Abstract
The Systematized Nomenclature of Dentistry (SNODENT) is an effort of the American Dental Association (ADA) to develop a controlled terminology that addresses the needs of clinical dentistry. The ADA, collaborating with the College of American Pathologists, developed and incorporated SNODENT as a microglossary of SNOMED. However, little evidence exists of the effectiveness of its clinical coverage. We extracted diagnostic terms from clinical case reports and evaluated SNODENT’s and other medical terminologies’ coverage of those terms.
Methods
We selected 5 publicly available case reports from a dental treatment planning textbook1 for testing clinical coverage of dental findings and diagnoses. The criteria for selecting the case reports were that they were available publicly, included adult cases only, and provided enough information to generate diagnoses similar to what would be encountered in the clinic. ICD-10 (which includes ICD-DA), ICD-10 AHLTA dental (Armed Forces Health Longitudinal Technology Application), MeSH, SNODENT, and SNOMED Clinical Terminology were the terminologies under review. The five cases explicitly provided lists of findings and diagnoses which were directly matched against the terminologies. The terms were classified as either findings or diagnoses, as well as dental or medical. Each term was matched using the methodology described by Chiang et al2: no match (0), partial match (1), complete match (2). Coverage percentage was obtained. Student’s t-test and Bonferroni correction were used where appropriate. We expected that SNODENT would have a higher coverage among the listed terminologies since this study used dental cases.
Results
From the cases we obtained thirty two concepts (N=32). Twenty two of the concepts were classified as findings. In terms of clinical coverage we found that SNOMED had the highest mean (mean=1.6563) and higher overall number of matches (matched concepts=31). The Table and Figure show complete results for all terminologies. Terms of dental importance that were not present in SNODENT were found in SNOMED.
Clinical Coverage of Concepts in 5 Dental Teaching Cases
| N | Mean | Std. Deviation | Std. Error Mean | |
|---|---|---|---|---|
| SNOMED | 32 | 1.6563 | .54532 | .09640 |
| SNODENT | 32 | 1.3750 | .83280 | .14722 |
| MESH | 32 | 1.4688 | .67127 | .11867 |
| ICD10 | 32 | 1.2500 | .84242 | .14892 |
| ICD10 | 32 | .6250 | .70711 | .12500 |
| AHLTA |
Discussion
SNODENT performed similarly to other terminologies; however it did not perform as expected. This finding reinforces previous studies which highlight deficiencies in SNODENT. This study addressed the possible impact in clinical coverage that these discrepancies may have. Future studies should increase the number of cases and concepts using similar methodologies.
Acknowledgement
Dr. Torres-Urquidy is supported by CONACYT grant #167967. The authors would like thank Dr. Michael Chiang for his comments
REFERENCES
- 1.Stefanac SJ, Nesbit SP, editors. Treatment Planning in Dentistry. C.V. Mosby; St. Louis, MO: 2001. [Google Scholar]
- 2.Chiang MF, Casper DS, Cimino JJ, Starren J. Representation of ophthalmology concepts by electronic systems: adequacy of controlled medical terminologies. Ophthalmology. 2005 Feb;112(2):175–83. doi: 10.1016/j.ophtha.2004.09.032. [DOI] [PubMed] [Google Scholar]