bias

Question 1

what can be done about bias

Answer 1

selection and information bias can only be controlled in the design and data collection phases

Question 2

define bias

Answer 2

any systematic error in a study that results in an incorrect estimate of the association between exposure and risk of disease

Question 3

to minimise bias at study start what two things do researchers have to do

Answer 3

at the planning stages of a study investigators must identify potential sources of bias in the proposed study, identify possible ways to minimise these potential biases

Question 4

what is selection bias

Answer 4

when there is systematic differences between the people chosen and not chosen for a study, or when the study groups are selected inappropriately or using different criteria

Question 5

3 important considerations in selection bias

Answer 5

who agrees to be in the study, how did we recruit them, do they all remain in the study

Question 6

how do we minimise loss to follow up

Answer 6

create main and alternative means of contact at the start of the study.
maintain regular updates like study newsletters
make several attempts to contact

Question 7

what are the magical bias sentences?

Answer 7

the measure of association is biased numerically upward/downward towards the null. therefore bias has resulted in an over/under estimation of the association between X and Y

Question 8

what are the steps of answering the bias questions

Answer 8

1: draw the MoA equation
2: answer the following, who does it affect, how does it affect them, what does this mean for the MoA
3: draw the numberline with true and biased MoA
4: state the three magical sentences

Question 9

what is information bias

Answer 9

observation and information bias result from systematic differences in the way data on the exposure or outcome are obtained from the various study groups

Question 10

two ways measurement error can occur

Answer 10

participants provide inaccurate responses, forget old exposures or over/under estimate an exposure.
or data is collected incorrectly or inaccurately, problem with measuring device
* E.g. person collecting the data doesn’t follow the same procedure for all participants

Question 11

what effect may effect might measurement error have in descriptive or analytic study

Answer 11

in descriptive study it could over or under estimate the prevalence,
in an analytic it could lead to misclassification

Question 12

define differential and non differential misclassification

Answer 12

when measurement error and any resulting missclassification occurs equally or inequally in the groups being compared

Question 13

examples of differential misclassification in cohort, cross sectional and case control studies

Answer 13

Cross sectional: people with the outcome might report the exposure differently to those without the outcome
cohort study: an interviewer aware of the exposure status may ask different questions between groups
case control: interviewer may conduct interview differently if they know who they are interviewing. or cases may more accurately remember their past exposures

Question 14

what is recall bias

Answer 14

systematic error due to differences in accuracy or completeness of recall to memory of past events or experiences

Question 15

how can we minimise recall bias

Answer 15

objective measures, validate the self reported data with other information, provide memory aids like dates

Question 16

define interviewer bias

Answer 16

if interviewer/observer knew the exposure status and examined the outcome differently for those in the exposed group compared with those in the comparison group

Question 17

how can we limit interviewer bias?

Answer 17

structured questionaires, training of interviewer, blinding, clearly defined study protocol and measures

Question 18

how does one limit information bias

Answer 18

Collecting information from participants, Validated survey instruments, Validate using an objective measure
Measurement instruments, Use standardised equipment, Use calibrated equipment

Question 19

what is publication bias

Answer 19

the result of the tendancy of authors to submit, organisations to encourage, reviewers to approve and editors to publish articles containing positive or new findings

Question 20

explain why “ how are people recruited is important”

Answer 20

are the people recruited representative of the population

Question 21

explain why “does everyone agree to partake” is important

Answer 21

are there differences in the type of people who agree to be in the study or not

Question 22

explain why “loss to follow up” is important

Answer 22

are the reasons of not remaining in the study related to the exposure and outcome