Variability, Predictive Value, and Uses of the Beryllium Blood Lymphocyte Proliferation Test (BLPT): Preliminary Analysis of the Ongoing Workforce Survey

Abstract

The beryllium blood lymphocyte proliferation test (BLPT) is used as a medical surveillance tool for assessment of persons at risk for developing clinical and subclinical chronic beryllium disease (CBD). Three laboratories, coded "A," "B," and "C," were used to perform the BLPTs, which involved two simultaneous tests on a split specimen. The intra-laboratory agreement analysis compared the first and the second test performed by the same laboratory. The interlaboratory agreement analysis compared test results performed by different laboratories on the same sample. The level of agreement was expressed as a kappa statistic. The positive predictive value (PPV) analysis compared BLPT results against the results of a bronchoscopy used to detect CBD. The data included 5483 records representing 3081 samples from 1510 persons. Intra-laboratory agreement was fair to moderate, with kappa values between 0.3 and 0.6. Inter-laboratory agreement was moderate (kappa = 0.5) for Labs A and B, moderate (kappa = 0.6) for Labs B and C, and poor (kappa = 0.2) for Labs A and C. A single unconfirmed abnormal test had a PPV for CBD of 39 percent, a confirmed (based on subsequent testing) abnormal test had a PPV for CBD of 45 percent, and a first-time double abnormal test had a PPV for CBD of 49 percent. Substantial inter- and intralaboratory disagreement exists between and within major laboratories that conduct this test.