Module 9 (Ch 10 and 11)

Question 1

what 3 things do you need to assess to determine the validity of a study?

Answer 1

• Bias
• Confounding
• Random Error

Question 2

________ is a systematic error that results in an incorrect (invalid) estimate of the measure of association

Answer 2

Bias is a systematic error that results in an incorrect (invalid) estimate of the measure of association

Question 3

Bias is primarily introduced by the investigator or study participants. True or False?

Answer 3

True

Question 4

How does bias affect association?

Answer 4

• Can create spurious association (bias away from the null)
• Can mask an association (bias towards the null)

Question 5

Bias occurs in the ________________ of a study; it can
be evaluated but not fixed in the ________ phase

Answer 5

Bias occurs in the design and conduct of a study; it can
be evaluated but not fixed in the analysis phase

Question 6

When the study population doesn’t represent the target population ____________ has occurred.

Answer 6

When the study population doesn’t represent the target population selection bias has occurred.

Question 7

When is selection bias most likely to occur?

Answer 7

Most likely to occur in case-control or
retrospective cohort, but can also occur in prospective cohort and experimental studies

Question 8

________ occurs when researcher doesn’t use same criteria for selection
controls as for selecting cases

Answer 8

Control selection bias occurs when researcher doesn’t use same criteria for selection
controls as for selecting cases.

The odds ratio from a study that suffers from selection bias will
incorrectly represent the relationship between exposure and disease
in the overall study population

Question 9

________ can occur if you have a case-control study that
doesn’t have high participation rates if the
decision to participate or not is related to the
exposure and the outcome

Answer 9

Self selection bias can occur if you have a case-control study that doesn’t have high participation rates if the
decision to participate or not is related to the
exposure and the outcome

Question 10

how to prevent self selection bias?

Answer 10

Of a population asked to participate in a
research study, you want at least 80% of
them to do so

Question 11

________ occurs due to differential surveillance,
diagnosis, or referral of cases

Answer 11

Surveillance bias occurs due to differential surveillance,
diagnosis, or referral of cases

Occurs most often in case-control studies
through hospitals and medical clinics. If exposed people have more clinic visits, then it is more likely that subclinical cases of the
outcome of interest will be identified in those
people.

Question 12

What are ways that bias can occur in cohort studies?

Answer 12

• The selection of exposed and
  unexposed people is related to their
  health outcome
• There is loss to follow-up

Question 13

What is non-differential loss to follow up?

Answer 13

This is when you lose study participants that have the disease or exposure, but it’s not only related to one group (doesn’t just occur in the exposed group or the unexposed group)

Question 14

What is differential loss to follow up?

Answer 14

This occurs when the loss is related to both the exposure and the disease

• This can bias the relative and absolute measures of association either toward or away from an association, depending on the affected group

Question 15

How can you avoid selection bias?

Answer 15

While designing the study:
* Use the same enrollment criteria
* Obtain all relevant subject records
* Obtain high participation rates
* Take into account diagnostic and referral patterns of disease

Once bias has occurred, nothing can be done to fix it