A proposal for evaluation of exposure data.

There is a growing need for transparency concerning ways in which existing exposure data are weighted for their relative value and quality. Currently, this evaluation is largely subjective and is dependent on the quality of the judgement of the individual assessor or expert group. In this paper some general guidelines are presented for a quality assessment procedure. Such a predetermined procedure potentially enhances the consistency among different assessors and assessments and facilitates harmonization of assessment procedures. The guidelines are presented in the context of a decision tree with four decision rules for data quality, i.e. 'availability of occupational hygiene information', 'variability and precision issues', 'internal validity' and 'external validity'. These methodological issues are considered to be the most important aspects of data quality and will be discussed in this paper. The decision tree eventually results in three quality classes, i.e. exposure data providing sufficient information, supplementary information and data which should be excluded from the exposure assessment process. The guidelines should not be used in a rigid manner but have to be interpreted in the light of the particular circumstances and purposes of the assessment.