Synthesis of Trials to Prevent Suicide Risk Behavior in Sexual and Gender Minorities

性和性别少数（SGM）青年的自杀企图年发生率为29%， 是其他年轻人的4.6倍。我们建议测试预防计划是否针对 直接自杀，或一系列增加自杀风险的因素（内化症状、酗酒、吸毒 使用，外化，欺凌和人际暴力）降低青少年自杀风险，以及是否 SGM青年或多或少地受益于这些方案。与广泛的预防试验网络合作 研究人员，我们已获准从21个随机预防试验中获得个人水平的数据 参与人数100，851人。所有试验都包括自杀风险的直接或替代措施， 关于SGM的情况尽管所有试验都测试并报告了干预对特定行为的影响， 健康结果，很少有人测试对SGM青年的具体影响，因为这些青年构成了相对 每项研究中的小部分样本。我们和其他人的工作表明， 跨试验将为检验关于自杀风险影响的几个关键假设提供更大的力量， SGM青年是否从这些干预措施中获得不同的好处。我们将使用在我们的 先前的工作是将这些数据集联合收割机，并利用SGM的两项测量开发研究的结果 和非SGM青年自杀风险，以指导协调结果措施的高级心理测量分析 并建立一个自杀风险的通用指数。我们将采用综合数据分析（IDA）方法， 解释缺失数据以检验三组假设。 目标1。测试不同类型的干预是否会降低自杀念头的风险， 在一般人群中年轻人的行为。我们假设预防性项目可以降低 不管主要目标是什么都有自杀倾向我们将使用IDA在21项试验（n= 100，851）中检验这一假设。 目标二。测试SGM青年的预防效果是否不同。鉴于其经常面临的独特风险， SGM青少年，我们假设预防计划将降低SGM青少年自杀的风险， 与其他年轻人相比，效果较差。我们将对5,232名SGM和95,619名非SGM青年进行适度测试。 目标3。测试对SGM青年的预防效果是否因干预类型而异。我们 假设针对欺凌和内化症状的干预措施将更有效地减少 SGM青年自杀风险与针对酒精和药物使用以及外部化的干预措施的比较 行为我们将使用IDA来比较干预措施对试验目标行为的预防效果（例如， 欺凌）和个体因素（例如，女性vs男性SGM青年）。 这项研究的发现将填补我们对当前预防计划的理解的主要空白。 影响自杀风险的一般和在SGM和非SGM青年，并将提供指导， 针对这一高危人群的下一代预防试验。

Sexual and gender minority (SGM) youth have annual prevalence rates of suicide attempts of 29%, up to 4.6 times higher than that of other youth. We propose to test whether prevention programs that target suicide directly, or a range of factors that increase risk for suicide (internalizing symptoms, alcohol abuse, drug use, externalizing, bullying and interpersonal violence) reduce suicide risk for youth in general, and whether SGM youth benefit more or less from such programs. Working with a broad network of prevention trials researchers, we have received permission to obtain individual-level data from 21 randomized prevention trials involving 100,851 participants. All trials include direct or surrogate measures of suicide risk, and information about SGM status. Although all trials have tested and reported intervention impact on specific behavioral health outcomes, few have tested specific impact on SGM youth, given that such youth make up a relatively small proportion of the samples in each study. Our work and that of others indicates that combining data across trials will provide far greater power for testing several key hypotheses about impact on suicide risk, and whether SGM youth receive differential benefit from these interventions. We will use methods developed in our prior work to combine these datasets, and utilize findings from two measurement development studies of SGM and non-SGM youth suicide risk to guide advanced psychometric analyses for harmonizing outcome measures and developing a common index of suicide risk. We will apply integrative data analysis (IDA) methods that account for missing data to test three sets of hypotheses. Aim 1. Test whether different types of interventions decrease risk for suicidal thought and behavior among youth in the general population. We hypothesize that preventive programs reduce risk for suicidality regardless of primary target. We will use IDA to test this hypothesis on 21 trials (n=100,851). Aim 2. Test whether preventive effects differ for SGM youth. Given the unique risks often faced by SGM youth, we hypothesize that prevention programs will reduce risk for suicidality for SGM youth, but will be less effective compared with other youth. We will test moderation on 5,232 SGM and 95,619 non-SGM youth. Aim 3. Test whether preventive effects on SGM youth vary by type of intervention. We hypothesize that interventions targeting bullying and internalizing symptoms will be more effective in reducing suicide risk for SGM youth compared with interventions targeting alcohol and drug use and externalizing behavior. We will use IDA to compare the preventive effects of interventions across trial target behaviors (e.g., bullying) and individual factors (e.g., female vs male SGM youth). Findings from this study will fill major gaps in our understanding of how current prevention programs impact suicide risk both generally and in SGM and non-SGM youth, and will provide guidance in directing the next generation of prevention trials targeted towards this at-risk population.