Searcharxiv⌕ Search

arXiv subjects

Isao Yokota

Publications and source records attributed to Isao Yokota.

3 recordsLinked to original sources

Stratified Randomization Incorporating Historical Control Data

Randomized controlled trials are often infeasible in rare diseases because of limited sample size, high costs, or lengthy durations. Hybrid control designs address this by combining a concurrent control arm with historical data. However, existing methods such as propensity score matching (PSM) borrow historical data only at the analysis stage, which limits power gains when covariate distributions differ between historical and concurrent data. We propose a stratified block randomization method for hybrid control trials. Within each stratum, more concurrent subjects are allocated to the treatment arm than to the concurrent control arm, and the difference is filled with historical controls so that the treatment and hybrid control arms are balanced 1:1; once the available historical controls in the stratum are used or a pre-specified overall cap is reached, allocation returns to 1:1. Historical controls are incorporated either by sampling or by weighting. We compared both methods with PSM in terms of type I error rate, power, and sample size via Monte Carlo simulations. The sampling method maintained the type I error rates when there was neither heterogeneity in the outcome model nor unmeasured covariates. It also had power comparable to or greater than that of PSM, particularly when historical controls were unevenly distributed across strata. The power of the weighting method increased with the pooled historical sample size, but its type I error rate was inflated when covariates beyond the stratification factors were present or when stratification factors were dichotomized from continuous covariates. The proposed method incorporates historical data at the allocation stage rather than only at the analysis stage, enabling more efficient use of historical data.

stat.ME↗

The Ghosh-Lin and Fine-Gray models for a mix of administrative and random censoring

Recurrent events or competing risks regression models are often applied in the bio-medical setting and both can be considered as marginal models. In presence of right-censoring, such models need to be adjusted to give consistent estimators. When censoring is administrative, marginal regression models are particularly easy to estimate. However, when censoring is instead acting randomly, inverse probability of censoring weighting (IPCW) adjustments are typically considered to obtain parameter estimates. This technique relies on a censoring-weights adjustment via a correct censoring model, but for administrative censoring the adjustment is done correctly simply by modifying the risk-set. In practice for large central registries or some clinical trials, the administrative censoring time will be known for all subjects, but there will typically also be a proportion of subjects that are censored at random. In this work, we consider two frequently used regression approaches, the Ghosh-Lin model for recurrent events with terminal events and the Fine-Gray model for competing events. For these two settings, when both administrative and random censoring are present, we demonstrate how to obtain correct estimation by dealing with the combination of the two different types of censoring relying on a minimum of modeling assumptions.

stat.ME↗

Exact sequential single-arm trial design with curtailment for binary endpoint

Due to ethical and economical reasons, sequential single-arm trial designs are used for assessing the therapeutic efficacy of new treatments in phase II trials. Simon's 2-stage design and Lan-DeMets' $α$-spending function method with O'Brien-Fleming type are widely recognized as the traditional methods for futility stopping and efficacy stopping, respectively. These methods have two practical problems, which are the difficulty of interpretation for stopping under staggered entry and the inflation of error rate due to a small-sample trial. In this research, we propose the exact sequential design making the threshold value for efficacy fixed, and compare with traditional designs in sample size. Since the maximum sample size and average sample number of the proposed design are generally smaller than those of traditional designs containing fixed design, the proposed design is expected to be enrolled fewer subjects. In addition, we evaluate several kinds of point estimators and confidence intervals at the end of trials in the proposed design. If one is concerned with bias, the bias-adjusted estimator may be better. As for a confidence interval, the mid-p approach will be a good choice.

stat.ME↗