The highly trained, nuanced, capable scientist you are describing is unfortunately only a small part of the bigger "system" I referred to, which also involves regulators, insurers, phone line operators, hospital staff etc (as described in the article). I don't think that the person who designs and evaluates the trial can control or correct for the biases introduced by all these additional factors. The only thing they can hope for is that the many errors average each other out, but the situation described by the article looks as though all errors point in the same direction...