Blasco-Fontecilla Hilario, Bella-Fernández Marcos
Instituto de Transferencia e Investigación (ITEI), Universidad Internacional de La Rioja (UNIR), 26006 Logroño, Spain; Department of Psychiatry, Emooti, 28010 Madrid, Spain; Center of Biomedical Network Research on Mental Health (CIBERSAM), 28029 Madrid, Spain.
Faculty of Health, Universidad Internacional de La Empresa (UNIE), 28015 Madrid, Spain; Department of Psychology, Universidad Pontificia de Comillas, 28108 Alcobendas, Spain.
Actas Esp Psiquiatr. 2025 May;53(3):546-556. doi: 10.62641/aep.v53i3.1883.
Attention Deficit/Hyperactivity Disorder (ADHD) is the most common neurodevelopmental disorder. The affectation of executive functions is stressed in the most recent research on ADHD, and many tests are used to assess it in ADHD, but they are usually time- and effort-consuming.
From a database of a total of 222 children with ADHD, 59 of them suffering executive dysfunction, we took the most widely used tests for executive functions in ADHD (Behavior Rating Inventory of Executive Function (BRIEF), Swanson, Noland, and Pelham (SNAP)-IV, and Conners' Parent Rating Scale (CPRS-R)) and applied several methods of test shortening: Item-total correlations from the Classical Test Theory, factor analysis and their subsequent factor loadings, elastic nets, and the Graded Response Model from the Item Response Theory models. Using the parameters or indicators provided by each of these methods, we selected the most discriminative items to develop a brief screening tool.
Our results show that different selection methods select different items. More importantly, we found that the shortened tests obtained this way are in general capable of discriminating between patients with and without ADHD. More precisely, all the shortened tests show high sensitivity, but relatively low specificity.
Shortened tests can be used for screening purposes without having to administer full test versions.
注意力缺陷/多动障碍(ADHD)是最常见的神经发育障碍。ADHD的最新研究强调了执行功能的影响,并且在ADHD中使用了许多测试来评估它,但这些测试通常既耗时又费力。
从总共222名患有ADHD的儿童数据库中,其中59名患有执行功能障碍,我们采用了ADHD中最广泛使用的执行功能测试(执行功能行为评定量表(BRIEF)、斯旺森、诺兰德和佩勒姆(SNAP)-IV以及康纳斯父母评定量表(CPRS-R)),并应用了几种测试缩短方法:经典测试理论中的项目总分相关性、因子分析及其后续因子载荷、弹性网络以及项目反应理论模型中的等级反应模型。使用这些方法中的每一种提供的参数或指标,我们选择了最具区分性的项目来开发一种简短的筛查工具。
我们的结果表明,不同的选择方法会选择不同的项目。更重要的是,我们发现通过这种方式获得的缩短测试总体上能够区分患有和未患有ADHD的患者。更确切地说,所有缩短测试都显示出高敏感性,但特异性相对较低。
缩短测试可用于筛查目的,而无需进行完整测试版本。