Presented at the 1999 APIII Conference Return to 1999 Abstract Index
EVALUATION OF 530,000 DIAGNOSES ENCODED IN SNOMED II
G.
William Moore, MD, PhD
Baltimore Veteran's Administration Medical Center
Pathology and Laboratory Medicine Service
Baltimore, Maryland
G. William
Moore, MD, PhD 2
Rudolf P. Baumann, MD 1
1 Neuchatel Institute of Pathologic Anatomy, Neuchatel,
Switzerland
2 Baltimore Veteran's Administration Medical Center
Background: The Neuchatel Informatics System was introduced in 1982 with the MUMPS language, and evolved into a client-server system using ORACLE.
Design: The functional components of the data-bank include: PATIENTS, SPECIMEN, EXPEDITERS, REPORT, and DIAGNOSES. There has been a chronologic evolution since 1986 of person-registration, and distribution of diagnoses by organ. We discuss the possibilities for combining various lesions in order to register a given lesion only once per patient. The efficiency and reliability of manual coding by the pathologist in the Systematized Nomenclature of Medicine (SNOMED II) is discussed.
Results: There are 1,500 autopsies with 80,000 diagnoses, 140,000 surgical specimens (264,000 diagnoses), and 180,000 cytology specimens (190,000 diagnoses). The topography distribution of diagnoses in surgical specimens indicates a progressive increase in diagnoses for the digestive system (51%); a decrease in specimens from gynecology (17%) and skin (9%) in the thirteen year period of observation; and a relatively constant frequency for urogenital, respiratory, and hemolymphatic organs (each about 5%) in the entire collection. The coding of all observed diagnoses enables one to perform analyses of combinations (e.g., Helicobacter pylori and concomitant lesions); or to examine the morphologic expression of important general disorders, such as alcoholism or diabetes. An important feature is the simple and very rapid retrieval of rare or extraordinary lesions (e.g., lymphoma of thyroid; metastasis of basal cell carcinoma).
Conclusion: The future integration of information into routine practice, and the enhanced exploitation of medical content for epidemiologic research and statistical analysis is discussed.