Question

1. Investigators who us nonrandomized controls in clinical trials often argue that their controls are satisfactory because the distribution of prognostic factors in the control and experimental groups is similar before therapy. It is better to randomize because

1. A placebo effect s easier to detect in randomized trials
2. Randomization tends to magnify the differences between placebo and treatment groups
3. Many important prognostic factors that were not considered may lead to bias
4. It is easier to maintain blinding in randomized trails
5. Compliance is better in randomized trials

Answer 1

Problem statement: Why is it better to randomise control group during clinical trials.

Solution: The answer is option (d)

Explanation: In non-randomised control design for clinical trials, either the subjects (patients) assign themselves into a group which they would like to get into or researcher assigns them into a particular group.

Since in non-randomised design the subjects already know which category they belong (whether they belong to control group or treatment group) they suffer from performance bias. For example, if subjects in control group know they are not receiving a particular treatment or therapy, then their behaviour might be very different compared to subjects in "treatment" group, resulting in a may be not so desirable outcome for the experiment. In randomised control trial, subjects are slept into groups randomly without letting them know which group receives "treatment" and which is "control" group, Thus reducing performance bias which subjects might have displayed if they knew whether they were in "treatment" group or not. This process not letting subjects know whether they are receiving "treatment" or placebo ("control) is called blinding. The concept of blinding is easier to maintain in randomised trials.