A recursively interpreted data structure for representing clinical entities

Clinical investigations generate a spectrum of data and data relationships that no longer fit simple data structures. To build a data management facility sufficiently flexible to deal with this spectrum of structures is thus an important problem. A general table-based data structure is presented that provides the clinical investigator with a means of tailoring the data structure to the experimentally derived data. The use of such a general structure appears limited to settings where deletion of instances of data is infrequent. This restriction poses no serious constraint in most clinical investigations.