[结核菌素试验判读中的观察者间一致性]

[Interobserver agreement in reading tuberculin tests].

作者信息

Guerra M, García de Blas F, Sanz M T, Morales M C, Pose B

机构信息

Medicina Familiar y Comunitaria, Centro de Salud Mendiguchía Carriche, Madrid.

出版信息

Enferm Infecc Microbiol Clin. 1993 Dec;11(10):531-5.

PMID:8142502

Abstract

BACKGROUND

The aim of the study was to evaluate the variability between four observers in the reading of the tuberculin test (TT).

METHODS

The study was carried out in the Centro de Salud Mendiguchía Carriche (Madrid) 72 subjects. By direct inspection, four observers read tuberculin determination (skin induration) independently in each subject. To assess variability, the kappa index (K) was calculated.

RESULTS

On considering the variable as a dichotomous one, with a cut-off point of 5 mm was 0.88; and of 10 mm, 0.78. Upon considering the index as qualitative in four categories the kappa index in 0 mm was 0.89; in 1-4 mm, 0.24; in 5-9 mm, 0.40 and greater than 9 mm, 0.79. In eight of the subjects studied, the variability observed in reading the test had conditioned further different clinico-epidemiologic attitudes.

CONCLUSIONS

The degree of concordance when reading TT is significantly lower at the 1-4 mm and 5-9 mm intervals.

摘要

背景

本研究旨在评估四名观察者在读取结核菌素试验（TT）结果时的变异性。

方法

该研究在马德里的门迪古希亚卡里切卫生中心对72名受试者进行。通过直接观察，四名观察者独立读取每名受试者的结核菌素测定结果（皮肤硬结情况）。为评估变异性，计算了kappa指数（K）。

结果

将变量视为二分变量时，截断点为5毫米时kappa指数为0.88；截断点为10毫米时，kappa指数为0.78。将指数视为四类定性指标时，0毫米时kappa指数为0.89；1 - 4毫米时为0.24；5 - 9毫米时为0.40；大于9毫米时为0.79。在八名研究受试者中，读取试验结果时观察到的变异性导致了进一步不同的临床流行病学态度。