In conventionally randomized controlled trials, adjustment for baseline prognostic information (BPI) is used to increase power. However, its performance hasn’t been formally characterized within the context of more flexible designs, such as Bayesian group sequential designs (BGS). BGS are sequentially randomized and allow for early stopping at interim analyses based on pre-defined stopping rules, which are typically a function of the posterior probability of the treatment effect. Adjustment for BPI at each interim analysis improves the posterior estimation of the treatment effect, so its use is shown to be beneficial for BGS. The present research investigates the impact of BPI adjustment on BGS with continuous, binary and time-to-event outcomes through a simulation study. Several scenarios for the interim analysis adjustment models are used. The impact of these adjustment models on power, expected sample size, probability of stopping the trial early, and bias is quantified.