What is the difference between a randomisation and code list?

A randomisation list is a way of choosing the next treatment group. A code list links treatment groups to kit codes. Both lists may be present in a blinded randomisation system, although the randomisation list can be replaced by another randomisation method (such as minimisation).

Here is a simple example of how the randomisation and code lists are used in a blinded trial:

1

One randomisation has already been carried out (to treatment A kit code JS6). Now we are going to randomise the next participant

2

The next treatment is taken from the randomisation list – in this case it is treatment A

3

Now we need to find a kit code in the code list that matches treatment A. This kit code could come from anywhere in the list, but we usually limit the choice according to a dispensing policy so that we don't jump around the code list too much. This chooses an unused kit at random from the lowest available kit block (kit block 1 in this example). In this case we choose kit code ER5 as it's the only available code left in kit block 1 for treatment A.

4

Later we want to randomise another participant. The next allocation in the randomisation list is treatment B. We choose a matching kit code at random from kit block 1. In this case we are going to choose kit WA6 rather than UR8.

Notice that we use the randomisation list in sequence order (from top to bottom), but we jump around at random within kit blocks in the code list. We still go through the code list in kit block order though (from low kit blocks to high). If the randomisation list is stratified, we use different parts of the list depending on the strata values of the participant we are randomising. But we still use the randomisation list in sequence order within strata. The code list is not stratified, but different sections of the code list can be allocated to different sites to assist with the logistics of shipping pre-labelled kits to sites.