Læknablaðið - 15.04.1980, Blaðsíða 17

LÆKNABLADID
79
anate) TDI, to cotton dust or to formaldehyde
rests upon symptoms and a decrease in
function such as forced expiratory volume
in 1 second (FEV]0) during exposure. In
contrast the first observable response to silica
or asbestos may be the development of an X-
ray pattern. Obviously even the knowledge
gained from a worker or two, observed
because of this complaints is valuable in
design of the survey procedure. Biochemical
immunological and other measurements may
be useful.
The design must be mindful of industrial
realities — that is minimally disruptive. It is
generally possible to do a complete job by a
team requiring only about two hours per
worker. Also, invasive methods or uncomfor-
table steps must be minimal. If these are
essential they are reserved for a third step
which is in-depth investigation in a hospital of
selected persons identified, that is screened,
by the field survey.
A statistical model and plan is as essential
as a smooth flow of workers through the
examination and testing stations. It is wasteful
to study more persons than are needed to
fulfill a statistical model. However, this is only
one consideration, often a cross-sectional stu-
dy including all workers is more acceptable to
management and labor organizations and a
limited or stratified sample gratification can
be done subsequently.
The most important consideration is hav-
ing a control or comparison group against
which to analyze the group exposed to the
toxic or suspected agent. Otherwise, because
of the limited responses possible by the lung
or other organs, effects may be attributed
wrongly to an agent or exposure. This is most
important when the searching for the princi-
pal responsible agent within a complex expo-
sure or for instance among nurses in a
hospital, farmers or handicrafters. The unex-
posed control group should be matched to the
exposure group for age, sex, cigarette smo-
king and frequency of asthmatics but may
require other matching as well. Pair matching
may be of help if groups are small and may
make differences stand out most clearly.
Essentially prospective studies use these
same measurements, repeated at intervals in
time. The methods are usually selected and
niade specific so as to economize effort.
Comparisons are then possible within measure-
ments of individuals which increase the preci-
sion of the work. Provocative testing is simply
application of this strategy within a brief
period of which there is deliberate exposure.
In a brief survey such as this, an encyclope-
dic description of clinical epidemiology is
neither appropriate or possible. However, it is
worthwhile to emphasize that by using these
simple techniques in a constrained and organ-
ized fashion aided by small or medium size
computers for data handling, elegant studies
of human response are being done every day.
Questions concerning not only causation but
intervention, effects of co-factors and influence
of preexisting disease can be answered. It is
important to design the procedure to answer
the most specific question or questions pos-
sible. To do this a preliminary feasibility study
of far broader scope but limited numbers of
workers may be essential. Then the population
study can be crisp, precise and economical of
team and worker’s effort.
Open records, with sharing of finding
within countries and between countries has
been very useful and should be encouraged.
Careful reading of the suspected agents re-
cord internationally is crucial to designing a
study and anticipating results. Otherwise, as
the philosopher George Santayana said,
»Those who do not read history are doomed
to repeat it«.
Once an agent’s effect has been identified,
a decision concerning abatement strategy
must be made. If the effect measured is minor
further surveillance such as a prospective
study may be begun to see whether effects
over time are more serious.
If health effects are serious thus important
a strategy must be developed to reduce the
levels in the air of workers. Many rawmaterial
and product handling factors can affect air
concentration and provide places to control it,
total material being processed, speed of
throughput, sources of air, type of cleaning,
wet vs dry, compresed air vs vacuuming, etc.
A major place to insert control is in the air
conditioning, heating or cooling system where
provision for filtration may exist. High volume
media filtration may be useful but enclosure of
dusty operations with seperate air masses
from human operators is generally more
effective. Clearly it requires medical industrial
hygiene input and monitoring added to engi-
neering expertise to devise and test the