Meijne W, van Neerbos K, Aufdemkampe G, van der Wurff P
Institute for Physical Therapy, Bussum, The Netherlands.
J Manipulative Physiol Ther. 1999 Jan;22(1):4-9. doi: 10.1016/s0161-4754(99)70098-9.
The purpose of this study was to determine the intraexaminer and interexaminer reliability of the Gillet test.
In a test-retest study the incidence of asymmetric or symmetric sacroiliac joint motion was investigated with the Gillet test.
Forty-one male subjects volunteered for this study (mean age, 23 +/- 2.24 years). Thirty-eight subjects were examined in the first test procedure, and at least 4 days later 37 subjects were examined again. The subjects were subdivided into symptomatic and asymptomatic groups on the basis of certain criteria.
To obtain the intraexaminer and interexaminer reliability values, Cohen's kappa, the percentage agreement, bias-adjusted kappa, and prevalence-adjusted bias-adjusted kappa were used. The mean Cohen's kappa did not exceed the value of 0.081. Only the percentage agreement of the symptomatic group did exceed the minimum level of 80%. When kappa was positive, the prevalence-adjusted bias-adjusted kappa was markedly higher than kappa; when kappa was negative, the prevalence-adjusted bias-adjusted kappa was only slightly higher than kappa. Only small differences were found between kappa and bias-adjusted kappa.
The Gillet test, as performed in this study, does not appear to be reliable.
本研究旨在确定吉列试验的检查者内和检查者间可靠性。
在一项重测研究中，采用吉列试验调查骶髂关节不对称或对称运动的发生率。
41名男性受试者自愿参与本研究（平均年龄23±2.24岁）。38名受试者在首次测试程序中接受检查，至少4天后，37名受试者再次接受检查。根据某些标准将受试者分为有症状组和无症状组。
为获得检查者内和检查者间可靠性值，使用了科恩kappa系数、一致性百分比、偏差调整kappa系数和患病率调整偏差调整kappa系数。科恩kappa系数的平均值未超过0.081。只有有症状组的一致性百分比超过了80%的最低水平。当kappa为正时，患病率调整偏差调整kappa系数明显高于kappa系数；当kappa为负时，患病率调整偏差调整kappa系数仅略高于kappa系数。kappa系数和偏差调整kappa系数之间仅发现微小差异。
本研究中所进行的吉列试验似乎不可靠。
