Background Early recognition of the signs and symptoms of clinical deterioration could diminish the incidence of cardiopulmonary arrest. The present study investigates outcomes with respect to cardiopulmonary arrest rates in institutions with and without rapid response systems (RRSs) and the current level of cardiopulmonary arrest rate in tertiary hospitals. Methods This was a retrospective study based on data from 14 tertiary hospitals. Cardiopulmonary resuscitation (CPR) rate reports were obtained from each hospital to include the number of cardiopulmonary arrest events in adult patients in the general ward, the annual adult admission statistics, and the structure of the RRS if present. Results Hospitals with RRSs showed a statistically significant reduction of the CPR rate between 2013 and 2015 (odds ratio [OR], 0.731; 95% confidence interval [CI], 0.577 to 0.927; P = 0.009). Nevertheless, CPR rates of 2013 and 2015 did not change in hospitals without RRS (OR, 0.988; 95% CI, 0.868 to 1.124; P = 0.854). National university-affiliated hospitals showed less cardiopulmonary arrest rate than private university-affiliated in 2015 (1.92 vs. 2.40; OR, 0.800; 95% CI, 0.702 to 0.912; P = 0.001). High-volume hospitals showed lower cardiopulmonary arrest rates compared with medium-volume hospitals in 2013 (1.76 vs. 2.63; OR, 0.667; 95% CI, 0.577 to 0.772; P < 0.001) and in 2015 (1.55 vs. 3.20; OR, 0.485; 95% CI, 0.428 to 0.550; P < 0.001). Conclusions RRSs may be a feasible option to reduce the CPR rate. The discrepancy in cardiopulmonary arrest rates suggests further research should include a nationwide survey to tease out factors involved in in-hospital cardiopulmonary arrest and differences in outcomes based on hospital characteristics.