INTRODUCTION
In 1971 a group of pathologists in The Netherlands decided to join forces in order to investigate the possible use of computers in pathology and founded the PALGA Institute. They wanted to achieve both a national database for support of patient care and research and a low-cost nationwide solution.
After experiments were completed the proposed solution was approved in 1980. The tests had been conducted by five pathology departments, among which two universities and the biggest Dutch laboratory that processes over seventy thousand specimens a year. The PALGA Institute became a permanent body in Dutch Health Care. Now all Dutch pathology departments are participants, both con- tributing to and profiting from PALGA.
The bureau of the PALGA Institute is now part of the SIG Institute at Utrecht as the department SIG-Amsterdam.
THE PATHOLOGY DEPARTMENT
In the pathology department investigations are done on tissue or cell material for the occurrence of disease. The treating physician takes the material from the patient to get for instance an answer on the question whether or not the patient suffers from cancer. The investigation is performed by looking at the tissue or (part of an) organ (the macroscopy) and if necessary by using a microscope (microscopy), then the pathologist formulates a diagnosis.
PALGA OVERVIEW
Essentially PALGA is computerization of the administrative workload in the laboratory. The system offers especially high speed data entry, an electronic archive of pathology reports, complete word processing facilities, and extensive quality checking.
Within the network the local system can be connected to the national database with standardized ABSTRACTS of the reports. The abstracts are submitted to the national database at regular, most daily, intervals. Within this process of data standardization and reduction special care is taken of the patients privacy. Vital items like birthname and birthdate are encrypted. Items that are less relevant for the computerized identification of a person, like current address, are omitted. The laboratory uses the central database for the (daily) retrieval of pathology patient histories, and (at weekly interval) for a compilation of reports, which contained a cancer diagnosis, and a reminder with respect to reports that, containing a diagnosis with suspicion of cancer, should have had a follow up.
A very important entry in the abstract is the diagnosis. It is used for the weekly compilations and for scientific (epidemiologic) purposes. Questions from pathologist, researchers or third parties, and relating to pathology diagnoses in a certain period with a distribution over The Netherlands can be answered with the information in the national database. A remarkable feature is that diagnoses are NOT encoded by people, but are entered as plain text, slightly formalized. Plausibility checking and coding into SNOMED or ICD-O classification codes is done automatically through thesauri (dictionaries). This also happens when queries are entered into the national database: they are entered in plain language. Therefore nobody but a special working group that maintains these thesauri has to be bothered by the properties and complexities of any classification and coding system.
CANCER SCREENING AND CANCER REGISTRY
In recent years the PALGA institute has closely cooperated with other national institutes and the Dutch Government in creating a Dutch Cancer Registry (based on input from regional Comprehensive Cancer Centers).
PALGA also developed an information system for the registration of cervical cancer screening (cervical smears) and participated in the development of the information system for breast cancer screening.
From 1986 PALGA delivers compilations of reports with diagnoses of cancer every week to the pathology departments in support of the work of the Comprehensive Cancer Centers. From 1989 PALGA runs the network and database that constitutes a vital part of the Dutch National Cancer Registry.
FACTS AND FIGURES
Local EDP is done with Intel based Unix systems and dedicated software, a harddisk (with enough capacity to keep five to ten years of reports on-line), printers, a modem and three to twenty- four terminals.
The network is established through public switched telephone and low-cost modems (300/2400 baud).
The central database is supported by a powerfull Unix system with over 15 Gbytes of disk and the relational database management system INGRES.
Early 1994 the database contained 16 million abstracts, that can all be accessed on-line. According to the identification (encryp- ted name, encrypted birthdate and sex) that is used these abstracts relate to 4.5 million persons.
Annual intake amounts to 2 million new abstracts.
After admission to the national system abstracts are filed as documents. They will not be converted or processed but will be interpreted and inverted into diagnostic, patient and other database tables. This enables PALGA to give various interpre- tations of the original data at later stages from any point of view without data loss through conversion.
The method of automatic coding of diagnoses results in an extre- mely low error rate in diagnostic data (less than 5 per ten thou- sand of all abstracts contain any error) and high flexibility. Also there is hardly loss of diagnostic detail, since the number of diagnostic statements and expressions is virtually unlimited.
GLOSSARY
EDP Electronic Data Processing.
ICD-O International Classification of Diseases for Oncology, WHO 1976.
PALGA Pathologisch Anatomisch Landelijk Geautomatiseerd Archief (Dutch National Pathology Information System).
SIG (Stichting) Informatiecentrum voor de Gezondheidszorg; (Information Center for Health Care Services).
SNOMED Systematized Nomenclature Of MEDicine, College of American Pathologists, 1979, updated through 1982.
INFORMATION
Please contact:
SIG-Amsterdam
Muntplein 10
1012 WR Amsterdam
The Netherlands
tel. 31-20-6271261
fax 31-20-6226247
Copyright © 1994-1996. Netherlands Cancer Institute.
Date last modified: