Transcript Overview of Social Epidemiology

Introduction to Social
Epidemiology
Illustration via a case study:
Type 2 Diabetes
The prevalence of diabetes varies widely among
people, apparently both according to race and
living circumstances:
• People of European
origin
–
–
–
–
In Britain 2%
Germany 2%
Australia 8%
USA 8%
• Native Americans
– Chilean Mapuche 1%
– US Hispanic 17%
– US Pima 50%
• People in New Guinea
– Rural 0%
– Urban 37%
• Aboriginal Australians
– Traditional 0%
– Westernized 23%
• Black Africans
– Rural Tanzania 1%
– Urban S. Africa 8%
– United States 13%
Source: Jared Diamond. Nature 2003;423:599
WHY?
Type 2 non-insulin dependent diabetes
• Etiologically heterogeneous; common feature is
high blood glucose due to altered insulin
secretion and insulin resistance
• Patients still produce insulin but are unable to
respond effectively to it
• Patients are typically obese
• Often a disease of socioeconomic disadvantage.
Genes and Diabetes
• Several genes implicated; presumably these
must have conferred a survival advantage at
some time
• For example, the hypothesized ‘thrifty gene’
enables a person carrying it to use food
efficiently in times of plenty in preparation for
famine conditions. (JV Neel. Am J Human Genet
1962;14:353-362)
• Perhaps this may be relevant to diabetes?
Genes interacting with lifestyle?
1. Diabetes involves genetic factors and lifestyle,
especially diet
2. Symptoms disappear under conditions of
starvation (e.g., siege of Paris, 1870)
3. Migrant populations see increases (immigrants
to Israel; Japanese moving to USA); perhaps
their diet changes when they migrate?
4. Rates fluctuate with economic conditions
5. = Lifestyle disorder seen in genetically
susceptible populations; environmental factors
associated with lifestyle unmask the disease.
The counter-arguments
1. 30 years of searching have not identified a
culprit gene
2. Obesity and Type 2 diabetes are responses to
late 20th century lifestyles, so it’s really a socialenvironmental issue
3. There is also a rival hypothesis of intra-uterine
exposure to hyperglycemia that has been
supported in cohort studies
4. Or, alternatively, hypothesis of early childhood
under-nutrition (see McDermott, Soc Sci Med
1998;47:1189-95)
Some remaining questions…
1.
2.
3.
4.
5.
Explanations are not merely scientific; they lead to blame
and action. Do we blame individuals for their unhealthy
diet, or do we blame their cultural heritage, or capitalism
for its marketing practices, or governments, or scientists,
or … ?
Will advances in bench science remove the need to worry
about the social context of illness?
What are the social implications of the current emphasis
on searching for genetic explanations?
More broadly, are the causes of individual cases the same
as the causes of incidence rates? (I.e., is it the same
factor that explains why one individual is diabetic, that
also explains racial differences?)
So, should we view a population as just an aggregation of
individuals, or is it somehow different?
Ways of thinking about
disease
Susser’s Eras in Epidemiology
Paradigm
Era
Analytic
approach
Prevention
Sanitary
Miasma
theory
Clustering of
mortality
Sanitation
Infectious
disease
Germ theory
Laboratory
Vaccination
Chronic
disease
Black box
Risk ratios
Host, agent,
environment
EcoSystems
epidemiology theory (?)
Determinants Contextual to
at many
molecular
levels
Source: M. Susser. Am J Public Health 1996;86:674-7.
Life-course human development view
• Health is a consequence of multiple deficits
• Health is an interaction between living context
and bio-behavioral regulatory systems
• Personal health trajectories reflect the effect of
many exposures; these cumulate over time
• The timing and sequence of the events is
important – there are periods of enhanced
susceptibility
– E.g., the weathering hypothesis: cumulative exposure
to stressors leads to vulnerability
Things the social epidemiologist
typically worries about
• Biological determinism, represented in the
human genome project; perception that we are
largely controlled by our genes
• Social Darwinism; sociobiology
• Implicit reductionistic & deterministic stance;
narrow focus on pathogenesis
• Treatment or early detection rather than
primordial prevention
• Denial of the agency of people and communities
Reactions to uncertainty
• There is much we do not understand.
DL Weed (1988) described three reactions to
scientific uncertainty:
– Belief (“retreat to commitment”). Implies cessation of
enquiry. Characteristic of the religious right
– Statistics and reference to probability. This does not
help us decide where to look for further evidence, or
what to ignore, nor when we have arrived
– Criticism. Will not make us certain, but helps to bring
weaknesses to the surface
• Fourth way may be to integrate disciplines; how
do we do this?
Conceptual Model for Social
Epidemiology
Starting Point:
Designing Multiple
Interventions
Biological
Processes
& Overall
Model
Inequalities
in Health
Explanations &
Causal Theory
Sociological
Explanations
Biological
Societal
Processes
Life
Events
PNI
Coping,
Vulnerability
& Resistance
Social
Support
Individual
Work
Behavioral
Theories
Stress
Theories
Personality