Using 'Big Data' and Precision Medicine to Assess and Manage Suicide Risk in U.S. Veterans

使用“大数据”和精准医学评估和管理美国退伍军人的自杀风险

• 批准号: 9842275
• 负责人: JEAN C. BECKHAM
• 金额: --
• 依托单位: DURHAM VA MEDICAL CENTER
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-01-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9842275
• 关键词: Address; Adoption; Age; Algorithms; Area; Big Data; Biological; Biology; Candidate Disease Gene; Clinical; Code; Data; Diagnostic; Enrollment; Event; Failure; Family; Feeling suicidal; Gene Expression; Genetic; Genetic Markers; Genetic Research; Genetic Risk; Genetic study; Genotype; Goals; Grief reaction; Heritability; Intervention; Lead; Mental disorders; Methylation; Military Personnel; Mission; Not Hispanic or Latino; Pain; Pathway interactions; Phenotype; Population; Prevention approach; Research; Sample Size; Self-Direction; Single Nucleotide Polymorphism; Suicide; Suicide attempt; Suicide prevention; Time; Twin Studies; United States Department of Veterans Affairs; Variant; Veterans; Violence; Work; active duty; administrative database; base; cohort; economic cost; genetic variant; genome wide association study; genome-wide; high risk; improved; innovation; novel; phenotypic data; precision medicine; programs; psychogenetics; reducing suicide; risk variant; screening; sex; statistics; suicidal behavior; suicidal morbidity; suicidal risk

Reducing suicide and suicidal behavior (i.e., self-directed violence) is a top priority for the Department of Veterans Affairs. Recent statistics indicate that, on average, 20 Veterans die by suicide in the U.S. each day. Family, adoption, and twin studies indicate that genetic factors account for 30-50% of the heritability in suicidal behavior. Numerous candidate gene and genome wide association studies (GWAS) have been conducted to identify variants associated with suicidal behavior; however, a major limitation of all prior genetic studies in this area of research is low statistical power due to small sample sizes and the infrequency with which suicidal behavior occurs. Another significant limitation concerns the failure of most prior genetic studies of suicidal behavior to include Veterans, despite the fact that Veterans are at significantly increased risk for suicide and suicidal behavior. The proposed research will address these limitations by leveraging the genetic and phenotypic data available through the Million Veteran Program (MVP) and other key administrative databases to perform the largest and most well-powered GWAS of suicidal behavior to date. The potential impact of identifying novel genetic markers that reliably predict suicidal behavior would be enormous. It could fundamentally shift current understanding of the biology of suicide, lead to new and improved approaches to suicide prevention for Veterans and civilians alike, and significantly improve VA's ongoing efforts to identify and intervene with high risk Veterans before they engage in suicidal behavior. Our long-term goal is to develop effective screening and intervention strategies to reduce the occurrence of suicide and suicidal behavior. The overall objective of this application is to discover novel genetic variants that increase Veterans' risk for suicidal behavior. The rationale for the proposed research is that identification of genetic variants that are reliably associated with suicidal behavior could lead to the discovery of novel, clinically-meaningful biological pathways that could, in turn, lead to new and improved suicide prevention approaches for Veterans. We will accomplish our overall objective by pursuing the following specific aims: In Aim 1, we will refine the phenotypes that we will use to define cases of suicidal behavior within MVP. In Aim 2, we will use GWAS to identify novel genetic variants associated with suicide attempts and suicidal ideation among Veterans in MVP. In Aim 3, we will replicate significant findings obtained from the MVP cohort in the Mid-Atlantic MIRECC and Army STARRS Cohorts. In Aim 4, we will explore whether the genetic findings obtained from MVP can be used to improve VA's ability to identify Veterans at risk for suicidal behavior.

减少自杀和自杀行为（即，自我导向的暴力）是该部的首要任务。 退伍军人事务部最近的统计数据显示，平均每天有20名退伍军人在美国自杀。 家庭、收养和双胞胎研究表明，自杀的遗传因素占30- 50 行为已经进行了许多候选基因和全基因组关联研究（GWAS）， 确定与自杀行为相关的变异;然而，所有先前的遗传研究在这方面的主要局限性是， 研究领域是低统计功率由于小样本量和频率与自杀 行为发生。另一个重要的局限性涉及大多数先前的自杀基因研究的失败， 行为，包括退伍军人，尽管事实上，退伍军人是在显着增加自杀的风险， 自杀行为 拟议的研究将通过利用现有的遗传和表型数据来解决这些限制 通过百万退伍军人计划（MVP）和其他重要的管理数据库， 最有力的自杀行为GWAS识别新的遗传基因的潜在影响 可靠地预测自杀行为的标记将是巨大的。它可以从根本上改变电流 了解自杀的生物学，导致新的和改进的方法来预防自杀， 退伍军人和平民一样，并显着改善VA的持续努力，以确定和干预高 在他们从事自杀行为之前， 我们的长远目标是发展有效的甄别和介入策略，以减少 自杀和自杀行为。本申请的总体目标是发现新的遗传变体， 增加退伍军人自杀行为的风险。拟议研究的理由是， 与自杀行为可靠相关的遗传变异可能导致新的发现， 具有临床意义的生物学途径，反过来可以导致新的和改进的自杀预防 为退伍军人服务。我们将通过实现以下具体目标来实现我们的总体目标： 在目标1中，我们将细化表型，用于定义MVP中的自杀行为病例。在Aim中 2，我们将使用GWAS来识别与自杀企图和自杀意念相关的新型遗传变异 在MVP的老兵中。在目标3中，我们将复制从MVP队列中获得的重要发现， 中大西洋MIRECC和陆军STARRS队列。在目标4中，我们将探讨基因发现是否 从MVP获得的数据可以用来提高VA识别有自杀行为风险的退伍军人的能力。

项目成果

Deep sequential neural network models improve stratification of suicide attempt risk among US veterans.

深度序列神经网络模型改善了美国退伍军人自杀未遂风险的分层。

• DOI: 10.1093/jamia/ocad167
• 发表时间: 2023
• 期刊: Journal of the American Medical Informatics Association : JAMIA
• 作者: Martinez,Carianne;Levin,Drew;Jones,Jessica;Finley,PatrickD;McMahon,Benjamin;Dhaubhadel,Sayera;Cohn,Judith;MillionVeteranProgram;MVPSuicideExemplarWorkgroup;Oslin,DavidW;Kimbrel,NathanA;Beckham,JeanC
• 通讯作者: Beckham,JeanC