BACKGROUND: In most jurisdictions, policies have been adopted to encourage the development of treatments for rare or orphan diseases. While successful as assessed against their primary objective, these policies have prompted concerns among payers about the economic burden that might be caused by an annual cost per patient in some cases exceeding 100,000 Euro. At the same time, many drugs for rare disorders do not meet conventional standards for cost-effectiveness or ‘value for money’. Owing to the fixed (volume-independent) cost of research and development, this issue is becoming increasingly serious with decreasing prevalence of a given disorder. METHODS: In order to critically appraise the problems posed by the systematic valuation of interventions for ultra-rare disorders (URDs), an international group of clinical and health economic experts was convened in conjunction with the Annual European ISPOR Congress in Berlin, Germany, in November 2012. Following this meeting and during subsequent deliberations, the group achieved a consensus on the specific challenges and potential ways forward. RESULTS: The group concluded that the complexities of research and development for new treatments for URDs may require conditional approval and reimbursement policies, such as managed entry schemes and coverage with evidence development agreements, but should not use as justification surrogate end point improvement only. As a prerequisite for value assessment, the demonstration of a minimum significant clinical benefit should be expected within a reasonable time frame. As to the health economic evaluation of interventions for URDs, the currently prevailing logic of cost-effectiveness (using benchmarks for the maximum allowable incremental cost per quality-adjusted life year gained) was considered deficient as it does not capture well-established social preferences regarding health care resource allocation. CONCLUSION: Modified approaches or alternative paradigms to establish the ‘value for money’ conferred by interventions for URDs should be developed with high priority.