Background: Korea has very high prevalence of penicillin- and macrolide-resistant Streptococcus pneumoniae in the world. The impact of such resistance has not been measured properly. This study tried to estimate the clinical outcome of initial treatment according to the choice of antibiotics in outpatient treatment of community-acquired pneumonia(CAP) caused by S. pneumoniae in South Korea. Methods : We employed the probabilistic model to calculate rates of initial treatment success and hospitalization for CAP in adult without major co-morbidities. Three 2-step strategies were considered: Initial treatment with (1) clarithromycin, (2) amoxicillin-clavulanate, or (3) telithromycin; and failures of initial treatment were treated with levofloxacin. We used surveillance data from the 1999-2000 PROTEKT study, and applied the proportions of susceptible organism to each antibiotics to calculate the failure rates. We assumed spontaneous resolution of 10% regardless of treatment strategy and treatment success of 90% if organisms were susceptible to the chosen antibiotics. We also assumed that 20% of initial treatment failures would be hospitalized, the others given a second prescription. All patients failing the second round of therapy were assumed to be hospitalized. Results : According to PROTEKT Korea, resistance of S. pneumoniae isolates to clarithromycin, amoxicillin-clavulanate, levofloxacin and telithromycin was 87.6 %, 30%, 3%, and 0%, respectively. The model calculated initial treatment success to be 15.6% for clarithromycin, 64.9% for amoxicillin-clavulanate and 90.5% for telithromycin. The hospitalization rate was 25.4% for clarithromycin/levofloxacin, 10.6 % for amoxicillin-clavulanate/levofloxacin, and 2.9% for telithromycin/levofloxacin. Results for telithromycin were not sensitive to breakpoint (range 0.5-4.0 mg/L). Conclusion : The modeling suggests that initial treatment with telithromycin --- the first ketolide antibacterial --- has the potential to increase the success of initial treatment to 90.5% comparing with clarithromycin(15.6%) and amoxicillin-clavulanate(64.9%) in out-patient with S. pneumoniae CAP in South Korea. It also reduce the percentage of hospitalization to 2.62% comparing with clarithromycin(73.85%) and amoxicillin-clavulanate(16.87%).