OBJECTIVES: The coverage with evidence development (CED) is a mode of reimbursement intended for highly innovative drugs (HID) in the P&R system in the Czech Republic. Cost-effectiveness analysis and budget impact analysis (BIA) are mandatory requirements for setting the reimbursement. The objective of the present study was to assess whether the drug costs stated in BIA matches to real costs. METHODS: Twelve HID obtained CED in 2013. The drug costs were identified in BIA of eleven HID. Real costs of General Health Insurance Company (VZP) were found out for each HID in the first year of therapy. As VZP holds 60% of the health insurance market, the data were extrapolated to the whole population. The differences between estimated and real drug costs were analysed and a correlation between this differences and drug characteristics justifying their HID status were investigated. RESULTS: The estimated costs were exceeded in five cases (overrun between 31-332%). In six cases real costs did not achieve the estimation (12-91% of estimated costs). Concerning effectiveness and safety characteristics of investigated drugs, within seven drugs granted as HID because of absence of an alternative drug, four exceeded the estimation. One of two drugs granted as HID because of adverse effects reduction compared with current treatment exceeded the estimation. Remaining two drugs did not achieve the estimation, one was granted HID because of mortality reduction, another because of higher effectiveness compared with current treatment. CONCLUSIONS: However  HID costs estimated in BIA are submitted in order to predict costs of public health insurance, the analysis did not prove their validity and contribution to a reasonable decision making. Despite the fact that BIA is a mandatory requirement and it is cost and time consuming, its role is formal only and its premises have neither practical impact nor they are subject of further review.