Facebook: A screening tool to identify alcohol use among female college freshmen

描述（由申请人提供）：具体目的：本试点项目的主要目标是测试通过分析公开Facebook （FB）资料来识别18至20岁女大学新生饮酒情况的预测有效性。背景和理由：大学生饮酒与显著的发病率和死亡率相关。对大学生饮酒者的研究使用了来自诊所的样本；然而，在线社区对这群人来说越来越重要。大约四分之三的威斯康辛大学麦迪逊分校的本科生报告拥有社交网站（SNS）的个人资料，70%的威斯康辛大学麦迪逊分校的大学新生在他们公开的FB个人资料中提到了饮酒情况。我们的试点研究已经确定了女大学新生饮酒与感知压力之间的关键联系，我们打算在未来的干预中针对这一关系。在未来的干预工作中，确定一种利用FB识别大学生酒精使用的创新方法具有巨大的潜力。初步研究：PI之前的工作包括4项描述性研究，使用社交媒体识别大龄青少年和大学生从事酒精使用等危险行为，与大学新生进行关于酒精使用诱发因素的焦点小组，以及一项使用社交媒体减少大龄青少年在线风险行为的试点干预研究。联合研究所的研究小组已经发表了14篇关于酒精筛查的研究文章。方法：本研究将通过facebook资料对170名威斯康星大学麦迪逊分校（UW Madison）新生进行识别：其中85名在facebook资料中有饮酒记录，85名没有。这些受试者将通过FB的邮件或电话联系，并邀请他们参加面对面的面试。本次访谈将使用经过验证的筛查工具来评估酒精使用情况，以及包括压力、精神疾病和其他风险行为在内的协变量。我们将通过28天的随访来比较FB上显示的酒精参考资料与访谈中报告的酒精使用情况，以评估FB档案编码的预测有效性。标准将是在过去一个月的任何采访中报告的酒精使用情况。意义：FB档案提供了一个机会来识别参考酒精使用的女大学新生，并调查可能导致酒精使用的协变量，如压力。针对酒精使用及其与压力的关系，包括减压在内的干预措施以及简短的医生干预措施可能对女大学生有吸引力。在进行此类干预之前，必须建立显示的酒精参考和访谈报告的酒精使用之间的联系。这将是第一个识别使用FB的大学生并进行面对面访谈来验证FB显示的健康信息的研究。这些信息可以提供一种创新的方法，在学生生活的“真实”和“虚拟”世界中与他们见面。

DESCRIPTION (provided by applicant): Specific aim: The primary goal of this pilot project is to test the predictive validity of analyzing public Facebook (FB) profiles to identify alcohol use in female college freshmen ages 18 to 20 years old. Background and justification: Alcohol use among college students is associated with significant morbidity and mortality. Studies of college student drinkers have been conducted using samples derived from clinics; however, online communities are of increasing importance to this population. Approximately three-fourths of University of Wisconsin Madison undergraduates report ownership of a social networking web site (SNS) profile, and 70% of UW Madison college freshmen reference alcohol use on their publicly available FB profiles. Our pilot studies have identified key links between alcohol use and perceived stress among female college freshmen, a relationship we intend to target in a future intervention. Identifying an innovative method using FB to identify alcohol use among college students has tremendous potential for future intervention work. Preliminary Studies: The PI's previous work includes 4 descriptive studies using SNSs to identify older adolescents and college students engaging in risk behaviors such as alcohol use, focus groups with college freshmen regarding precipitators of alcohol use and a pilot intervention study using SNSs to reduce online risk behaviors in older adolescents. The Co-I's research group has published 14 research articles focused on alcohol screening. Methods: The proposed study will identify 170 female UW Madison freshmen via FB profiles: 85 with references to alcohol use on their FB profiles and 85 without these references present. These subjects will be contacted by FB email or phone and invited to participate in a face-to-face interview. This interview will use validated screening tools to assess alcohol use, as well as covariates including stress, mental illness and other risk behaviors. We will compare displayed alcohol references on FB to alcohol use reported during interviews via a 28 day time line followback to assess the predictive validity of FB profile coding. The criteria standard will be any interview-reported alcohol use in the past month. Significance: FB profiles offer an opportunity to identify female college freshmen who reference alcohol use, and investigate covariates that may contribute to alcohol use such as stress. An intervention encompassing stress reduction as well as brief physician intervention regarding alcohol use and its relationship to stress may be compelling to female college students. Before such an intervention can take place, the link between displayed alcohol references and interview-reported alcohol use must be established. This would be the first study to identify college students using FB and conduct in-person interview to validate the displayed health information from FB. This information could provide an innovative method to meet students in both the "real" and "virtual" world in which they live. PUBLIC HEALTH RELEVANCE: This study will pilot test the use of a social networking Web site (SNS) to identify female college freshmen who display references to alcohol use on their Facebook (FB) web profiles. The results of this study have the potential to establish links between self-displayed health information on SNS profiles and self-reported health information via interview. Further, the results of this evaluation have the potential to impact public health practice by demonstrating how to expand identification of college student alcohol use to the Internet using public Web sites, a strategy that could be easily translated to colleges around the country.