Hazard Assessment and Contaminated Sites - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 538

12.4.3 Types of Epidemiological Study - An Overview

Broadly speaking, epidemiological activity can be either “descriptive” (reporting
and describing the distribution of exposure and effect) or “analytical” (designed
to analyse and understand the degree of association between exposure and effect).
Descriptive studies include case reports, case series and cross-sectional surveys.
Cross-sectional surveys measure exposure and effect in an individual at the same
point in time and thus are unable to support causal inference.
In practical terms in environmental epidemiology there are four main categories
of analytical study (Moolgavkar et al. 1999 ):

• cohort (longitudinal) studies;

• case-control studies;

• cross-sectional studies;

• ecological studies (including a special subgroup known as time-series studies).

Cohort, cross-sectional and case control studies differ from ecological studies
in that information on exposure and disease is available on an individual basis.
With ecological studies this information is only available on a group basis, so the
community or region is the unit of analysis.
In case-control studies, exposure and other attributes of cases of the disease under
investigation are compared with those from a suitable control or comparison group
of persons unaffected by the disease, and analysed to yield effect estimates. The
selection of appropriate controls to avoid bias is a significant challenge with case-
control studies. They are relatively inexpensive, ideal for studying rare diseases and
useful for investigating multiple, different exposures (Gregg 1996 ).
Cross-sectional studies measure prevalence of disease and measure exposure and
effect at the same time. They are relatively easy and economical to conduct and
are particularly useful for measuring fixed characteristics of individuals such as
socioeconomic status (Beaglehole et al. 1993 ).
Cohort studies follow cohorts or groups of individuals, defined in terms of their
exposures, over time to see if there are differences in the development of new cases
of the disease of interest (or other health outcome) between the groups with and
without exposure. Such studies can be carried out by either reviewing past records
(retrospective) or by tracking people into the future (prospective cohort). The essen-
tial feature of these longitudinal studies is that for each individual prior exposure
information can be related to subsequent disease experience (Breslow and Day
1987 ).
Ecological studies involve the investigation of a group of people such as those
living within a geographical area such as a region or state. For example, place and
time of residence may be used to create surrogate measures of the real exposure of
interest (Elliott et al. 1992 ). Rates of disease and average exposure levels to a par-
ticular contaminant are determined independently, and on a group basis. This may
give rise to spurious apparent correlation, called the ecological fallacy. Because
it is not ascertained whether individuals who have been exposed to the contami-
nant are the same individuals who developed the disease, statements about causal