Your privacy, your choice

We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media.

By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection.

See our privacy policy for more information on the use of your personal data.

for further information and to change your choices.
Skip to main content
Fig. 2 | BMC Medical Research Methodology

Fig. 2

From: Forced randomization: the what, why, and how

Fig. 2

The randomization flow for a “toy” example. Description for Figure 2: We have a 1:1 RCT with the permuted blocks of 4 as the method of randomization. The randomization list for the first two blocks is |AABB|ABBA|. Initially, there are two kits of drug A and three kits of drug B onsite. Patients are enrolled sequentially and must be randomized to treatment immediately. For patients 1 and 2, the randomized treatment assignment will be made consistently with the randomization list with all four IRT configurations. Subsequently, the randomized assignments will depend on the IRT configuration. If the IRT system is configured as FR0a, then patients 3, 4, and 5 will be sent home without being randomized because no drug A is available on site (therefore, randomization transactions are not authorized). The drug has been re-supplied at patient 6’s entry, and patients 6, 7, and 8 are randomized to B, B, and A, as scheduled. If the IRT system is configured as FR0b, then patients 3 and 4 are randomized to drug B as scheduled because drug B is available on site, and randomization transactions are authorized. Patient 5 will be sent home without being randomized because of the lack of drug A supply on site. The drug has been re-supplied at patient 6’s entry, and patients 6, 7, and 8 are randomized to B, B, and A, as scheduled. If the IRT system is configured as FR1a, then patients 3 and 4 are randomized to drug B as scheduled, but patient 5 cannot be randomized to drug A because no drug A is available on site. Instead, patient 5 is “forced-randomized” to the next free number in the randomization list that corresponds to a treatment that is available (drug B in our example), and the fifth allocation (A) is crossed out to show that it can no longer be assigned. The drug has been re-supplied at patient 6’s entry, and patients 6, 7, and 8 are randomized to treatments that appear consecutively on the randomization list. If the IRT system is configured as FR1b, then patients 3 and 4 are randomized to drug B as scheduled, patient 5 is “forced-randomized” to receive drug B (because it is available), and the fifth allocation (A) will be provisioned for the next patient. The drug has been re-supplied at patient 6’s entry, and patient 6 is “forced-randomized” to drug A to backfill the previously unused position in the randomization list. Patients 6 and 7 are randomized to B and A, as scheduled

Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us