Week 9 - Confounding

Question 1

What are disease determinants?

Answer 1

Most diseases are caused by high exposure to certain risk factors and/or low
exposure to certain protective factors.
These factors are collectively called disease determinants and differ from
disease to disease.

Question 2

What does disease determinants include?

Answer 2

-Genetic factors
-Biological/physiological factors (e.g. ageing, other disease status,
biochemical factors)
-Lifestyle factors (e.g. diet, physical activity, smoking, alcohol consumption,
sleeping patterns)
-Psychosocial factors (e.g. stress, social isolation)
-Sociodemographic (e.g. socioeconomic position, educational attainment,
urbanization)
-Wider environmental factors (e.g. air pollution, impure water, toxic agents,
radiation)

Question 3

What are the disease determinants of Cardiovascular disease?

Answer 3

Question 4

What are the disease determinants of Type 2 diabetes?

Answer 4

Question 5

What are the disease determinants of Alzheimer’s diabetes?

Answer 5

Question 6

What are the disease determinants of breast cancer?

Answer 6

Question 7

What is cofounding?

Answer 7

A parameter that is linked to the multifactorial nature of disease.
Similar to chance and bias, may introduce error.
A third factor (confounder) that explains all or part of the
association between an exposure and an outcome

Question 8

What criteria does a potential cofounder need to fulfil?

Answer 8

1. Has to be associated with the outcome of interest
2. Has to be associated with the exposure of interest
3. Should not lie in the causal pathway between exposure and
   outcome

Question 9

Note!

Answer 9

• Note 1: If a factor is associated with the outcome but not with the exposure
  (or vice versa), then this factor cannot be a confounder in the specific
  association!
• Note 2: If a factor is associated with both the outcome and the exposure, but
  lies in the causal pathway between the two, then this factor is not called a
  confounder but a mediator!
  -Note 3: Adjusting for mediators may have exactly the same effects on an
  exposure-outcome association as adjusting for a confounder (i.e. association
  becomes weaker, stronger, etc.). The only difference is in the interpretation!

Question 10

What is the 3-step procedure for cofounding?

Answer 10

1. Identify potential confounders
2. Adjust for potential confounders
3. Compare crude and adjusted estimates

Question 11

What is statistical adjustment?

Answer 11

-The process of statistical adjustment aims at eliminating or reducing the confounding effect of potential confounders in any exposure-outcome
association
-In other words adjustment aims at removing any effects of the potential
confounder on the outcome, thus providing a more ‘clean’ estimate of the
exposure-outcome association
-After adjusting (also called ‘controlling’) our estimates for a potential confounder, these estimates are then said to be adjusted for
that confounder (i.e. age-adjusted estimates, smoking-adjusted, etc.)

Question 12

How do we apply statistical adjustment?

Answer 12

What statistical adjustment is doing is keeping the potential confounder
constant (i.e. making it a non- variable) and re-calculating the estimates for the
exposure-outcome association (i.e. Odds Ratios, mean difference, regression
coefficient, etc.)

Question 13

Explain the process of statistical adjustment.

Answer 13

1. One way of adjusting for a given confounder is to stratify the analysis based on
   the categories of the confounder.
   This process of stratification involves performing the analysis for the exposure-outcome association separately in the categories of the confounder (i.e. for
   smokers and non-smokers if smoking is the confounder).
2. This will therefore give us 2 different (stratified) estimates (assuming that the
   confounder had 2 categories) for the exposure-outcome association, one for
   each category of the confounder variable
3. These stratified estimates will then need to be combined, in order to get a
   combined (adjusted) estimate.
   This adjusted estimate is ‘clear’ of any confounding effect from the specific
   confounder, since the analysis was performed separately in its categories and the
   results were then combined.
4. Note that if the exposure-outcome association is substantially different in the
   categories of the confounder, then the estimates cannot be combined and we
   present them separately (this is known as effect modification).

Question 14

What is a crude estimate?

Answer 14

Crude estimates are simply estimates for exposure-outcome associations (Odds
Ratio, mean difference, etc.) before applying any adjustment.

Question 15

What is an adjusted estimate?

Answer 15

Adjusted estimates are estimates for exposure-outcome associations after
applying statistical adjustment for any potential confounder
It aims to answer the question: “ What would the exposure-outcome association be if everyone had the same cofounder value?”

Question 16

When can we combine estimates?

Answer 16

If two estimates are similar they can be combined.

Question 17

How do we combine two estimates?

Answer 17

The two estimates are combined in a weighted manner, based on the sample size of
each category of the confounder variable.

Question 18

How can we determine if something is a cofounder?

Answer 18

Compare crude and adjusted estimates.

Question 19

What effects can adjustments have on estimates?

Answer 19

1. The association becomes weaker
2. The association disappears
3. The association appears
4. The association becomes stronger
5. The association is reversed
6. The association remains unaffected

Question 20

What is overadjustment?

Answer 20

If a factor which lies in the causal pathway between exposure and outcome (i.e.
a mediator) is treated like a confounder and adjusted for, this may lead to an
underestimation of the true exposure-outcome association, a phenomenon called
overadjustment.

Question 21

What is residual cofounding?

Answer 21

If the researchers adjusted for some but not all potential
confounders, or in case the confounders adjusted for where not accurately
measured, we may have an overestimation of the true exposure- outcome
association, a phenomenon called residual confounding.