Background : Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) are major pathogens in community and hospital. And they sometimes cause the outbreak in hospital in the immunocompromised patients. Pulsed-field gel electrophoresis (PFGE) has been regarded as a standard method for genotyping in epidemiologic studies, but it is laborious and time-consuming. Infrequent restriction site-polymerase chain reaction (IRS-PCR), a new genotyping methods, was performed to compare the applicability with PFGE. Methods: We performed PFGE and IRS-PCR on S. aurues (n=120) and E. coli (n=117) which were collected clinically in 4 different hospitals, We assessed each method in terms of discriminatory power, quality, and efficiency. Results: In E. coli, the discriminatory power of IRS-PCR was 46.7~86.7%, and that of PFGE was 88.9~96.7% according to hospital. But in S. aurues, the discriminatory power of IRS-PCR was 20~56.7%, and that of PFGE was 40~90% according to hospital. The typablity and reproducibility of IRS-PCR were 100% of each. PFGE needed four days to complete the procedure, but IRS-PCR could be performed within one day, IRS-PCR showed better resolution than PFGE. Conclusion: In case of gram negative bacteria (like E coli), IRS-PCR could be a reliable altemative for epidemiologic typing due to better efficiency and comparable discriminatory power, But in the case of gram positive bacteria (like S. aureus), IRS-PCR does not seem to be suitable for the strain-to-strain differentiation. More trials and changes of restriction enzymes or primers could reveal the efficacy of IRS-PCR in the field of molecular typing.