Blinding refers to concealing or masking group allocation in controlled trials. For example, participants are ignorant of which test or intervention they are subject to: “single blind”. In “double blind” studies, the researchers are blind to which test or intervention is being offered. Blinding is also used as in “blind ratings” when a rater is unaware of things that might bias their ratings and in “blinded” analyses the statistician doing the analyses of the data does not know which group in the study is which (this last is sometimes termed “triple blind”).
We don’t think this has anything to do with the sayings “blinding, as in ‘science and blinding'” or “blinded by science”!
Detail #
Blinding is hugely important in research design where it can be achieved as we know that human expectations bias even what we think we see and expectations in participants in trials change outcomes, even mortality outcomes.
Blinding is very important where it is genuinely possible to use a placebo or other control intervention as is quite often the case in pharmacological trials; however, it seems largely tokenistic if that primary blinding can’t be achieved.
Try also … #
Trials
Causality
Hierarchy of evidence
Placebo effect
Nocebo
Randomised Controlled Trials (RCTs)
Randomisation
Double Blind Controlled Trials (DBRCTs)
Blinding/blindness
NICE
Cochrane
ESTs (Empirically Supported Treatments)
Chapters #
Chapter 10.
Further reading #
Nothing here yet!
