SNOMED-encoded surgical pathology databases: a tool for epidemiologic investigation

Area of Science:

• Pathology informatics
• Medical coding systems
• Health data management

Background:

• Pathology departments have long worked on coding anatomic pathology reports.
• Vast amounts of electronically coded surgical pathology data exist, rich in information.
• Previous attempts to use Systematized Nomenclature of Medicine (SNOMED) for population studies were hindered by non-uniform databases.

Purpose of the Study:

• To highlight the importance of coded surgical pathology databases for research, teaching, administration, and public health.
• To explain the functional differences between coded databases and free-text data.
• To provide guidelines for ensuring automatic coding or recoding of surgical pathology reports.

Main Methods:

• Description of a computer program for automatic SNOMED coding of surgical pathology reports.
• Analysis of a large surgical pathology SNOMED database.
• Explanation of principles for ensuring report codability.

Main Results:

• Automatic coding offers uniformity and completeness to SNOMED databases.
• Coded databases enable customized recoding using various nomenclatures and algorithms.
• Simple guidelines can ensure future surgical report files are automatically coded or recoded.

Conclusions:

• Uniform and complete coded surgical pathology databases are crucial for diverse applications.
• Automatic coding overcomes limitations of manual or non-uniform coding methods.
• Adherence to simple guidelines by departments and vendors can optimize the utility of pathology data.