Source document- Inception phase Document v1.0 in Word format. 

Open

Amendment History

Review Timetable

© International Health Terminology Standards Development Organisation 2009. All rights reserved.

SNOMED CT® was originally created by the College of American Pathologists.

This document forms part of the International Release of SNOMED CT® distributed by the International Health Terminology Standards Development Organisation (IHTSDO), and is subject to the IHTSDO’s SNOMED CT® Affiliate Licence. Details of the SNOMED CT® Affiliate Licence may be found at http://www.ihtsdo.org/our-standards/licensing/.

No part of this document may be reproduced or transmitted in any form or by any means, or stored in any kind of retrieval system, except by an Affiliate of the IHTSDO in accordance with the SNOMED CT® Affiliate Licence. Any modification of this document (including without limitation the removal or modification of this notice) is prohibited without the express written permission of the IHTSDO.

Any copy of this document that is not obtained directly from the IHTSDO [or a Member of the IHTSDO] is not controlled by the IHTSDO, and may have been modified and may be out of date.  Any recipient of this document who has received it by other means is encouraged to obtain a copy directly from the IHTSDO [or a Member of the IHTSDO. Details of the Members of the IHTSDO may be found at http://www.ihtsdo.org/members/].

1 Glossary

1.1 Domain Terms

2 Introduction

2.1 Purpose

The purpose of this project addresses to Issues with Site, especially to review disorders of blood vessels and nerves.

2.2 Audience

The audience for this document includes all standards terminology leaders, implementers and users.

Special cases may be Health Care Providers from all kinds of medical domains (internal medicine or surgical specialist disciplines), as disorders of blood vessels and nerves appear multi-disciplinary.

The audience includes also terminology modellers and computer scientists who deal with modelling problems affecting use of SNOMED CT hierarchies.

3 Detailed Problem Statement

3.1 Background

3.2 Statement of Problem

It has been stated, that the purpose of this project addresses to Issues with Site, especially to review disorders of blood vessels and nerves.

3.2.1 Summary of problem as reported

There has been very limited information provided what the core of the problem is, or what should be the intention of the demanded review. As there are no attachments, associations or dependencies provided in the tracker, this makes it at least more difficult.

In the content project tracker, it has been stated that:

414086009|Embolism inferred SubClassOf 19660004|Soft tissue disorder|

Assertion: Blood vessel structure SubClassOf Soft tissue structure.

This issue affects all disorders of blood vessels and nerves. Some of the consequences are surprising.

Above that, two documents have been noted as a reference:

Getting the foot out of the pelvis: modeling problems affecting use of SNOMED CT hierarchies in practical applications.

Rector AL, Brandt S, Schneider T.

J Am Med Inform Assoc. 2011 Apr 21. [Epub ahead of print]

PMID: 21515545

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3128394/ 

Notes for IHTSDO on Issues on SNOMED Inferred Hierarchies Raised in Recent Paper and Investigations

Rector AL, & the Manchester team

Notes in confidence

These documents can be used for a deeper understanding of the problem.

3.2.2 Summary of requested solution

At this preliminary stage it is possible to take a closer look at 414086009|Embolism inferred SubClassOf 19660004|Soft tissue disorder| as well as Blood vessel structure SubClassOf Soft tissue structure and to regard these concepts from the problems described in the reference literature documents.

3.2.3 Statement of problem as understood

Issues with Site shall be addressed, with a focus on the review of disorders of blood vessels and nerves.

3.2.4 Detailed analysis of reported problem

In the beginning, it may be of interest what “site” means in SNOMED CT. The definition is that it is an unapproved attribute. Otherwise, f. e. “Finding site” is an important Concept Model Attribute. At least, these informations are useful but they don´t seem to bring anyone closer to the core of the problem.

As we get out of the literature article by Rector et al. 2011, there are description logic modelling problems in the context of hierarchies in SNOMED CT. One of the examples for a problem type is “Issues with site and resulting interferences”. Referring to that, there are the related examples “Hypertension is kind of soft tissue disorder” and “Diabetes is kind of abdominal disorder”.

