Background: It has been suggested that decisions about which drugs should be reimbursed by the Norwegian National Insurance Administration (NIA) are ad hoc and made without explicit criteria for evaluating applications.
Material and methods: We assessed all documents that we were able to retrieve from the NIA for a sample of 31 applications to add new drugs to the drug benefit program, mainly in the 1990s. The assessment was done with respect to two questions. First, to what extent were different factors explicitly evaluated, such as treatment effects, side effects, cost-effectiveness and reimbursement costs? Second, to what extent did these factors affect the decisions that were made?
Results: We found documents for 19 of the 31 drugs. For the 19 drugs and nine factors that we considered as potentially important in making a decision (a total of 19 x 9 = 171 possible assessments), we found a total of only eight explicit written evaluations. In 10 out of 19 cases costs to the NIA and control of these costs, or use of the drug, appeared to have an important impact on the decision that was reached.
Interpretation: Based on the documents to which we have had access, the NIA's evaluations have not been systematic or transparent for the vast majority of drugs.