What is per-policy data submission?
The CMI has been consulting with data contributors on a fundamental review of the data submitted to the mortality investigations. Please click here for more information on this and for the latest coding guide.
In the coding guide, a field ‘Date of amount review’ is required. Is this date only applicable if a contractual review is built into the contract or is it the date of each contractual benefit change?
The "Date of amount review" field is intended to apply to regular, contractual changes to benefits as listed in the "Type of increment/decrement" field (see section 5.33) - for example annual RPI increases. For non-contractual or irregular changes this field should not be used. For example, consider a reviewable critical illness product, reviewable every 5 years, say. Changes to the sum assured may or may not occur, therefore such changes are irregular. In this case, changes to benefits would better be recorded as "Alterations", as described in section 4.5 of the coding guide. However, if there is no change to the benefit but a change to premiums then no alteration is required as premium information is not collected by the CMI.
Where policies are amended from joint life to single life (or vice versa), how should these be recorded?
For the example of a joint life to a single life, 2 records should be submitted, showing a date of exit as the date of change from joint to single for each life and a type of exit of 'A'. The single life is also included with a movement on date of the date of change from joint to single with an entry into current status of 'A'. The date of policy commencement and benefit commencement for all records should be the original policy dates. As for subsequent submissions, the "Entry into Current Status" field should be coded as 'I' rather than 'A' (assuming no more alterations occur). 'A' refers to alterations in the current year rather than previous years, as per section 4.5.
What if a policy cannot be submitted according to the coding guide or I need to make some assumptions about the data, e.g. the date a particular event occurred?
One of the aims of moving to a per-policy basis is to make it easier for offices to submit CMI data (after the initial pain!). Hence it is hoped that we will capture more data which can be included in the CMI investigations. If you cannot adhere to the coding guide then please contact us so that we find a way for you to submit the data rather than excluding it. If any data is then found to fall outside the scope of the current investigations then it may not be used straight away, but new investigations covering such data can be introduced, if sufficient data is received from other providers. If assumptions need to be made, then again, please contact us.
If you have any questions about the CMI please email