Furthermore it is stated in the article that, if we take the example of diabetes, the SNOMED CT policy implies that if we have an endocrine disorder it is necessary to give the site of the organ responsible for secreting the responsible hormone, even if there are systemic effects to expect. Above that it has been said that specific sites are also given to systemic disorders and that this can lead to unexpected classifications. A proposal for improving the hierarchy may be an alternative parent concept that focuses more on the affected system (and not on the “organ of origin”).

In the diabetes example we have the situation that “Diabetes is kind of abdominal disorder” in SNOMED CT. The improvement would be to have it as “Diabetes is kind endocrine system disorder” (see Figure 1).

Open

Figure 1: (A) Diabetes upwards hierarchy as is in the Systematized Nomenclature of Medicine showing classification as a disorder of the abdomen. (B) Alternative Diabetes type 1 upwards hierarchy with link to pancreas removed (and represented by other means not shown)                                                                           Rector et al. 2011

Another impressive example is the one with the hypertension presented earlier in this chapter. There, hypertension is defined as a kind of soft tissue disorder. It has been proposed that it would make more sense to provide a definition like “hypertension is kind of cardiovascular system disorder” (see Figure 2).

Open

Figure 2: (A) Existing upwards classification of Hypertension (fully specified name ‘Hypertensive disorder, systemic arterial’). (B) Hypertensive disorder site reduced to ‘Entire cardiovascular system’.                           

Rector et al. 2011

It is possible to apply the gained knowledge on the problem stated in the tracker as a next step. The constellation is:

414086009|Embolism| inferred SubClassOf 19660004|Soft tissue disorder|

For that we take a look at the specific description logic modelling in SNOMED CT (see Figure 3):

Open

Figure 3: Hierarchical connections of | 414086009 | Embolism (disorder) |

We see, that 414086009|embolism (disorder)| is defined as a 239953001|soft tissue lesion (disorder)| and the latter is defined as 19660004|disorder of soft tissue (disorder)|. 414086009|embolism (disorder)| has 20 direct child concepts (e.g. fat embolism, air embolism etc.) with a lot further descendants.

To get closer to the point why embolism may be a “grand-child-concept” of disorder of soft tissue, we take a look at definitions of the terms “embolism” and “soft tissue” given in literature.

An embolism is a condition where the blood flow in an artery is blocked by a foreign body. Embolisms may cause life threatening situations, e.g. lung embolism. There is a great variance of causing agents for embolism, like blood clots, air or gas bubbles, cholesterol or amniotic fluid[3].

As one of the definition of soft tissue is that it includes all kinds of tissues inside the body (except bone), so blood vessels are also meant[4].

In SNOMED CT there is also the distinction between 414086009|embolism (disorder)| and 55584005|embolus (morphologic abnormality)|.

The first concept describes embolism as a kind of disorder, so to say that it is the cause of a disturbance in the regular functions of the soft tissue (in this context the blood vessels).

The second concept describes the embolus itself as a morphologic abnormality and it is defined as a mechanical lesion. This means for example a kind of pathological structure located somewhere in the body.  

Referring to the upper insights, we state that the problem with embolism and disorder of soft tissue seems to be an analogous problem as “the one with the hypertension”. To say that embolisms develop in blood vessels, and blood vessels are soft tissues, this may lead to the inference and wrong conclusion that embolism is a soft tissue disorder.

For the other problem stated in the tracker the constellation is:

Blood vessel structure SubClassOf Soft tissue structure

We take a look at the specific description logic modelling in SNOMED CT as well (see Figure 4):

Open

Figure 4: Hierarchy of 87784001|blood vessel structure (body structure)|

In this case, the inference is more difficult to observe. It seems to be close to the phenomenon Rector et al. 2011 describe, as the ambiguity of the term “soft tissue”. It literature, soft tissue is defined as any type of tissue, except bone. In SNOMED CT, there is broad definition of soft tissue, it implies f. e. blood vessels and nerves but not internal organs. This lead to wrong description logic modelling, like the example “aortic aneurysms are disorders of soft tissue”.                                                                         

3.2.5 Subsidiary and interrelated problems

This artefact addresses issues with site, with focus on disorders of blood vessels and nerves.

