Basic↦Statistic Methodology↦Study Design↦Randomization
What is it? Why is it important?
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
Stratification helps to ensure, that during the randomisation process, the different treatment groups remain comparable by having an equal representation of specific characteristics (e.g. age, co-morbidities).
By ensuring treatment groups are comparable in terms of stratification factors, the risk of bias is minimised (i.e. thereby avoiding having differences between the control and intervention group caused by something other than the intervention).
What do I need to do?
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. A given study can have 0, 1, or even more than 1 stratification factors.
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.
Where can I get help?
Your local CTU↧ can support you with experienced staff regarding this topic
Basel, Departement Klinische Forschung, CTU, dkf.unibas.ch
Lugano, Clinical Trials Unit, CTU-EOC, www.ctueoc.ch
Bern, Clinical Trials Unit, CTU, www.ctu.unibe.ch
Geneva, Clinical Research Center, CRC, crc.hug.ch
Lausanne, Clinical Research Center, CRC, www.chuv.ch
St. Gallen, Clinical Trials Unit, CTU, www.kssg.ch
Zürich, Clinical Trials Center, CTC, www.usz.ch
References
ICH Topic E9 statistical Principles for Clinical Trials – see in particular
- 2.3.2 Randomisation
ICH Topic E8(R1) on 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..