Health care policy and regional epidemiology: international comparisons and a case-study from The Netherlands.

In a number of countries, regional levels of administration have assumed responsibilities for the health care sector. This paper presents a specification of epidemiological information that could be relevant for health care policy towards and within regions, as well as a review of British and Scandinavian attempts at using routinely available mortality data for these purposes. A Dutch case study is then presented. To determine whether regional mortality data by cause of death can be used as indicators of need for health care, and consequently be used as criteria for the allocation of financial resources, the correspondence between provincial disease-specific mortality, hospital admissions and new cases of disability benefits was analyzed. the findings were not convincing: for a number of diseases a reasonable correspondence between mortality and hospital admissions could be shown; for other conditions, however, especially ischaemic heart disease, none at all. It is concluded that: in the context of the formal allocation of health care resources to administrative units, a strong confidence in mortality data by cause of death is unwarranted; in the (broader) context of developing and evaluating regional health care policy, mortality data by cause of death can be useful for providing a focus on population health status and a starting point for further studies.