Reimbursement Status and Recommendations Related to Orphan Drugs in European Countries

Objective: To review the reimbursement recommendations issued by selected European health technology assessment agencies for orphan drugs and the reimbursement status of these drugs; to assess the relationship between the type of recommendation and reimbursement status. Methods: The list of orphan drugs to be included in the analysis was obtained from the European Medicines Agency and Orphanet. Seven European states were included in the analysis: Belgium, England, France, Germany, Poland, Scotland, and Spain. For all identified orphan drugs, relevant data on the reimbursement status and type of recommendation were collected for each country. The relationship between the type of recommendation and reimbursement status was evaluated separately for each considered country, using Cohen's kappa coefficient for the measurement of agreement; sub-analyses for oncology and metabolic drugs were performed. Results: Most reimbursement recommendations for orphan drugs were positive (71%), while approximately 17% were negative and almost 13% were conditional. The highest percentage of positive reimbursement recommendations was observed in Spain (97%) and France (95%) and the highest percentage of negative reimbursement recommendations was revealed for Poland (49%). On average, 65% of the 163 analyzed orphan drugs were reimbursed from public funds. The highest number of reimbursed orphan drugs was observed in Germany (n = 148), while the lowest, in Poland (n = 41). Considering all analyzed drugs, the highest agreement between recommendations and reimbursement status was observed for Spain (value of 1), and the lowest, for Germany (κ = -0.03). Conclusions: On average, more than 60% of identified orphan drugs were reimbursed from public funds in the included countries, and the majority of reimbursement recommendations were found to be positive. The agreement between reimbursement recommendations and reimbursement status differed between the countries, but overall, it did not show any patterns, as it ranged from -0.03 to 1 (κ coefficient).status: publishe