BASIS 32--A CONSUMER MENTAL HEALTH OUTCOME MEASURE

拟议项目的广泛目标是加强衡量不同心理健康和药物滥用服务接受者的客户自我报告的临床和功能性心理健康状况。遏制心理健康和药物滥用服务费用上升的努力突出了评估临床结果的重要性。32项行为与症状识别量表（BASIS-32）是专为结果评估而设计的工具。然而，它是在一个地点的同质住院病人群体中发展起来的。要在国家一级发挥作用，一项措施需要在更多不同的样本和多个地点进行测试和验证。建议的研究将使用定性（可读性分析，认知测试程序）和定量方法（心理测量测试和项目反应理论建模）来完善，现场测试和验证改进的和更广泛适用的结果评估工具。此外，这项研究将制定和实施一种方法，以制定经风险调整的结果基准。这项工作将通过在美国四个人口普查区的每个住院病人（N=6,000）的精神健康和药物滥用服务提供地点抽样治疗的客户来完成。数据分析将包括评估站点和客户的代表性、数据质量、因素结构/子量表组成、可靠性、有效性和随时间变化的敏感性。评估仪器的心理测量特性将基于经典测试理论和项目反应理论的概念。分层模型将用于创建风险调整后的结果基准。

The broad objective of the proposed project is to enhance the measurement of client self-reported clinical and functional mental health status among diverse mental health and substance abuse service recipients. Efforts to stem the rising costs of mental health and substance abuse services have highlighted the importance of assessing clinical outcomes. The 32-item Behavior and Symptom Identification Scale (BASIS-32) is an instrument designed specifically for outcome assessment. However, it was developed on a homogeneous population of inpatients at one site. To be useful on a national level a measure needs to be tested and validated on more diverse samples and at multiple sites. The proposed research will use qualitative (readability analysis, cognitive testing procedures), and quantitative methods (psychometric testing and Item Response Theory modeling), to refine,, field test and validate an improved and more widely applicable outcome assessment instrument. In addition, the research will develop and implement a methodology for creating risk-adjusted outcome benchmarks. This work will be accomplished by sampling treated clients from inpatient (N=6,000) mental health and substance abuse service delivery sites in each of the four US census regions. Data analysis will include assessment of representativeness of sites and clients, data quality, factor structure/ subscale composition, reliability, validity, and sensitivity to change over time. Assessment of the instrument's psychometric properties will be based on concepts from both Classical Test Theory and Item Response theory. Hierarchical modeling will be used to create risk-adjusted outcome benchmarks.