Association pour le développement de la neuropsychologie appliquée (ADNA),25000 Besançon, France.
Clinical Methodology Center, University Hospital of Besançon, 25000Besançon, France; UMR6249 CNRS, Chrono-environnement Laboratory, University of Bourgogne Franche-Comté, 25000Besançon, France.
Rev Neurol (Paris). 2018 Jan-Feb;174(1-2):44-55. doi: 10.1016/j.neurol.2017.05.010. Epub 2017 Jun 30.
Cognitive evaluation of young subjects is now widely carried out for non-traumatic diseases such as multiple sclerosis, HIV, or sleep disorders. This evaluation requires normative data based on healthy adult samples. However, most clinicians use a set of tests that were normed in an isolated manner from different samples using different cutoff criteria. Thus, the score of an individual may be considered either normal or impaired according to the norms used. It is well established that healthy adults obtained low-test scores when a battery of tests is administered. Thus, the knowledge of low base rates is required so as to minimize false diagnosis of cognitive impairment. The aim of this study was twofold (1) to provide normative data for RAPID-II battery in healthy adults, and (2) estimate the proportion of healthy adults having low scores across this battery.
Norms for the 44 test scores of the RAPID-II test battery were developed using the overall sample of 335 individuals based on three categories of age (20 to 29, 30 to 39, and 40 to 49 years) and two educational levels: Baccalaureate or higher educational degree (high educational level), lower than baccalaureate (low educational level). The 5th, 25th, 50th, and 75th percentiles were calculated from the six age and education subsamples and used to define norms. The frequency of low scores on the RAPID-II battery was calculated by simultaneously examining the performance of 33 primary scores. A low score was defined as less than or equal to the 5th percentile drawn from the six age and education normative subsamples. In addition, the percentages of low scores were also determined when all possible combinations of two-test scores across the RAPID-II were considered in the overall normative sample.
Our data showed that 59.4% subjects of the normative sample obtained at least one or more low score. With more than 9 test scores, this percentage was equal to 0% in the normative sample. Among all combinations of two-test scores, 96% had a false positive rate<2%.
Low scores are very common in young healthy subjects and are more obvious when simultaneously analyzing test scores across a battery of tests and are thus not necessarily indicative of cognitive impairment. The combinations of two-test scores can be a useful tool to improve the interpretation of low scores.
目前，对于多发性硬化症、HIV 或睡眠障碍等非创伤性疾病，人们广泛地对年轻患者进行认知评估。这种评估需要基于健康成年人样本的规范数据。然而，大多数临床医生使用的是一组单独从不同样本中制定的测试，使用不同的截断标准。因此，根据使用的规范，个体的分数可能被认为是正常的或受损的。众所周知，当一组测试被实施时，健康成年人的测试分数较低。因此，需要了解低基率的知识，以尽量减少认知障碍的误诊。本研究的目的有两个方面：(1)为 RAPID-II 电池在健康成年人中的使用提供规范数据；(2)估计整个电池中分数较低的健康成年人的比例。
根据三个年龄类别（20 至 29 岁、30 至 39 岁和 40 至 49 岁）和两个教育水平（学士学位或更高教育程度[高教育水平]，低于学士学位[低教育水平]），对 RAPID-II 测试电池的 44 个测试分数的规范数据进行了开发。使用来自六个年龄和教育子样本的中位数、第 25 百分位数、第 50 百分位数和第 75 百分位数来定义规范。通过同时检查 33 个主要分数的表现来计算 RAPID-II 电池的低分数频率。将小于或等于从六个年龄和教育规范子样本中得出的第 5 百分位数定义为低分数。此外，当在整个规范样本中同时考虑 RAPID-II 所有可能的两个测试分数组合时，也确定了低分数的百分比。
我们的数据显示，规范样本中 59.4%的受试者至少有一个或多个低分数。在规范样本中，当有超过 9 个测试分数时，这个百分比等于 0%。在所有两个测试分数的组合中，96%的假阳性率<2%。
低分数在年轻健康受试者中非常常见，当同时分析电池测试中的测试分数时更为明显，因此不一定表明认知障碍。两个测试分数的组合可以成为提高低分数解释的有用工具。
