Secondary Mental Disorder (ICD11)

It appears that there is a common concept in ICD10, ICD11, and DSM5 that represents the idea of a mental disorder due to a biomedical disorder. The idea was given a different name in each nosology:

• ICD10 uses the term 'organic mental disorder'
• ICD11 uses the term 'secondary mental disorder'
• DSM5 uses the term 'mental disorder due to another medical condition'

There is a different concept (different idea) used in early versions of DSM for the idea of mental disorder due to either a biomedical disorder OR  substance. This idea was given the same name in early versions of DSM ('organic mental disorder') that ICD10 gave to a more specific concept.

• DSM-III uses the term 'organic mental disorder' to refer to a mental disorder that is due either to a general medical condition or a substance

Because the description 'organic mental disorder' appears to have two different meanings over time, and because the term itself is not very descriptive,  should concepts in SNOMED having the expression 'organic mental disorder' and similar (e.g., organic mood disorder, organic anxiety disorder) be inactivated and new concepts created as replacement concepts with more appropriate FSNs?

• Create one set of concepts with new FSNs that represents the idea expressed in ICD10, ICD11, DSM5 by the term 'organic mental disorder'
• Create one set of concepts with new FSN that represents the idea expressed in earlier versions of DSM by the term 'organic mental disorder'
  • Does this concept need to be retained for 'backwards compatibility' for disorders historically captured based on DSM-IV and earlier editions?

Currently in SNOMED:

• It appears the concept 'organic mental disorder'  in SNOMED may have the meaning defined in earlier versions of DSM, as many of the substance-induced mental disorder concepts are child concepts of organic X mental disorder.
• There are a mix of concepts for mental disorders with FSNs using the expressions
  • 'Secondary X disorder' (ICD11 term) 
  • 'X disorder due to a general medical condition' (similar to DSM5 term)
  • 'Organic X disorder (ICD10, DSM-IV and earlier term)

Here are the verbatim definitions from the nosologies:

Specific input requested:

• How many discrete meanings are there in each of the above terms?
• Should the existing 'organic  mental disorder' concepts be retained, or should they be recreated with a more specific FSN (v. a nosology-specific FSN) to which each of the appropriate nosology-specific terms can be assigned?

Wow - an impressive summary!

The high-ranking returns from Googling 'dsm iv organic mental disorder' suggest that this topic has been under debate since at least the early 90's (including several papers by Michael), so I'm not expecting this to be easy to unpick.

Not sure I can add a lot in response to your 'How many discrete meanings...' question - you seem to have teased these out pretty comprehensively in the table above. The only possible additional thing I can see are multiple/shifting meanings for 'primary' and 'secondary' as ICD 10 moved to 11. In the text of the ICD 10 definition we are told that 'primary' and 'secondary' are used distinguish between the causal conditions (brain only or multi-system respectively). In ICD 11 'secondary' is used to refer to the caused mental disorder itself (contrasted with primary appearing in phrases such as 'primary psychotic disorder'). Not a big deal, but might be worth noting.

As a non-specialist, it appears that two things are changing simultaneously in the move from ICD 10 to 11 and from DSM III/IV to V:

• A change in labelling - organic becomes 'secondary' in ICD-speak, and 'due to another  medical condition' for DSM.
• A change in definition - as you explain, from the highly inclusive DSM III form to a more restrictive form where the substance-related conditions are classified elsewhere. Given the absence of a text definition the intensional definition can only be from the extension. It's notable that 111479008 | Organic mental disorder has subsumed the substance-related conditions (in a DSM III way) for 20 years in SNOMED CT without appearing to cause problems (looking back as far as I can it's always been happily mapped to ICD 10's F06.9). In part this may be because of the 'context-free assumption' in the ICD 10 mapping heuristics.

I don't know how helpful this might be, but I wonder if the ISO/IEC 25012 Data quality model (summarised here - maybe SI can source a full copy if this is needed) can help guide our thinking. Attempting to satisfy certain quality characteristics is likely to result in trade-offs: changes made in the name of 'credibility' and 'currentness' might include removal the substance-related conditions as descendants (motivated by a desire to align SNOMED CT with ICD 11's definition), as well as changes to the terms used ('organic' -> 'secondary' and/or ->'due to...'). Such changes, without further explanation, are likely to put pressure on 'understandability' - not least on understanding by the non-specialist MH community.

