Association, Causal Inference & Causality

Question 1

Define Cause

Answer 1

a precursor event, condition, or characteristic required for the occurrence disease/outcome

most causal relationships involve multiple-component factors

Question 2

Epidemiological studies yield __________ between exposure and disease

Answer 2

(statistical) associations

Question 3

What are the 3 main types of association delineated during a study?

Answer 3

Artifactual (aka False) Associations

Non-Causal Associations

Causal Associations

Question 4

Define Artifactual Associations

Answer 4

they can arise from bias and/or confounding

Question 5

Define Non-Causal Associations

Answer 5

Occurs in one of two ways

The disease may cause the exposure (rather than the other way around)

The disease and exposure are both associated with a 3rd factor (confounding)

Question 6

Define Causal Associations

Answer 6

There is a direct link between the exposure and the outcom

Question 7

What are the 3 types of Causal relationships?

Answer 7

Sufficient Cause

Necessary Cause

Component Cause

Question 8

Define Sufficient Cause. what is an example of this?

Answer 8

a set of minimal conditions that WILL cause disease

the cause always precedes the disease in this case

ex. genetic conditions

Question 9

Define Necessary Cause. What is an example of this?

Answer 9

A cause that precedes the disease and MUST be present for the disease to occur

the cause may be present and the disease not occur in this case

ex. Mycobacterium tuberculosis

Question 10

Define Component Cause. What is an example of this?

Answer 10

(aka risk factor) a factor that if present/active it increases the likelihood of a particular disease

ex. High HDL levels increasing he risk of AMI

Question 11

Define synergism. Give an example of this

Answer 11

when 2 or more component causes interact and their combined measure of effect is greater than the sum of the individual effects

ex. genetic predispositions to depression and a stressful life event causing a synergistic effect on the risk of depression

Question 12

Define Parallelism. Give an example of this

Answer 12

When 2 or more component causes interact in such a way that the measure of effect is greater if EITHER component is present

Question 13

Multiple component causes can work together to collectively become ____________.

Answer 13

Sufficient Causes

Question 14

What is used in the process of causal inference determination? give an example of this

Answer 14

Inductively-Oriented Criteria ; Hill’s Criteria (guidelines)

Question 15

Explain Hill’s criteria in layman’s terms

Answer 15

the criteria that must be met to determine an observed association to be a verdict of causation

Question 16

List the categories of Hill’s Criteria (guidelines)

Answer 16

1. Strength
2. Consistency
3. Temporality
4. Biologic Gradient
5. Plausibility

Mnemonic: “Still Can’t Take Blue Pills”

Question 17

Define Strength as it pertains to Hill’s Criteria

Answer 17

refers to the SIZE of the measure of association (RR/OR/HR)

the greater (below or above) 1, the more convincing the association may be causal

Question 18

Define Consistency as it pertains to Hill’s Criteria

Answer 18

the repeated observations of an association in different populations.

the length of time an association is found to be consistent also matters

Question 19

Define Temporality as it pertains to Hill’s Criteria

Answer 19

reflects that the cause precedes the effect/outcome in time

smoking causes lung cancer if you smoke prior to being diagnosed.

Question 20

Define Biologic Gradient as it pertains to Hill’s Criteria

Answer 20

refers to the presence of a gradient of risk associated with a degree of exposure

can be a good or bad risk (smoking or exercise)

Question 21

Define Plausibility as it pertains to Hill’s Criteria

Answer 21

refers to the biological feasibility to the association

has to be a logical cause

Question 22

A strong association is neither necessary nor _____ for causality.

Answer 22

sufficient

the weakness of an association is neither necessary or sufficient for ABSENCE of causality

Question 23

Compare proximate cause and distant cause when referring to temporality

Answer 23

Proximate cause: short-term interval

Distant Cause: long-term interval

Question 24

True or false: consistency will alway be a sign of causality in a study

Answer 24

False

consistency may still obscure truth (ex. postmenoposal estrogen study)