Introduction to SNOMED CT

1
Introduction to SNOMED CT

• Jo Oakes
• Training Consultant
• Data Standards Products

2
Agenda

• History of Coding
• What is SNOMED CT?
• What are the benefits of SNOMED CT?

3
History of Coding
4
A Background to Coding
17th Century John Gaunts Weekly Bills of
Mortality
5
London Bills of Mortalityevery Thursday from
1603 until the 1830s
6
Aggregated Statistics 1665
7
Manchester MercuryJanuary 1st 1754

• Executed 18
• Found Dead 34
• Frighted 2
• Kill'd by falls and other accidents 55
• Kill'd themselves 36
• Murdered 3
• Overlaid 40
• Poisoned 1
• Scalded 5
• Smothered 1
• Stabbed 1
• Starved 7
• Suffocated 5

Aged 1456 Consumption 3915 Convulsion
5977 Dropsy 794 Fevers 2292 Smallpox 774 Teeth
961 Bit by mad dogs 3 Broken Limbs 5 Bruised
5 Burnt 9 Drowned 86 Excessive Drinking 15
List of diseases casualties this year 19276
burials 15444 christenings Deaths by centile
8
Quotation

• I am fain to sum up with an urgent appeal for
  adopting some uniform system of publishing the
  statistical records of hospitals. There is a
  growing conviction that in all hospitals, even in
  those which are best conducted, there is a great
  and unnecessary waste of life In attempting to
  arrive at the truth, I have applied everywhere
  for information, but in scarcely an instance have
  I been able to obtain hospital records fit for
  any purposes of comparison If wisely used,
  these improved statistics would tell us more of
  the relative value of particular operations and
  modes of treatment than we have means of
  ascertaining at present.

9
Florence Nightingale

• Notes on Hospitals, London Longman, Green,
  Roberts, Longman and Green, 1863.

10
Florence Nightingale 1856

• Outcome groups
• Relieved
• Not Relieved
• Died

11
Dr William Farr 1855

• Developed the International Listing of Causes of
  Death
• Adopted by World Health Organisation (WHO) in
  1948

12
History of the Clinical Terms

• 1985 4 byte set introduced
• 1990 Version 2 introduced
• 1990 Purchase by Department of Health
• 1992 Terms Projects
• 1994 Version 3 introduced
• 1999 SNOMED Collaboration

13
What is SNOMED CT?
14
First things first

• Systematized Nomenclature of Medicine
• SNOMED CT- Clinical Terms
• Developed by NHS Information Authority and the
  College of American Pathologists (CAP)
• Merges the content of the NHS Information
  Authoritys Clinical Terms Version 3 (The Read
  Codes) with CAPs SNOMED Reference Terminology
  (SNOMED RT)
• For the front end users (clinicians)
• SNOMED is not about the codes, but the terms it
  is unlikely clinicians will see the codes

15
SNOMED CT overview

• Comprehensive clinical terminology that is used
  to
• Code
• Retrieve, and
• Analyze clinical data
• Comprises of
• Concepts
• Terms
• Relationships
• All necessary to precisely represent clinical
  information across the scope of health care.

16
Basic Elements of SNOMED CT

• Concepts
• The basic units of SNOMED CT
• Descriptions
• These relate terms that name the concepts to the
  concepts themselves. Each concept has at least
  two Descriptions.
• Hierarchies
• Concepts are organized into twenty SNOMED CT
  hierarchies (in UK extension). Each hierarchy has
  sub-hierarchies within it.
• Relationships
• Relationships are the connections between
  concepts in SNOMED CT.

17
Concepts

• Represent distinct clinical meanings
• Identified by a unique numeric identifier
  (Concept ID) that never changes and a unique
  human readable name (Fully Specified Name)
• Associated with each concept is a set of
  relationships (the logical definition) and a
  set of names or terms
• Differing levels of granularity
• There are currently around 400,000 terms in
  SNOMED CT

18
Descriptions

• Concept descriptions relate the terms or names of
  a SNOMED CT concept to the concept itself.
• Term in this context means a phrase used to
  name a concept. A single description associates a
  single term with a single Concept ID.
• Each of these descriptions has a unique
  Description ID, but all of these descriptions are
  associated with a single concept (and a single
  Concept ID).
• Descriptions are an important interface property
  because they give end users the flexibility to
  use terms that they are familiar with.
• The Concept ID ties terms with the same meaning
  together to aid consistent interpretation and
  retrieval.

