Association and Causation Flashcards
What is the primary goal of epidemiological studies?
To identify causes of diseases or health issues to guide prevention, diagnosis, and treatment.
What is the difference between association and causation?
Association means two variables occur together more often than by chance, while causation means one variable directly affects the other.
Define a ‘necessary cause’ in the context of causation.
A necessary cause is a factor that must be present for a disease to occur, but it may not be sufficient on its own to cause the disease.
Define a ‘sufficient cause’ in the context of causation.
A sufficient cause is a factor that alone can produce or initiate an outcome.
What is a spurious association, and why does it occur?
A spurious association is a false connection due to chance or bias, often caused by flaws in study design.
What is a statistical association?
A relationship where two variables are linked due to a shared underlying factor, not causation.
Provide an example of a non-causal statistical association.
High altitude and goiter: they appear associated, but the true cause is iodine deficiency common in high-altitude areas.
What type of association results from selection bias?
Spurious or false association due to differences in groups that are not accounted for in the study design.
What is a one-to-one causal relationship? Provide an example.
A model where the presence of a specific factor always leads to a particular disease. Example: Measles virus causing measles.
What is multifactorial causation, and why is it important in epidemiology?
A model where multiple factors contribute to disease development, often seen in chronic diseases like cardiovascular disease.
In the multifactorial model, what is the difference between independent and cumulative causation?
Independent causation: factors act alone to cause disease; cumulative causation: multiple factors interact to produce disease.
What is the purpose of the Bradford Hill Criteria?
To determine whether an association between two variables is likely to be causal.
What does ‘temporal association’ mean in the Bradford Hill Criteria?
The suspected cause must occur before the observed effect for the association to be considered causal.
How does ‘strength of association’ support causality?
A stronger association (higher relative risk) increases the likelihood that the relationship is causal.
Why is ‘specificity of association’ challenging to establish in epidemiology?
Because many diseases have multiple causes, and a single cause can lead to multiple diseases, making one-to-one relationships rare.
What does ‘biological plausibility’ mean in the context of causation?
The association aligns with known biological mechanisms, making the causal relationship more credible.
Give an example of ‘consistency of association’ in epidemiological studies.
The link between smoking and lung cancer, consistently observed across many studies worldwide.
What role does ‘coherence’ play in judging causality?
Coherence ensures that the association does not contradict established knowledge and historical data trends.
What is the ‘web of causation,’ and why is it significant in multifactorial diseases like ischemic heart disease (IHD)?
The web of causation illustrates how multiple factors interact in complex ways to cause diseases, which is crucial in multifactorial diseases.
What is a ‘confounding variable,’ and how does it affect association interpretation?
A confounding variable is an external factor linked to both the exposure and outcome that can create a misleading association if not accounted for.
Explain how selection bias might lead to a false association using the example of hospital vs. home births.
Higher-risk pregnancies are more likely to occur in hospitals, creating a misleading association that hospital births are riskier.
How is the correlation coefficient used in determining associations, and what are its limitations?
The correlation coefficient (ranging from -1 to +1) measures the strength of association but does not prove causation since it lacks a temporal sequence.
Why can cigarette smoking and sugar intake show a spurious association in relation to coronary heart disease (CHD)?
Smoking is linked with increased hot drink and sugar intake; however, only smoking, not sugar, is causally related to CHD.
What does it mean for a cause to be ‘necessary but not sufficient’ in the case of tuberculosis?
The tubercle bacilli are necessary for tuberculosis but are not sufficient alone, as additional factors like poor living conditions increase disease risk.
What example from the Bradford Hill Criteria illustrates ‘dose-response relationship’ with smoking and lung cancer?
Increased lung cancer risk correlates with the number of cigarettes smoked daily, showing a dose-response relationship.
Describe ‘coherence’ in the context of lung cancer and smoking trends in men and women.
Historical coherence is seen as smoking rose first in men, followed by women, with lung cancer trends reflecting this shift.
Why is ‘specificity of association’ often considered the least applicable Bradford Hill criterion?
Specificity is challenging because most diseases result from multiple causes, and single exposures can lead to various outcomes.
According to the Bradford Hill Criteria, why is ‘biological plausibility’ important, and give an example related to skin cancer.
Biological plausibility requires that the association makes sense biologically, such as UV exposure leading to skin cancer, which fits known biological mechanisms.
