012 Society of Hospital Medicine