This study aimed to estimate the impact of rare disease (RD) drugs on Bulgaria's National Health Insurance Fund's (NHIF) total drug budget for 2011-2014. While standard budget impact analysis is usually used in a prospective way, assessing the impact of new health technologies on the health system's sustainability, we adopted a retrospective approach instead. Budget impact was quantified from a NHIF perspective. Descriptive statistics was used to analyse cost details, while dynamics was studied, using chain-linked growth rates (every period preceding the accounting period serves as a base). NHIF costs for RD therapies were expected to increase up to 74.5 million BGN in 2014 (7.8% of NHIF's total pharmaceutical expenditure). Greatest increase in cost per patient and number of patients treated was observed in conditions, for which there were newly approved for funding therapies. While simple cost drivers are well known - number of patients treated and mean cost per patient - in real-world settings these two factors are likely to depend on the availability and accessibility of effective innovative therapies. As RD were historically underdiagnosed, undertreated and underfunded in Bulgaria, improved access to RD drugs will inevitably lead to increasing budget burden for payers. Based on the evidence from this study, we propose a theoretical framework of a budget impact study for RD. First, a retrospective analysis could provide essential health policy insights in terms of impact on accessibility and population health, which are significant benchmarks in shaping funding decisions in healthcare. We suggest an interaction between the classical prospective BIA with the retrospective analysis in order to optimise health policy decision-making. Second, we recommend budget impact studies to focus on RD rather than orphan drugs (OD). In policy context, RD are the public health priority. OD are just one of the tools to address the complex issues of RD. Moreover, OD is a dynamic characteristic and compromises the consistency and comparability of the calculated budget indicators.