The effect of non-asbestos fibres and sample fibre density on the agreement between laboratories evaluating membrane filter samples of airborne dust.

During an exchange of mounted membrane filter samples between six British laboratories, some of the participants counted as asbestos all fibres which fitted a geometric definition, whilst others evaluated only those fibres subjectively considered to be asbestos. This was found to result in different counting levels depending both upon the type of sample and the method used to identify asbestos fibres. Once counting criteria had been agreed, however, no significant difference in counting level was detected between three of the laboratories. The exchange also highlighted the difficulty in obtaining good agreement between different laboratories when counting samples which have a fibre density on the filter of less than 20 fibres mm−2.