Purpose: Emergency physicians are usually more experienced in emergency situations; therefore, a cardiopulmonary resuscitation team with an emergency physician as a leader would be operated effectively. The aim of this study is to evaluate the effectiveness of a cardiopulmonary resuscitation that includes an emergency physician. Methods: A retrospective analysis for in-hospital arrests that occurred in the general ward was conducted based on the in-hospital cardiopulmonary resuscitation registry of a tertiary care university hospital in Korea from January 1, 2005 through December 31, 2010. We compared outcomes of cardiopulmonary resuscitation performed by a team that included an emergency physician with those by a cardiopulmonary resuscitation team that included a nonemergency physician. Results: Survival rates at discharge were 29.6% for the emergency physician team and 17.7% for the non-emergency physician team. The good neurologic outcome rates at discharge were 20.6% and 10.6%, respectively. In multivariate analysis with adjustment for pre-arrest patient condition and arrest variables, survival rate did not differ significantly between the two groups. However, the good neuroneurologic outcomes showed an association with the emergency physician team. Conclusion: For in-hospital cardiac arrest, outcomes for patients who were rescued by the emergency physiciandirecting CPR team might be comparable or better, compared with those by the non-emergency physician team.