How to Configure Randomization in a Study.

Modified on Thu, 7 Aug at 8:52 PM

Setting up randomization directly within EasyMedStat allows your study team to allocate patients to treatment groups in a compliant, auditable, and blinded way, without relying on external tools.

This guide explains how to configure the randomization protocol in the study settings.


Who can configure randomization?

Only users who meet both of the following conditions can configure randomization:

  • They have the "Series settings" permission (typically study administrators)
  • They have the "Randomization" permission (either at the user level or the series level)


Where to find the randomization settings?

  1. Go to the Series Settings page of your study
  2. In the left-hand menu, click on “Randomization”
  3. Enable randomization by selecting a randomization method


Step-by-step: Configuring Randomization


1. Select the Randomization Method

You can choose one of the following methods:

  • Simple randomization
  • Block randomization
  • Stratified block randomization
  • Minimization

Each method will display its relevant parameters.


2. Define the Number of Arms

Specify how many treatment groups your study includes (between 2 and 4 arms) and their respective allocation ratios (for example 1:1 or 2:1:1). EasyMedStat will create a dedicated randomization variable based on this configuration.


3. Set the Expected Number of Patients

Enter the total number of patients you expect to include in the study. This number will be used to generate a randomization table that slightly exceeds the planned sample size, to account for unexpected inclusions or dropouts.

The system will validate this number against your selected block sizes and strata. For example, if your study uses stratified block randomization with 3 centers and block size 12, a minimum of 72 patients may be required.


4. Choose the Block Sizes (for Block or Stratified Block Randomization)

Select one or more block sizes from the predefined list. The system enforces compatibility rules:

  • Each block size must be a multiple of the sum of arm ratios
  • Each block size must be approximately 2 to 4 times the number of arms

Examples:

  • For 2 arms (1:1), valid sizes include 4, 6, 8, 10, 12
  • For 3 arms with ratios 2:1:1, the sum is 4, so valid sizes would be 8, 12, etc.

You can also choose a combination of block sizes to introduce randomness and reduce predictability.


5. Add Stratification Variables (only for Stratified Block Randomization)

If you selected "Stratified block randomization" in step 1, you can choose up to 4 stratification variables.

Conditions:

  • Only categorical variables (Yes/No or List) are allowed
  • The randomization variable cannot be used as a stratification factor
  • A warning will be displayed if more than 2 variables are selected (this may lead to complex strata with limited benefit)

The system will automatically compute all strata combinations based on the selected variables.


6. Download a Preview Table (Optional)

If your configuration is valid, you can download a preview of the randomization table. This table is provided for illustration purposes only and will not be used during patient randomization.

It includes:

  • Randomization numbers
  • Block numbers
  • Treatment groups
  • Stratification values (if applicable)

No patient data is included in the preview.


7. Save the Randomization Design

Once the configuration is complete, click on “Save randomization design”.

  • In test mode, the configuration remains editable. You can delete and redefine it as many times as needed.
  • Once the study is switched to production mode, the configuration becomes locked. It can no longer be modified, even if no patients have been randomized yet.

Switching to production will also regenerate the randomization table with a new seed.


Notes

  • The randomization setup does not support importing custom randomization tables through the interface.
  • The randomization variable is protected and hidden from patients (ePRO), bulk updates, and any unauthorized users.
  • It cannot be edited after randomization, either manually or via API.

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article