Chapter 13: Bias in Research

Question 1

What is a bias in research?

Answer 1

It is a form of systematic error in the research process that consistently shifts results in one direction.

Question 2

When can bias occur in research?

Answer 2

It can occur in all phases of research:
• Planning;
• Data collection and
intervention;
• Data analysis and publication.

Question 3

What is the consequence of bias in a study?

Answer 3

• A study loses validity according to the degree of bias.

Question 4

Why is it essential to understand how bias might influence a study’s results and conclusions?

Answer 4

To understand how bias might influence a study’s results and conclusions is essential for evidence-based practice.

Question 5

What are the biases that can occur in planning the research?

Answer 5

• Design bias

- Selection bias (Self-selection, convenient sample, non-response)

Question 6

What are the biases that can occur in data collection/intervention in a study?

Answer 6

• measure bias
• interviewer bias
• recall bias
• compliance bias
• attrition bias

Question 7

What are the biases that can occur in data analysis/publication in a study?

Answer 7

• confounding bias

- publication bias

Question 8

What are the two types of study designs?

Answer 8

• observational

- experimental

Question 9

What are the types of study designs that fall under observational studies?

Answer 9

• cross-sectional
• cohort (longitudinal)
• case-control (longitudinal)
• case-report series (longitudinal)

Question 10

What are the types of study designs that fall under experimental studies?

Answer 10

• Randomized controlled trials (RCT)

Question 11

What is design bias?

Answer 11

flawed study design

Question 12

How to avoid design bias?

Answer 12

Select the adequate study design for the research question.

Question 13

What type of study design should I use if I am interested in prevalence?

Answer 13

cross-sectional

Question 14

What type of study design should I use if I am interested in incidence?

Answer 14

cohort

Question 15

What type of study design should I use if I am interested in cause (in order of reliability)?

Answer 15

• cohort
• case-control
• cross sectional

Question 16

What type of study design should I use if I am interested in prognosis?

Answer 16

cohort

Question 17

What type of study design should I use if I am interested in treatment effect?

Answer 17

RCT

Question 18

What is selection bias and what are the types?

Answer 18

Occurs during identification of study population, when some individuals are more likely to be selected for the study than others.

• Self-selection (volunteer participation)
• Convenient Sample
• Non-response

Question 19

What is self-selection (volunteer participation) and convenient sample bias?

Answer 19

characteristics of those who volunteer (or are referred) to participate in a study are often different from the non-volunteers (not referred).

Question 20

What is a non-response bias?

Answer 20

those who do not respondent/participate often have characteristics from those who respondent/participate.
different

Question 21

How to avoid selection bias?

Answer 21

Randomly selection of the sample: each individual is chosen entirely by chance and each member of the study population has an equal chance of being included in the sample.

Question 22

How to avoid non-response bias?

Answer 22

• Provide incentives for completing survey
• Explain why survey is important
• Keep survey short

Question 23

What is measure bias? Give an example.

Answer 23

Occurs when the measurement tool is not sensitive enough to detect the exposure or outcome of interest.

Example: a study aims to assess vitamin D intake but the FFQ does not include all major sources of vitamin D.

Question 24

How to avoid measure bias?

Answer 24

Using validated measurement tools that have a high sensitivity and specificity to minimize bias

Question 25

What is interviewer bias? Give an example.

Answer 25

Occurs when there is a systematic difference between how information is asked, recorded, or interpreted. Interviewer may probe for certain risk factors. Example: A case-control study aims to verify if the consumption of dairy products increases the risk for respiratory allergy. If the interviewer is aware that the respondent is one of the “cases” (individuals with respiratory allergy) he/she may give more emphasis on the risk factor investigated, by asking "Are you sure you rarely consume dairy products?”, than when interviewing one of the “controls” (individuals without respiratory allergy).

Question 26

How to avoid interviewer-bias?

Answer 26

* Standardizing protocols for data collection * Training interviewers * Blinding the interviewers to the study hypothesis.

Question 27

What is recall bias? Give an example

Answer 27

Occurs when measuring the outcome or the exposure requires individuals to remember past events. Example: Diet assessment methods • Food Frequency Questionnaire (FFQ) • 24-hour recall • Dietary history

Question 28

How to avoid recall bias?

Answer 28

* Using a well-structured questionnaire that has been validated against direct measures. * Giving the participants enough time to reflect before answering. * Verifying the reported-data by using a reference criterion (e.g. medical records, data from another family member).

Question 29

What is compliance bias? Give an example

Answer 29

Occurs when some subjects do not adhere to the planned treatment (experimental studies). Example: A RCT is investigating the efficacy of a low-carb diet on weight loss. However, some subjects allocated in the intervention group did not follow the diet prescribed by the researchers.

Question 30

How to avoid compliance bias?

Answer 30

* Designing protocols that are simple and easy to follow and comply with. * Identifying subjects who are unable or unmotivated to comply (with the intervention) during the enrollment process. * Contacting subjects frequently to maintain interest and motivation.

Question 31

What is attrition bias?

Answer 31

Occurs when subjects who leave the study (drop- outs) differ significantly from those that remain (common in cohorts and RCT).

Question 32

How to avoid attrition bias?

Answer 32

* Using protocols to contact participants by email, post, telephone, etc. * Providing incentives (monetary, gifts). * Keeping the interviews short.

Question 33

What is confounding bias?

Answer 33

Occurs when an estimated association between exposure and outcome is biased by a third factor associated with both exposure and outcome.

Question 34

How to avoid confounding bias?

Answer 34

* Identified confounders can be controlled with study design (case control or randomization) and during data analysis (regression analysis). * Unidentified confounders can only be controlled with randomization during participants selection and enrollment.

Question 35

What is publication bias?

Answer 35

Occurs when researchers and research sponsors are unwilling to publish unfavorable results. Positive results are more likely to be reported and accepted for publication in scientific journals, than negative results.