Medical classifications and terminologies can be difficult to exploit.
Encoding data into medical repositories is not a simple procedure. Finding and selecting the proper classification codes in attempt to entangle clinical information - interpreted from patient records - can be difficult.
Likewise, selection of relevant codes for queries in medical repositories can be complicated, but is a prerequisite for medical statistics analysis, quality issues or financial reimbursement purposes.
The ability to select appropriate "code sets" for these purposes requires knowledge about classifications, data dictionaries and information models as well as some medical skills and insight into the primary interpretation process (see box).
Medical classifications are comprehensive - but often complicated to navigate due to their size and structure. Some classifications are partly indexed - but a considerable insight into the classifications is nevertheless often required for adequate exploitation.
These challenges have not diminished after the advent of large medical terminologies like SNOMED CT. However, SNOMED CT's internal relationships can be considered to be a kind of indexation.
The purposes of this site is to investigate the possibilities to utilize SNOMED CT - especially by means of the relationships - and develop a similar systematic indexing of some medical classifications.
Systematic indexations of medical classifications are apparently missing.
Thorough indexations of medical classifications could probably aid and align both the encoding and data retrieval of medical repositories.
Curiously however, and contrary to SNOMED CT, little attempts have been made towards systematic indexing of most medical classifications.
About the browsers: Be aware the the search mechanism on this site mainly focuses on "context-less" searches in SNOMED CT and classifications.
Context aided, "smart" searches and "post-coordination" in SNOMED CT, that are relevant in e.g. EHR systems, are investigated in another project: The EHR
• The SNOMED CT Browser ©
Read about SNOMED CT and the browser here
and try The SNOMED CT Browser © based on Release Format 2 (RF2).
• The SNOMED CT procedure lookup page:
The SNOMED CT procedure lookup page confines the possible search of concepts to procedures - with focus on surgical procedures. This focus is sustained by selection of specific attribute value pairs.
• The NCSP lookup page:
The NCSP lookup page works in the same way as the SNOMED CT procedure lookup page.
It is not a map - the classification codes are indexed to SNOMED CTs supporting hierarchies. Read more about NOMESCO's classifications here.
Try to compare your search results in The NCSP lookup page with the results in The SNOMED CT procedure lookup page. Note that the search facilities/functions is not exhaustive - they are experimental and under development.
Medical classifications have been in use for more than 100 years. Until the 1980s they have mainly been used in connection with local and national registries in order to aid the generation of local and nationwide medical statistics. Medical statistics has mainly been of "administrative" character (e.g. number of admitted patients) or for the purpose of e.g. general disease surveillance.
Through the last quarter of a century the information in the medical registries has increasingly been used in connection with DRG grouping and reimbursement. With a parallel increase for the interest of the "value" of particular classifications codes.
The use of medical classifications are, however, not simple and the phrase "Garbage In, Garbage Out" - or
GIGO reflects an issue that has been put forward in relation to the content of medical registries. The staff that interpret the clinical information in health records and transform this information into encoded data is rarely the same as the people, who use the content of medical repositories for analysis.
If the basic interpretation of the classification codes or their use differ amongst these groups, the value of the outcome of statistical, quality and financial analysis can be challenged.
A systematic indexation of classifications that rests on ontological or terminological principles might be the first step to overcome this gap.