Objective To investigate the effects of sepsis bundles in China.Method An observational study of 43 patients with severe pneumonia and septic shock admitted to the respiratory intertive care unit(1/11/2006-31/12/2007)was carried out.The selection criteria were in accordance with criteria set by International Conference On Sepsis in 2001.Implementation of 6 hours and 24 hours sepsis bundles was divided into 3 continu-ous phases consisting of education,trial,and application phase.A cohort of 43 patients with matched disease his-tory(1/1/2004-31/10/2006)was enrolled as control group.The percentages for categorical variables and mean±SD for continuous variables were reported.Chi-Square test.unpaired Student's t -test.paired-samples t test,univariate and multivariate logistic regression models were used.Statistical significance was defined as P＜0.05.Results There were very little significant differences in basic characteristics of patients between the two groups.Compared with control group,the differences in serrum lactate,fluid resuscitation and fluid volume infused within 6 hours and blood glucose control in shock subgroup were significant(P values were 0.024,0.009,0.045,and 0.000,respectively).Compared with control group,the differences in respiratory rate and oxygenation index of bundles group at 72 hours later were significant(P values were 0.033 and 0.041,respectively).Compared with control group,the differences in APACHE Ⅱ score and predicted mortality in shock subgroup of bundles were sig-nificant(P values were 0.017 and 0.040,respectively).Compared with control group,the reduction in absolute mortality was 23.30% in bundles group(P=0.019).Conclusions Implementation of sepsis bundles con-tributes noticeably to the significant reduction in mortality of patients with severe pneumonia and septic shock.