Ferguson A C, Murray A B
CMAJ. 1986 Jun 15;134(12):1365-8.
The predictive value and post-test probability of disease were compared for skin prick tests and radioallergosorbent tests (RASTs) in 168 children suspected of clinical allergy to dogs and cats. The skin tests included negative and positive (histamine) controls. The results of RASTs with the same allergen extracts were expressed in relation to the results with allergen-specific pooled reference serum. All the tests were performed blind. The predictive values of positive test results were comparable and low (53% to 76%), whereas the predictive values of negative test results were comparable and high (88% to 95%). The post-test probability of clinical allergy to dog or cat allergen, based on the prevalence rates in the referral population (15.1% and 22.5% respectively), increased to between 46% and 67% for positive test results and decreased to between 4% and 8% for negative results, which suggests that the primary role of skin prick tests and RASTs is in eliminating the diagnosis of clinical allergy.
对168名疑似对犬猫临床过敏的儿童进行了皮肤点刺试验和放射变应原吸附试验（RAST），并比较了两种试验对疾病的预测价值和验后概率。皮肤试验包括阴性和阳性（组胺）对照。使用相同变应原提取物的RAST结果与变应原特异性混合参考血清的结果相关联。所有试验均采用盲法进行。阳性试验结果的预测价值相当且较低（53%至76%），而阴性试验结果的预测价值相当且较高（88%至95%）。根据转诊人群中的患病率（分别为15.1%和22.5%），对犬或猫变应原临床过敏的验后概率，阳性试验结果增加到46%至67%之间，阴性结果降低到4%至8%之间，这表明皮肤点刺试验和RAST的主要作用是排除临床过敏的诊断。
