OBJECTIVES: Budget impact analysis (BIA) is a tool used to predict and understand the potential financial impact of introducing a new health care intervention into a health care system that has finite financial resources. Czech laws only prescribe duty to attach BIAs to the applications for new drug reimbursements without specifying guidelines how to prepare BIA. We analyzed the differences between the BIA in the applications and the real expenditures of the public health insurance. METHODS: We have selected 3 applications of new drugs (romiplostin, lenalidomid, bevacizumab) submitted in year 2009 (or an established drug in new indication) and compared submitted BIA estimates for year 2010 with the real expenditures in the same year. We also compared the methods in the submitted BIAs with the Impact Analysis Guidelines published by Patented Medicine Prices Review Board (Canadian National Drug Reimbursement Authority) to identify potential reasons of differences. RESULTS: We found differences in the predicted number of patients, average cost of drug application and total impact on the public health insurance. CONCLUSIONS: The comparison with standard guidelines identified the key areas to be addressed in the future Czech legislative to improve the quality of submitted BIAs. The inaccuracies were mainly caused due to a) Lack of data sources and their transparency; b) Inaccurate or misapplied assumptions; c) Inappropriate choice of comparators; and d) Overall quality, e.g. false interpretation of referenced studies conclusions.