文章基本信息

标题：Bayesian Small Area Estimates of Diabetes Incidence by United States County, 2009
本地全文：下载
作者：Lawrence E. Barker ; Theodore J. Thompson ; Karen A. Kirtland 等
期刊名称：Journal of Data Science
印刷版ISSN：1680-743X
电子版ISSN：1683-8602
出版年度：2013
卷号：11
期号：2
页码：249-267
出版社：Tingmao Publish Company
摘要：In the United States, diabetes is common and costly. Programs toprevent new cases of diabetes are often carried out at the level of the county,a unit of local government. Thus, ecient targeting of such programs requirescounty-level estimates of diabetes incidence􀀀the fraction of the nondiabeticpopulation who received their diagnosis of diabetes during the past12 months. Previously, only estimates of prevalence􀀀the overall fraction ofpopulation who have the disease􀀀have been available at the county level.Counties with high prevalence might or might not be the same as countieswith high incidence, due to spatial variation in mortality and relocation ofpersons with incident diabetes to another county. Existing methods cannotbe used to estimate county-level diabetes incidence, because the fraction ofthe population who receive a diabetes diagnosis in any year is too small.Here, we extend previously developed methods of Bayesian small-area estimationof prevalence, using diuse priors, to estimate diabetes incidence forall U.S. counties based on data from a survey designed to yield state-levelestimates. We found high incidence in the southeastern United States, theAppalachian region, and in scattered counties throughout the western U.S.Our methods might be applicable in other circumstances in which all casesof a rare condition also must be cases of a more common condition (in thisanalysis, \newly diagnosed cases of diabetes" and \cases of diabetes"). If appropriatedata are available, our methods can be used to estimate proportionof the population with the rare condition at greater geographic specicitythan the data source was designed to provide.
关键词：ayesian estimates; diabetes; small area estimates.