Abstract: Objective To find out the problems which needed to resolve during the design phase of clinical trials via questionnaire survey.Methods Clinical doctors who have taken part in clinical trials in the bases of Drug's Clinical Trials were surveyed by questionnaire.The data were recorded by software Epidata 3.0 and analyzed by software SAS8.0 Results Forty doctors who have finished protocols of clinical trials were assigned in this survey.We can draw a conclusion that there are many limitations of statistical design of clinical trials, such as improper calculation of sample size, the insufficient utilization of randomization and blinding, the inappropriate use of control drugs.Conclusion There are some problems in the design of clinical trials, which affected the quality of clinical trials.Therefore designer of clinical trials should pay more attention to the problems mentioned above in order to improve the quality of clinical trials.

Key words: clinical trial, cross-sectional study, statistical design