Lewis G, Wessely S
General Practice Research Unit, Institute of Psychiatry, London.
Br J Psychiatry. 1990 Dec;157:860-4. doi: 10.1192/bjp.157.6.860.
The specificity and sensitivity of the HAD, 12-item GHQ and CIS were calculated by comparing the scores of dermatological patients on these tests with a criterion measure of disorder. Since psychiatry, along with many other branches of medicine, does not have an error-free criterion, it was assumed that the criterion was an underlying latent construct which was measured by all of the tests and could be derived by factor analysis from the scores on them. No differences were found between the two questionnaires (HAD and GHQ) in their ability to detect cases of minor psychiatric disorder although they were somewhat less reliable than the CIS.
通过将皮肤病患者在这些测试中的得分与疾病的标准测量值进行比较,计算出医院焦虑抑郁量表(HAD)、12项一般健康问卷(GHQ)和临床访谈量表(CIS)的特异性和敏感性。由于精神病学与许多其他医学分支一样,没有无误差的标准,因此假定该标准是一种潜在的隐性结构,所有测试都对其进行了测量,并且可以通过对这些测试的得分进行因子分析得出。尽管这两份问卷(HAD和GHQ)在检测轻度精神障碍病例的能力上没有差异,但它们的可靠性略低于CIS。