PREDICTORS OF DIRECT COSTS IN ALZHEIMERS DISEASE

职业发展计划和机构资源摘要。 职业发展计划包括流行病学和卫生服务研究方面的五年重点教学。通过课程工作和个人辅导相结合，增强研究技能将在三个核心领域发展：卫生经济学和成本分析;统计建模的先进方法;以及介绍行政和国家调查数据的使用。 这些新技能将为痴呆症的独立研究生涯提供基础，卫生服务研究侧重于卫生保健利用率，成本和对自然发生人群干预措施的反应的决定因素。 密歇根州立大学在流行病学和卫生服务研究方面有着悠久的传统。 这种环境将提供频繁的互动机会，将具有类似研究兴趣的同事进行职业指导，正在进行的研究的批评，以及新的研究思路的发展。研究计划摘要。 阿尔茨海默病（Alzheimer's disease，AD）给患者、家庭和社会带来了巨大的经济负担。 目前正在开发新的AD治疗方法，以改变疾病的临床特征，但这些临床特征与直接医疗保健费用之间的关系尚不清楚。 我们将建立一个数据集，用于调查AD人群直接医疗保健费用的决定因素。 卫生服务使用行为模式将为数据收集和分析提供一个概念框架。 我们将通过重复的调查测量和与HCFA管理数据的联系，在一个由300名可能的AD患者组成的具有良好特征的队列中测量四年内的直接医疗保健费用。 我们将研究直接成本和六个“需要”因素之间的关系，假设是AD的直接医疗保健成本的重要决定因素。 这六个“需要”因素（即，AD患者的认知、精神和运动障碍的严重程度、共病医学状况、“依赖性”和护理者“感知负担”）将使用经验证的临床测量工具进行测量。 我们将使用多元回归分析技术，包括路径分析，结构方程模型和一般纵向模型来测试研究假设。 该项目将开发AD患者直接医疗保健成本的解释性和预测性统计模型，有助于关注和评估这种疾病的未来干预措施。

Career Development Plan and Institutional Resources Summary. The career development plan consists of five years of focused instruction in epidemiology and health services research. Through a combination of course work and individual tutorials, enhanced research skills will be developed in three core areas: health economics and cost analysis; advanced methods in statistical modeling; and introduction to the use of administrative and national survey data. These new skills will provide a foundation for an independent research career in dementia, health services research focused on determinants of health care utilization, costs and response to interventions in naturally occurring populations. Michigan State University has a strong tradition in both epidemiology and health serviceS research. This environment will provide frequent opportunities for interaction will colleagues with similar research interests for career guidance, critique of ongoing research, and development of new research ideas. Research Plan Summary. Alzheimer's disease (AD) imposes a great economic burden on patients, their families and society. New treatments for AD are being developed to modify clinical features of the disease, but the relationships between these clinical features and direct health care costs are not known. We will build a data set for investigating determinants of direct heath care costs in an AD population. The Behavioral Model of Health Services Use will provide a conceptual framework for data collection and analysis. We will measure direct health care costs in a well-characterized cohort of 300 probable AD patients over four years, by repeated survey measurements and linkage to HCFA administrative data. We will examine the relationships between direct costs and six "need" factors hypothesized to be important determinants of direct, health care costs in AD. These six "need" factors (ie, severity of the AD patient's cognitive, psychiatric, and motor impairment, co-morbid medical conditions, "dependency" and caregiver "perceived burden") will be measured with validated, clinical measurement instruments. We will use multiple regression analysis techniques, including path analysis, structural equation modeling, and general longitudinal modeling to test study hypotheses. This project will develop explanatory and predictive statistical models of direct health care costs in AD patients, that can help focus and evaluate future interventions in this disease.