IMED2003 Study Guide - Midterm Guide: Smoking Age, Selection Bias, Effect Size
12 Jan 2019
School
Department
Course
Professor
Document Summary
System for assigning study subjects to treatment groups that are independent of the individual participants" characteristics. Each participant should have an equal chance of being assigned to any group. Removes systematic differences between the treatment and control groups. Since it produces comparable groups, it may eliminate known and unknown confounders. Groups should have a similar spread of other characteristics. Once a patient is randomised they should be analysed in the treatment group to which they were randomised to. Even if they actually received the alternative treatment or do not comply with the treatment. Participants who adhere may be different from those who don"t in ways that are related to the outcome. Effect size will reflect what likely to be seen in real-world" practice. Usually results in an underestimation of the treatment effect. Need to be developed objectively and prior to recruitment. Randomization removes the bias in allocation to study groups. Blinding and use of placebos reduces information bias.
