0

Nadiyah Ali Alshehri

King Abdulaziz Medical City, Saudi Arabia

Title: Sources of bias in randomised controlled trials

Abstract

Despite the randomized controlled trial being, the best research method it can lead to biased results. Care must be taken to avoid as many biases as possible. Forms of bias include Subversion bias, technical bias, attrition bias, consent bias, ascertainment bias, and dilution bias. One of the main strengths of the RCT design in health care is its ability to mitigate selection bias. When done correctly, randomization can ensure that study groups are as homogeneous as possible at the outset, allowing investigators to isolate and measure the effect of the interventions being studied. No other design gives us the ability to balance unknown predictors at the outset. Random allocation, however, does not shield RCTs from other types of biases. Most biases associated with RCTs can be accounted for by common sense. However, in recent years, significant research efforts have focused on RCTs instead of the tool of research. These efforts are typically focused on generating empirical evidence to enhance the design, report, disseminate, and use of RCTs across the health care system. The results of these studies have confirmed the prevalence of many types of biases across the lifespan of RCTs. While randomization increases the likelihood of a study being free from allocation bias, it has no impact on other important biases.

Biography

To be Added..