| Literature DB >> 6516584 |
Abstract
This article summarizes the results of a recoding study, the design of which has been presented elsewhere (ref. 4). Results on underlying cause of death are presented as validity (comparison with coding of a medical expert) and reliability (comparison of repeated FSO internal codings). Validity problems occur chiefly with cardiopathies and cerebrovascular disease and in old age. Valitity of coding of cancer is generally satisfactory, although an estimated 7% overcoding occurs. An estimated 63% of deaths fall into reliably coded cause-of-death categories, but again problems occur with cardiopathies and old agegroups. For the analysis of trends in cardiopathies, ne needs to consider simultaneously the evolution of the numbers of deaths with cancer of breast and prostate, hypertension and arteriosclerosis, and diseases of connective tissue, skin and skeleton.Entities:
Mesh:
Year: 1984 PMID: 6516584 DOI: 10.1007/bf02078258
Source DB: PubMed Journal: Soz Praventivmed ISSN: 0303-8408