What are the Purposes of Air Pollution and Health Research?
The first and obvious answer to this question is that the researcher should
try to determine whether or not there is a causal relationship between
an environmental exposure and an adverse effect on health. This is much
more difficult than most lay people believe. Several criteria
must be fulfilled. You may wish to consult a page on asthma
and air pollutants as an example. However, even if one establishes
that a particular specified hazard can be responsible for some unwanted
health effect, this level of information on its own is of limited practical
use. What is really needed is a quantitative measure of the likelihood
of an adverse health effect (i.e. the risk) for a given level of exposure.
Another necessity is to determine the exposure-response relationships
for mixed exposures, and the interaction between these and other risk factors.
Having established causation, and exposure-response relationships (no mean
task), it may be important to translate these into health economic terms.
In other words one might need to respond to questions such as:
What degree of environmental change, and at what cost would
produce what extent of health benefit?.
This information is very important for standard setting. Any interventions,
especially costly ones, should be properly evaluated so as to ensure that
the resource is being well utilised.
Some Difficulties with Research
Difficulties pervade study design, implementation and interpretation. For
example in relation to the issue of "causation", three important difficulties
need to be considered:
Bias. Essentially this results from not comparing like with like - for
example it may be that people who respond and participate in an epidemiologic
study, might be more likely to be concerned about ill health effects, and
perhaps suffer from them, than individuals who decline to participate.
Confounding. Rather than present an epidemiologic definition of this,
consider the following scenario:
Assume a study shows that in a particular city, on days
with a high level of pollution, there are more admissions to hospital of
elderly people with fractures of the hip. Should one conclude therefore
that in some way or another pollution causes hips to fracture? The alternative,
more likely explanation is that in certain bitterly cold weather conditions,
some pollutant concentrations are more likely to be increased because of
"temperature inversion". However, those same weather conditions are also
likely to be associated with slippery iced ground, and folk are therefore
more likely to slip and injure themselves.
Chance. If a study is small, it may result in a false negative effect because
the numbers of subjects are too small, to show an increase in risk, for
example in relatively rare conditions like leukaemia. On the other hand,
if a large number of statistical tests are carried out, especially in a
relatively indiscriminate fashion, it is likely that by chance a number
of them will appear to be "statistically significant" - a false positive
interpretation may then result. |
