7.2: Information Bias (Measurement Error) ✅

Question 1

Systematic error

Answer 1

Selection bias

Information bias

Question 2

Information bias

Answer 2

Systematic error arising from mistakes during measurement of key study variables

-from wrong/inaccurate assessment of exposure/outcome

-researcher part (unintentionally) or from participants (unintentionally/intentionally)

-instrument bias (fault of the instrument)

Question 3

Assessor bias

Answer 3

Wrong diagnosis due to clinical error

“Not blinded” to exposure/outcome status of participants

Inaccurate measurements due to faulty instruments

Poor training of assessor

Mistakes during recording of the data and transferring data

Question 4

Response bias

Answer 4

Wrong/inaccurate answers from participants:
-due to misinterpretation
-intentionally

Wrong/inaccurate answers due to :
-poor recall
-sensitive issue

Question 5

Recall bias

Answer 5

Participants with a particular outcome or exposure may remember events more clearly or amplify their recollections

Very common in case-control studies
-the primary difference arises more from under-reporting of exposures in the control group
(rather than over-reporting in the case group)

Question 6

Interviewer bias

Answer 6

A researcher’s knowledge may influence:
-the structure of questions and
-the manner of presentation

Which may influence responders (in any study design)

Question 7

Observer bias

Answer 7

Researchers may have preconceived expectations of what they should find in an examination

Question 8

Hawthorne effect

Answer 8

Participants act differently if they know they are being watched

Question 9

Surveillance bias

Answer 9

The group with the known exposure or outcome may be followed more closely or longer than the comparison group

Question 10

Misclassification bias

Answer 10

Errors made in classifying either disease or exposure status

Occur in exposure, outcome and covariates

Question 11

Minimizing bias

Answer 11

Be purposeful in the study design to minimise the chance for bias
-e.g. Use more than one control group

Define a priori who is a case or what constitutes exposure so that there is no overlap
-define categories within groups clearly

Set up strict guidelines for data collection
-train observers or interviewers to obtain data in the same fashion
-optimize questionnaire

Question 12

How does information bias affect study results

Answer 12

To investigate an association you have to measure the factors accurately

Information bias can occur in the exposure and outcome so association between them will be distorted

More likely to arise from participants or through medical records

Affects mainly studies that rely on self-reports
- Outcomes can be double-checked
- Exposures require using reliable instruments

Question 13

What must we always assume if a study solely relies on self report?

Answer 13

There is information bias

This compromises the validity of the study results

Question 14

Validity

Answer 14

The extents to which an assessment tool measures accurately what is intended to measure

Question 15

Criterion validity

Answer 15

Most common used in medical research

Results from assessment tool are compared to the results of an established assessment tool

Question 16

Reliability

Answer 16

Aka reproducibility or repeatability

Overall consistency of a measure as regards to producing the same results as when administered under the same conditions in the same group of people

Question 17

Reliability type

Answer 17

Inter-observer reliability

Intra-observer tool

Question 18

Inter-observer reliability

Answer 18

The degree of agreement between the results when:

two or more researchers (observers) administer the assessment tool :
-on the same people
-under the same conditions

Question 19

Intra-observer reliability

Answer 19

Describes the agreement between results when the assessment tool is used by the same researcher (observer) on 2 or more occasions (under the same conditions and in the same test population)

Question 20

How does bias affect study results

Answer 20

False positive and false negative findings can arise in situations where information bias affects the outcome of interest

If both exposure and outcome are affected by information bias, then results are expected to be particularly biased (erroneus)

Also, all sorts of bias (including selection bias) as well random error can give rise to false positive and false negative findings

Question 21

Internally valid

Answer 21

Findings of the study were not due to one of the 3 sources of error

The conclusions reached are likely to be correct for circumstance

(Systematic error, assessor bias and response bias)

Question 22

What must you not do in the goal of generalization?

Answer 22

Do not compromise internal validity

As an invalid result cannot be generalized