Objective To investigate the application effect of Problem based learning (PBL) combined with role play teaching method in the training of emergency internal medicine.Methods Totally 60 medical students were randomly divided into two groups.The students in the observation groop were trained with PBL teaching method while those in control group were trained with traditional teaching method.The practical performances of the students in both groups were evaluated by virtual experience of dealing with patients in the emergency department through the role play teaching method.The grasp of basic knowledge was assessed by writing tests.Finally all the students in the observation group answered a survey about the teaching methods.Results There was uo difference in the scores of the basic knowledge test between the two groups ( P ＞ 0.05 ).The practical ability of the observation group was much better than that of the control group in the role play practice ( P ＜ 0.05 ).Almost all the students in the observation group accept the PBL teaching method.Conelusion PBL combined with role play teaching method is effective based on the characteristics of emergency medicine.

Objective To assess the risk factors of the in-hospital mortality of acute type A aortic dissection after operation. Method From January 2003 to June 2008,185 patients, 144 males and 41 females, with acute type A aortic dissection operated on were enrolled. The average age of patients was (49.46 ± 11.04 ) years old.The patients' demographics, history, clinical features, and some laboratory examinations were reviewed. Univariate and multivariate analysis followed by logistic regression analysis were carried out to identify the predictors of inhospital mortality. Results The in-hospital mortality rate was 9.1%. The results of univariate and multivariate analyses as follows: pre-operation positive neurological symptom (Univariate OR = 5.084,95%CI:1.792 -14.426, P = 0.002; Multivariate OR = 5.538,95%CI: 1.834 - 16.721, P = 0.002, respectively), hypotension (Univariate OR = 6.986,95%CI:1.510- 32.323,P =0.013; multivariate OR = 1.998,95%CI:0.315-12.679,P = 0.463, respectively) and renal failure (Univariate OR = 3.594,95%CI:1.237 - 10.438,P =0.019; Multivariate OR = 3.254,95%CI:1.034- 10.242, P= 0.044, respectively). Conclusions There are two predictors, pre-operation positive neurological symptom and renal failure, of pre-hospital mortality found in current analyses. Our results may improve the regimen made by cardiac surgeons and emergency doctors so as to help patients and their relatives to make correct decision.