RAND Corporation, Pittsburgh, PA 15213, USA.
J Public Health Manag Pract. 2012 Mar-Apr;18(2):156-9. doi: 10.1097/PHH.0b013e318223b138.
Public health systems vary by degree of centralization, reflecting the distribution of authority, responsibility, and effort between state and local public health agencies. We analyzed data from the 2008 National Association of City and County Health Officials Profile of Local Health Departments survey, and propose an improved composite measure of centralization that can be computed for all local health departments within a state, as opposed to a single state respondent, as done in 1998. While most states' structures (79.5%) are decentralized, the new measure presents a continuum from highly decentralized to highly centralized. The measure was internally consistent (Cronbach α = .87) and correlated somewhat strongly with the centralization classification from the 1998 survey (Kendall's τ correlation = .62, P < .001), suggesting that a stable centralization construct can be reliably determined. This new centralization variable can facilitate more nuanced studies of public health systems, and inform policy design and implementation.
公共卫生系统的中央集权程度各有不同,这反映了州和地方公共卫生机构之间权力、责任和努力的分配情况。我们分析了 2008 年全国城市和县卫生官员协会地方卫生部门概况调查的数据,并提出了一个改进的中央集权综合衡量标准,该标准可用于计算一个州内所有地方卫生部门,而不是像 1998 年那样只针对一个州的答复者。虽然大多数州的结构(79.5%)是分散的,但新的衡量标准呈现出从高度分散到高度集中的连续体。该衡量标准具有内部一致性(Cronbach α =.87),与 1998 年调查的中央集权分类有一定的强相关性(Kendall 的 τ 相关系数 =.62,P <.001),这表明可以可靠地确定一个稳定的中央集权结构。这个新的中央集权变量可以促进对公共卫生系统更细致的研究,并为政策设计和实施提供信息。