Abstract
The first generation of computerized medical records stored the data as text, but
these records did not bring any improvement in information manipulation. The use of
a relational database management system (DBMS) has largely solved this problem as
it allows for data requests by using SQL. However, this requires data structuring
which is not very appropriate to medicine. Moreover, the use of templates and icon
user interfaces has introduced a deviation from the paper-based record (still existing).
The arrival of hypertext user interfaces has proven to be of interest to fill the
gap between the paper-based medical record and its electronic version. We think that
further improvement can be accomplished by using a fully document-based system. We
present the architecture, advantages and disadvantages of classical DBMS-based and
Web/DBMS-based solutions. We also present a document-based solution and explain its
advantages, which include communication, security, flexibility and genericity.
Keywords
Computerized Medical Record - Structured Documents - Software Architecture - Genericity