• 1 January 1996
    • journal article
    • No. 86,p. 29-35
Abstract
This article describes the implementation of a computerised system for coding the cause of death from death registrations in England and Wales since the beginning of 1993, and the impact this has had on national cause of death statistics. Overall the automated system has been effective and reliable at deriving ICD-9 codes for the underlying cause of death and other conditions mentioned on the certificate. Automation increases the consistency and international comparability of cause of death coding. However, it has led to discontinuities in time trends between 1992 and 1993, which are explored here. The largest effect was related to expected changes in the application of ICD-9 selection rule 3. However there were other effects which were not expected. The automated system does not deal adequately with external causes of death, which are certified after coroner's inquest, and ONS has reverted to coding these clerically. Lessons for the implementation of ICD-10 for mortality have been learnt.

This publication has 0 references indexed in Scilit: