module 5 lec 26 information bias

what is information bias

what is information bias

results from systemic differences in the way data on exposure or outcome are obtained from the various study groups

2 ways data can be collected

by participants - self-report
collected or measured by someone else

how can measurement error occur

participants may provide inaccurate responses (subjective)
data is collected incorrectly/inaccurately (objective)

measurement error

can be ransom or systematic

measurement error in cross-sectional studies

could over or under estimate prevalence

misclassification

non-differential - not different between groups
differential - different between groups

non-differential misclassification

when measurement error and any resulting misclassification occur equally in all groups being compared

differential misclassification examples

interviewers asking different questions to the different groups
in case-control studies people may more accurately recall events when they are a case

differential misclassification in case control studies

susceptible to recall bias

recall bias

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

minimising recall bias

objective measures, validate self reported measures with other information
memory aids

differential misclassification in cohort studies

misclassification of exposures and outcomes - have both exposure and outcome been correctly classified

interviewer/observer bias

if classification depends on outcome
if classification of outcome depends on exposure
e.g. interviewer examining outcomes differently for different groups

minimising interviewer/observer bias

clearly defined study protocol measures
structured questionnaire and standard prompts
training of interviewers
blinding

bias in RCTs

if knowledge of treatment/exposure influences assessment of the outcome
- ensure there is blinding
if measures are undertaken differently for different treatment groups

how to minimise information bias - collecting information

validate survey instruments or using objective measures

measurement instruments minimising information bias

useing standard equipment and using calibrated equipment

minimising information bias - collecting info via interviewers

ensure blinding, use objective measures, use structured interviews and standardised prompts, training of interviewers

publication bias

tendency of authors to submit and organisations to publish etc to publish articles containing positive findings especially new results