Exposure-response relation for vibration-induced white finger: influence of different estimates of daily exposure...

It is analyzed whether differences in how the daily usage time is determined in a population exposed to hand-transmitted vibration are the cause of discrepancies between exposure-response models constructed for vibration-induced white finger (VWF). The predictions of the models are then used to evaluate the one in ISO 5349-1:2001.

Pooled analyses employing data from epidemiologic studies of vibration-exposed occupations deemed the most reliable, together with published ratios between daily tool or machine usage times determined by questionnaire or interview and measured values, are used to construct exposure-response relations. These models are compared to those in which the daily usage time is not compensated.

Of 17 data points, for 7 the usage time ratio is greater than 1, and for the rest the ratio is equal to 1. Adjusting daily vibration exposures, expressed as A(8)-values, by means of such ratios, results in reductions of varying magnitude. However, the exposure-response models show minor changes, and the scatter in the data points previously observed remains.

Methods for estimating 10% prevalence of VWF from the prevalence recorded in a population study are found to have more effect on the models than different methods for estimating the daily usage time. Yet, the spread and clustering in the data remain and point to issues in the constituents of the A(8)-value. The ISO relation is generally consistent with the lower 95-percentile confidence limit of the models and so provides a conservative exposure limit at least for A(8)-values greater than about 3 [Formula: see text].