Abstract
Medical records of in-patients have been stored in the computer ever since the hospital affiliated with the Shimane Medical University was established in 1979. The central processing unit of the computer system comprised the MUMPS language-operated PANAFACOM U-1500 having 320kB main memory and two 40MB magnetic discs as external memory units. Medical data input is given through four CRT display units and the output is obtained through CRT displays or a line printer.
A new medical record abstracting method designated as the POS matrix has been developed. The POS chart contains problems (#), clinical data (S, P, L, T, O/A/B), the duration thereof and the description of the relationship between the problems (diagnoses) and clinical data. These relations are indicated by symbol (+) and arranged as a matrix.
Patient data abstracted under the POS matrix method are put into and stored in the CPU in the “tree-structure” composition, and these construct data base known as the “POS data base (POSDB)”. Useful lists of all sort are prepared through retrieval of data in the POS data base, but the most significant lists comprise the following two: One is the list for designating some combination of clinical data as key word for retrieval and accumulation of diagnoses, while the other retrieves clinical data on certain specified diseases and rearranges them in the order of frequency.
Of greater importance is that the POS matrix succeeds to the spirit of the original POS (Problem-Oriented System) advocated by Weed and that describing of medical records has educational effect on the medical personnel.