There has been no example for a disorder of nerves where the “issue with site” problem occurs.

For that we are searching SNOMED CT for a suspicious case:

A matching example would be 7916009|alcoholic polyneuropathy SubClass of 75478009|poisoning.

Open

Figure 5: Hierachy of 7916009|alcoholic polyneuropathy

Of course, this example does not focus on the site of the disorder but the problem seems to be almost transferrable (and that it occurs in other issues than “site” as well).

In the example, it seems that alcoholic polyneuropathy is at least a poisoning, that is obviously wrong.  

In the context of nerve disorders, the Rector et al. paper describes the inference problems with the concept “neuropathy”. In this context, injuries of nerves are describes as neuropathies. To avoid this problem, the authors suggest to create a new hierarchic connection: To model “Neuropathy” as a child concept of “Disorder of nerve”. “Neuropathy” should cover just the functional disorders of the nerves (no other disorders like injuries or other damages).

The artf222308 may also have a dependency to artf221327 “Defining meanings of soft tissue and modelling of different meanings”. This topic addresses the different interpretations or meanings of the term “soft tissue”:  It has been said that "Soft tissue" does not always mean the same thing and cannot be simple referred to a single anatomical category. Three different meanings have been provided in the document:

Meaning 1: Soft tissue can mean a category for tumors in the framework types of neoplasms.

Meaning 2: Soft tissue can mean a category for sites of non-bone disorders.

Meaning 3: Soft tissues can mean a category for structures identified in images, including everything except for mineralized bone tissue and teeth.

It seems that meaning 2 and 3 have a significant overlap (soft tissue disorder and imaging soft tissue). Meaning 1 seems to differ from the other two in the form that soft tissue focuses on tumors and a concrete structure, which is a defined restriction.

This artefact shows that problems have occurred with the term “soft tissue” and it´s meaning as well. It seems that the term “soft tissue” is highly flexible and includes almost every part of tissue inside of the body except bone. This means that “soft tissue” is applicable on a great variety of body structures and it should be questioned whether it makes sense or provides any benefit of having subconcepts e.g. of disorders of soft tissue.

4 Risks / Benefits

4.1.1 Risks of not addressing the problem

By not addressing this problem, there will be a risk that the description logic modelling problems in the context of hierarchies in SNOMED CT cause misunderstandings and an inadequate use of the terminology, which can lead to unwanted effects like miscoding etc.

These “circumstances” may not meet the needs of the stakeholders in case of an inadequate content structuring.

If we don’t align with the representation that the clinical stakeholders are expecting, they will have less incentive to use SNOMED CT, and it will be less likely to recommend the adoption of the terminology.

4.1.2 Risks of addressing the problem

At least, it is a very ambitious task to gather information from a multidisciplinary stakeholder group to provide feedback about the appropriateness of the content adjustments.

As this issue affects all sites of disorders of blood vessels it will be very time-consuming and a high effort to determine all the hierarchical relationships where the described problem occurs and to work on these issues.

5 Requirements: criteria for success/completion

5.1 Criteria for success/completion

Successful completion of this project will include the revision of the “Issues with site” focusing on the disorders of blood vessels and nerves.

5.2 Use cases

The biggest challenge seems to work on this project within sensible limits (just think of an uncountable number of issues with site-problems in SNOMED CT that may exist).

In this context it seems to be appropriate to focus on the problem types that have been described by Rector et al. to take a strategic problem solution and to learn from the gain of knowledge. For that, the use cases would be f. e. “soft tissue” and “neuropathy”. 

In terms of neuropathy, it would be an approach to find all non-functional disorders of nerves (like injury, lesion etc.) and model them as children of “disorder of nerves” – not neuropathy as before. 

Possible use cases could be using these codes in clinical records at a primary stage. Other potential use cases belong to the secondary use, such as analysis of disorders related to soft tissue, blood vessels and nerves.

5.2.1.1 Fit with IHTSDO strategy

This project supports IHTSDO Strategic Priority 1:

• Improve understanding of SNOMED CT use in order to inform future directions

• Develop implementation guidance and resources in order to make appropriate use of SNOMED CT easier and more valuable, as well as to reduce risk.

