Department of Psychiatry and Behavioral Sciences, University of Washington School of Medicine, Seattle, Washington.
School of Community and Global Health, Claremont Graduate University, Claremont, California.
J Stud Alcohol Drugs. 2018 Jul;79(4):658-663. doi: 10.15288/jsad.2018.79.658.
The Rutgers Alcohol Problem Index (RAPI) has been used extensively as a measure of alcohol-related problems experienced by adolescents and young adults. The present study aimed to comprehensively examine the psychometric profile and criterion-related validity of an 18-item RAPI adapted to measure negative consequences resulting from alcohol and other drug use in an at-risk adolescent population.
Categorical confirmatory factor analysis and the graded response model were used for evaluations of the latent factor structure, item properties, test information, and item invariance across gender, age groups (middle vs. late adolescence), and drug use profiles (readily available drugs vs. other illicit drugs), as well as computation of the item response theory scale scores for each of 617 alternative high school students (44% female).
A reduced set of 16 RAPI items provided plausible evidence of unidimensionality and good measurement precision at a relatively wide range of the latent trait continuum. All but two items were invariant across the drug use profiles. The other illicit drug group had higher levels of the latent substance-related problems than the readily available drug group. Criterion-related validity was supported using a measure of past-year polysubstance use.
The RAPI can be effectively modified to measure negative consequences resulting from alcohol and other drug use and reflects a sufficiently wide range of the latent continuum of problem use.
罗格斯酒精问题指数（RAPI）已被广泛用于衡量青少年和年轻人经历的与酒精相关的问题。本研究旨在全面检查经过改编的 18 项 RAPI 的心理测量学特征和效标关联效度，以衡量处于危险中的青少年群体因酒精和其他药物使用而产生的负面后果。
类别验证性因子分析和渐进反应模型用于评估潜在因子结构、项目属性、测试信息以及性别、年龄组（中期与晚期青春期）和药物使用模式（易得药物与其他非法药物）之间的项目不变性，并计算 617 名替代高中生（44%为女性）的项目反应理论量表得分。
一套简化的 16 项 RAPI 提供了单一维度的合理证据，并在潜在特质连续体的较宽范围内提供了良好的测量精度。除了两个项目之外，所有项目在药物使用模式上都是不变的。其他非法药物组的潜在物质相关问题水平高于易得药物组。使用过去一年多物质使用的衡量标准支持效标关联效度。
RAPI 可以有效地修改，以衡量因酒精和其他药物使用而产生的负面后果，并反映出足够广泛的潜在问题使用连续体。