19
Description types

• Preferred Term
• The most common word or phrase used by clinicians
  to name a concept
• The Fully Specified Name
• Provides an unambiguous way to name a concept
• Synonyms
• The rest of the names that may be used for a
  concept

20
Example of components

• Some of the descriptions associated with
  ConceptID 22298006
• Fully Specified Name Myocardial infarction
  (disorder)
• DescriptionID 751689013
• Preferred term Myocardial infarction
• DescriptionID 37436014
• Synonym Cardiac infarction
• DescriptionID 37442013
• Synonym Heart attack
• DescriptionID 37443015
• Synonym Infarction of heart
• DescriptionID 37441018

21

• I don't know what you mean by "glory," Alice
  said.
• Humpty Dumpty smiled contemptuously. Of course
  you don't -- till I tell you.
• I meant "there's a nice knock-down argument for
  you!"
• But "glory" doesn't mean "a nice knock-down
  argument," Alice objected.
• When I use a word, Humpty Dumpty said in rather
  a scornful tone, it means just what I choose it
  to mean -- neither more nor less.

22
Words and semantics

• People behave like Humpty Dumpty
• Its not what you say, its what you mean
• The meanings of words and phrases change
• In context
• Between places
• Between disciplines and specialties
• Between different times
• SNOMED CT encodes meaning, not just words

23
The problem with words and meaning

• What is a pyogenic granuloma?
• Pyogenic pus forming
• Granuloma a collection of inflammatory cells of
  a particular type
• But
• Pyogenic granuloma a benign tumor of small
  blood vessels of the skin
• It is neither pyogenic nor a granuloma.

Combinations are frequently very different from
the sum of their parts
24
The problem with words and meaning

• Does the leg mean the same as the lower limb?
• Thats what the average person assumes
• Health professionals also often use the word
  leg in this way
• But medical dictionaries take a different view
• Stedmans the segment of the inferior limb
  between the knee and the ankle
• Dorlands that section of the lower limb
  between the knee and ankle
• Ordinary dictionaries recognise both meanings

Some formal definitions conflict with ordinary
usage
25
SNOMED is not the language police

• SNOMED CT seeks to
• Establish how language is used by clinicians
• To represent meaning as faithfully as possible.
• SNOMED CT declares what it thinks a phrase means
• To reflect current usage
• To minimise ambiguity
• Not to shape or control the way a phrase is used
• If you want someone to demand that clinicians
  change what they call pyogenic granuloma
• That may be someone elses job
• Its not something SNOMED CT is trying to do

26
Practical Tip

• Dressing (oneself)
• Dressing (e.g. a bandage)
• Dressing (assisting the person to dress)
• Dressing (of wound)

• Dressing (observable entity) parent personal
  care activity
• Dressing, device (physical object)
• Dressing patient (procedure)
• Dressing of wound (procedure)

27
Top-level hierarchies (July 2006)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
Relationships

• The connections between concepts
• Every SNOMED CT concept has at least one
  relationship to another concept
• Relationships characterize concepts and give them
  their meaning
• The list of relationships for a particular
  concept makes up the logical definition of that
  concept
• 2 types
• IS-A (sub-type/super-type)
• Attribute

36
Relationships cont

• Is a relationship also known as
  supertype-subtype or parent child relationships
• A relationship is assigned only when that
  relationship is always known to be true

37
IS-A relationships

• SNOMED CTs hierarchies consist entirely of IS-A
  relationships
• IS-A relationships in the Inflammatory disorder
  hierarchy
• Lumbar discitis IS-A Discitis
• Discitis IS-A Arthritis
• Arthritis IS-A Inflammatory disorder
• Inflammatory disorder IS-A Disease
• Some concepts have more than one IS-A
  relationship. These concepts have parent concepts
  in more than one hierarchy
• Lumbar discitis IS-A Discitis
• Lumbar discitis IS-A Disorder of back

