English  Français

Application of Multiple Cause of Death Information to Eliminate Garbage Codes

Agnieszka Fihel, Institut national d'etudes démographiques / ICMigrations
Magdalena Muszynska-Spielauer, Wittgenstein Centre (IIASA, OeAW, Univ. Vienna)

International comparisons of mortality largely depend on the quality of data. In Poland, for example, with more than 20% of deaths annually assigned to ill-defined cardiovascular conditions, mortality level due to well-defined causes of death is under-registered. The objective of this work is to demonstrate on the Polish case how the information from Multiple Causes of Death (MCoD) data can be applied to reclassify cardiovascular Garbage Codes (GCs) into well-defined causes to further to approximate mortality due to well-defined causes of death. In this particular example, based on the unique MCoD dataset for Poland, death counts due to cardiovascular GCs were re-assigned to well-defined underlying causes with the use of coarsened exact matching. We asses accuracy of the applied approach with by comparing age-specific death rates and age-standardized rates for large groups of causes before and after redistribution to those of other Eastern European countries with relatively good data quality. More than 86 thousand deaths initially assigned to GCs were reclassified mainly into well-defined cardiovascular diseases, cancers, endocrine, nutritional and metabolic diseases and respiratory diseases. Cardiovascular mortality structure by large groups of causes became similar to structure registered in other Eastern European countries characterized by a low prevalence of GCs.

Keywords: Civil Registration and Vital Statistics (CRVS), Applied demography

See extended abstract.

  Presented in Session P15.