Skip to main content
    Path to Science

    Thinking Dictionary

    Blinding

    What it is

    Blinding means keeping people in a study from knowing who got the real treatment and who got the comparison. If nobody knows, hopes and expectations can't tilt the results.

    It isn't about hiding anyone's name. It's about hiding which group each person is in. Several different people can be blinded:¹

    • the participants, who receive the treatment,
    • the doctors or carers who give the treatment,
    • the people who measure the results, often called assessors, and
    • the people who analyse the data.

    You will often see the term double-blind. It means more than one of these groups was kept in the dark. The term alone doesn't tell you which groups, so it's best when a study report says exactly who was blinded.

    The Placebo effect and Randomized controlled trial entries mention blinding briefly. This entry explains who can be blinded, and how much it matters.

    Why it matters

    People see what they expect to see. That includes doctors and researchers, not just patients. Confirmation bias is part of the reason. If the person measuring a result knows who got the new treatment, their hopes can nudge the measurement, often without them realising.

    What the evidence shows

    A trial with two sets of eyes. In one trial of treatments for multiple sclerosis, a disease of the nervous system, every patient was examined by two neurologists. One knew which treatment the patient had received. The other didn't. Only the neurologists who knew recorded an apparent benefit from one of the treatments. The blinded neurologists didn't. Blinding stopped the trial from reaching a false conclusion.²

    Across many trials. A review looked at 21 trials that measured the same outcome twice: once by assessors who knew the treatment, and once by assessors who didn't. On average, the assessors who knew made the treatment look about a third more effective. That happened even though they judged only a small share of patients differently, a median of 3%.³

    Where it matters most. A study of more than 1,300 trials found that a lack of blinding exaggerated the benefits of treatments when outcomes were subjective, meaning they depend on someone's judgement, like pain scores. When outcomes were objective, like death, there was little sign of bias.⁴

    When blinding isn't possible

    Some things can't be hidden. You can't easily hide whether someone had an operation or took a pill, or whether they did an exercise programme. In those cases, blinding the assessors still helps. They can measure the results without knowing who had which treatment.

    And a study without blinding isn't automatically worthless. It depends on what was measured. Objective outcomes are much less affected.⁴

    Blinding was also used in a famous test of astrology, described in the Falsifiability entry, and in the SAMSON statin trial, described in the Nocebo effect entry.

    In everyday life

    • Taste tests. If you know which drink is the famous brand, that knowledge could colour your verdict. The Try it below lets you check.
    • Reading a study. Was it blinded? And who exactly was blinded? These are good questions to ask about any trial.

    The common trap

    The first trap is thinking only patients need to be blinded. In the multiple sclerosis trial, it was the assessors who made the difference.²

    The second trap is thinking all outcomes need blinding equally. Pain scores and other judgements are much more easily swayed than outcomes like death.⁴

    Some ways out:

    • When you read about a trial, check who was blinded.
    • Be most cautious about unblinded studies with subjective outcomes.
    • Don't dismiss an unblinded study with clear, objective outcomes too quickly.

    Try it

    Do a blind taste test. Ask a friend to pour two drinks, such as a supermarket's own-brand cola and a well-known brand, into cups labelled A and B. You shouldn't know which is which. Taste both and rate them. Then do it again, this time knowing which is which. Did knowing the brand change your ratings?

    Related concepts

    Placebo effect · Randomized controlled trial · Nocebo effect · Confirmation bias · Falsifiability

    Sources

    1. Schulz, K. F., & Grimes, D. A. (2002). Blinding in randomised trials: Hiding who got what. The Lancet, 359(9307), 696–700. https://doi.org/10.1016/S0140-6736(02)07816-9

    2. Noseworthy, J. H., Ebers, G. C., Vandervoort, M. K., Farquhar, R. E., Yetisir, E., & Roberts, R. (1994). The impact of blinding on the results of a randomized, placebo-controlled multiple sclerosis clinical trial. Neurology, 44(1). https://doi.org/10.1212/WNL.44.1.16

    3. Hróbjartsson, A., Thomsen, A. S. S., Emanuelsson, F., Tendal, B., Hilden, J., Boutron, I., Ravaud, P., Brorson, S., et al. (2012). Observer bias in randomised clinical trials with binary outcomes: Systematic review of trials with both blinded and non-blinded outcome assessors. BMJ, 344, e1119. https://doi.org/10.1136/bmj.e1119

    4. Wood, L., Egger, M., Gluud, L. L., Schulz, K. F., Jüni, P., Altman, D. G., Gluud, C., Martin, R. M., et al. (2008). Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: Meta-epidemiological study. BMJ, 336(7644), 601–605. https://doi.org/10.1136/bmj.39465.451748.AD