To compare and contrast the recorded causes of death with the underlying cause of death for all non-neonatal LS members

Chris Marshall and Andy Sloggett, London School of Hygiene & Tropical Medicine

[Project number 30077]

The objective of this project is to ascertain whether there is a difference between the recorded underlying cause of death and other recorded contributory causes, and to establish if using these contributory causes might alter the pattern of broader cause of death groupings. We would also wish to see if changes from ICD8 to ICD9 and then ICD10 altered the recorded cause of death patterns.

Several LS projects have used the Underlying Cause of Death to group LS members into disease and other classifications. In all projects only the underlying cause of death variables have been used, as a result of which the true incidence of some causes of death which have only been recorded as 'contributory' may have been missed.

The 'Guidance for doctors certifying cause of death' document (http://www.gro.gov.uk/Images/certifiers_guidance_v2_tcm69-21289.pdf) states that 'the disease or condition thought to be the underlying cause should appear in the lowest completed line of part I' of the form. Hence it is possible, for example, that a subject who dies from a heart attack brought on by the stress of the extreme effort, due to arthritis in hips and knees, needed to cross a road quickly, will have an 'underlying cause of death' recorded as 'Arthritis' or 'Joint disease' and the heart disease/problem will be missed if only the underlying cause of death is considered.

This study will look at the patterns of both the underlying causes of death and the additional contributory causes and consider if a different approach should be recommended in the future. (n.b. currently the period December 1986 to Jan 1993 will be excluded from the study as the appropriate variables are not available within the LS.)