38
Procedure
IS_A
Procedure on lymph node
IS_A
Biopsy of lymph node
IS_A
Surgical biopsy of lymph node
IS_A
Excisional biopsy of lymph node
39
SNOMED CT Concept
Disease
Disease of respiratory system
Infectious disease
Disease of lung
Pneumonia
Infective pneumonia
Bacterial infectious disease
Bacterial pneumonia
40
Attribute Relationships

• Characterize and specify concepts
• An example of an attribute is FINDING-SITE, which
  is used to further specify Disease concepts

e.g. part of the logical definition of the
concept Pneumonia in SNOMED CT is Concept
Pneumonia Attribute FINDING-SITE Value of
attribute Lung structure
41
Attribute Relationships
IS A ...
IS A ...
finding site
lower limbstructure
IS A ...
IS A ...
finding site
ATTRIBUTE
42
Why is SNOMED CT relevant?
Cross Boundary Universal Language The Spine
43
SNOMED CT Delivers Information

• Most clinicians should neither know nor care
• How many digits are in a concept identifier
• How many concepts are in SNOMED CT
• How descriptions and relationships are tied to
  concepts
• David Markwell, Chair of SNOMED Concept Model
  Working Group

44

• What clinicians care about is
• Clinical records they use meet their needs
• Record system components that work together
  reliably
• Being able to easily express the information they
  wish to enter in a patient record
• Being able to retrieve information to support
  delivery of care or for secondary uses with a
  minimum of extra effort
• David Markwell, Chair of SNOMED Concept Model
  Working Group

45
What are the benefits of SNOMED CT?
46
SNOMED CT The Benefits

• The benefit of recording information in a
  standard terminology such as SNOMED CT is linked
  to the benefits of the electronic care record and
  the benefits of recording clinical information in
  a structured form

47
Benefits of an Electronic Record

• Reduced storage costs
• Can be accessed from many places
• Can be transferred quickly
• It is legible

48
Benefits of a Structured Record

• Can display logical progression of clinical data
• Can retrieve clinical data based on situation or
  author
• Allows clinical data items to be transmitted
  longitudinally through a patients record

49
Benefits of using a Clinical Terminology

• Controlled vocabulary
• Point of care uses-
• The ability to search records for clinical
  information
• Identification of patients who match a given set
  of criteria
• Provision of decision support

50
Benefits of using a Clinical Terminology

• Aggregation uses-
• Public health monitoring
• Outcomes analysis
• Performance analysis

51
Benefits of using SNOMED CT

• Provides a consistent terminology across all care
  domains
• SNOMED CT allows precise recording of clinical
  information
• SNOMED CT has an inherent structure
• SNOMED CT is a developing international standard

52
What does it mean for clinicians?

• It will be legible!
• May encourage more detailed documentation
• Decrease duplication of recording patient
  information
• Less ambiguous/better clarity of clinical
  information
• More communicable
• Consistency leads to better sharing of
  information
• Improved granularity for research purposes

53
The Future for Clinical Coders
I see changes!!
54
What does this mean for Clinical Coders?

• The current system of ICD-10 and OPCS-4
  classification is not going to disappear
  overnight
• The cross-mapping of SNOMED CT to ICD-10 or
  OPCS-4 can only ever be a semi automatic process
• Coders will continue to be responsible for the
  clinical classification coding and trust payment
• Coders will be supported by new tools that help
  with some of the simpler translations of SNOMED
  CT to ICD-10 and OPCS-4

55
What does this mean for Clinical Coders? (2)

• NHS CFH Informatics and the NHS Classifications
  Services are assessing the training and support
  coders require under the Education, Training and
  Development unit with CFH
• Support for clinical uses of SNOMED CT and
  maintenance of SNOMED CT subsets is potentially
  an enormous opportunity for people who understand
  the issues surrounding coded medical information
  and are skilled at performing complex data
  analysis
• If the process of populating HRGs from SNOMED CT
  can be proved, there will be additional benefits
  in rationalising the many coding schemes
  currently used in medicine.

56
Is SNOMED CT Perfect?

• The man who makes no mistakes does not usually
  make anything.
• E J Phelps (1822 - 1900)
• so SNOMED CT is not perfect
• The goal is fitness for purpose not perfection

57
Any Questions?