Your Easy Guide to Clinical Studies

Randomisation is the act of randomly allocating study participants to different treatment groups (e.g. study intervention versus control group). The allocation is entirely by chance with no regard to potential preference of the researcher (i.e. SP-INV, Site-INV) or a participant’s first choice.

 

In order to decide on the study’s allocation sequence, the following factors must be defined:

• The allocation ratio (e.g. in a 1:1 ratio, half the participant are assigned to the intervention and the other half to the control group)
• Stratification factor(s). Stratification helps to ensure a balanced representation of a specific characteristics (i.e. the stratification factors) in different treatment groups (e.g.  age, co-morbidities).

 

By ensuring that the treatment groups are comparable with respect to important stratification factors, stratification helps minimize the risk of bias (i.e. systematic differences between the control and intervention groups that are unrelated to the intervention itself).

More

As a SP-INV, consult a statistician to discuss the randomisation process of your study. Aspects to consider include:

• The required allocation ratio
• Applicable stratification factor(s)

 

Stratification factor(s) are selected based on their relevance to the research question, and their potential to impact treatment outcome. These variables must be identified before the trial starts and formulated in the study protocol and statistical analysis plan. To avoid strata with few participants, the number of strata should be kept at a minimum. 

 

Example of a stratified randomization

• The study: assesses the effect of a new drug on cancer survival.
• Potential bias: Previous work has shown a strong correlation between cancer severity and survival. Hence, an un-balanced distribution of severe cancer cases between the control and the intervention group could be a potential bias, resulting in a higher mortality rate in the group with more severe cancer cases.
• Stratification factor: in order to minimize bias, the randomisation process will be stratified using cancer severity as a stratification factor.

SCTO Research Tools & Resources

• Randotools Package

Useful External Links

• Randomization Simulator - Help in deciding number of stratification factors

References

ICH GCP E6(R3) Guideline – see in particular:

• Glossary: Definition randomisation
• 2.11 Randomisation Procedures and Unblinding
• C3v Essential trial records
• Essential records table: Master randomisation list

ICH Topic E9 statistical Principles for Clinical Trials – see in particular

• 2.3.2 Randomisation

ICH Topic E8(R1) General considerations for clinical studies - see in particular

• 5.5 Methods to reduce bias

Publication PubMed – see in particular

• PMID: 20332511 David Moher et. al. CONSORT 2010 Explanation and Elaboration: Updated Guidelines for Reporting Parallel Group Randomised Trials.

Swiss Law

ClinO – see in particular article

• Art. 2b Definition of intervention