Halbert L. Dunn

Evolving Mechanisms for the Production of International Health Statistics

Abstract * Presented at the Annual Meeting of the American Statistical Association, Boston, December 27–29, 1951.

The Evaluation of the Effect upon Mortality Statistics of the Selection of the Primary Cause of Death

F OR ALMOST a century the nations of the civilized world have been working to develop a uniform reporting system for causes of death. As late as 1893, no two countries had attained complete uniformity of technique in their methods of statistical classification of causes of death. Since that time, the International List of Causes of Death has been established and adopted in most of these countries. More than one-half of the death certificates in the United States have on them two or more titles for cause of death. The tabulation of the multiple causes of death involving such a high proportion of the death certificates constitutes a major technical problem. Consequently, methods of procedures supplementary to the International List have been developed by this country to aid in the selection and tabulation of the primary or principal cause of death. At the 1929 meeting in Paris of the International Commission charged with the decennial revision of the International List of Causes of Death, it was recommended that the United States Government, which had developed rules for codification and selection of such primary causes of death, be asked to take measures for the establishment of a code of general instruction which could be adopted by the countries using the International Nomenclature. Thus this country was given the opportunity to assume the leadership in this particular problem. The problems involved with the selection of the primary cause of death are peculiarly complex and bewildering in character. The rules themselves, as expressed in the Manual of the Joint Causes of Death, have resulted from certain general concepts. These concepts in turn have arisen from the principles of nomenclature. Since there is no sharp line of demarcation between the problems involved with the selection of the primary cause of death and those inherent in the nomenclature of causes of death, the principles of nomenclature must be reviewed in a study of this subject. Cause of death nosology is the basis of mortality statistics. Mortality nosology consists of four major divisions: definition of the disease