This paper describes an end-user oriented data base system POD (Patient Oriented Data Base for Medical Research) which has been developed to provide the most convenient tool for medical research activities. By means of end-user languages, DDL, STORE and QUERY, a researcher physician can easily define his data base, maintain it and use the data in it himself for his research purposes.
The DDL is a data definition language to define the data structure of the data base by means of filling out a Data Definition Table. The STORE is a language to generate the data base and to update it, which includes four commands : CREATE, ADD, DELETE and UPDATE. The QUERY is a query language to retrieve the data stored in the data base and to analyze the basic statistics of the data, which includes eight commands : DEFINE, SELECT, SORT, LIST, HISTOGRAM, TABLE, MEAN and DTEST which were carefully chosen and designed on the basis of several years of experience at the University of Tokyo Hospital. The QUERY is designed to meet about 95 percent of physicians' requirements for statistical analysis. More complicated calculations can be achieved using a FORTRAN interface.
The basic data structure of a patient's record consists of two levels of hierarchy : an ID-record, a higher level record, including non-chronological data of the patient and DATA-records, lower level records, including chronological data. Each record has a bit-pattern to compress blank elements, which results in a remarkable reduction of the file volume without any complication of the data structure or increase of the processing time.
The processing system is written in FORTRAN language for machine independence, so that the system can run on most computers with little modification.
抄録全体を表示