Walsh Kate, Short Nicole, Ji Yin Yao, An Xin Ming, Witkemper Kristen D, Lechner Megan, Bell Kathy, Black Jenny, Buchanan Jennie, Ho Jeffrey, Reed Gordon, Platt Melissa, Riviello Ralph, Martin Sandra L, Liberzon Israel, Rauch Sheila A M, Bollen Kenneth, McLean Samuel A
Department of Psychology, University of Wisconsin-Madison, Madison, WI, USA; Department of Gender & Women's Studies, University of Wisconsin-Madison, Madison, WI, USA.
Institute for Trauma Recovery, University of North Carolina at Chapel Hill, North Carolina, USA; Department of Anesthesiology, University of North Carolina at Chapel Hill, North Carolina, USA; Department of Psychology, University of Nevada, Las Vegas, NV, USA.
J Psychiatr Res. 2024 Jun;174:54-61. doi: 10.1016/j.jpsychires.2024.04.013. Epub 2024 Apr 9.
This study aims to develop and validate a brief bedside tool to screen women survivors presenting for emergency care following sexual assault for risk of persistent elevated posttraumatic stress symptoms (PTSS) six months after assault. Participants were 547 cisgender women sexual assault survivors who presented to one of 13 sexual assault nurse examiner (SANE) programs for medical care within 72 h of a sexual assault and completed surveys one week and six months after the assault. Data on 222 potential predictors from the SANE visit and the week one survey spanning seven broadly-defined risk factor domains were candidates for inclusion in the screening tool. Elevated PTSS six months after assault were defined as PCL-5 > 38. LASSO logistic regression was applied to 20 randomly selected bootstrapped samples to evaluate variable importance. Logistic regression models comprised of the top 10, 20, and 30 candidate predictors were tested in 10 cross-validation samples drawn from 80% of the sample. The resulting instrument was validated in the remaining 20% of the sample. AUC of the finalized eight-item prediction tool was 0.77 and the Brier Score was 0.19. A raw score of 41 on the screener corresponds to a 70% risk of elevated PTSS at 6 months. Similar performance was observed for elevated PTSS at one year. This brief, eight-item risk stratification tool consists of easy-to-collect information and, if validated, may be useful for clinical trial enrichment and/or patient screening.
本研究旨在开发并验证一种简短的床边工具，用于筛查遭受性侵犯后前来接受急诊护理的女性幸存者，以评估其在遭受性侵犯六个月后创伤后应激症状（PTSS）持续升高的风险。参与者为547名顺性别女性性侵犯幸存者，她们在遭受性侵犯后72小时内前往13个性侵犯护士检查（SANE）项目之一接受医疗护理，并在袭击后一周和六个月完成调查。来自SANE就诊和一周调查的222个潜在预测因素的数据，涵盖七个广义定义的风险因素领域，是筛选工具纳入的候选因素。袭击后六个月PTSS升高定义为PCL-5>38。将LASSO逻辑回归应用于20个随机选择的自抽样样本，以评估变量的重要性。由前10、20和30个候选预测因素组成的逻辑回归模型在从80%的样本中抽取的10个交叉验证样本中进行测试。最终的工具在其余20%的样本中进行验证。最终的八项预测工具的AUC为0.77，Brier评分为0.19。筛查器上的原始分数为41对应于6个月时PTSS升高风险为70%。在一年时PTSS升高也观察到类似的表现。这种简短的八项风险分层工具由易于收集的信息组成，如果得到验证，可能对临床试验富集和/或患者筛查有用。
