Patient registries are a valuable research tool for understanding disease and can support new therapies. Diverse registry models, sponsors, and data requirements complicate efforts for the standardisation of registry Questions And Answer Sets (QAS). Bottom-up standardisation can be achieved by facilitating the re-use of existing questions through the use of searchable metadata coded using controlled terminologies. This paper describes the Patient Registry Item Specifications and Metadata for Rare Diseases (PRISM) library of registry questions representing a variety of rare diseases. PRISM uses the Dublin Core (DC) and the Systematised Nomenclature of Medicine Clinical Terms (SNOMED CT).