Cause of death as a contemporary problem.

Four characteristics have been identified that affect the contemporary analysis of cause of death: (1) the increasing use of cause-of-death data for public policy; (2) the growing disjuncture between cause of death and cause of poor health; (3) the problem of specifying cause of death for the elderly, who constitute a growing proportion of the population and, accordingly, of mortality; and (4) the impact of technology and medical science on the temporal comparability of mortality statistics. With regard to the first issue, statistical data in general are being used with increasing sophistication throughout society. The public is exposed to all types of data by the media,and is increasingly educated in how to interpret statistics. Government at all levels increasingly uses data to help to help identify priorities, to choose among policy options, and to evaluate outcomes of decision making. Cause-of-death data are but a specific subset of information that are being used to understand social, economic, and health issues and to make more informed private and public choices regarding these issues. Increasing use of any data set is likely to result in better data simply because greater use results in closer scrutiny of data quality with resultant emphasis on evaluation and systematic efforts to improve quality. This is happening to cause-of-death data in the United States. The National Center for Health Statistics (NCHS) initiated major efforts through two national workshops in 1989 and 1991 to improve the quality of medical certification of death through better training of physicians, medical examiners, and coroners. While there is no clear evidence that the initiatives resulting from these workshops alone resulted in data quality improvement, the indicators used by NCHS to measure quality of medical certification point to continued gradual improvement in completeness and specificity in cause-of-death data. The gap between causes of morbidity and causes of mortality is growing because advances in medical science has essentially eliminated many infectious and communicable diseases that previously caused premature death and have considerably reduced the lethality of chronic conditions that previously caused death at earlier ages. As acute conditions have moved into the background, so to speak, a variety of disabilities and impairments have moved forward as sources of illness,pain, and diminished quality of life. As a consequence, the previously clear etching in mortality statistics of a society's health has become blurred by advances in medicine, healthcare, and prevention. Mortality data continue to have value per set and are important to social and medical knowledge because of their availability as a uniquely continuous data set. However, increasingly, a complete and contemporary profile of the health of a society requires data on both morbidity and mortality as complimentary indicators of health and social development. Not only do we need to know of what causes people to die, but also what health conditions aggravate and diminish their lives and, at an aggregate level, that conditions impose the greatest health costs on the community. Increasingly, these are debilitating conditions of the elderly that may diminish life but may not cause death. Medical description of the cause of death of an elderly person has probably never been easy; however, the contemporary extension of life has made the problem more visible and substantial than in earlier times. It is likely that the relatively few octogenarians of centuries past were also afflicted by concurrent medical conditions, and that even younger persons were prematurely afflicted by conditions that are now identified with advanced age. For these population groups, the description of cause of death, as a single and singular initiating event, is problematic.