10. Association and Causation Flashcards
variable
any observable event that can vary
age weight
variables are either
associated or they are not
if not they are independent
if they are they are dependent
positively associated
they both increase and decrease together
negatively associated
they increase and decrease inversely
Types of association
no association = X and Y independent
associated = X does not cause Y (non-causally) X causes Y (causally)
independent variable
factor that stands alone and isn’t changed by other factor you are trying to measure
dependent variable
factor that is influenced or changed by another factor
Independent vs dependent
independent variable causes a change in the dependent variable and it is not passible that dependent variable could cause a change in independent variable
confounding variable
is interference bt third factor that distorts the association within a study of two primary variables
What are the illustrations of association
scatter plots
dose response curves
epidemic curve
contingency table
importan thing about causation
association does not prove causation
threshold
to the lowest dose at which a particular response occurs
epidemic curve
graphic plotting of the distribution of cases by time of onset
contingency table
tabular method of demonstrating association
central concern of epidemiology
one of the central concerns of epidemiology is to be able to assert that a causal association exists between an exposure factor and disease
causality criteria
consistency
strength
specificity
temporality
coherence
consistency
association has been observed repeatedly ideally by different observers
strength of association
refers to magnitude of relative risk or odds ratios from observed studies
specificity
one particular exposure produces one specific outcome
temporality
the exposure or factor must precede the outcome or disease
coherence
synonymous with biological plausability
cause and effect should not conflict the generally known facts
multifactor causality
many types of causal relationships involve diseases with more than one causal factor
two models of multifactorial causality
epidemiological triangle
web of causation
web of causation
many points not just the three
smokeing
ethnicity
diet
excersise
lots of factors can cause it and pull it one what or another
How can chance be ruled out
we can never completely rule out chance
how to minimize chance
minimize bias
work through causality criteria
risk factor
exposure that is associated with a disease morbidity mortality or adverse health outcome
risk assessment
methodology used to provide quantitative measurements of risk to health
probability vs odd
probability = chance or risk of occuring
odds= ration of probability of an event occurring to the probability of an even not occurring
odds= P/(1-p)