The latter (terming) changes also expose a tension between 'currentness' and 'completeness': if SNOMED CT to be used in settings where ICD 10 and ICD 11-based recording (or thinking) happen concurrently, it may be that, for example, ICD 10's 'Organic dissociative disorder' and ICD 11's 'Secondary dissociative syndrome' should both be available - possibly for a protracted period of parallel running. If we are confident they are the same thing then this might be achieved by adding new terms to existing concepts, but if not we may need distinct active concepts for both forms. In either case 'understandability' will require a considerable amount of 'metadata' and implementation guidance to avoid confusion (once again, in particular, to assist the non-specialist MH community).

Hope this is useful - Ed.

If we are to bring this content up to date, then I would be reluctant to base any changes on an outdated classification such as DSM-III.  This is especially true in light of the fact that there was no attempt in DSM-III to align with ICD, whereas both IV and 5 are closely aligned with ICD.  Another consideration is what is meant by "another medical condition".  It is conceivable that this would, in fact, incorporate substance use and misuse, which makes the interpretation of both DSM-III and DSM-5 much more similar. SNOMED has taken the view that the FSN of a concept should be specific and explicit to the point of removing any ambiguity, while the PT may be more "accessible" to users.  This would lead us to the DSM-5 terming, but would as suggested require a more clear understanding of what is included under "another medical condition".

I would concur that we should not be harkening back to DSM-III, which came out in 1980, in terms of aligning wording with old concepts. 

My understanding, which Michael First could confirm, was that DSM-IV used the phrase "general medical condition" and that DSM-5, just changed that to "another medical condition".  My recollection of the logic was that the term "organic" was viewed as outmoded and problematic in multiple respects but that some distinction needed to be made between that psychiatric conditions (including substance use disorders) and non-psychiatric conditions.  Since there was a desire to emphasize that psychiatric conditions were biomedical disorders like any other disorder and the mental vs. physical distinction was problematic, the phrase "general medical condition" was used.  But in actual use, no one knew what it meant and it sounded bizarre, which led to the change to "another medical condition". 

The description in DSM-5TR in the section Other Mental Disorders and Additional Codes notes: "This chapter provides diagnostic codes for psychiatric presentations that are mental disorders (i.e., symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning) but that do not meet diagnostic requirements for any of the mental disorders in the prior Section II chapters. These codes allow for the documentation and coding of these otherwise unclassified mental disorders. .... The categories 1) other specified mental disorder due to another medical condition and 2) unspecified mental disorder due to another medical condition are for presentations for which it has been determined that the psychiatric symptoms (e.g., dissociative symptoms) are a direct physiological consequence of another medical condition but do not otherwise meet diagnostic criteria for any of the prior Section II mental disorders due to another medical condition. For the diagnosis of other specified or unspecified mental disorder due to another medical condition, it is necessary to code and list the medical condition first (e.g., B20 HIV disease), followed by the applicable code for either other specified or unspecified mental disorder due to another medical condition."

Section II includes the substance use disorders but each of the chapters includes both Substance/Medication Induced Disorders (e.g., Substance Medication Induced Anxiety Disorder) as well as Disorder due to Another Medical Condition (e.g., Anxiety Disorder due to another medical condition).  This makes it seem as though the substance related and other causes are viewed as distinct. 

Clinically, at least, I think most people would view the word "secondary" as being synonymous with "due to"

The ICD-11 info above specifically excludes delirium as a secondary mental disorder, so there should be caution used in that regard as well since delirium was part of the Organic Brain Syndromes in DSM-III. 

We must also remember that SNOMED CT is not a classification and thus the exclusions specified in classifications do not apply.  SNOMED also does not use "other specified" or "other unspecified" residual categories.  The need for these types of grouping concepts need to be justified.  The separation of substance use disorders are handled in the modeling of SNOMED concepts using the CAUSATIVE AGENT relationship, while the mental disorders causally associated with another medical condition are modeled with a DUE TO relationship.  

If the relevant sources of truth have eliminated the use of "organic" in describing mental disorders, then any attempt to bring the mental disorder terminology up to date should do likewise, with inactivation of outdated terms and replacement with current terminology.  Historical relationships in SNOMED CT would allow for traceability.

There are 37 concepts in the clinical finding hierarchy with 'Organic' in the FSN. All but 4 of these concepts appear to refer to mental disorders.  

Excel spreadsheet with conceptId and FSN attached.