Bias

Question 1

Non-random assignment to participate in a study group: bias?

Answer 1

Selection bias

Question 2

A study that looks only at inpatients: bias?

Answer 2

Berkson bias (subtype of selection bias)

Question 3

What kind of bias is risked in studying a disease with early mortality?

Answer 3

Loss to follow-up (subtype of selection bias)

Question 4

Disproportionate loss to follow-up btwn exposed and unexposed groups: bias?

Answer 4

Attrition bias (subtype of selection bias)

Question 5

What kind of bias arises when subjects can self-select to be in the study?

Answer 5

Volunteer bias: often the people are healthier than the general population. Could also lead to groups who need free health care, want money offered from the study, etc.

Question 6

What kind of bias commonly occurs in retrospective studies when patients who are now aware of a disorder have altered memory of risk factors, etc?

Answer 6

Recall bias

Question 7

What kind of bias arises when subjects in different groups are treated differently?

Answer 7

Procedure bias

Question 8

What kind of bias arises when the way that information is gathered distorts that information?

Answer 8

Measurement bias

Question 9

What kind of bias occurs when groups who know they’re being studied behave differently than they would normally?

Answer 9

Hawthorne bias

Question 10

What kind of bias arises when a researcher’s belief in the efficacy of a treatment changes the outcome of that treatment?

Answer 10

Observer-expectancy bias; aka Pygmalion effect

Question 11

What kind of bias arises when a factor is related to both the exposure and outcome BUT is not on the causal pathway?

Answer 11

Confounding

Question 12

What kind of bias happens when early detection is confused with increased survival?

Answer 12

Lead-time bias

Question 13

What type of bias arises from non-random sampling of a target population?

Answer 13

Sampling bias (subtype of selection bias)

Question 14

What type of bias occurs when case and control groups differ due to recruitment or admission policies?

Answer 14

Referral bias

Question 15

What type of bias occurs when the presence of a risk factor itself means that that population is more likely to be worked-up for the disease, increasing the chance that the disease is found?

Answer 15

Detection bias

Question 16

What type of bias arises when patients are non-randomly assigned to different arms of a study?

Answer 16

Allocation bias

Question 17

Severely ill cancer patients are more likely to enroll in experimental trials than moderately ill cancer patients are. This leads to results that wouldn’t really apply to patients with milder disease. What type of bias is this?

Answer 17

Sampling bias

Question 18

A smoker is more likely to go get CXRs to screen for lung cancer. What type of bias could this lead to?

Answer 18

Detection bias

Question 19

Pathologists who know that a patient has SLE are more likely to dx a renal bx as DPGN. What kind of bias is this?

Answer 19

Observer bias

Question 20

What kind of bias occurs when the investigator’s decision is affected by prior knowledge of the exposure status?

Answer 20

Observer bias

Question 21

How can you reduce selection bias?

Answer 21

Randomization

Question 22

How can you reduce recall bias?

Answer 22

Decrease time from exposure to follow-up

Question 23

How can you reduce measurement bias?

Answer 23

Blinded placebo control groups

Question 24

How can you reduce procedure bias?

Answer 24

Blinded placebo control groups

Question 25

How can you reduce observer-expectancy bias?

Answer 25

Blinded placebo control groups

Question 26

3 ways to reduce confounding bias?

Answer 26

Matching
Multiple/repeated studies
Crossover studies

Question 27

How can you reduce lead-time bias?

Answer 27

Measure “back-end” survival (adjust survival according to severity of disease at the time of dx)