Using a pre-defined randomisation list

There are several advantages to using a pre-defined list with our simple randomisation service. Using a list allows you to:

  • Stratify e.g. by age group or centre
  • Have more than two treatment groups
  • Name your treatment groups, e.g. Placebo and Aspirin rather than A and B
  • Randomise to a code, e.g. 1001, 1002,… rather than A and B. This makes the system more suited to blinded trials where the link between the codes and group is kept secret
  • Alter the allocation ratio from 1:1, e.g. can be 2:1

Cost

Using a randomisation list incurs a charge - see pricing for details.

When you create a trial with a randomisation list it will initially be limited to 5 randomisations. Contact us to arrange payment and extend this limit.

How do I set up a trial using a randomisation list?

When you create your trial you should upload the randomisation list using the file upload button on the trial creation form.

Your list must be a CSV file in the format described below. We have a randomisation list creation tool that you can use to generate a list in the correct format, but as long as the list matches the specified format you can generate it however you want.

CSV file format

Line 1 of the CSV file should be the header:

block identifier, block size, sequence within block, treatment, [strata 1, strata 2, ...] [code]

The remaining lines should be the randomisations and should contain the fields:

block identifier
Block number (integer number)
block size
Size of block (integer number)
sequence within block
Sequence number within the current block (integer number)
treatment
Name of treatment group or randomisation code (string)
strata 1, strata 2, …
Optional stratification variables (string). The header is used as the stratification label so you should capitalise, e.g. Age-group.
code
Optional column containing unique randomisation code (string).

About randomisation codes

If your trial is double-blind, you should use randomisation codes. You can put these into the treatment column of your CSV or use a separate code column. Either way, on randomisation the user will see the randomisation code and not the underlying treatment group. You will need to keep the file linking the codes to the treatment groups to allow the results to be analysed when the trial is unblinded.

Warning! Even if you upload both treatment groups and codes, we do not store the treatment groups - only the codes! Make sure you keep a copy of your original list.

Example

This is a randomisation list with a block size of 4, 1:1 randomisation to Intervention or Placebo, and stratification by age-group (under 50 vs. 50 or over) and sex (male vs. female).

block identifier, block size, sequence within block, treatment, Age-group, Sex
1, 4, 1, Intervention, Under 50, Male
1, 4, 2, Placebo, Under 50, Male
1, 4, 3, Intervention, Under 50, Male
1, 4, 4, Placebo, Under 50, Male
2, 4, 1, Placebo, Under 50, Female
2, 4, 2, Intervention, Under 50, Female
2, 4, 3, Placebo, Under 50, Female
2, 4, 4, Intervention, Under 50, Female
3, 4, 1, Intervention, 50 or over, Male
3, 4, 2, Intervention, 50 or over, Male
3, 4, 3, Placebo, 50 or over, Male
3, 4, 4, Placebo, 50 or over, Male
4, 4, 1, Intervention, 50 or over, Female
4, 4, 2, Placebo, 50 or over, Female
4, 4, 3, Placebo, 50 or over, Female
4, 4, 4, Intervention, 50 or over, Female
Page updated 28 Aug 2014 09:17