Identifying Unique Neighborhood Characteristics to Guide Health Planning for Stroke and Heart Attack: Fuzzy Cluster and Discriminant Analyses Approaches
نویسندگان
چکیده
BACKGROUND Socioeconomic, demographic, and geographic factors are known determinants of stroke and myocardial infarction (MI) risk. Clustering of these factors in neighborhoods needs to be taken into consideration during planning, prioritization and implementation of health programs intended to reduce disparities. Given the complex and multidimensional nature of these factors, multivariate methods are needed to identify neighborhood clusters of these determinants so as to better understand the unique neighborhood profiles. This information is critical for evidence-based health planning and service provision. Therefore, this study used a robust multivariate approach to classify neighborhoods and identify their socio-demographic characteristics so as to provide information for evidence-based neighborhood health planning for stroke and MI. METHODS AND FINDINGS The study was performed in East Tennessee Appalachia, an area with one of the highest stroke and MI risks in USA. Robust principal component analysis was performed on neighborhood (census tract) socioeconomic and demographic characteristics, obtained from the US Census, to reduce the dimensionality and influence of outliers in the data. Fuzzy cluster analysis was used to classify neighborhoods into Peer Neighborhoods (PNs) based on their socioeconomic and demographic characteristics. Nearest neighbor discriminant analysis and decision trees were used to validate PNs and determine the characteristics important for discrimination. Stroke and MI mortality risks were compared across PNs. Four distinct PNs were identified and their unique characteristics and potential health needs described. The highest risk of stroke and MI mortality tended to occur in less affluent PNs located in urban areas, while the suburban most affluent PNs had the lowest risk. CONCLUSIONS Implementation of this multivariate strategy provides health planners useful information to better understand and effectively plan for the unique neighborhood health needs and is important in guiding resource allocation, service provision, and policy decisions to address neighborhood health disparities and improve population health.
منابع مشابه
Defining Neighborhood, Analysis of Two Different Approaches: Expert-oriented Approach of Theorists and Perceptual Approach of Residents
The different offered definitions of “Neighborhood Unit” in various approaches and viewpoints by different specialized views, on the one hand; and excessive attention to expert-oriented and reduced viewpoints as well as neglecting residents’ perception of the neighborhood on the other hand necessitate providing a complete and exact definition of neighborhood which can cover all approaches a...
متن کاملIMNTV-Identifying Malicious Nodes using Trust Value in Wireless Sensor Networks
Security is the major area of concern in communication channel. Security is very crucial in wireless sensor networks which are deployed in remote environments. Adversary can disrupt the communication within multi hop sensor networks by launching the attack. The common attacks which disrupt the communication of nodes are packet dropping, packet modification, packet fake routing, badmouthing atta...
متن کاملA Fuzzy Decision-Making Methodology for Risk Response Planning in Large-Scale Projects
Risk response planning is one of the main phases in the project risk management and has major impacts on the success of a large-scale project. Since projects are unique, and risks are dynamic through the life of the projects, it is necessary to formulate responses of the important risks. The conventional approaches tend to be less effective in dealing with the impreciseness of risk response p...
متن کاملKernel Local Fuzzy Clustering Margin Fisher Discriminant Method Faced on Fault Diagnosis
In order to better identify the fault of rotor system,one new method based on local fuzzy clustering margin fisher discriminant (LFCMFD) was proposed. For each point on manifold, the farthest point in local neighborhood and the nearest point outside local neighborhood usually constituted the local margin. LFCMFD introduced fuzzy cluster analysis algorithm, eliminated the influence of pseudo-mar...
متن کامل