Mendelian inheritance in man: diagnoses in the UMLS

Area of Science:

• Medical Informatics
• Genetics
• Bioinformatics

Background:

• Geneticists face challenges in translating rare inheritance disease codes into standardized biomedical terminologies.
• Existing coding schemes may lack the specificity required for detailed genetic diagnoses.

Purpose of the Study:

• To evaluate the utility and limitations of the Unified Medical Language System (UMLS) Metathesaurus for genetic diagnoses.
• To provide recommendations for enhancing the UMLS for the coding of common genetic disorders.

Main Methods:

• Translated 110 common Mendelian Inheritance in Man (MIM) disorders into Medical Subject Headings (MeSH), International Classification of Diseases (ICD), and Systematized Nomenclature of Medicine -- Clinical Terms (SNOMED).
• Assessed mapping success rates across different coding schemes (Meta-1.2).

Main Results:

• Commonly occurring genetic diseases and those with confirmed inheritance patterns were more successfully mapped.
• Approximately one-third of diagnoses could not be mapped across all three coding schemes.
• The ICD coding scheme was deemed too broad for genetic diagnosis and epidemiology.
• MeSH and SNOMED require enhanced specificity and completeness, with the inclusion of new SNOMED versions.

Conclusions:

• The UMLS Metathesaurus has potential for genetic diagnosis but requires significant improvements.
• Enhancements in MeSH and SNOMED specificity and completeness are crucial.
• The ICD coding system is inadequate for detailed genetic diagnostic and epidemiological use.