Choosing the Unit of Randomization — Individual or Cluster?

Randomized trials are widely regarded as providing the most robust evidence for guiding patient management and informing public health policy. A typical individual-randomized trial (IRT) randomly assigns individual participants into treatment and control groups and then compares their outcomes to estimate the intervention effect.1 However, sometimes individual randomization may not be feasible or optimal, prompting the consideration of a cluster-randomized trial (CRT), where groups of individuals are randomly assigned to different treatment conditions.2 When aiming to assess the broad population effects of an intervention, a cluster-randomized design is often used. In certain scenarios, the intervention is designed to be delivered to groups of individuals.