Background: Quality accessment is crucial in systematic review, because it affects the credibility of the conclusion. Presently there are many quality assessment standards for randomized control trial (RCT) and they are widely applied to cross-over trials. This way of accessment is inexpedient because cross-over trial is different from RCT. Though there is no definite quality assessment standard for cross-over trial so far, Cochrane handbook recommended that four points should be considered: appropriate cross-over design, randomized treatments order, carry-over effects and unbiased data. Besides, some universal issues between corss-over trial and RCT, such as blinding and lost of follow up, also affect the quality of cross-over trial. Objective: To evaluate the limitation of the current quality assessment for cross-over trial. Method: We performed a computer search for Cochrane Systematic Reviews (CSR) which included cross-over trials in Cochrane Library (2009, issue 4). Part of CSRs were randomly sampled and the cross-over trials included were reassessed with a nine-items standard, which was composed by the four recommended points and five universal items. The former quality assessment results in CSRs were extracted. Result: A total of 637 CSRs which involved cross-over trials were found. Only one of them considered the particularity of cross-over design. 155 cross-over trials were included in the 50 samples and 128 of them were reassessed. Risk of bias in carry-over effects, appropriate cross-over design and unbiased data existed to different degrees (see table1), but hardly considered in the former assessment. Conclusion: The current quality assessment standards for RCT is not comprehensive enough to assess cross-over trials and serious defects existed. We recommend that the quality accessment standard should be improved when cross-over trials are involved.