Cause of death statistics-how to avoid misinterpretation of mortality data

Abstract

National mortality registers provide important data for monitoring population health. Analyses of cardiovascular mortality in particular-and especially mortality from coronary heart disease-are frequently the basis for conclusions about the quality of healthcare and prevention. To be able to interpret disease-specific mortality differences between countries and changes in mortality over time, it is necessary to know the basics of monocausal cause-of-death statistics and the associated limitations in comparative statements.Conclusions from disease-specific mortality data can be problematic due to the highly subjective selection of diseases that are entered on adeath certificate after apost-mortem examination. In death certification, unawareness of the different "roles" of specific diseases-known at the time of death-within the causal chain leading to death can result in incomplete death certificates as adata basis for mortality statistics. Comparisons of disease-specific mortality rates are difficult due to different proportions of non-informative-including unknown-causes of death and due to country-specific preferences for which diseases are recorded on adeath certificate-especially in the prevalence of multimorbidity. Apopulation'smorbidity is therefore only reflected to alimited extent by mortality rates. Conceptual ambiguities with regard to the concepts of lethality and mortality when describing mortality rates can also lead to erroneous conclusions.Training of medical staff on how to complete adeath certificate and the introduction of an electronic death certificate can improve the quality of mortality data. Irrespective of this, knowing potential pitfalls when analyzing mortality data will improve the quality of health reporting.