Table 4 Summary of concerns regarding randomisation methodology

Simple randomisation

Each allocation is determined entirely randomly. In a two-arm trial with 1:1 allocation this is equivalent to flipping a coin for each allocation

Stratified block randomisation

Stratified randomisation involves separate lists being produced for different factors of a variable to obtain balance. The lists are typically compromised of ‘blocks’ of varying sizes that ensure the target allocation ratio is maintained

For example, in a two-arm trial with 1:1 allocation to groups A and B, a block of size 6 would be made up of a sequence of three As and three Bs in a random order

Minimisation

Minimisation involves an algorithm that considers characteristics of previously recruited participants and aims to allocate the next participant so as to minimise the imbalance in these factors between groups. This will typically involve a “random element” where the allocation will, with a predetermined probability, allocate at random instead