Dr. Scott Hamilton, principal biostatistician at Bracket, told us that using PBD distinguishes Bracket as the only IRT provider that has developed – and now provides – PBD technology to all its clients.
“PBD reduces selection bias and predictability in permuted block randomization,” explained Hamilton. “Additionally, the use of PBD does not add additional time nor cost to the creation of the randomization list.”
Hamilton explained PBD is different from similar technologies, as it reduces potential for extreme block pattern distributions, and strings of repeated block patterns compared to standard permuted block randomization
To date, Bracket has used PBD in four separate clinical trials.
“All the statisticians involved were favorable and two of them have requested their next randomization list to utilize PBD,” said Hamilton.
How does Precision Block Design technology work?
To get a better understanding of how PBD technology works, Hamilton provided us with the following situation:
Imagine a study with 2 treatment groups {A, B} where the patients are randomly assigned to treatment using a permuted blocked list stratified by enrolling center with a blocksize of 2. There would be 2 permutations within this list [A, B] and [B, A]. Suppose by chance a stratum was allocated all of one permutation so that the list for that center looked like this: [A, B, A, B, A, B, A, B, …].
Because of the perfectly alternating pattern of treatment allocation there would be a high probability of guessing the next treatment, thus, a high selection bias.
One way to reduce the selection bias would be to increase the blocksize to four. However, in many studies maintaining a high level of treatment group balance requires a blocksize of 2. The other method to reduce selection bias is to ensure all the permutations are represented in a relatively uniform distribution within each stratum.
The statistics group at Bracket checks every randomization list for uniformity of permutation distribution within the strata. When our criterion is not met, we re-generate the list.
When we receive a list generated by a sponsor we apply our criterion, and if it is not met we advise the sponsor so they can decide whether to generate a new list.
A direct approach to the problem of selection bias is Bracket’s novel method called Precision Block Design, which guarantees uniform block patterns within strata.
To illustrate using a block size of 4, we would sample from the 6 possible block permutations without replacement until no more are available, then sample 6 more, and repeat until we have reached the desired number of blocks within each stratum.
This method produces a uniform distribution of block patterns within each stratum. Therefore, an acceptance test is not required.
There are certain randomization designs that lend themselves to the use of Precision Block Design:
- A high number of strata
- A low number of slots