What is the Unified Medical Language System and Why Should We Use It?

Introduction

We will show a way of using existing tools to easily code an entire medical patient encounter with one code set, translate into any other code set, give us metrics that will relate symptoms to diagnoses and eliminate medical coding completely. This could be used to streamline and automate the entirety of health insurance and all of medicine with the exception of actually providing patient care.

Definition

The Unified Medical Language System (UMLS) is a compilation of various medical nomenclatures. Commissioned in 1986 by Donald A.B. Lindberg, then director of the US National Library of Medicine, it is intended for use by developers for medical informatics. Betsy Humphries oversaw the actual project and production of the database. UMLS provides a single way to search the various constituent nomenclatures and a built-in way to translate between them.

Uses

The purpose of the UMLS is to provide access to the language of medical documents without searching those documents, and to facilitate the development of computer systems that understand biomedical language. This is accomplished by putting the information into a single, searchable, hierarchical schema that relates similar concepts.

Licensing

The users of the UMLS are required to sign a “UMLS Agreement” and to provide annual usage reports.

Knowledge Sources

Metathesaurus

The UMLS Metathesaurus contains millions of concepts and concept names like SNOMED_CT, LOINC, RxNORM, CPT, ICD and many others. This Metathesaurus is arranged by concept, and all the concepts are housed in the same structure to make them easily searchable. The concepts are ranked according to their hierarchy, that means that ‘appendectomy’ has this hierarchy: Procedure/Surgical procedure/Operation on trunk/Operation on abdominal region/Operation on intestine/Operative procedure on large intestine/Operation on appendix. We can also search for related concepts like the 33 concepts directly related to Colonoscopy.

Figure 1 - 33 SNOMED_CT Concepts Related to Colonoscopy

External Sources

As mentioned above there are currently any of 189 sources and source languages available for the UMLS

Why does the UMLS Matter

Interoperability

For the first time (since 1986 at least) we have the ability to automatically translate between various nomenclatures and code sets and retrieve related and hierarchical concepts. This means that your old, evil, legacy health insurance companies that only speak ICD and CPT can be translated seamlessly into UMLS (SNOMED_CT specifically) allowing users to use insurance data as if it were native. Even better, you can go the other direction and translate UMLS language into ICD and CPT codes. Notice that it is UMLS Language. The UMLS uses codes, but there is no need to code a patient encounter. The codes are only used internally in the database to find related, hierarchical or translated concepts.

Metrics

Since the entire patient encounter can be coded in UMLS we can study the relationship between symptoms, diagnoses and treatments, finding the most efficacious treatments and diagnosing tough diseases. Every episode of ‘House’ would become 35 seconds long. “Hey computer, my patient presents with these symptoms, what is wrong?” it isn’t great television, but it is great medicine.

No Codes

Since the entire patient encounter can be documented in UMLS, and UMLS speaks English or French or German or whatever, AND can automatically translate your documentation into any of the 189 available code sets, you never have to code a patient encounter again. No coding, no coders, no additional expenses, with the caveat that the evil, legacy insurance companies require some procedure (CPT) codes to have a corresponding diagnosis code (ICD) in order to convince them to pay. The coders can help you with this, but it wouldn’t be a big stretch to create a report that looks for those codes that require a certain diagnosis code and alert you when those aren’t present. Even better, we could create a new kind of insurance company that hosts your medical records system and just pays for the work practitioners do. Let’s repeat that. If a new insurance company provided the medical records system, and it worked on UMLS, they could pay for procedures performed in real time. That is what we are working on at Sentia Health.

Complications

The UMLS is large, complicated and difficult. It was created and is maintained by lexicographers, not technologists. What is needed is to import that data into a Relational Database Management System (RDBMS). The one we have chosen is Microsoft’s SQL Server. Then we indexed it for performance, and wrote several custom queries (create, read, update, delete and search are all generated by our tools) that give us a breadcrumb that shows the ancestors of any passed concept, all the related concepts and translations to and from the old, legacy ICD and CPT codes. So yes, this is hard, but doable, and in fact, done.

Conclusions

We have shown a way to use existing tools to easily code an entire medical patient encounter with one code set, translate into any other code set, give us metrics that will relate symptoms to diagnoses and eliminate medical coding completely. We have solved these problems and are working on the user interface to implement this solution.

If you liked what you read, please like and subscribe, click on the notification icon, subscribe to our newsletter, and follow us on all our social media and blog sites.

We have built a comprehensive health information system to keep the patient healthy and on the right track with the ability to incentivize healthy living. Implementing this system should be fairly simple and will completely revolutionize the way healthcare is paid for, saving countless lives. We have shown a way to use this system to make the best healthcare system in the world also the most efficacious and the most affordable.