Lecture 26 - Information Bias

Question 1

Why is systematic error important in research?

Answer 1

• Critically appraising scientific literature
• In evidence-based practice
• Considering studies reported in the media
• Undertaking research

Question 2

How can selection and information bias can be controlled?

Answer 2

The can only be controlled during the design and data collection phases of study.
- Identify potential sources of bias
- Identify ways to minimise bias

Question 3

What is Information bias?

Answer 3

Information bias occurs when the are systematic differences in how data on exposure or outcome are collected from various study groups

Question 4

How is data typically collected in a study

Answer 4

• By participants e.g. self report
• Collected or measured by someone else

Question 5

How can measurement error occur

Answer 5

• Participants may provide inaccurate responses (e.g. past exposures or under or over estimate of their exposure)
• Data collection methods may be flawed (e.g. incorrect measurement or inconsistent procedures)

Question 6

Effect of measurement error in descriptive study?

Answer 6

It could leat to an overestimation or underestimation of prevalence

Question 7

Measurement error

Answer 7

Measurement error can be random or systematic

Question 8

Effect of measurement error in analytic study

Answer 8

• can lead misclassification
  people without the exposure may be classified as having the exposure and vice versa

Question 9

Two types of misclassification

Answer 9

• Non-differential misclassification
• Differential misclassification

Question 10

What is non-differential misclassification

Answer 10

It occurs when misclassification of exposure or outcome happens equally in all study groups

Question 11

Different misclassification

Answer 11

It occurs when misclassification of exposure or outcome happens differently between study groups

Question 12

Examples of differential misclassification

Answer 12

• In a cross sectional study, people with the outcome might report exposure differently to those without the outcome

Question 13

What is recall bias

Answer 13

Systematic error due to differences in accuracy or completeness of recall to memory of past events or experiences

Question 14

Minimising recall bias

Answer 14

• Objective measures
• Validate self-reported measures with other information
• Memory aids

Question 15

Minimising interviewer/observer bias

Answer 15

• Clearly defined study protocol and measure
• Structure questionnaire and standard prompts
• Training of interviewers
• Blinding

Question 16

Cohort studies: Different misclassifications

Answer 16

• If classification of exposures depends on outcome
• If classification of outcome depends on the exposure (interviewer bias)

Question 17

RCT: Bias occurs

Answer 17

Bias could occur if knowledge of treatment/exposure category influences the assessment of the outcome
- Ensure there is blinding
Bias can occur due to the measurement being undertaken differently for different treatment groups
- Ensure measurements undertaken in the same way

Question 18

Minimising information bias

Answer 18

• Collecting information from participants e.g. validated survey instruments
• Measurements instruments e.g. use standardised equipment
• Collecting information via interviewers/observers
  e.g. ensure blinding, use objective measure, clearly defined study protocols

Question 19

Publication bias

Answer 19

Publication bias occurs when studies with positive findings are more likely to be published than those with negative or non significant findings