Inference in Experiments With Matched Pairs

Regression-based multiple treatment effect estimation under covariate-adaptive randomization

Covariate-adaptive randomization methods are widely used in clinical trials to balance baseline covariates. Recent studies have shown the validity of using regression-based estimators for treatment effects without imposing functional form requirements on the true data generation model. These studies have had limitations in certain scenarios; for example, in the case of multiple treatment groups, these studies did not consider additional covariates or assumed that the allocation ratios were the same across strata. To address these limitations, we develop a stratum-common estimator and a stratum-specific estimator under multiple treatments. We derive the asymptotic behaviors of these estimators and propose consistent nonparametric estimators for asymptotic variances. To determine their efficiency, we compare the estimators with the stratified difference-in-means estimator as the benchmark. We find that the stratum-specific estimator guarantees efficiency gains, regardless of whether the allocation ratios across strata are the same or different. Our conclusions were also validated by simulation studies and a real clinical trial example.

Testing weak nulls in matched observational studies

We develop sensitivity analyses for the sample average treatment effect in matched observational studies while allowing unit-level treatment effects to vary. The methods may be applied to studies using any optimal without-replacement matching algorithm. In contrast to randomized experiments and to paired observational studies, we show for general matched designs that over a large class of test statistics, any procedure bounding the worst-case expectation while allowing for arbitrary effect heterogeneity must be unnecessarily conservative if treatment effects are actually constant across individuals. We present a sensitivity analysis which bounds the worst-case expectation while allowing for effect heterogeneity, and illustrate why it is generally conservative if effects are constant. An alternative procedure is presented that is asymptotically sharp if treatment effects are constant, and that is valid for testing the sample average effect under additional restrictions which may be deemed benign by practitioners. Simulations demonstrate that this alternative procedure results in a valid sensitivity analysis for the weak null hypothesis under a host of reasonable data-generating processes. The procedures allow practitioners to assess robustness of estimated sample average treatment effects to hidden bias while allowing for effect heterogeneity in matched observational studies.

Statistical Detection of Differentially Abundant Proteins in Experiments with Repeated Measures Designs and Isobaric Labeling

Repeated measures experimental designs, which quantify proteins in biological subjects repeatedly over multiple experimental conditions or times, are commonly used in mass spectrometry-based proteomics. Such designs distinguish the biological variation within and between the subjects and increase the statistical power of detecting within-subject changes in protein abundance. Meanwhile, proteomics experiments increasingly incorporate tandem mass tag (TMT) labeling, a multiplexing strategy that gains both relative protein quantification accuracy and sample throughput. However, combining repeated measures and TMT multiplexing in a large-scale investigation presents statistical challenges due to unique interplays of between-mixture, within-mixture, between-subject, and within-subject variation. This manuscript proposes a family of linear mixed-effects models for differential analysis of proteomics experiments with repeated measures and TMT multiplexing. These models decompose the variation in the data into the contributions from its sources as appropriate for the specifics of each experiment, enable statistical inference of differential protein abundance, and recognize a difference in the uncertainty of between-subject versus within-subject comparisons. The proposed family of models is implemented in the R/Bioconductor package MSstatsTMT v2.2.0. Evaluations of four simulated datasets and four investigations answering diverse biological questions demonstrated the value of this approach as compared to the existing general-purpose approaches and implementations.