Neurocognitive disorders in ICD‐11: a new proposal and its outcome

The appropriate classification of diseases involving neurocognitive impairment has been an area of professional dissent between psychiatry and neurology for the past several years. This has been reflected in the groupings of “neurocognitive disorders” in chapter 6 on mental disorders and “disorders with neurocognitive impairment as a major feature” in chapter 8 on diseases of the nervous system in the draft version of the ICD‐11.

The disagreement about the placement of dementias in ICD‐11 has been settled by international consensus after the intervention of several scientific associations in the mental health field1, with the dementia categories being included in chapter 6 and their underlying causes being represented in chapter 8, following the logic of ICD‐10.

In August 2018, a group of neurologists posted on the ICD‐11 website a “complex hierarchical changes proposal” to replace “vascular dementia” with “vascular cognitive impairment” (VCI)2. Referring to the publication of a “primer”3, VCI was defined as “the contribution of vascular pathology to any severity of cognitive impairment, ranging from subjective cognitive decline and mild cognitive impairment to dementia” . In essence, the proposal argued that the term “vascular dementia” in chapter 6 had become “obsolete” and should be replaced by VCI in chapter 8.

The stated rationale3 was that “vascular pathology is common in the elderly with and without cognitive decline… mostly caused by a mixture of degenerative brain pathology in association with ischemia…; this requires detailed neurological and imaging workup and will forever preclude the diagnosis of ‘vascular dementia' without proper investigations” . The proposal concluded: “We also suggest to our colleagues dealing with the mental health chapter 6 to reconsider their definition of vascular dementia. It is crucial that ICD‐11 truly reflects modern 21st century thinking and practice”.

After consultation with the World Health Organization (WHO), a roundtable discussion with invited psychiatric experts was organized at the World Congress of Psychiatry in Mexico City in September 2018. A consensus was reached to post a critical commentary after endorsement by a larger group of experts representing a variety of national and international scientific psychiatric associations4.

The commentary, posted on October 19, 2018, stated that:

• Even if it would be appropriate to include “cognitive impairment” as a clinical manifestation of diseases or disorders, its proper location would be under “neurocognitive disorders” in chapter 6. Moreover, it remains unclear why “vascular” should be attached to cognitive impairment, since the authors rightly claim that the vascular one is almost never the exclusive aetiology of that impairment.

• Additionally, the VCI proposal refers to a patho‐clinical continuum of cognitive impairment adopted from current Alzheimer research models5. It lacks a clear classificatory concept, a convincing definition, an explicit operationalization of the “cognitive” profile, as well as a valid severity grading of “impairment” . Hence, its diagnostic and classificatory relationship with subjective (preclinical), mild or severe forms (dementia) remains unclear and not consistently developed for application in ICD‐11.

• Concerning the predominantly vascular forms of neurocognitive disorders, neither the close similarity of the terms “vascular cognitive impairment” and “vascular dementia” nor the latter's existing option for post‐coordination with the detailed category of “cerebrovascular diseases” in chapter 8 are reflected in the proposal. Hence, the proposal to relocate, rename or replace vascular dementia by VCI is neither consistent with current classification principles6 nor ready for implementation.

• Accordingly, using the term VCI and proposing pure vascular cognitive “impairment” as a separate category is not convincing. Moreover, “vascular” as a collective term refers to very different cerebrovascular diseases, which may interact with other aetiologies, and whose role may change over lifetime. Therefore, “vascular” should not be used as a fixed combination in a broad‐spectrum term like VCI, spanning several diagnostic stages and aetiologies of cognitive impairment.

Given the scientific state of the art3, 5, the classificatory rules of ICD‐116, and the existing ICD‐11 classification and coding of neurocognitive disorders across chapters 6 and 81, the following modifications were proposed:

• For “vascular dementia” , a coding note says that “this category should never be used in primary tabulation” . By post‐coordination, “6D81 Vascular dementia” optionally could already be “associated with” various “cerebrovascular diseases” from chapter 8, with “6D86 Behavioural or psychological disturbances in dementia” , and with an additional severity code. “6D80.2 Alzheimer disease dementia, mixed type, with cerebrovascular disease” already provides an opportunity to code mixed etiological forms of dementia as suggested in the above proposal. In case of multiple aetiologies, all that applies could be coded.

• For classificatory consistency, however, vascular dementia should be reformulated as “dementia due to cerebrovascular disease” following the pre‐coordinated formulation (“dementia due to…”) of other dementia categories in chapter 6 and should mandatorily be post‐coordinated with the respective category of cerebrovascular diseases in chapter 8.

• A related issue is the aetiological underpinning of “6D71 Mild neurocognitive disorder” . Post‐coordination offers an opportunity to add as causing conditions a number of “diseases classified elsewhere” , from chapter 8 and others. However, the option for also adding “cerebrovascular diseases” or multiple conditions is missing. This should be corrected.

• Together with these proposed modifications, the current ICD‐11 version of vascular related neurocognitive disorders would already allow coding for the mild and severe stages of vascular or mixed neurocognitive disorders.

In conclusion, the implementation of a new category of VCI in chapter 8 seems premature and not acceptable from the perspective of: a) the underdeveloped status of the classificatory concept of this entity, and b) its lack of adaptation to the present structure and coding options of ICD‐11 neurocognitive disorders.

On October 20/21, 2018, the authors of the VCI proposal posted an agreement7 with the above proposals and renounced the introduction of VCI in chapter 8. After being conveyed to responsible WHO bodies, the debate's outcome and resulting actions were officially endorsed at the WHO Family of International Classifications ICD‐11 conference in Seoul.

Since December 18, 2018, the proposed changes are implemented both in the frozen and the maintenance version of ICD‐11 (https://icd.who.int/browse11/l‐m/en).