Why all randomised controlled trials produce biased results.

Area of Science:

• Public Health Research
• Clinical Trial Methodology
• Evidence-Based Policy

Background:

• Randomised controlled trials (RCTs) are considered the gold standard for informing public health and social policy.
• They are typically perceived as providing the most rigorous evidence of treatment effectiveness, free from significant assumptions, biases, and limitations.

Purpose of the Study:

• To critically assess the hypothesis that RCTs provide unbiased evidence.
• To examine the 10 most cited RCT studies globally to identify potential biases and limitations.

Main Methods:

• Identification of the 10 highest-cited RCTs published in any journal up to June 2016.
• Systematic review of these selected RCTs to evaluate assumptions, biases, and limitations.

Main Results:

• The most influential RCTs demonstrate biased results, with poor distribution of participant traits affecting outcomes.
• Trials often neglect alternative factors influencing reported outcomes and may suffer from partial or complete unblinding.
• Novel assumptions, biases, and limitations in trial design, implementation, and analysis were identified.

Conclusions:

• RCTs are subject to significant assumptions, biases, and limitations not fully discussed in current literature.
• The study highlights that even highly cited RCTs inevitably produce bias due to inherent complexities in trial processes.
• There is a need for greater awareness among researchers and policymakers regarding these trial limitations and the integration of RCTs with other research methodologies.