SNOMED-CT is a huge coding system that uniquely identifies over 300,000 clinical concepts. These concepts may be diseases, procedures, labs, drugs, organs, infectious agents, infections, symptoms, clinical findings, and more. Additionally, SNOMED-CT defines over 1.3 million relationships between these concepts. SNOMED-CT is maintained by the National Institutes of Health (NIH) and is a subset of an even larger coding system, SNOMED, which includes concepts not relevant to clinical practice. The NIH has a software program called MetaMap (https://metamap.nlm.nih.gov/), which can be used to tag clinical concepts appearing in text, making it useful for natural language processing in healthcare.

While coding systems cannot uniquely identify every clinical concept with all of their variations and nuances, they come reasonably close, and, in so doing, make certain activities in medicine (particularly billing and analytics) much easier. In Chapter 7, Making Healthcare Predictive Models, we will use some coding systems to make predictive models for healthcare.

Now that we've covered healthcare fundamentals, in the next section, we will present frameworks for thinking specifically about healthcare analytics.