bias penelitian

8/10/2019 Bias Penelitian

1/27

Evaluating the Role of Bias


2/27

Bias slide #2

Definition of Bias

Bias is a systematic error that results in anincorrect (invalid) estimate of the measure ofassociation

A. Bias can create spurious association whenthere really is none (bias away from the null)

B. Bias can mask an association when there

really is one (bias towards the null)

C. Bias is primarily introduced by the

investigator or study participants


3/27

Bias slide #3

Definition of Bias(contd)

D. Bias does not mean that the investigator isprejudiced.

E. Bias can arise in all study types: experimental,

cohort, case-control

F. Bias occurs in the design and conduct of a

study. It can be evaluated but not fixed in the

analysis phase.

G. Two main types of bias are selection and

observation bias.


4/27

Bias slide #4

Selection Bias

A. Results from procedures used to select

subjects into a study that lead to a result

different from what would have been obtained

from the entire population targeted for study

B. Most likely to occur in case-control or

retrospective cohort because exposure andoutcome have occurred at time of study

selection


5/27

Bias slide #5

Selection Bias in a Case-Control Study

A. Occurs when controls or cases are more

(or less) likely to be included in study if

they have been exposed -- that is,

inclusion in study is not independent of

exposure


6/27

Bias slide #6

Selection Bias in a Case-Control Study

B. Result: Relationship between exposure and

disease observed among study participants is

different from relationship between exposure and

disease in individuals who would have been

eligible but were not included.

The odds ratio from a study that suffers from

selection bias will incorrectly represent therelationship between exposure and disease in the

overall study population


7/27

Question: Do PAP smears prevent cervical cancer? Casesdiagnosed at a city hospital. Controls randomly sampledfrom household in same city by canvassing the

neighborhood on foot. True relationship:CervicalCancerCases

Controls

Had PAPsmear

100 150

Did nothave PAP

smear

150 100

Total 250 250

OR = (100)(100) / (150)(150) = .44 There is a 54% reduced risk of

cervical cancer among women who had PAP smears vs. women

who did not. (40% of cases had PAP smears versus 60% of controls)


8/27

Bias slide #8

Recall: Cases from the hospital and controls

come from the neighborhood around the

hospital.

Now for the bias: Only controls who were at

home at the time the researchers camearound to recruit for the study were actually

included in the study. Women at home were

less likely to work and less likely to have

regular checkups and PAP smears.

Therefore, being included in the study as a

control is not independent of the exposure.


9/27

Bias slide #9

Cervical

CancerCases

Controls

Had PAP

smear

100 100

Did not

have PAPsmear

150 150

Total 250 250

OR = (100)(150) / (150)(100) = 1.0

There is no association between PAP smears and the

risk of cervical cancer. Here, 40% of cases and 40% of

controls had PAP smears.

The resulting data are as follows:


10/27

Bias slide #10

Ramifications of using women who were at

home during the day as controls:

These women were not representative of

the whole study population that produced

the cases. They did not accuratelyrepresent the distribution of exposure in

the study population that produced the

cases, and so they gave a biased estimateof the association.


11/27

Bias slide #11

Selection Bias in a Cohort Study

Selection bias occurs when selection of

exposed and unexposed subjects is notindependent of outcome (so, it can only

occur in a retrospective cohort study)


12/27

Bias slide #12

Selection Bias in a Cohort Study

Example:

A retrospective study of an occupational exposure

and a disease in a factory setting.

The exposed and unexposed groups are enrolled on

the basis of prior employment records.The records are old, and many are lost, so the

complete cohort working in the plant is not available

for study.

If people who did not develop disease and were

exposed were more likely to have their records lost,

then there will be an overestimate of association

between the exposure and the disease.


13/27

Bias slide #13

Diseased Non-

diseased

Total

Exposed 50 950 1000

Un-

exposed

50 950 1000

RR = (50/1000) / (50/1000) = 1.00

True relationship, if all records were

available


14/27

Bias slide #14

Diseased Non-diseased Total

xposed 50 750 800

n-exposed 50 950 1000

RR = (50/800) / (50/1000) = 1.25

If more records were lost in this category (exposed

subjects who did not get the disease), the bias would be

even greater.

200 records were lost, all among exposed who

did not get the disease


15/27

Bias slide #15

Selection Bias: What are the solutions?

Little or nothing can be done to fix this biasonce it has occurred.

You need to avoid it when you design andconduct the study by, for example, using thesame criteria for selecting cases andcontrols, obtaining all relevant subject

records, obtaining high participation rates,and taking in account diagnostic and referralpatterns of disease.


16/27

Bias slide #16

Observation Bias

An error that arises from systematicdifferences in the way information onexposure or disease is obtained from the

study groups

Results in participants who areincorrectly classified as either exposed orunexposed or as diseased or notdiseased


17/27

Bias slide #17

Observation Bias

Occurs after the subjects have entered

the study

Several types of observation bias: recall

bias, interviewer bias, loss to follow up,

and differential and non-differentialmisclassification


18/27

Bias slide #18

Observation Bias

Recall bias - People with disease

remember or report exposures differently

(more or less accurately) than those

without disease.

Can result in over- or under-estimate ofmeasure of association.


19/27

Bias slide #19

Observation Bias

Solutions: Use controls who are

themselves sick; use standardized

questionnaires that obtain complete

information, mask subjects to study

hypothesis


20/27

Classic recall bias:Cases underreport exposure

TRUTH OBSERVEDSTUDY DATA

Case Control Case Control

Exposed 40 20 Exposed 30 20

Unexposed 60 80 Unexposed 70 80

Total 100 100 100 100Odds Ratio:

2.7

Odds Ratio:

1.7


21/27

Bias slide #21

Observation Bias

Interviewer bias- Systematic differencein soliciting, recording, interpretinginformation.

Can occur whenever exposureinformation is sought when outcome isknown (as in case-control), or whenoutcome information is sought whenexposure is known (as in cohort study).


22/27

Bias slide #22

Observation Bias

Interviewer bias

Solutions: mask interviewers to study

hypothesis and disease or exposure

status of subjects, use standardized

questionnaires or standardized methods

of outcome (or exposure) ascertainment


23/27

Bias slide #23

Observation Bias Loss to follow up - A concern in cohort and

experimental studies if people who are lost tofollow up differ from those that remain in the

study.

Bias results if subjects lost differ from those that

remain with respect to both the outcome and

exposure.

Solution: Since that information cannot be known,

you must achieve high and equal rates of follow

up for the exposed and unexposed groups.


24/27

Bias slide #24

Misclassification - Subjects exposure or

disease status is erroneously classified.

Two types of misclassification: non-differential

and differential. We will cover only the more

common form: non-differential

misclassification.

Observation Bias


25/27

Bias slide #25

Non-differential misclassification

Inaccuracies with respect to disease

classification are independent of exposure.

Or, inaccuracies with respect to exposure are

independent of disease. Will bias towards the

null if the exposure is has two categories.

Non-differential misclassification makes the

groups more similar.

Observation Bias


26/27

Bias slide #26

Misclassification-

Example: Study of vaginal spermicides and

congenital disorders (Jick et al., 1981).

Solutions: Use multiple measurements,

most accurate source of information

Observation Bias


27/27

Bias slide #27

When interpreting study results,

ask yourself these questions

Given conditions of the study, could bias

have occurred?

Is bias actually present? Are consequences of the bias large enough

to distort the measure of association in an

important way? Which direction is the distortion?Is it

towards the null or away from the null?

bias penelitian

Documents