The Landscape and Lexicon of Blinding in Randomized Trials

• Randomized controlled trials
• Single-blind method
• Double-blind method
• Observer variation
• Selection bias

Blinding in medical research possesses a rich history spanning a couple of centuries (1). Most researchers and readers grasp its meaning. Unfortunately, beyond that general understanding lies confusion. In addition to terms such as “single blind” and “double blind” meaning different things to different people, some steadfastly refuse to use the term “blinding” and insist instead on the term “masking.” Others confuse blinding with other methodologic precautions, such as concealment of allocation during the process of creating comparison groups. Still others consider that randomization is of little use unless accompanied by “double-blinding,” thus revealing that they have not understood that these separate aspects of methodology address separate sources of bias.

A recent survey addressed whether the process historically termed “blinding” should be termed “masking” (2). The survey revealed a lack of accord on that question and inconsistencies concerning other blinding terminology. Although many resources address the lexicon of blinding, including clinical trial textbooks (3-5), clinical trial dictionaries (6, 7), and a recently released epidemiology dictionary (8), these sources do not entirely clear the lexicographic fog. Indeed, a recent study found that investigators, textbooks, and published articles all varied greatly in their interpretations of single-, double-, and triple-blinding (9). In other words, terminologic tangles abound with blinding. We delve into the landscape and lexicon of blinding in randomized trials in the hope of untangling some of that terminology.