术后认知功能障碍测试量表的重测信度分析

摘要

Objective To investigate the test-retest reliability of the scale of postoperative cognitive dysfunction. Methods Fifty healthy elderly people were randomly selected and tested by the scale of postoperative cognitive dysfunction, involving visual verbal learning (VVL), concept shifting task (CST), stroop colour word test (SCWT) and symbol digit substitution coding (SDSC). After the first test (T1), they were tested three days (T2) and seven days (T3) later. The test results were recorded each time and the test-retest reliability was calculated. Results Compared with T1, the scores of VVL and SDSC on T2 and T3 were increased, but the scores of SCWT were decreased, and there were no significant changes in CST. The split half reliability of two parts of the total scale was greater than 0.5. The test-retest reliability of VVL, SDSC and SCWT was high (r=0.564-0.951,P<0.01). The test-retest reliability of the time part in CST was high (r=0.725-0.779,P<0.01), but was low in the number of errors in CST (r=0.000-0.316,P>0.05). Conclusion The measurement scale of postoperative cognitive dysfunction has some learning effects, and the test-retest reliability is high, so it can be used to evaluate postoperative cognitive dysfunction.%目的 探讨术后认知功能障碍测试量表的重测信度.方法 随机选取50例健康老年人,采用术后认知功能障碍测试量表进行神经心理学测试,包括视觉语言测试(visual verbal learning,VVL)、概念转移测试(concept shifting task,CST)、STROOP色词测试(stroop colour word test,SCWT)和符号数字转化测试(symbol digit substitution coding,SDSC),然后在第1次测试(T1)后第3天(T2)和第7天(T3)分别进行第2、3次测试,记录3次测试结果 并计算重测信度.结果与T1时比较,T2、T3时VVL和SDSC得分升高,SCWT得分降低,CST无明显变化.总量表半信度分析各部分信度α均>0.5.VVL、SDSC和SCWT的重测信度较高(r=0.564~0.951,P<0.01),CST的时间重测信度较高(r=0.725~0.779,P<0.01),但错误个数重测信度较差(r=0.000~0.316,P>0.05).结论 术后认知功能障碍测试量表具有一定的学习效应,且重测信度较高,可以用来评价术后认知功能障碍.