Page 53 - ESHRE2019
P. 53
Sources of bias
Selection Performance Detection Attrition Reporting
Target population Allocation
Intervention group
Outcome assessment
Control group
Outcome assessment
Blinding of participants & personnel
3. Blinding: Participants and personnel
• If not blinded consider the actual risk of bias on affecting outcomes • Pre-define assessment criteria
• Different treatment of the groups by staff/doctors
• Different participant expectations, e.g. resentful demoralisation
• Important: just because you can’t blind doesn’t mean that it doesn’t lead to bias!
• Do not take terms like “single blind” “double blind” to mean anything: who was actually blinded? The doctor? The janitor?
• Most inconsistent/subjective domain (66% inconsistent = same as chance)
Publication of study outcomes
48
PRECONGRESS COURSE 17 I VIENNA, AUSTRIA – 23 JUNE 2019 51
