Lecture 13 (Causality)

Question 1

Importance of the “Web odf Causation”:

Answer 1

• lists out all of the possible causes of a disease/adverse outcome
  
  • some are modifiable, some are not. Modify the modifiable causes, and decrease rates of disease/adverse outcomes.
• YOU DO NOT HAVE TO UNDERSTAND ALL THE CAUSAL MECHANISMS TO INTERVENE AND PREVENT

Question 2

Sufficient definition:

Answer 2

• the cause will lead to the effect regardless of other factors

Question 3

Necessary definition:

Answer 3

• the effect will not occur without both the cause and other factors

Question 4

If you need other cofactors in addition to the cause to manifest disease, the cause if classified as:

Answer 4

neccesary

Question 5

Koch’s Postulates:

Answer 5

• microorganism in diseased, not in the healthy
• microorganism can be isolated
• isolated microorganism can cause disease in healthy host
• microorganism can be re-isolated from newly infected healthy host

Question 6

The 8 Bradford Hill Guidlines:

Answer 6

1. temporality
2. strength association
3. dose response
4. reversibility
5. consistency
6. biological plausibility
7. specificity of association
8. analogy

Question 7

What is the only required criteria for causality?

Answer 7

temporality

• exposure to cause precedes disease

Question 8

Temporality:

Answer 8

• exposure to cause precedes disease

ONLY REQUIRED CRITERIA FOR CAUSALITY

Question 9

Strength of Association:

Answer 9

• magnitude of association between exposure and disease

Question 10

Dose-response:

Answer 10

• disease rate increases as exposure rate increases
  
  • can be linear or U-shaped correlation

Question 11

Threshold:

Answer 11

• important for dose-response
• lowest dose at which response (disease) occurs

Question 12

Reversibility:

Answer 12

• remove exposure, disease rate decreases
  
  • may not always be true if the pathogenic process is irreversible

Question 13

Consistency:

Answer 13

• results across different study types and populations are similar

Question 14

Biological plausibility:

Answer 14

• mechanism of how cause initiates disease must be plausible

Question 15

Specificity:

Answer 15

• exposure/cause is associated with one specific disease outcome
  
  • watch for exposures that cause multiple outcomes

Question 16

Analogy:

Answer 16

• is exposure/outcome relationship similar to other known causal relationships

Question 17

What are the two weakest Bradford-Hill criteria regarding cause?

Answer 17

1. reversibility
   
   • pathogenic process may not be reversible
2. specificity
   
   • cause may have more than one adverse outcome

Question 18

Ecological study design:

Answer 18

• analyze groups rather than individuals
  
  • i.e. compare average smoking levels with average lung cancer rates in a county/state
• stat: correlation
• good at generating hypotheses

Question 19

What is the problem with ecological studies?

Answer 19

• you cannot determine if the individuals who had the exposure developed the adverse outcome
  
  • YOU JUST HAVE RATES

Question 20

What is ecologic fallacy?

Answer 20

• attribute overall group characteristics to individual members of that group

Question 21

Meta-analysis:

Answer 21

• a study of studies - what is the overall result?
• take the results of all studies and combine

Question 22

Pooled analysis:

Answer 22

• combine data from a lot of studies, re-analyze
• increases power and precision
• good with similar study protocols

Question 23

What are the two forms of quantitative review?

Answer 23

1. meta-analysis
2. pooled-analysis

Question 24

Is disease randomly distributed?

Answer 24

• No.
• factors that determine disease are knowable and modifiable.

MODIFICATION WILL LEAD TO PREVENTION OR CONTROL