Implemented and live
Snow Owl MQ is a scalable Big Data software platform for Searching and browsing health records; Grouping patients that share the same characteristics into cohorts; Inspecting patient records to identify trends and correlations; and, Statistical analysis of patient cohorts to test and verify clinical hypotheses
Descriptive statistics for clinical characterization and casemix analysis help care providers identify groups of patients to improve outcomes. Typical questions are around natural history, understanding which patients have certain diseases and received different clinical procedures or medications.
Quality improvements are also a part of this, where the proportion of patients experiencing a specific outcome can be understood. Population-level estimation makes use of advanced statistical features and machine learning algorithms.
This helps clinical researchers use observational health data to understand whether a treatment is causing an outcome. It also helps identify associations between drugs and clinical events and evaluate the comparative effectiveness of different treatments.
How SNOMED CT will be used
Although we support over 70 terminologies, we use SNOMED CT as our reference terminology.
Why SNOMED CT will be used
SNOMED CT particularly excels at storing and retrieving clinical information contained in electronic health records. This is especially true for analytics.
The most important reason for us is that SNOMED CT is not a classification but an ontology, and that means it has encapsulated semantics that we can utilize when we would like to create queries and retrieve knowledge from patient data. This enables the meaning-based retrieval of clinically relevant facts. This is because SNOMED CT can ensure the semantic richness and the fine granularity that allows capturing all kind of clinical information, even very detailed types of data.
SNOMED CT is a comprehensive set of clinical concepts covering everything that needs to be recorded during a clinical encounter. It contains codes and concepts for diseases, procedures, observables, body sites, for administrative concepts, as well as for substances and pharmaceuticals.
Besides that, SNOMED CT is mapped to a number of other terminologies or classifications, such as ICD-10, ICD-10-CM, ICD-O, ICD-9, and ICD-9-CM which opens up the field for even more categorization options.
SNOMED CT uses multi-axial hierarchies. Concepts can have multiple categories that they can belong to—so they can have multiple parent concepts that represent different aspects of the concept’s meaning.
Finally, SNOMED CT's extensibility means that there are many existing extensions from governments, organizations, and healthcare providers that can be easily incorporated. In addition to these extensions, there are 100s of subsets from different countries and organizations.