Selection bias

Question 1

What is bias?

Answer 1

Bias is any systematic error in an epidemiological study that results in an incorrect estimate of the association between exposure and risk of disease

Question 2

Random vs systematic error

Answer 2

Random error - the larger the sample size or more times you do the experiment you can reduce random error

Systematic error is something that will not change no matter how many times you take a sample

Question 3

Why is systematic error important?

Answer 3

Findings could be over estimated, under estimated or not affected
Important when critically appraising scientific literature, in evidence based practise, undertaking research

Question 4

What can we do about bias?

Answer 4

Once data has been collected, if systematic error is there it cannot be changed/removed, we can consider the impact it may have had but cannot control it
When a study is being designed researchers need to consider what bias could occur and how they could limit potential bias

Question 5

What is selection bias?

Answer 5

Selection bias occurs when there is a systematic difference between the people who are included in a study and those who are not, or when study and comparison groups are selected inappropriately or using different criteria

Question 6

How to minimise loss of follow up

Answer 6

Use alternative contact details obtained at the start of the study, maintain regular contact e.g. study newsletters, making several attempts to contact people
If loss to follow up is related to both the exposure and the outcome this can lead to bias

Question 7

Considering selection bias in cross sectional studies

Answer 7

Sample population is who enters study
Ideally want a sample representing of the source population (not a group of people who are convenient to the researcher)
We want a high response rate (unlikely to get 100%)
Is prevalence over or underestimated?

Question 8

Case control studies - selection bias

Answer 8

Participants are selected on the basis of their outcome status, if this selection is in some way dependent on their exposure status than bias can occur
Selection of cases and controls must be independent of exposure status

Question 9

Potential biases in selection of cases

Answer 9

If the identification or participation of cases is related to exposure status there can be bias
- Ensure high participation
- Clearly defined population of interest
- Reliable way of ascertaining all cases or a representative sample of cases

Question 10

Three steps to consider

Answer 10

1. Who does it affect
2. How does it affect them
3. What effect does it have on the measure of associated (MOA)
   Is MOA under or over estimated and specify direction in relation to null value (MOA been biased towards or away from null)