FOI reference: FOI-2026-2605
You asked
On 7 January 2022, ONS released its latest version on the comparability ratios for underlying cause of death. In the documentation, there are two matrices included, one for Muse chapter 5.5 and 5.8 for leading cause of death and another one for the same chapters related to underlying cause of death.
In the leading cause of death matrix, the net change rate for Suicide and injury/poisoning of undetermined intent is 0% (both Muse 5.5 and 5.8 totalling 402 cases) used in X60 to X84 and Y10 to Y34.
In the matrix for underlying cause of death, the entry for chapter V Mental and behavioural disorders (F00-F99) has a net change rate of 0% (nine cases comparing Muse5.5 to Muse 5.8 outcomes on a total of just over 4,000).
Since the publication of that version in 2022, there has been a significant change which impacts Chapter V Mental and behavioural disorders (F00 to F99), namely, the NHS has since retired all the entries for F64.0 to F64.9 Gender Identity Disorders.
Volume 2 of ICD-10 is clear in not excluding F64 as a potential underlying cause of death. Furthermore, it can also be linked at causes of multimorbidity.
As a result, the removal of these codes by the NHS has potential consequences in no longer being able to determine gender identity disorders as an underlying cause of death in case of suicide.
Considering the amount of entries from coroners reports where the absence of access to appropriate healthcare is clearly referenced as a major contributing factor of mental ill health leading to death, please could you provide me with the following:
Has ONS been informed of the code change by the NHS regarding the retiring of ICD-10 code F64.0 to F64.9?
Has a risk assessment been performed to determine the impact of this change in terms of statistics, specifically those related to suicide?
If a risk assessment was performed, were any remedial actions identified and, if yes, please provide the detail of the determined actions.
Why has the dataset Indicative comparability ratios for underlying cause of death by ICD-10 chapter and leading cause of death between MUSE 5.5 and MUSE 5.8 to date not been revised to determine the impact on this dataset by the removal of those codes?
Given that the direct consequence of retiring the codes F64.0 to F64.9 is that no patient with gender identity disorder can be identified within the global population of suicides, this change appears to erase the entire subpopulation from the national statistics. If this statement is incorrect, please provide evidence thereof.
If no communication has taken place between the NHS (or NHS Digital, in this case) and the ONS on this particular topic, please provide with the normal procedure to be informed of major changes to data models which may directly impact the work of the ONS for calculating statistics.
Did members of the ONS and member responsible for the NCMD (National Children's Mortality Database) have any meetings on this topic since 2022? If yes, please provide with the agreed-upon actions from those meetings and whether they have been implemented.
We said
Thank you for your request.
We do not hold recorded information suggesting that NHS England or NHS Digital notified the ONS of a decision to retire International Classification of Diseases, Tenth Revision (ICD-10) codes F64.0 to F64.9 from use in NHS data. We have also not held meetings or had engagement with NHS England or the National Child Mortality Database on this specific topic since 2022.
Our mortality statistics are based on information collected at death registration, including all conditions recorded on the medical certificate of cause of death. We do not use NHS operational datasets to produce suicide statistics.
All conditions mentioned on the death certificate are coded using the ICD-10. From these conditions, an underlying cause of death is selected using ICD-10 coding rules.
The World Health Organization defines the underlying cause of death as:
a) the disease or injury that initiated the train of events directly leading to death, or
b) the circumstances of the accident or violence that produced the fatal injury
Where information relevant to ICD-10 codes F64.0 to F64.9 is recorded on the medical certificate of cause of death, we will continue to code that information in line with ICD-10 rules. Therefore, any change to the use of these codes within NHS data would not prevent us from assigning these codes where they are mentioned on the death certificate.
As we are not excluding ICD-10 codes F64.0 to F64.9 from cause of death coding, we do not hold a risk assessment on the impact of NHS retirement of these codes. For the same reason, we do not hold recorded information on remedial actions arising from such a risk assessment.
Our Indicative comparability ratios for underlying cause of death by ICD-10 chapter and leading cause of death between MUSE 5.5 and MUSE 5.8 dataset has not been revised for this reason, as the continued use of ICD-10 codes F64.0 to F64.9 in ONS mortality coding means that the asserted NHS data change does not affect the coding approach used for death registration data.
We are preparing for the transition to the International Classification of Diseases, Eleventh Revision (ICD-11). Under ICD-11, gender incongruence is classified within code block HA60.
For further information, please contact health.data@ons.gov.uk.