POPH192 Lecture 25 - Selection Bias

Question 1

what is bias in epidemiology?

Answer 1

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

Question 2

why does bias matter?

Answer 2

because it can result in an incorrect estimate

Question 3

the findings can be what when there is a systematic error?

Answer 3

finds could be

• over-estimated
• under-estimated
• not affected

Question 4

when is the effect of a systematic error important?

Answer 4

important when

• critically appraising scientific literature
• in evidence-based practise
• considering studies reported in the media
• undertaking research

Question 5

when is the only time when bias can be controlled?

Answer 5

selection and information bias can only be controlled during the design and data collection phases of a study

Question 6

what is selection bias?

Answer 6

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

Question 7

how do researchers recruit people into epidemiological studies?

Answer 7

• through advertisements

- researchers can randomly select people from the population

Question 8

what is a case?

Answer 8

people with the outcome

Question 9

what is a control?

Answer 9

people without the outcome

Question 10

what is the relationship between selection of cars and controls and exposure status?

Answer 10

selection of cases and controls must be independant of the exposure status

Question 11

how can you minimise selection bias in the selection of cases?

Answer 11

• ensure high participation
• clearly defined population of interest
• reliable way of ascertaining all cases or a representative sample of cases

Question 12

what do you need to ensure for in controls to avoid potential bias in selection of controls?

Answer 12

ensure controls are from the same defined population as the cases over the same time period

Question 13

what do you need to ensure for in inclusion/exclusion criteria to avoid potential bias in selection of controls?

Answer 13

sam elusion/exclusion criteria for caesura and controls

Question 14

what does it mean to specify the direction of bias in relation to the null value?

Answer 14

specify if the measure of association has been biased towards or away from the null value (numerically upwards or downwards)

Question 15

how can loss to follow up lead to selection bias?

Answer 15

if loss to follow-up is related to both the exposure and the outcome this can lead to bias