Environmental effects on health


Hierarchy of systems that influence population health.
Health emerges from a highly complex interaction
between factors intrinsic to the patient and his or her
environment. Many factors within the environment
influence health, including aspects of the physical
environment, biological environment (bacteria, viruses),
built environment and social environment, but these
also encompass more distant influences such as the
global ecosystem (Fig.). Environmental changes
affect many physiological systems and do not respect
boundaries between medical specialties. The specialty
of ‘public health’ in the UK is concerned with the
investigation and management of health in communities
and populations, but the principles apply in all
Exposure to infectious agents is a major environmental
determinant of health . This chapter describes the approach to
other common environmental factors that influence
The hierarchy of systems – from molecules to ecologies
When assessing a patient, a clinician subconsciously
considers many levels at which problems may be occurring,
including molecular, cellular, tissue, organ and
body systems. When the environment’s influence on
health is being considered, this ‘hierarchy of systems’
extends beyond the individual to include the family,
community, population and ecology. Box 5.1 shows an
example of the utility of this concept in describing


determinants of coronary heart disease operating at each
level of a hierarchy.
Interactions between people and their environment
The hierarchy of systems demonstrates that the clinician
should not focus too quickly on the disease process
without considering the context. Health is an emergent
quality of a complex interaction between many determinants,
including genetic inheritance, the physical
circumstances in which people live (e.g. housing, air
quality, working environment), the social environment
(e.g. levels of friendship, support and trust), personal
behaviour (smoking, diet, exercise), and access to money
and the other resources that give people control over
their lives. Health care is not the only determinant – and
is usually not the major determinant – of health status
in the population.
These systems do not operate in isolation in separate
communities. When one group responds to ill health
by manipulating its environment, the consequences may
be global. For example, an Afghan farmer who starts
growing opium for money in order to feed his children
influences the environment of a teenager in Europe; in
turn, drug misuse in Europe has fostered higher prevalence
of blood-borne infectious diseases such as human
immunodeficiency virus/acquired immunodeficiency
syndrome (HIV/AIDS); in turn, these have spilled out
into sexually transmitted disease. This process contributes
significantly to the tragedy of the epidemic of
The life course
The determinants of health operate over the whole
lifespan. Values and behaviours acquired during childhood
and adolescence have a profound influence on

educational outcomes, job prospects and risk of disease.
Attributes such as the ability to form empathetic
relationships or assess risk have a strong influence on
whether a young person takes up damaging behaviour
like smoking, risky sexual activity and drug misuse.
Influences on health can even operate before birth.
Individuals with low birth weight have been shown
to have a higher prevalence of conditions such as hypertension
and type 2 diabetes as young adults and of cardiovascular
disease in middle age. It has been suggested
that under-nutrition during middle to late gestation permanently
‘programmes’ cardiovascular and metabolic
This ‘life course’ perspective highlights the cumulative
effect on health of exposures to episodes of illness,
adverse environmental conditions and behaviours that
damage health. In this way, biological and social risk
factors at each stage of life link to form pathways to
disease and health.
Investigations in environmental health
Incidence and prevalence

The first task is to establish how common a problem is
within the population. This is expressed in two ways
• If the problem is a continuing condition (e.g.
enlarged spleen due to malaria), then prevalence is
the appropriate measurement and is calculated by
dividing the number of people with the condition at
a specified time by the number of people in the
population at risk at that time. Prevalence tends to
be higher if the problem is common (many new
cases) and/or if it is of longer duration.
• If the problem is an event that occurs at a clear
point in time (e.g. fever due to malaria), then
incidence is used. Incidence is a measure of the rate
at which new cases occur (e.g. confirmed pyrexia

with malaria parasites on a blood film) in the
population at risk during a defined period of time.
Variability by time,
person and place
The next task is to establish how the problem varies in
terms of time, person and place. The incidence may fluctuate
throughout the year; for example, malaria occurs
in the wet season but not the dry. Observation over
longer periods establishes whether a problem is becoming
more or less common: malaria may re-emerge due
to drug resistance. The next questions are, who are the
victims? Are males or females more commonly affected?
What is the age pattern? What are the occupations and
social positions of those affected? In this example, symptomatic
malaria is more common in poorer, ruraldwelling
children. Finally, there is the question of
variability by place: the prevalence of malaria is dictated
by the distribution of Anopheles mosquitoes.
Measuring risk
Epidemiology is also concerned with the numerical
estimation of risk. This is best illustrated by a simple
example. In a rural African town with a population of
5000, disease ‘d’ is under investigation. The majority of
the cases of disease ‘d’ (300 out of 360) occurred among
women and children who use the river, which recently
had its flow of water reduced because of a new irrigation
scheme. A formal experiment is established to measure
risk. The 1000 women and children who use the river
are followed up for 1 year and compared to a cohort
with a similar age and sex distribution who use standpipes
as their source of water.
The incidence (new cases) of disease ‘d’ in the 1000
exposed to risk ‘r’ (river water) was 300. The incidence
(new cases) of disease ‘d’ in the 1000 not exposed to risk
‘r’ was 60. The relative risk is the incidence in the
exposed population (300 per 1000 per year) divided by
the incidence in the non-exposed population (60 per
1000 per year); 300/60 = 5, meaning that those exposed
to the river water are 5 times more likely to contract the
disease – their relative risk is 5. The attributable risk of
exposure ‘r’ for disease ‘d’ is the incidence in the exposed
population (300) minus the incidence in the non-exposed
population (60), which is 240 per 1000 per year. The
fraction, or proportion, of the disease in the exposed
population which can be attributed to risk (r) is called
the attributable fraction, in this case (300−60)/300 = 0.8.
This means that 80% of the disease can be attributed to
exposure to river water.
Establishing cause and effect
Associations between a risk factor and a disease do not
prove that the risk factor causes the disease. In the
northern hemisphere, both multiple sclerosis and blue
eyes are more common but it is implausible that having
blue eyes is the cause of multiple sclerosis. Cause and
effect can only be proven by more detailed investigation.
In the above example, further investigation of the river
water will be needed, using the criteria for causation
defined in Koch’s postulates (for infectious agents,                                                                                                                                                   ) or the more generic Bradford Hill criteria.

Preventive medicine
There are many examples of epidemiological associations
defining causative factors in disease, e.g. the association
between cigarette smoking and lung cancer
. However, as illustrated above, the complexity
of the interactions between physical, social and economic
determinants of health means that successful prevention
is often difficult. Moreover, the life course
perspective illustrates that it may be necessary to intervene
early in life or even before birth, to prevent important
disease in later life. Successful prevention is likely
to require many interventions across the life course and
at several levels in the hierarchy of systems. The examples
below illustrate this principle.