Mapping Lab terminologies

There can be anywhere from 5000 to 15,000 labs code at a given institution. The task of mapping these codes to a standard code set to another institutions codes for the purposes of use in Health Information Exchanges, EHRs for trending lab results can be daunting.

This task cannot be done by someone who understands just the bits and bytes. Considerable domain knowledge is required in order to this efficiently as well as accurately. It is common knowledge within this domain that far fewer than the 5000-15,000 codes account for the vast majority of the commonly ordered tests. And among this subset of commonly ordered, a smaller number of codes are actually important for the purposes of trending.

For example, in the in patient setting, each morning during review of patients lab during rounds, the most commonly trended lab observations are the lab elements contained within the "Comprehensive and Basic Metabolic" profiles, and the CBC. The CPK and ESR rates are trended, but usually for a limited time duration within a patients hospital stay. Test such as the Rheumatoid Factor or ANA on the other hand do not typically require a trended analysis for clinical decision making.

This paper by Daniel Vreeman et al (Regenstrief Institute, Inc. and Indiana University, Indianapolis, IN) makes the following conclusion:

"Given limited mapping resources, our findings support the strategy of focusing the effort on the small subset of observations that account for the majority of volume. Mapping the observation codes that cover 99% of the reported results would ensure that all of the results for more than 99% of patients would be mapped. Mapping even the few (49 to 68) observation codes accounting for 80% of reported results would cover all results for 91–98%
of patients."

Also see: Automated Mapping of Observation Codes Using Extensional Definitions