Bias

Question 1

Is the study valid?

Answer 1

• -if the result of a study remains close to the truth then the study is said to have internal validity
• if the results of the study are applicable to the clinician seeking the evidence, then the study is said to have external validity

Question 2

Validity

Answer 2

• may be compromised by three factors
  1. random error (chance)
  2. systematic error (bias)
  3. confounding

Question 3

Random error

Answer 3

• due to the play of chance
• all studies are prone to random error
• the probability of this is assessed using statistical measures such as p-values and confidence intervals
• random error can be reduced by repeating/replicating the study elsewhere

Question 4

Systematic error

Answer 4

• bias
• error in the way we select our patients, measure our outcomes or analyse our data
• bias may result in inaccurate results
• this is predictable and repetition doesnt help

Question 5

Negative bias

Answer 5

-bias can be negative and yield estimates closer to the null value ( no difference between two groups)

Question 6

Positive bias

Answer 6

-higher magnitudes of estimates than the true ones

Question 7

When can bias occur?

Answer 7

1. Selection of participants- selection of bias
2. measurement of variables- measurement bias
3. analysis of data- analysis bias

Question 8

Selection bias

Answer 8

• selected groups in a study differ in important factors other than the study variables
• adequate randomisation helps
• Berkson bias
• Neyman bias
• Response bias
• Unmasking bias
• lead-time bias
• referral bias
• diagnostic purity bias
• membership bias

Question 9

Berkson bias

Answer 9

• admission rate bias
• results from a difference in the rates of admission of cases and controls due to influence of exposure
• e.g strength and ability to be well enough for trial

Question 10

Neyman Bias

Answer 10

• incidence-prevalence bias

- one must look for an association between a risk factor and incidence not prevalence

Question 11

Response bias

Answer 11

-persons who respond to an invitation to participate in a study differ systemically from those who do not respond

Question 12

Unmasking bias

Answer 12

-said to exist when a risk factor unmasks rather than causes an event which it is suspected to cause

Question 13

Lead-time bias

Answer 13

• lead time is defined as the difference in time between the date of diagnosis with screening and the date of diagnosis without screening
• if the lead time is not accounted for then the survival time should not be compared to an unscreened control group of patients

Question 14

Referral bias

Answer 14

-refers to variation in concentration of rare exposures or diseases between patients in primary and secondary care

Question 15

Diagnostic purity bias

Answer 15

-refers to the exclusion of comorbidities resulting in a non-representative sample

Question 16

Membership bias

Answer 16

-refers to case identification using members of patient’s organisations leading to systematic differences and non-representativeness
e,g local self-help group

Question 17

Measurement Bias

Answer 17

• when data are not collected in a uniform fashion
• e.g when cases are interviewed in person while controls are interviewed over the phone
• this type of bias is minimised by blinding

Question 18

Recall bias

Answer 18

• type of measurement bias
• subjects often recall risk factors differently depending on their disease status
• case control studies are particularly vulnerable to this type of bias

Question 19

Reporting bias

Answer 19

-results when a larger percentage of either case or control subjects are reluctant to report an exposure due to attitudes, perceptions or other concerns

Question 20

Observer bias

Answer 20

-can occur whenever a researcher either knowingly or unknowingly evaluates a variable depending on the status of the individual under study

Question 21

Surveillance bias

Answer 21

-disease may be better ascertained in a monitored population than in the general population

Question 22

Work up bias

Answer 22

• aka verification bias
• during assessment of validity of a diagnostic test, the execution of the gold standard test may be influenced by the results of the assessed new instrument
• reference test may be less frequently performed when the test result is negative

Question 23

Misclassification bias

Answer 23

• in extreme cases measurement bias may lead to misclassification
• cases may be misclassified as controls or ‘exposed group’ may be misclassified as ‘non-exposed’
• this misclassification amounts to bias only if it is differential ie. one sided
• errors in measurement instruments may lead to non-dfferential misclassification (both sides are affected equally)

Question 24

Desirability bias

Answer 24

-patients may choose socially desirable answers to provide during data collection, distorting the true picture- reporting bias

Question 25

Hawthorne effect

Answer 25

• refers to the observed respondents minimising perceived deviation form the norm
• occurs especially in cross-sectional surveys using questionnaires

Question 26

Analysis bias

Answer 26

• said to occur when participants change group (contamination bias) or are lost to follow up (attrition bias) during the study
• this can be minimised by doing ‘intention to treat analysis’.