Dr David McGrath

Spine Physician

MB BS (Hons) FAFOM, RACP, FAFMM
Master of Pain Medicine

Cause & Effect (Part 1)

Bottom Line:
When an observation "A" is associated with a disease "D"
1. An upstream input can cause A+D
2. D can cause A
3. A can cause D
Or in simple terms, "A" can be the cause of D,an effect of D or associated through upstream input.

Cause and Effect are often Confused. We notice association, and wrongly believe that there is a causal connection between the two phenomena.
There are three types of associations, Positive and Negative and Neutral.
1. A positive association means A and D are always observed together.
2. A negative association means A and C are never observed together.
3. A neutral association means A and E are associated 50% of the time
The negative association can be simply restated as a positive.
3. A positive association means A and (-C) are always observed together

The usual interpretation of association (1) is:
1.Clouds cause rain
But an equally likely statement, with the current information base is:
2.Rain causes clouds
But, we can also have:
An unknown factor "X" causes both rain and clouds (rain and clouds share an upstream input)

Of course an important observation is, clouds are always seen BEFORE rain occurs. Can we now legitimately state that "clouds cause rain", or can we? Factor "X" may still cause clouds and rain, but clouds first. Water vapour is a cause of clouds and rain. But, this is not causal unless, water vapour always passes through a "cloud" phase before rain. If this is so, we could again state that clouds cause rain, in the sense that, no clouds mean no rain.
From this logic, we could define a cause A of D if:
1. Without A, there is no D (no cloud,no rain)
2. With A, D may be present (cloud and rain can occur together )
Or in logic words. A is essential for D,but may or may not be sufficient (other inputs needed)

In medicine we are often lack the full information to help us decide,if cause and effect are present.
Our associations are sometimes confused with cause.

Looking at Cholesterol as an Example
High cholesterol (A) is thought to be the cause of heart disease (D)
Certainly, point (2) is fulfilled. Heart disease can be present with high cholesterol. Point (1) is not satisfied however. Heart disease, occurs with equal frequency in people with low cholesterol.
For cholesterol to remain a cause under these circumstances, we would need to do some mental gymnastics. Cholesterol would have to be a cause, when it is high, while not be a cause, when it is low. (some other factor takes over) This unlikely. There is in fact evidence, that cholesterol is protective. High cholesterol can be associated with a true cause of heart disease, and hence falsely implicated.