Send to

Choose Destination
See comment in PubMed Commons below
Nihon Koshu Eisei Zasshi. 2004 May;51(5):347-56.

[Health care resources and mortality as assessed by the "empirical Bayes estimate of standardized mortality ratio": results for Fukuoka Prefecture].

[Article in Japanese]

Author information

Department of Technology Assessment and Biostatistics, National Institute of Public Health.



The standardized mortality ratio (SMR) is frequently used to compare health status among different populations. However, it may be biased when based upon communities with small population sizes such as towns, cities, and wards. Thus, comparison of SMRs among such small communities is not appropriate. But the "empirical Bayes estimate of standardized mortality ratio" (EBSMR) is, in contrast, a useful index. The objective of the present study was to use the EBSMR to clarify the relationships between health care resources and mortalities in 109 communities in Fukuoka Prefecture.


Data for health care resources (number of physicians, number of general clinics, number of general sickbeds in hospitals, number of emergency hospitals, and proportion of elderly outpatients within their resident' community) and socioeconomic factors (birth rate, inflow or outflow population, aged households, marital status, taxable income per individual under taxes duty, unemployment, primary, secondary, tertiary industrial employment and criminal offense records) were obtained from officially published reports. EBSMRs for all causes, cerebrovascular disease, heart disease, malignant neoplasms, and acute myocardial infarction were calculated from the 1993-1997 vital statistic records. Multiple regression analysis with stepwise variable selection was used to examine the relationships between EBSMRs and the five variables representing health care resources, considering the eleven socioeconomic factors as covariates. Some of the variables were log-transformed to normalize the distribution.


Some of the EBSMRs were inversely related to the numbers of physicians per person (acute myocardial infarction in males (P=0.047) and females (P=0.012)), emergency hospitals per person (acute myocardial infarction in females: P=0.001), and general sickbeds per person (all causes in females: P<0.001, cerebrovascular disease in females: P=0.007, heart disease in females: P<0.001, malignant neoplasms in females: P=0.049). In contrast, when the higher the number of clinics per person, the higher the EBSMR in females for all causes (P=0.025), as well as acute myocardial infarction (P=0.006).


The results suggest that an appropriate distribution of hospital care resources such as physicians, general sickbeds, and emergency hospitals is an important factor related to mortality in a community.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Loading ...
    Support Center