Above that it supports Strategic Priority 2:

To produce robust terminology standards that are “fit for purpose” for priority use cases

6 Envision Possible Technical Approaches

6.1 Indicative Technical Approaches

In the present investigation we found out that there needs to be an improved description logic modelling to avoid inference effects in the future.

For that we take the tracker example

414086009|Embolism inferred SubClassOf 19660004|Soft tissue disorder|

The current description logic modelling in SNOMED CT is (see Figure 5): 

Open

Figure 5: Hierarchy of 414086009 | Embolism (disorder) |

This leads to the wrong conclusion that embolism is a disorder of soft tissue.

To improve the modelling, we take a look at the definition of “embolism”.

 An embolism is the lodging of an embolus, which may be a blood clot, fat globule, gas bubble or foreign material in the bloodstream.[5]

According to this definition and with the intention to include the “site aspect”, it seems more appropriate (or rather an option) to retire the hierarchical connection 414086009 | Embolism (disorder) | as a child concept disorder of soft tissue and to create a new hierarchical connection like the example presented in Figure 6:

Open

Figure 6: Modelling option for 414086009 | Embolism (disorder)|

6.2 End user Impact

This approach can make SNOMED CT more functional with regard to the use cases.

If you take the “embolism example” it will considerably be a huge effort to filter all the concepts that seem to have an “issue with site”-problem in combination with blood vessels and nerves..

As proposed, it seems to be the best option to limit the content improvement on one or two defined use cases, like f. e. “soft tissue” or neuropathy”.

7 Indicative Project Plan

7.1 Scope

7.2 Skills required

7.2.1 Solution Specification (Elaboration)

7.2.2 Implementation

7.2.2.1 Outline of work packages

7.2.3 Preventing recurrence of problem

7.3 Project size, lifecycle, duration and resource requirements

7.3.1 Expected project resource requirement

The project resource requirement is classed as MEDIUM.

7.3.2 Expected project impact and benefit

The project impact and benefit are expected to be high. There will be an improvement by creating correct hierarchical connections and to avoid unwanted effects, like f. e. misunderstandings and an inadequate use of the terminology. At least it is a contribution to quality assurance by providing improved terminology modelling.

7.3.3 Indicative resource estimates for elaboration, construction, transition and maintenance:

Elaboration phase:              2 person month effort, determine all the relevant issues with site contexts on blood vessels and nerves

Construction and transition phase: Development of definitions and guidance on usage (dependent on the outcome of the evaluation).  Ca. 3 person months´ effort from primary developer

Maintenance phase:           An evaluation of the content change should be considered regarding different use cases to test the general validity and practical applicability.

8 Appendices

8.1 Appendix One: Source materials

This artefact document is referencing the papers:

Rector AL et al. (2011). Getting the foot out of the pelvis: Modeling problems affecting use of SNOMED CT hierarchies in practical applications. J Am Med Inform Assoc. 2011 Jul-Aug; 18 (4):432-40

Notes for IHTSDO on Issues on SNOMED Inferred Hierarchies Raised in Recent Paper and Investigations. Rector AL, & the Manchester team

Notes in confidence

Another reference is artf221327 “Defining meanings of soft tissue and modelling of different meanings”.

https://jira.ihtsdotools.org/browse/IHTSDO-723?jql=text%20~%20%22artf221327%22

[1] NCI dictionary of cancer terms (2015). Soft tissue. http://www.cancer.gov/publications/dictionaries/cancer-terms?CdrID=45882 [25.07.2015]

[2] Dorland's (2012). Dowland's Illustrated Medical Dictionary (32nd edition). Elsevier. p. 606

[3] NHS (2015): Embolism. http://www.nhs.uk/Conditions/Embolism/Pages/Introduction.aspx [04.01.2016]

[4] NCI dictionary of cancer terms (2015). Soft tissue. http://www.cancer.gov/publications/dictionaries/cancer-terms?CdrID=45882 [25.07.2015]

[5] Dorland's (2012). Dowland's Illustrated Medical Dictionary (32nd edition). Elsevier. p. 606.