Statistical Methods in Medical Research最新文献

Sequential multiple assignment randomized trial design is becoming increasingly used in the field of precision medicine. This design allows comparisons of sequences of adaptive interventions tailored to the individual patient. Superiority testing is usually the initial goal in order to determine which embedded adaptive intervention yields the best primary outcome on average. When direct superiority is not evident, yet an adaptive intervention poses other benefits, then non-inferiority testing is warranted. Non-inferiority testing in the sequential multiple assignment randomized trial setup is rather new and involves the specification of non-inferiority margin and other important assumptions that are often unverifiable internally. These challenges are not specific to sequential multiple assignment randomized trial and apply to two-arm non-inferiority trials that do not include a standard-of-care (or placebo) arm. To address some of these challenges, three-arm non-inferiority trials that include the standard-of-care arm are proposed. However, methods developed so far for three-arm non-inferiority trials are not sequential multiple assignment randomized trial-specific. This is because apart from embedded adaptive interventions, sequential multiple assignment randomized trial typically does not include a third standard-of-care arm. In this article, we consider a three-arm sequential multiple assignment randomized trial from an National Institutes of Health-funded study of symptom management strategies among people undergoing cancer treatment. Motivated by that example, we propose a novel data analytic method for non-inferiority testing in the framework of three-arm sequential multiple assignment randomized trial for the first time. Sample size and power considerations are discussed through extensive simulation studies to elucidate our method.

A great deal of literature has been established for regression analysis of longitudinal data and in particular, many methods have been proposed for the situation where there exist some change points. However, most of these methods only apply to continuous response and focus on the situations where the change point only occurs on the response or the trend of the individual trajectory. In this article, we propose a new joint modeling approach that allows not only the change point to vary for different subjects or be subject-specific but also the effect heterogeneity of the covariates before and after the change point. The method combines a generalized linear mixed effect model with a random change point for the longitudinal response and a log-linear regression model for the random change point. For inference, a maximum likelihood estimation procedure is developed and the asymptotic properties of the resulting estimators, which differ from the standard asymptotic results, are established. A simulation study is conducted and suggests that the proposed method works well for practical situations. An application to a set of real data on COVID-19 is provided.

Diagnostic accuracy studies assess the sensitivity and specificity of a new index test in relation to an established comparator or the reference standard. The development and selection of the index test are usually assumed to be conducted prior to the accuracy study. In practice, this is often violated, for instance, if the choice of the (apparently) best biomarker, model or cutpoint is based on the same data that is used later for validation purposes. In this work, we investigate several multiple comparison procedures which provide family-wise error rate control for the emerging multiple testing problem. Due to the nature of the co-primary hypothesis problem, conventional approaches for multiplicity adjustment are too conservative for the specific problem and thus need to be adapted. In an extensive simulation study, five multiple comparison procedures are compared with regard to statistical error rates in least-favourable and realistic scenarios. This covers parametric and non-parametric methods and one Bayesian approach. All methods have been implemented in the new open-source R package cases which allows us to reproduce all simulation results. Based on our numerical results, we conclude that the parametric approaches (maxT and Bonferroni) are easy to apply but can have inflated type I error rates for small sample sizes. The two investigated Bootstrap procedures, in particular the so-called pairs Bootstrap, allow for a family-wise error rate control in finite samples and in addition have a competitive statistical power.

Cross-validation is the most common way of selecting tuning parameters in penalized regression, but its use in penalized Cox regression models has received relatively little attention in the literature. Due to its partial likelihood construction, carrying out cross-validation for Cox models is not straightforward, and there are several potential approaches for implementation. Here, we propose a new approach based on cross-validating the linear predictors of the Cox model and compare it to approaches that have been proposed elsewhere. We show that the proposed approach offers an attractive balance of performance and numerical stability, and illustrate these advantages using simulated data as well as analyzing a high-dimensional study of gene expression and survival in lung cancer patients.

In preclinical investigations, for example, in in vitro, in vivo, and in silico studies, the pharmacokinetic, pharmacodynamic, and toxicological characteristics of a drug are evaluated before advancing to first-in-man trial. Usually, each study is analyzed independently and the human dose range does not leverage the knowledge gained from all studies. Taking into account all preclinical data through inferential procedures can be particularly interesting in obtaining a more precise and reliable starting dose and dose range. Our objective is to propose a Bayesian framework for multi-source data integration, customizable, and tailored to the specific research question. We focused on preclinical results extrapolated to humans, which allowed us to predict the quantities of interest (e.g. maximum tolerated dose, etc.) in humans. We build an approach, divided into four steps, based on a sequential parameter estimation for each study, extrapolation to human, commensurability checking between posterior distributions and final information merging to increase the precision of estimation. The new framework is evaluated via an extensive simulation study, based on a real-life example in oncology. Our approach allows us to better use all the information compared to a standard framework, reducing uncertainty in the predictions and potentially leading to a more efficient dose selection.

We compared methods to project absolute risk, the probability of experiencing the outcome of interest in a given projection interval accommodating competing risks, for a person from the target population with missing predictors. Without missing data, a perfectly calibrated model gives unbiased absolute risk estimates in a new target population, even if the predictor distribution differs from the training data. However, if predictors are missing in target population members, a reference dataset with complete data is needed to impute them and to estimate absolute risk, conditional only on the observed predictors. If the predictor distributions of the reference data and the target population differ, this approach yields biased estimates. We compared the bias and mean squared error of absolute risk predictions for seven methods that assume predictors are missing at random (MAR). Some methods imputed individual missing predictors, others imputed linear predictor combinations (risk scores). Simulations were based on real breast cancer predictor distributions and outcome data. We also analyzed a real breast cancer dataset. The largest bias for all methods resulted from different predictor distributions of the reference and target populations. No method was unbiased in this situation. Surprisingly, violating the MAR assumption did not induce severe biases. Most multiple imputation methods performed similarly and were less biased (but more variable) than a method that used a single expected risk score. Our work shows the importance of selecting predictor reference datasets similar to the target population to reduce bias of absolute risk predictions with missing risk factors.