reference method for Method 7400. Therefore, the consensus mean is the “true” value and the interlaboratory results effectively define the method accuracy. Under the heading “Evaluation of Method, B. Interlaboratory comparability,” Method 7400 provides a means of calculating the confidence limits on a single analysis result (Equations 3 and 4). From Equation 3, the interlaboratory variability at the LOD is such that the upper 95% confidence limit on a measured value is 300% greater than (or 4 times) the measured value. Using the upper confidence limit, the equation in Section 21 in Method 7400 can be used to estimate the sampling volume.
With the appropriate values inserted, the equation becomes
Solving this equation for sampling volume gives 1080 L. This is the minimum volume that will give a result allowing a single sample to indicate compliance with the 0.01 fiber/mL limit with 95% confidence. It requires that the sample give a result less than or equal to the LOD or 5.5 fibers per 100 fields. A higher fiber count may still indicate that the concentration meets the target level, but not with the same level of confidence. This is likely to be a conservative estimate of concentration and additionally ensure compliance with the standard because the fiber concentration is low and, as indicated above, low fiber loadings are usually overestimated. However, the background concentration of non-fibrous dust on the filter also must be low to ensure that fibers are not obscured. f. Other techniques Since fiber counting by human analysts produces relatively high biases and variability, several researchers have attempted to develop automated counting systems. With the increases in computer power over the last 25 years, it has been tempting to assume that fiber counting is a solvable problem and significant efforts have been made to develop such a system. The most intensive effort to produce a fiber counting system was carried out by Manchester University in collaboration with the Health and Safety Executive in the UK [39]. The Manchester Asbestos Program (MAP) was able to give reasonably good agreement with human counters for certain types of samples. It was used as a reference analyst for the US and UK reference sample programs for several years. Eventually, the MAP was dropped as the reference because it was not sufficiently consistent for all types of samples.
3/15/03
150
NIOSH Manual of Analytical Methods
