Beach Scott R, Liu Pi-Ju, DeLiema Marguerite, Iris Madelyn, Howe Melissa J K, Conrad Kendon J
a University Center for Social and Urban Research , University of Pittsburgh , Pittsburgh , Pennsylvania , USA.
b Institute for Health and Aging , University of California, San Francisco , San Francisco , California , USA.
J Elder Abuse Negl. 2017 Aug-Oct;29(4):229-253. doi: 10.1080/08946566.2017.1338171. Epub 2017 Jun 7.
Improving the standardization and efficiency of adult protective services (APS) investigations is a top priority in APS practice. Using data from the Elder Abuse Decision Support System (EADSS), we developed short-form measures of four types of elder abuse: financial, emotional/psychological, physical, and neglect. The EADSS data set contains 948 elder abuse cases (age 60+) with yes/no abuse substantiation decisions for each abuse type following a 30-day investigation. Item sensitivity/specificity analyses were conducted on long-form items with the substantiation decision for each abuse type as the criterion. Validity was further tested using receiver-operator characteristic (ROC) curve analysis, correlation with long forms and internal consistency. The four resulting short-form measures, containing 36 of the 82 original items, have validity similar to the original long forms. These short forms can be used to standardize and increase efficiency of APS investigations, and may also offer researchers new options for brief elder abuse assessments.
提高成人保护服务(APS)调查的标准化和效率是APS实践中的首要任务。利用来自老年人虐待决策支持系统(EADSS)的数据,我们开发了四种老年人虐待类型的简短测量方法:财务、情感/心理、身体和忽视。EADSS数据集包含948起老年人虐待案件(年龄60岁及以上),在进行30天调查后,针对每种虐待类型都有虐待确认与否的决定。以每种虐待类型的确认决定为标准,对长篇项目进行了项目敏感性/特异性分析。使用接受者操作特征(ROC)曲线分析、与长篇的相关性和内部一致性进一步测试了有效性。由此产生的四种简短测量方法包含了82个原始项目中的36个,其有效性与原始长篇相似。这些简短形式可用于标准化和提高APS调查的效率,也可能为研究人员提供用于简短老年人虐待评估的新选项。