PREDICTORS OF ADOLESCENT ECSTASY USE IN THE NATIONAL SURVEY OF PARENTS AND YOUTH

描述（由申请人提供）：在过去的十年中，摇头丸的使用在美国和其他国家的流行率有所增加。尤其是在青少年和年轻人中。预测青少年未来使用摇头丸的潜在决定因素尚未得到充分研究。具体来说，缺乏纵向研究来控制时间序列，并估计哪些因素预测使用摇头丸的风险增加的程度。为了解决这些差距，本研究的目的是：1)测试青少年使用摇头丸的患病率和发生率是否随时间的人口和地理变量而变化；2)检验青少年既往的异常行为、既往的高感觉寻求分数和既往的其他药物使用是否能预测随后的摇头丸使用；3)检验先前低水平的父母监控和高水平的同伴使用摇头丸是否能预测青少年随后的摇头丸使用；4)测试青少年对摇头丸使用的态度和信念的变化，并测试接触禁毒媒体运动是否可能影响青少年对摇头丸使用的态度。来自全国父母和青少年调查（NSPY）的纵向数据，这是一项代表美国青少年人口（9-18岁）的流行病学研究，将用于解决这些目标。在初步的探索性数据分析之后，我们将通过横断面和纵向逻辑回归模型来分析数据，以解决每个具体目标。首先，我们将估计摇头丸使用的流行率和发生率与人口统计学（年龄、性别、种族/民族）的关联，然后，我们将使用交替逻辑回归来检验是否有证据表明全国各地区青少年使用摇头丸。使用纵向回归边际模型，我们将检验先前的异常行为、高感觉寻求分数和高水平的先前其他药物使用是否预测青少年随后的摇头丸使用，以及检验较低水平的父母监控和高水平的同伴摇头丸使用是否预测青少年随后的摇头丸使用。逻辑回归模型将测试青少年对使用摇头丸的态度和信念的变化，并测试接触禁毒媒体运动是否影响青少年对使用摇头丸的态度。最后，我们相信我们的研究将为更好地理解摇头丸使用的潜在决定因素带来公共卫生贡献。

DESCRIPTION (provided by applicant): During the past decade the prevalence of ecstasy use has increased in the U.S. and in other countries. particularly among adolescents and young adults. Potential determinants that predict future ecstasy use in adolescence are understudied. Specifically, there is a lack of longitudinal research designed to control temporal sequencing and estimate the extent to which factors predict increased risk of ecstasy use. To address these gaps, the proposed study aims to: 1) test if the prevalence and incidence of ecstasy use in adolescence vary in relation to demographic and geographic variables over time; 2) test whether previous deviant behaviors, previous high sensation-seeking scores and prior other drug use predict subsequent ecstasy use among adolescents; 3) test whether previous low levels of parent monitoring and high levels of peer ecstasy use predict subsequent ecstasy use among adolescents; 4) test for changes in attitudes and beliefs towards ecstasy use among adolescents and test whether exposure to an anti-drug media campaign might have influenced adolescents' attitudes towards ecstasy use. Longitudinal data from the National Survey of Parents and Youth (NSPY), an epidemiologic study representative of the adolescent U.S. population (9-18 yeas old) will be used to address these aims. After initial exploratory data analyses, we will analyze data through cross-sectional and longitudinal logistic regression models to address each of the specific aims. First, we will estimate associations of prevalence and incidence of ecstasy use with demographics (age, gender, race/ethnicity), then, we will use alternating logistic regression to test whether there is evidence of clustering of adolescent ecstasy use by region of the country. Using longitudinal regression marginal models, we will test whether previous deviant behaviors, high sensation-seeking scores and high levels of prior other drug use predict subsequent ecstasy use among adolescents, as well as examine if lower levels of parent monitoring and high levels of peer ecstasy use predict subsequent ecstasy use among adolescents. Logistic regression models will test for changes in attitudes and beliefs towards ecstasy use among adolescents and to test whether exposure to an anti-drug media campaign influenced adolescents' attitudes towards ecstasy use. Finally, we believe that our study will bring public health contributions to better understand potential determinants of ecstasy use.