23. SYSTEMATIC ERRORS

Question 1

1. What are the two main types of bias in Research?

Answer 1

1. Selection Bias
2. Information Bias

Question 2

2. What is Selection Bias?

Answer 2

• Systematic Errors arise from the mistakes that are
  conducted during the SELECTION of the study sample

Question 3

3. What is Information Bias?

Answer 3

• Systematic Errors arise from the mistakes conducted
  during the measurement of the key study variables

KEY STUDY VARIABLES:
- Exposure
- Outcome

Question 4

4. What leads to the rise in Information Bias?

Answer 4

• wrong and inaccurate assessments of either the:
  
  • exposure variables
  • or the outcome variables

THESE KINDS OF MISTAKES CAN ARISE EITHER:
- unintentionally from the researcher’s part
- intentionally or unintentionally from the participant’s
part

Question 5

5. What is Instrument Bias?

Answer 5

• this is when the instrument is faulty
• this is an error on the Researcher’s Part

Question 6

6. What is the title we give to any Information bias that arises from the Researcher’s Actions?

Answer 6

• Assessor Bias

Question 7

7. What are the 5 types of Assessor Bias?

Answer 7

1. A WRONG AND INACCURATE DIAGNOSIS
   
   • this is due to a Clinical Error
2. WHEN THE RESEARCHERS ARE NOT “BLINDED”
   
   • they are aware of the exposure or outcome status of
     the participants
3. WRONG AND INACCURATE MEASUREMENTS
   
   • can be a result of faulty instruments and machines
4. WRONG AND INACCURATE MEASUREMENTS
   - can be due to poor training of the assessor
5. MISTAKES CAN HAPPEN DURING
   - the recording of the data
   - and the transferring of the data from paper into
   electrical form

Question 8

8. What is the title we give to any Information bias that arises from the Participant’s Actions?

Answer 8

• Misreporting
• it is mostly called Response Bias

Question 9

9. What are the 5 types of Misreporting Bias?

Answer 9

1. WRONG AND INACCURATE ANSWERS
   - from the participants
   - this is due to the misinterpretation of the question
2. WRONG AND INACCURATE ANSWERS
   - from the participants
   - this is due to a sensitive issue relating to the question
3. WRONG AND INACCURATE ANSWERS
   - from the participants
   - this is due to poor recall
   - this is known as Recall Bias
4. WRONG AND INACCURATE ANSWERS
   • can come from the participants intentionally

Question 10

10. What are the 6 types of Information Bias?

Answer 10

1. Recall Bias
2. Interviewer Bias
3. Observer Bias
4. Hawthorne Effect
5. Surveillance Bias
6. Misclassification Bias

Question 11

11. What is Recall Bias?

Answer 11

THERE ARE PARTICIPANTS THAT HAVE:
- a particular outcome or exposure
- they may remember events more clearly
- this may amplify their recollections

Question 12

12. In which cases is Recall Bias very common?

Answer 12

• it is very common in case-control studies

Question 13

13. What leads to the Primary Differences that are seen in Recall Bias?

Answer 13

THE PRIMARY DIFFERENCE ARISES MORE FROM:
- under-reporting of exposures in the control group
- rather than from over reporting in the case group

Question 14

14. What is Interviewer Bias?

Answer 14

A RESEARCHER’S KNOWLEDGE MAY INFLUENCE:
- the structure of the questions
- and the manner of presentation

THIS COULD INFLUENCE THE RESPONSES FROM THE INDIVIDUALS:
- this is applicable to any study design
- this is especially true if the interviewer is not blinded
to the exposures of the individuals

Question 15

15. What is Observer Bias?

Answer 15

THE RESEARCHERS
- may have preconceived expectations of what they
should find in the examination

THIS IS ESPECIALLY TRUE:
- if they are not blinded to exposures or medical history

Question 16

16. What is the Hawthorne Effect?

Answer 16

• the participants will act differently if they know that
  they are being watched

Question 17

17. What is the Surveillance Bias?

Answer 17

• the group with the known exposure or outcome may
  be followed more closely
• this group can be followed longer than the comparison
  group
• this is an example of Researcher’s Bias

Question 18

18. What is the Misclassification Bias?

Answer 18

THESE ARE THE ERRORS THAT ARE MADE:
- in either classifying the disease
- or in classifying the exposure status (instrument)

Question 19

19. The errors in which fields of Research can lead to Misclassification Bias?

Answer 19

MEASUREMENT ERRORS IN THE ASCERTAINMENT OF:
- exposure
- outcomes (diseases)
- covariates (cofounders)

Question 20

20. List some examples of Misclassification Biases?

Answer 20

• using cases that are incorrectly classified as controls
• the controls are incorrectly classified as cases
• the exposed groups are incorrectly classified as the
  non-exposed groups
• the non-exposed groups are incorrectly classified as
  the exposed groups

Question 21

21. What are four ways to minimise Bias in Research?

Answer 21

1. BE PURPOSEFUL IN THE STUDY DESIGN
   - this minimises the chance for bias
   - EG: using more than one control group
2. DEFINE A PRIORITY
   - define who is considered to be a case
   - define what constitutes to an exposure
   - ensure that there is no overlap
3. DEFINE THE CATEGORIES WITHIN THE GROUPS
   CLEARLY
   
   • such as age groups
     - collections of person years
4. SET UP STRUCT GUIDELINES FOR DATA COLLECTION

Question 22

22. Define methods in which we can set up strict guidelines for Data Collection.

Answer 22

• the observers should be trained to obtain data in the
  same fashion
• the interviewers should be trained to obtain data in the
  same fashion
• use more than one observer or interviewer
• do not use too many observers or interviewers
• this would make it difficult to train them all in an
  identical manner
• Optimise the questionnaire