Dr. Nelson and Dr. Kansagara are with the Department of Medical Informatics and Clinical Epidemiology and Dr. Denneson, Dr. O'Neil, and Dr. Teo are with the Department of Psychiatry, all in the School of Medicine at Oregon Health and Science University, Portland. Dr. Denneson, Dr. O'Neil, Dr. Kansagara, and Dr. Teo are also with the Center to Improve Veteran Involvement in Care, Portland Department of Veterans Affairs Medical Center, Portland, Oregon, where Ms. Low is affiliated. Mr. Bauer is with the Department of Psychology, University of Southern Mississippi, Hattiesburg.
Psychiatr Serv. 2017 Oct 1;68(10):1003-1015. doi: 10.1176/appi.ps.201600384. Epub 2017 Jun 15.
Suicide rates in veteran and military populations in the United States are high. This article reviews studies of the accuracy of methods to identify individuals at increased risk of suicide and the effectiveness and adverse effects of health care interventions relevant to U.S. veteran and military populations in reducing suicide and suicide attempts.
Trials, observational studies, and systematic reviews relevant to U.S. veterans and military personnel were identified in searches of MEDLINE, PsycINFO, SocINDEX, and Cochrane databases (January 1, 2008, to September 11, 2015), on Web sites, and in reference lists. Investigators extracted and confirmed data and dual-rated risk of bias for included studies.
Nineteen studies evaluated accuracy of risk assessment methods, including models using retrospective electronic records data and clinician- or patient-rated instruments. Most methods demonstrated sensitivity ≥80% or area-under-the-curve values ≥.70 in single studies, including two studies based on electronic records of veterans and military personnel, but specificity varied. Suicide rates were reduced in six of eight observational studies of population-level interventions. Only two of ten trials of individual-level psychotherapy reported statistically significant differences between treatment and usual care.
Risk assessment methods have been shown to be sensitive predictors of suicide and suicide attempts, but the frequency of false positives limits their clinical utility. Research to refine these methods and examine clinical applications is needed. Studies of suicide prevention interventions are inconclusive; trials of population-level interventions and promising therapies are required to support their clinical use.
美国退伍军人和军人中的自杀率很高。本文回顾了评估方法识别自杀风险增加个体的准确性的研究，以及与美国退伍军人和军人相关的降低自杀和自杀企图的卫生保健干预措施的有效性和不良影响。
在 MEDLINE、PsycINFO、SocINDEX 和 Cochrane 数据库（2008 年 1 月 1 日至 2015 年 9 月 11 日）、网站和参考文献列表中搜索与美国退伍军人和军事人员相关的试验、观察性研究和系统评价。调查人员提取并确认了纳入研究的数据，并对偏倚风险进行了双重评价。
19 项研究评估了风险评估方法的准确性，包括使用回顾性电子记录数据和临床医生或患者评定工具的模型。大多数方法在单一研究中显示出≥80%的敏感性或曲线下面积值≥.70，包括两项基于退伍军人和军事人员电子记录的研究，但特异性不同。在八项人群干预的观察性研究中有六项报告了自杀率降低。仅有两项个体心理治疗的试验报告了治疗与常规护理之间的统计学显著差异。
风险评估方法已被证明是自杀和自杀企图的敏感预测因子，但假阳性的频率限制了其临床应用。需要研究来改进这些方法并检查临床应用。自杀预防干预的研究尚无定论；需要进行人群干预和有前途的治疗方法的试验，以支持其临床应用。
