Critical illness investigation

Brief history

The current CMI critical illness investigation started in 1999 with the first data collected in 2000. The first data collected pertained to claims settled during 1998 and 1999. The investigation has continued to collect data since. The 1998 data was significantly lower in volume than subsequent years, so has not been analysed further.

Data, methodology and results

Data are collected for both ‘stand alone’ and ‘full accelerated’ (where the cover includes death as well as critical illness) business.

In the early years of the investigation, critical illness data was collected annually from insurance companies (also termed contributing offices) on both policies in force and claims settled during an investigation year. The original analysis methodology is described in detail in Working paper 14. The results compared actual settled claims against expected diagnosed claims, calculated using CIBT93 (a population-based table which was originally contained in A Critical Review, presented to SIAS in 2000). These results have subsequently been referred to as "unadjusted results".

In July 2008, a new methodology for analysing critical illness experience was presented in Working paper 33. This produces results that compare actual settled claims with expected settled claims, referred to as “adjusted results”. These results properly match claims to exposure, but are difficult to interpret particularly in terms of duration.

Starting with the submission of 2006 data, we have also received data submitted in accordance with the 'per policy' format which additionally requires information on other exits and also encompasses the CMI’s life office mortality investigation. Further information on the 'per policy' initiative is available here.

Slow progress in the transition to ‘per policy’ data led to the launch of a special initiative to capture 2007-2011 data, again in conjunction with the life office mortality investigation. Provided the greater coverage of dates of diagnosis in the 2003-2006 data is repeated in the newer, the Committee intends moving to a more conventional analysis methodology - producing results that compare actual diagnosed claims with expected diagnosed claims - as set out in Working paper 67.

Critical illness diagnosis rate tables

The first set of insured lives accelerated critical illness diagnosis rates, based on 1999-2004 data, were presented in Working paper 43. Working paper 50 updated these to 2003-2006 experience and formally presented tables of rates known as AC04. It is anticipated that these will be introduced as a comparison basis in future results. Working paper 52 presented cause-specific diagnosis rates, again based on 2003-2006 data, and Working paper 58 describes supplementary analyses designed to aid understanding of the AC04 Series rates.