Overview

Permuted Block Design (also called Random Permuted Blocks or Blocked Randomization) is the most widely used algorithm for simple parallel-arm clinical trials. It creates balanced blocks of treatment assignments that are internally shuffled to maintain allocation concealment while guaranteeing balance within each block.

How it works

1. A block of size n is generated with treatment assignments in the correct allocation ratio.
2. Within each block, the assignments are randomly permuted (shuffled).
3. Subjects are assigned in sequence — when one block is exhausted, a new block is generated.

Example — Block size 4, 1:1 allocation (TRT : PBO):

Each block contains 2 TRT and 2 PBO assignments, shuffled randomly:

Block 1: [TRT, PBO, TRT, PBO]
Block 2: [PBO, PBO, TRT, TRT]
Block 3: [TRT, PBO, PBO, TRT]

At any point within a block, the number of TRT and PBO subjects will not differ by more than block_size / 2.

When to use

Use Permuted Block Design when:

• The trial has a single site (or balance at the site level is not required).
• Stratification by subject characteristics is not needed.
• You want a simple, well-understood algorithm accepted by all major regulatory agencies.

For multi-site trials or when subject-level stratification is needed, use Stratified Permuted Block Design instead.

Configuration

In the Randomization app, select Permuted Block Design as the algorithm, then configure:

Setting	Description
Study Groups	One row per treatment arm: Label, Code, Weight, Description
Block size	Number of subjects per block. Must be a multiple of the total weight sum. For 1:1 allocation, use 4 or 8.
Generate randomization list	Real-time (recommended) — assign at the moment a subject is randomized. Or pre-generate the list in advance.
Level of stratification	By Study — generates a single list for all subjects regardless of site

Block size guidance

Allocation ratio	Minimum block size	Common choices
1:1	2	4, 6, 8
1:1:1	3	6, 9
2:1	3	6, 9
1:2:1	4	8

Security note: Small block sizes (2 or 4 for 1:1) make it easier to guess the next assignment toward the end of a block. For unblinded trials, use larger block sizes or variable block sizes (if supported) to maintain allocation concealment.

Limitations

• No site-level balance — if enrollment is uneven across sites, some sites may finish with imbalanced arms.
• No subject-level stratification — prognostic factors can still be unbalanced between arms by chance (more of a concern in small trials).

Regulatory acceptance

Permuted Block Design is accepted by FDA, EMA, PMDA, and ICH E9. Reference ICH E9 §3.4 for guidance on randomization in clinical trials.