Causal Inference for Mediation Models in Substance Abuse Prevention Research

药物滥用预防研究中中介模型的因果推断

• 批准号: 7779398
• 负责人: Donna Lynn Coffman
• 金额: $ 6.24万
• 依托单位: PENNSYLVANIA STATE UNIVERSITY, THE
• 依托单位国家: 美国
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-04-01 至 2012-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7779398
• 关键词: Address; Adolescent; Affect; Alcohol or Other Drugs use; Charge; Computer software; Confidence Intervals; Data; Data Analyses; Development; Diagnostic; Drug abuse; Educational workshop; Evaluation; Goals; Individual; Intervention; Intervention Studies; Leisure Activities; Life; Literature; Measures; Mediating; Mediation; Mediator of activation protein; Methodology; Methods; Modeling; Outcome; Population; Prevention; Prevention Research; Prevention program; Preventive Intervention; Process; Publications; Randomized; Reporting; Research; Research Personnel; Resistance; Risk Behaviors; Scientist; South Africa; Statistical Methods; Work; base; cost; design; effective intervention; innovation; interest; intervention effect; journal article; prevent; programs; public health relevance; simulation; skills; skills training; substance abuse prevention; substance use prevention; symposium

DESCRIPTION (provided by applicant): This R03 project proposes to examine the feasibility of new methods for drawing causal inferences in mediation models of substance use preventive interventions. Mediation occurs when an independent variable (e.g. preventive intervention) has an effect on the dependent variable (e.g. substance use) through a third variable (e.g. resistance skills). Traditional methods for assessing mediation allow for causal inferences between the independent variable and the mediator variable but not between the mediator and the dependent variable because individuals are not randomly assigned to levels of the mediator. Although the new methods, which are based on potential outcomes, have been developed in the statistical literature, they have not been implemented in applied prevention research. Using simulation studies, the new methods will be compared with the traditional methods to determine whether they are an improvement over the traditional methods. The new methods will then be applied to a secondary data analysis of a randomized substance use prevention program for adolescents. The new methods will be implemented in software programs and macros, which will be made available free of charge to applied researchers. Methods and findings will ultimately inform the design of more cost-effective interventions for substance use prevention by allowing applied researchers to draw more valid causal inferences. PUBLIC HEALTH RELEVANCE: This project will (a) generate new statistical methods and software to enable applied researchers to draw more valid causal inferences about mediation processes, and (b) apply these methods in secondary analysis of a randomized substance use prevention intervention for adolescents. Methods and findings will ultimately inform the design of more cost-effective interventions for substance use.

描述（由申请人提供）：本R03项目旨在研究在物质使用预防性干预措施的调解模型中进行因果推断的新方法的可行性。当一个自变量（如预防性干预）通过第三个变量（如抵抗技能）对因变量（如药物使用）产生影响时，就会发生调解。传统的中介评估方法允许自变量和中介变量之间的因果关系推断，但不允许中介变量和因变量之间的因果关系推断，因为个体不是随机分配到中介变量的水平。虽然新的方法，这是基于潜在的结果，已经在统计文献中开发，他们还没有在应用预防研究中实施。通过模拟研究，将新方法与传统方法进行比较，以确定它们是否优于传统方法。然后，新方法将应用于青少年随机物质使用预防计划的二级数据分析。新方法将在软件程序和宏中实现，这些程序和宏将免费提供给应用研究人员。方法和研究结果将最终通知设计更具成本效益的干预措施，预防物质使用，使应用研究人员得出更有效的因果关系推断。公共卫生相关性：该项目将（a）产生新的统计方法和软件，使应用研究人员能够对调解过程进行更有效的因果推断，（B）将这些方法应用于青少年随机药物使用预防干预的二次分析。方法和调查结果将最终为设计更具成本效益的药物使用干预措施提供信息。