Causal Inference

Question 1

Definition of Cause, Causality, & Causal Inference

Answer 1

1. Cause is a factor that gives rise to an effect (event).
2. Causality is the potential that a specified change in a factor (cause) produces a predictable change in an event (effect)
3. Causal inference is a reasoning process or the thought process, methods, and evidence used to support or refute a relationship as one of cause and effect

Question 2

Definition of inference

Answer 2

Generalizing to a larger target group based on observations made in a subset of that group

Question 3

T/F: Causal inference is a deductive reasoning process

Answer 3

False. It is an inductive reasoning process (from small to large)

Question 4

T/F: Randomized Trail is the best choice for “causality” since we have control over cause/exposure.

Answer 4

True

Question 5

Assigning Evidence Levels for Causation (3 levels)

Answer 5

Question 6

General categories of observed associations (3)

Answer 6

1. Chance (interval estimation, hypothesis testing)
2. Spurious (artifactual due to bias)
3. True (causal or noncausal [confounding] ?)

If after dealing with confounding, the association is gone, then there’s no causality here. The observed association is solely due to the confounding.

Question 7

T/F: Single factor model is reasonable to be used when studying chronic disease

Answer 7

False. It is reasonable to be used when studying infectious disease. Multifactorial model is used when studying chronic disease.

Question 8

2 types of causal relationships in modern concepts

Answer 8

1. Component (necessary): a factor that is required, but not singly responsible, for disease occurrence (also called component cause)
2. Sufficient (required): a set of minimal conditions (component causes) that operate collectively to produce disease (they are actually causal pathway)

When all the component causes occur, sufficient cause is formed.

Intervening on a component cause may effectively lower disease by rendering the other components insufficient.

Question 9

Conceptual scheme of 3 sufficient causal paths for a hypothetical disease

Answer 9

Question 10

Guidelines for evaluating the evidence for causality

Answer 10

consistency (in other studies)

temporality (exposure precedes disease)

biologic gradient (dose-response)

Coherence