Page 409 - Industrial Ventilation Design Guidebook

364 CHAPTER 6 TARGET LEVELS

of environmental monitoring in existing workplaces may be compared to en-
sure that, as far as the current state of knowledge permits, control is adequate
to protect health. They may also be used for design purposes, to ensure that
new plants and process are engineered in such a way that exposures can be
controlled at levels that will not damage health. They should not be used as a
basis for assessing the acceptability of nonoccupational exposure, for compar-
ing the toxicity of one substance with that of another, in the evaluation or
control of community air pollution nuisances, or as proof or disproof of an
existing disease or physical condition.

6.2.2 Setting OELs
The next topic to address is the process of setting OELs. The rationale for set-
ting OELs has no basis in absolute information, and the procedure to be fol-
lowed may differ from country to country and from substance to substance.
Scientific criteria for health protection should be used in combination with
considerations of their technical and economic feasibility in a dynamic process
in which the development of scientific knowledge underlies rapid changes en-
tailing the need to periodically review the data.
In general, it requires a multidisciplinary approach that takes data on tox-
icological and other relevant properties of chemicals developed in the fields of
industrial hygiene, chemistry, toxicology, epidemiology, engineering, occupa-
tional medicine, etc., so that interrelated data from all sources combines to
give the most rational input upon which to base OELs, even though adequate
documentation is not available for the majority of chemical agents.
Scientific information for the process of establishing OELs may come
from human or animal data obtained using different methods, from studies of
acute, subacute, and chronic toxicity through various routes of entry. Human
data, which is usually the best source, is not easily available, and frequently it
is incomplete or inadequate due to poor characterization of exposure and
clear dose-response relationships. Human data falls into one of the following
categories:

» Epidemiological and clinical studies of workers and other long-term
case-control studies
» Experimental studies with volunteers, especially for situations involving
the acute effects of an identified adverse agent
• Literature from industrial experience, especially individual case reports
showing relationship between exposure to a chemical agent and specific
adverse effects
Data coming from laboratory studies with animals offer good characteriza-
tion of exposure, adequate use of controls, extensive pathological investigations,
and the potential to give clear indications of dose-response relationships. How-
ever, extrapolating this data to human exposure is not straightforward and may
lead to mistakes. Therefore, studies of this type should be used primarily to com-
plement other studies, in order to check certain aspects or to guide the study.
The crucial objective is to identify what effects can be produced by ex-
posure to chemical agents and to select which effects should be considered