Assessing the Development of Drug Markets Using Bayesian Space-Time Models

描述(由申请人提供)：拟议的研究将使用档案数据来测试17年来(1990-2006年)整个加州毒品市场和相关问题的增长和蔓延的统计模型。我们将研究与非法毒品有关的市场和执法活动的广泛框架，但重点是研究甲基苯丙胺市场(这一市场在此期间已遍及整个加州)。这项研究将评估社区和邻里风险因素(如贫困)在多大程度上加速毒品市场和相关问题(如吸毒过量、暴力)的发展。贝叶斯时空疾病模型将应用于不同级别的地理分辨率(例如，城市和邮政编码地区)。该项目将回答三个具体的研究问题：1)有证据表明毒品市场活动随时间的空间扩散；2)毒品市场的增长率是当地人口“脆弱”的函数吗？3)当地毒品市场活动指标与经常与毒品市场有关的社会问题之间的空间和时间关系是什么？为了回答这些问题，该项目将把来自内部专有数据库--加州索引位置数据库(CILD)的数据与表明毒品市场、执法活动和相关问题的数据结合起来。伴随这三个具体研究问题的是三个方法论步骤：1)定义和应用具有空间相关性的分层广义线性模型。2)定义和应用药物流行病学的变系数模型。3)定义和应用网络位置分布的空间分析方法。这项工作的短期目标是提供必要的数据和方法，以评估随着时间的推移，在广泛不同的生态环境中毒品市场活动的发展。长期目标是建立影响美国各州社区毒品市场动态演变的社会条件。更好地了解使毒品市场得以发展的更广泛的社会人口条件，将有助于制定有效的全州毒品预防战略。

DESCRIPTION (provided by applicant): The proposed research will use archival data to test statistical models of the growth and spread of drug markets and related problems across California over 17 years (1990-2006). We will examine the broad frame of market and enforcement activities related to illegal drugs, but emphasize studies of the methamphetamine market (a market that has spread across California during this time). This research will assess the degree to which community and neighborhood risk factors (e.g., impoverishment) accelerate the growth of drug markets and related problems (e.g., overdoses, violence). Bayesian space-time disease models will be applied at different levels of geographic resolution (e.g., cities and zip code areas). The project will answer three specific research questions: 1) Is there evidence for the spatial diffusion of drug market activities over time? 2) Is the rate of growth of drug markets a function of local population "frailties?" 3) What are the spatial and temporal relationships between local indicators of drug market activities and social problems often associated with drug markets? In order to answer these questions, the project will couple data from an in-house proprietary database, the California Index Locations Database (CILD), to data that indicates drug markets, enforcement activities and related problems. Accompanying the three specific research questions are three methodological steps: 1) Define and apply hierarchical generalized linear models with spatial correlation. 2) Define and apply variable coefficient models for drug epidemiology. 3) Define and apply methods for the spatial analysis of location distributions on a network. The short-term goal of this work is to provide the data and methods necessary to assess the development of drug market activities across broadly different ecological settings over time. The long-term goal is to establish the societal conditions that affect the dynamic evolution of drug markets across communities within states in the US. Greater understanding of the broader sociodemographic conditions that enable the development of drug markets will enable the formulation of effective state-wide drug prevention strategies.