arXiv subjects

Scott Evans

Publications and source records attributed to Scott Evans.

6 recordsLinked to original sources

arXiv2026

A Unified Causal Inference Framework for the Desirability of Outcome Ranking Paradigm in Benefit-Risk Evaluation

We developed a unified covariate-adjusted causal inference framework for estimating the desirability of outcome ranking (DOOR) probability for benefit-risk evaluation in randomized trials and observational studies. The framework expresses the DOOR probability as a bilinear functional of the marginal ordinal outcome distributions under the two treatment strategies, estimates conditional ordinal distributions through sequential risk-set hazards, and derives the efficient influence function (EIF) of the DOOR probability. The point-estimation simulations compared G-computation, normalized inverse probability weighting (IPW), augmented IPW (AIPW), and targeted maximum likelihood estimation (TMLE), with nuisance functions estimated using generalized linear models or Super Learner (SL). TMLE-SL showed the strongest and most consistent point-estimation performance, with AIPW-SL ranking second. EIF-based inference was then evaluated for AIPW-SL and TMLE-SL, with and without cross-fitting, across settings varying in overlap, treatment-effect heterogeneity, and treatment allocation. CVTMLE-SL showed the strongest overall performance across DOOR-scale bias, recovery of the underlying ordinal distributions, standard-error accuracy, and confidence-interval coverage. We illustrate the methodology using data from the multidrug-resistant organism network of the Antibacterial Resistance Leadership Group.

stat.ML↗

arXiv2026

On Cluster Randomized Trials with the Desirability of Outcome Ranking (DOOR) Endpoints

Cluster randomized trials are widely used when individual randomization is logistically infeasible or when correlations between observations cannot be ignored, especially in fields such as ophthalmology, infectious disease, vaccine research, and sociology. The desirability of outcome ranking (DOOR) framework evaluates patient-centric benefit-risk using an ordinal outcome and a Wilcoxon-Mann-Whitney statistic-based approach to compare outcome distributions between interventions. We propose a suite of new methods to extend DOOR to cluster trials based on properties of U-statistics and influence functions to estimate within-cluster and between-cluster treatment effects. These approaches can be applied in different scenarios, including mixtures of clusters with two treatment groups and clusters with only one group, and both small and large numbers of clusters. Simulations demonstrate that the proposed methods perform well under various scenarios regarding the number of clusters and cluster sizes. As an illustration, we apply the proposed methods to a cluster randomized crossover trial comparing delayed cord clamping and umbilical cord milking for newborns.

stat.ME↗

arXiv2026

Navigating the Landscape of Hierarchical Multi-Component Strategies: GPC, DOOR, and MOST

There is a growing recognition of the importance to involve patients in every stage of drug development. This shift acknowledges that patients' perspectives, experiences, and preferences are essential for ensuring that treatments meet real-world needs. In this context, a new body of statistical literature has emerged, focusing not only on the simultaneous consideration of multiple outcomes that reflect patients' overall experiences, but also on their structured prioritization. We refer to this class of approaches as hierarchical multi-component statistical methods. Among these, two influential frameworks - generalized pairwise comparisons (GPC) and desirability of outcome ranking (DOOR) - have emerged in the last decade, each aiming to offer a comprehensive approach to evaluating treatment effects. A new methodology, referred to here as the Markov ordinal state transition model (MOST), has recently been introduced without focusing on an explicit link with GPC nor DOOR. This paper seeks to fill this gap by offering a comprehensive and comparative analysis of the three approaches. Through examples and an exploration of the structural and philosophical differences between the methods, our aim is to provide guidance and encourage lines of research in the rapidly-evolving landscape of hierarchical multi-component statistical methodologies.

stat.ME↗

arXiv2026

Doubly Robust Estimation of Desirability of Outcome Ranking (DOOR) Probability with Application to MDRO Studies

In observational studies, adjusting for confounders is required if a treatment comparison is planned. A crude comparison of the primary endpoint without covariate adjustment will suffer from biases, and the addition of regression models could improve precision by incorporating imbalanced covariates and thus help make correct inference. Desirability of outcome ranking (DOOR) is a patient-centric benefit-risk evaluation methodology designed for randomized clinical trials. Still, robust covariate adjustment methods could further expand the compatibility of this method in observational studies. In DOOR analysis, each participant's outcome is ranked based on pre-specified clinical criteria, where the most desirable rank represents a good outcome with no side effects and the least desirable rank is the worst possible clinical outcome. We develop a causal framework for estimating the population-level DOOR probability, via the inverse probability of treatment weighting method, G-Computation method, and a Doubly Robust method that combines both. The performance of the proposed methodologies is examined through simulations. We also perform a causal analysis of the Multi-Drug Resistant Organism (MDRO) network within the Antibacterial Resistant Leadership Group (ARLG), comparing the benefit:risk between Mono-drug therapy and Combination-drug therapy.

stat.ME↗

arXiv2025

Desirability of outcome ranking (DOOR) analysis for multivariate survival outcomes with application to ACTT-1 trial

Desirability Of Outcome Ranking (DOOR) methodology accounts for problems that conventional benefit:risk analyses in clinical trials ignore, such as competing risks and the trade-off relationship between efficacy and toxicity. DOOR levels can be considered as a multi-state process in nature, as event-free survival, and survival with side effects are not equivalent and the overall patient trajectory requires recognition. In monotone settings where patients' conditions can only decline, we can record event times for each transition from one level of the DOOR to another, and construct Kaplan-Meier curves displaying transition times. While traditional survival analysis methods such as the Cox model require assumptions like proportional hazards and suffer from the challenge of interpreting a hazard ratio, Restricted Mean Survival Time (RMST) offers an alternative with greater intuitiveness. Therefore, we propose a combination of the two domains to develop estimation and inferential procedures that could benefit from the advantages of both DOOR and RMST. Particularly, the area under each survival curve restricted to a time point, or the RMST, has clear clinical meanings, from expected event-free survival time, expected survival time with at most one of the events, to expected lifetime before death. We show that the nonparametric estimator of the RMSTs asymptotically follows a multivariate Gaussian process through the martingale theory and functional delta method. There are alternative approaches to hypothesis testing that recognize when patients transition into worse states. We evaluate our proposed method with data simulated under a multistate model. We consider various scenarios, including when the null hypothesis is true, when the treatment difference exists only in certain DOOR levels, and small-sample studies. We also present a real-world example with ACTT-1.

stat.ME↗

arXiv2019

Causal Inference via Conditional Kolmogorov Complexity using MDL Binning

Recent developments have linked causal inference with Algorithmic Information Theory, and methods have been developed that utilize Conditional Kolmogorov Complexity to determine causation between two random variables. We present a method for inferring causal direction between continuous variables by using an MDL Binning technique for data discretization and complexity calculation. Our method captures the shape of the data and uses it to determine which variable has more information about the other. Its high predictive performance and robustness is shown on several real world use cases.

cs.LG↗