News Release

Medical jargon is not unequivocal

Peer-Reviewed Publication

Netherlands Organization for Scientific Research

Patient data are increasingly recorded in an electronic form instead of on paper. This makes it much easier to retrieve certain information. Yet if one patient with diabetes is given the label 'diabetes mellitus type II' and the other patient 'DM type 2', a search query for 'diabetes mellitus' will not produce a list containing all diabetes patients.

Terminology systems have been developed in order to introduce structure to the medical jargon in patient files. These are lists of terms from a certain medical area, such as diagnoses, causes of death or surgical interventions. Synonyms are linked to each other and underlying terms are often provided with a definition.

Cornet's research focused on the methods for evaluating the quality of these terminology systems. These methods examine the quality of the content, its accuracy and how complete the term lists and definitions are. By carrying out these quality assessments, a hospital department can establish how useful and accurate a system is. Then they do not need to blindly trust the developer of the system.

If everyone uses the same medical terms, it is much easier to use the data of all patients for improving patient care or for epidemiological research. Then a search query on the basis of a certain term will no longer incorrectly omit data. In the United States and the United Kingdom unequivocal term lists are already in use. But according to Cornet it will not be long before the Netherlands follows suit.

###

Cornet's research was part of a larger project entitled 'Making semantics explicit: fulfilling terminology needs of Computer Patient Record users'.

Ronald Cornet's research was funded by NWO.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.