Confounding: Guilty By Association Flashcards

1
Q

What is a confounding factor?

A

Relationship between an exposure and outcome is via a shared relationship. It must have a significant impact on the adjusted risk ratio when compared to the crude ratio

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are examples of confounding factors?

A

Osteoarthritis and chronic knee pain are linked by a high BMI.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How can confounding factors be addressed in a study design?

A

Restriction
Matching
Stratification
Multiple variable regression
Standardisation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How are confounding factors restricted?

A

They set limitations on the participants to increase their similarity in order to reduce in the confounders based on factors like BMI or age.

However, this is difficult if there is more than one confounder, reduces generalisability and the number of eligible participants.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How are confounding factors matched?

A

Putting those with the same confounding factor in the same group and creating a comparison group for the confounder. This is typically used for case control, however it is expensive and difficult to carry out.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How are confounding founders stratified?

A

All participants are sorted into subgroups based on the confounding factors and measure the exposure-outcome association. The final step is to combine the results to achieve a measure of average effect in each strata.

If it is different from te crude measure of effect but similar to each other, this is evidence of confounding.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the coefficient variable?

A

Calculated by the gradient of a graph, which indicates the relationship between the dependent and independent variable.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

When is multiple variable regression graphs used in confounding?

A

When there are too many strata to be calculated, therefore the coefficient of the graph is used to measure values such as odds or risk ratios.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is a crude mortality rate?

A

The proportion of deaths in a total population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is standardised mortality ratio?

A

Uses a standardised mortality rate to compare to a specific population’s mortality rate. This is dependent on age, population size and sex specific rates on the standard population and the study population. calculated by:

Observed rate in study population/ Standardised death rate= x per 100,000 deaths

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is indirect standardisation?

A

This is a method used to determine the mortality rate of our study population when age specific rates are unknown, using the values from the population structure of a standard population.

The standardised mortality ratio is then used to measure the mortality.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is direct standardisation?

A

Requires that the specific mortality rates of subtypes of the study population like age are available and obtain a weighted average of each strata and compared to a defined comparison standard population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How is direct standardisation calculated?

A

Comparative mortality ratio where observed specific death rate in our study population that is measured per 100,000 deaths will be multiplied by the rate of death in the standard population.

(X/ 100,000) x the standard population= expected death rate in specific strata

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are the requirements for direct standardisation?

A

A standard population is defined
AND
Age specific rates for populations are available

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is a limitation of indirect standardisation?

A

Requires the total number of cases for death in the study population, however this does not take into account the confounding factors like population structure.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

How is indirect standardisation calculated?

A

Standard mortality ratio.

17
Q

What are the consequences of not adjusting for confounding?

A

Inaccurate measurement of true association between exposure and outcome

May change the direction of the observed effect