Hartmaier S L, Sloane P D, Guess H A, Koch G G
Department of Epidemiology, School of Public Health, University of North Carolina at Chapel Hill.
J Am Geriatr Soc. 1994 Nov;42(11):1173-9. doi: 10.1111/j.1532-5415.1994.tb06984.x.
We report the development and validation of an MDS-based cognitive index, the MDS Cognition Scale (MDS-COGS), by evaluating it against two popular dementia rating scales, the Global Deterioration Scale (GDS) and the Mini-Mental State Examination (MMSE).
A Cross-sectional study.
Two hundred nursing home residents.
Each study participant was assessed on the GDS and the MMSE by trained medical students. At the same time, but independent of the medical students, a geriatric nurse completed the Minimum Data Set (MDS) instrument on each participant.
The Cognitive Performance Scale (CPS), a categorical measure of cognition that uses MDS items, was compared with the GDS in 133 subjects, using a split-sample technique. The GDS was found to be more appropriate as a 4-stage than a 7-stage scale, with GDS stages 1-4 acting as a single stage. The CPS showed very poor percent agreement with GDS stages 5 and 7 (50% or less) and, therefore, was revised by adding other MDS predictors. The new instrument, designated MDS-COGS, is a 0-10 point scale generated from eight MDS cognitive items. The MDS-COGS was then validated against the 4-stage GDS and MMSE in the remaining 67 nursing home residents. Chance-corrected agreement (kappa) between the MDS-COGS and GDS in the 67 subjects was 0.80 (95% CI = 0.70, 0.88), and percent agreement with GDS stages 5 and 7 was 60% and 85%, respectively. The MDS-COGS was also demonstrated to be a valid measure of cognitive impairment as defined by the MMSE, with sensitivity, specificity, chance-corrected agreement (kappa), and area under the ROC curve, all above 0.80.
The MDS Cognition Scale, the MDS-COGS, provides a valid measure of the presence and severity of cognitive impairment in nursing home residents using items from the Minimum Data Set.
通过与两种常用的痴呆评定量表——总体衰退量表（GDS）和简易精神状态检查表（MMSE）进行对比评估，报告一种基于最低数据集（MDS）的认知指数——MDS认知量表（MDS - COGS）的开发与验证情况。
一项横断面研究。
200名养老院居民。
由经过培训的医学生对每位研究参与者进行GDS和MMSE评估。同时，由一名老年护理护士独立于医学生完成每位参与者的最低数据集（MDS）工具评估。
使用MDS项目的认知表现量表（CPS）这一认知分类测量方法，采用拆分样本技术，在133名受试者中与GDS进行了比较。结果发现，GDS作为一个4阶段量表比7阶段量表更合适，GDS的1 - 4阶段作为一个单一阶段。CPS与GDS的5期和7期的一致性百分比非常低（50%或更低），因此，通过添加其他MDS预测指标对其进行了修订。新工具MDS - COGS是一个由八个MDS认知项目生成的0至10分的量表。然后在其余67名养老院居民中，将MDS - COGS与4阶段GDS和MMSE进行了验证。67名受试者中，MDS - COGS与GDS之间的机遇校正一致性（kappa）为0.80（95%置信区间 = 0.70, 0.88），与GDS的5期和7期的一致性百分比分别为60%和85%。MDS - COGS也被证明是MMSE所定义的认知障碍的有效测量方法，其敏感性、特异性、机遇校正一致性（kappa）和ROC曲线下面积均高于0.80。
MDS认知量表（MDS - COGS）使用最低数据集的项目，为养老院居民认知障碍的存在和严重程度提供了一种有效的测量方法。
