What is Randomisation?
Randomisation is the process of assigning participants to different treatment arms, rather than by choice or preference.
Its main goal is to reduce bias and ensure that the groups being compared are as similar as possible at the start of a study.
By distributing participants randomly, researchers can be confident that any differences observed between groups at the end of the study are due to the treatment itself, not to pre-existing differences between participants.
In practice, randomisation helps ensure:
Scientific validity: Each group is comparable, making results more reliable.
Fairness: Every participant has an equal chance of being assigned to any treatment.
Transparency: The process is predefined, documented, and reproducible.
In Datacapt, randomisation can be configured in just a few minutes, from simple 1:1 assignments to complex multi-arm or cross-over studies without any custom scripts.
Core concepts
Randomisation: The process of assigning participants to treatment arms/groups, to minimize bias and strengthen the study’s validity.
Randomised Participant: A participant who has been successfully assigned to a treatment arm.
Randomisation ID: A unique, incremental identifier automatically generated after randomisation (e.g., RD-0001).
Randomisation Allocation: The specific action of assigning an eligible participant to a treatment arm once all inclusion criteria are confirmed.
Treatment Arms: The different treatment or intervention groups within a study to which participants are assigned.
Treatment Arm Weight/Ratio: The intended proportion of participants across treatment groups (e.g., 1:1, 2:1). In Datacapt, this is configured via the Treatment Arms setting (e.g., entering 1-1 for two equal groups).
Randomization Methods & Lists
Static randomization: randomization based on a randomized list,
Dynamic randomization (Pocock and Simon): randomization based on an algorithm.
Import: randomization based on external randomization list
Static Randomization (Block Randomization)
Static randomization uses a pre-generated list of assignments, arranged in permuted blocks according to your predefined ratios (for example 1:1 or 2:1).
Each new participant is assigned to the next available slot in that list.
This method ensures predictable group sizes and is ideal when you want simple, reproducible, and controlled allocation.
Dynamic Randomization (Minimization, Pocock–Simon)
Dynamic randomization does not rely on a predefined list. Instead, Datacapt calculates each assignment in real time, using an algorithm that minimizes imbalance across chosen stratification factors (centers and/or questions).
For each new participant, the system simulates all possible allocations, computes an imbalance score per treatment arm, and then assigns the participant to the arm that best maintains balance with a small degree of randomness controlled by a probability p (biased-coin method).
Import Randomization
Import randomization allows you to use an external randomization list prepared outside Datacapt. Once uploaded, Datacapt follows this list exactly as provided, ensuring full alignment with the sponsor’s pre-approved plan.
This option is most useful when the sponsor or statistician has already generated a validated sequence for example, to reproduce a legacy study or comply with regulatory documentation requirements.
Summary
Method | Logic | Inputs |
Block (static list) | Pre-generated list in permuted blocks, next subject takes the next slot. | Arms, ratio, block sizes. |
Dynamic (minimization, Pocock–Simon) | Allocation computed in real time to minimize imbalance between groups using chosen distance method. | Arms, ratio, distance method (Range/Variance/Max), probability p. |
Import | External list uploaded and used as-is. | Predefined list file in CSV/XLSX format. |
Randomization Type
Randomization can be configured as Open Label (Single-Blind) or Double-Blind, depending on how treatment information should be displayed to study users.
Open Label (Single-Blind)
All users can see the treatment arm assigned to each participant directly after randomization.
Double-Blind
In blinded studies, treatment arm details are hidden by default from all users except those granted the appropriate unblinding permissions. Both participants and investigators are blinded. Only users with the “Unblind access” permission can view allocation details.
For emergency situations, Datacapt provides a specific “Emergency unblinding” permission that allows revealing allocation for a single participant.
This configuration ensures that the randomization workflow respects your study’s blinding requirements while maintaining full auditability.
Scope and Stratification
Datacapt allows you to define the scope of randomization and optionally stratify to maintain balance across key study factors.
Scope per Center
Each center has its own randomization list or algorithm instance, maintaining local balance between treatment arms.
This option is preferred for multicentric studies where enrollment rates or participant profiles may vary by site.
Stratification by Questions
You can further balance allocations based on up to 10 specific variables (radio or dropdown items from the eCRF).
Once a variable is used for stratification, its answer options become locked in the eCRF builder to ensure consistency with the randomization configuration.
By combining scope and stratification, you can tailor randomization precision from global balance across all sites to fine-tuned control per center and per key factor.
👉🏻 Next step: