**Core Concept:** Randomisation is a fundamental principle in clinical research, particularly in randomized controlled trials (RCTs), which aim to minimize bias and increase the internal validity of study results. In order to eliminate bias, randomisation ensures that subjects are allocated to study arms (intervention or control) at a 50:50 ratio, without any preferential assignment based on observable or unobservable characteristics.

**Why the Correct Answer is Right:**

Randomisation is useful to eliminate bias. Bias occurs when there is an unequal distribution of certain factors in the study groups, leading to inaccurate conclusions. By randomly assigning participants to either the intervention or control group, researchers can prevent the influence of confounding variables on the study results.

In this case, the correct answer refers to eliminating **confounding bias**, which occurs when an observed association between an exposure and an outcome is due to an unmeasured or unknown factor (confounder) that is associated with both the exposure and outcome. By randomly allocating participants, researchers can ensure that any observed association is due to the effect of the intervention itself, rather than an unmeasured confounder.

**Why Each Wrong Option is Incorrect:**

A. **Confounding bias**: This type of bias arises when an observed association between exposure and outcome is due to an unmeasured or unknown factor (confounder) that is associated with both exposure and outcome. However, this option is not mentioned in the question or correct answer, as confounding bias is the intended target of randomisation.

B. **Selection bias**: This type of bias occurs due to an unequal distribution of participants between groups based on observable characteristics. Randomisation helps to eliminate this type of bias, as it ensures an equal distribution of participants based on prespecified criteria.

C. **Performance bias**: This type of bias occurs when participants or those delivering the intervention are aware of which group the participant belongs to. Randomisation helps to eliminate this type of bias, as it ensures that both participants and treating healthcare professionals remain unaware of the treatment allocation, reducing the risk of performance bias.

D. **Detection bias**: This type of bias occurs when the outcome assessor is not blinded to the treatment allocation. Randomisation helps to eliminate this type of bias, as it ensures that the outcome assessor is unaware of the treatment allocation, reducing the risk of detection bias.

**Clinical Pearls:**

1. **Blinding**: Blinding refers to the concealment of information related to the treatment allocation from participants, healthcare professionals, or outcome assessors. By using randomisation, blinding can be achieved, thereby reducing the risk of bias.

2. **Equipoise**: In medical research, equipoise refers to the state of uncertainty about the best treatment option. Researchers should only initiate a study when they are equipoised about the treatment they are comparing. This helps ensure that the study results are meaningful and reliable.

3. **Random sequences**: Random sequences can be generated using various methods, such as computer-generated random numbers, block randomization, stratified randomization, or permuted blocks. Each method helps to minimize the risk of bias by ensuring that participants, healthcare professionals, and outcome assessors are unaware of the