Biostatistics最新文献

Nonlinear causal effects are prevalent in many research scenarios involving continuous exposures, and instrumental variables (IVs) can be employed to investigate such effects, particularly in the presence of unmeasured confounders. However, common IV methods for nonlinear effect analysis, such as IV regression or the control-function method, have inherent limitations, leading to either low statistical power or potentially misleading conclusions. In this work, we propose an alternative IV framework for nonlinear effect analysis, which has recently emerged in genetic epidemiology and addresses many of the drawbacks of existing IV methods. The proposed IV framework consists of up to three key "S" elements: (i) the Stratification approach, which constructs multiple strata that are sub-samples of the population in which the IV core assumptions remain valid, (ii) the Scalar-on-function model and Scalar-on-scalar model, which connect local stratum-specific information to global effect estimation, and (iii) the Sum-of-single-effects method for effect estimation. This framework enables study of the effect function while avoiding unnecessary model assumptions. In particular, it facilitates the identification of change points or threshold values in causal effects. Through a wide variety of simulations, we demonstrate that our framework outperforms other representative nonlinear IV methods in predicting the effect shape when the instrument is weak and can accurately estimate the effect function as well as identify the change point and predict its value under various structural model and effect shape scenarios. We further apply our framework to assess the nonlinear effect of alcohol consumption on systolic blood pressure using a genetic instrument (ie Mendelian randomization) with UK Biobank data. Our analysis detects a threshold beyond which alcohol intake exhibits a clear causal effect on the outcome. Our results are consistent with published medical guidelines.

Gaussian graphical models are widely used to study the dependence structure among variables. When samples are obtained from multiple conditions or populations, joint analysis of multiple graphical models are desired due to their capacity to borrow strength across populations. Nonetheless, existing methods often overlook the varying levels of similarity between populations, leading to unsatisfactory results. Moreover, in many applications, learning the population-level clustering structure itself is of particular interest. In this article, we develop a novel method, called Simultaneous Clustering and Estimation of Networks via Tensor decomposition (SCENT), that simultaneously clusters and estimates graphical models from multiple populations. Precision matrices from different populations are uniquely organized as a three-way tensor array, and a low-rank sparse model is proposed for joint population clustering and network estimation. We develop a penalized likelihood method and an augmented Lagrangian algorithm for model fitting. We also establish the clustering accuracy and norm consistency of the estimated precision matrices. We demonstrate the efficacy of the proposed method with comprehensive simulation studies. The application to the Genotype-Tissue Expression multi-tissue gene expression data provides important insights into tissue clustering and gene coexpression patterns in multiple brain tissues.

Major depressive disorder (MDD), a leading cause of years of life lived with disability, presents challenges in diagnosis and treatment due to its complex and heterogeneous nature. Emerging evidence indicates that reward processing abnormalities may serve as a behavioral marker for MDD. To measure reward processing, patients perform computer-based behavioral tasks that involve making choices or responding to stimulants that are associated with different outcomes, such as gains or losses in the laboratory. Reinforcement learning (RL) models are fitted to extract parameters that measure various aspects of reward processing (e.g. reward sensitivity) to characterize how patients make decisions in behavioral tasks. Recent findings suggest the inadequacy of characterizing reward learning solely based on a single RL model; instead, there may be a switching of decision-making processes between multiple strategies. An important scientific question is how the dynamics of strategies in decision-making affect the reward learning ability of individuals with MDD. Motivated by the probabilistic reward task within the Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) study, we propose a novel RL-HMM (hidden Markov model) framework for analyzing reward-based decision-making. Our model accommodates decision-making strategy switching between two distinct approaches under an HMM: subjects making decisions based on the RL model or opting for random choices. We account for continuous RL state space and allow time-varying transition probabilities in the HMM. We introduce a computationally efficient Expectation-maximization (EM) algorithm for parameter estimation and use a nonparametric bootstrap for inference. Extensive simulation studies validate the finite-sample performance of our method. We apply our approach to the EMBARC study to show that MDD patients are less engaged in RL compared to the healthy controls, and engagement is associated with brain activities in the negative affect circuitry during an emotional conflict task.

Immune response decays over time, and vaccine-induced protection often wanes. Understanding how vaccine efficacy changes over time is critical to guiding the development and application of vaccines in preventing infectious diseases. The objective of this article is to develop statistical methods that assess the effect of decaying immune responses on the risk of disease and on vaccine efficacy, within the context of Cox regression with sparse sampling of immune responses, in a baseline-naive population. We aim to further disentangle the various aspects of the time-varying vaccine effect, whether direct on disease or mediated through immune responses. Based on time-to-event data from a vaccine efficacy trial and sparse sampling of longitudinal immune responses, we propose a weighted estimated induced likelihood approach that models the longitudinal immune response trajectory and the time to event separately. This approach assesses the effects of the decaying immune response, the peak immune response, and/or the waning vaccine effect on the risk of disease. The proposed method is applicable not only to standard randomized trial designs but also to augmented vaccine trial designs that re-vaccinate uninfected placebo recipients at the end of the standard trial period. We conducted simulation studies to evaluate the performance of our method and applied the method to analyze immune correlates from a phase III SARS-CoV-2 vaccine trial.

Randomized controlled trials evaluating the diagnostic accuracy of a marker frequently collect information on baseline covariates in addition to information on the marker and the reference standard. However, standard estimators of sensitivity and specificity do not use data on baseline covariates and restrict the analysis to data from participants with a positive reference standard in the intervention arm being evaluated. Covariate-adjusted estimators for marginal treatment effects have been developed and been advocated for by regulatory agencies because they can improve power compared to unadjusted estimators. Despite this, similar covariate-adjusted estimators for marginal sensitivity and specificity have not yet been developed. In this manuscript, we address this gap by developing covariate-adjusted estimators for marginal sensitivity and specificity of a diagnostic test that leverage baseline covariate information. The estimators also use data from all participants, not just participants with a positive reference standard in the intervention arm being evaluated. We derive the asymptotic properties of the estimators and evaluate the finite sample properties of the estimators using simulations and by analyzing data on lung cancer screening.

Individualized treatment rule (ITR) is a stepping stone to precision medicine. To ensure validity, ITRs are ideally derived from randomized trial data, but the use cases of ITRs extend beyond these trial populations. Transferring knowledge from experimental data to real-world data is of interest, while experimental data with selective inclusion criteria reflect a population distribution that may differ from the real-world target. In well-designed experiments, granular information crucial to decision making can be thoroughly collected. However, part of this may not be accessible in real-world scenarios. We propose a learning scheme for ITR that simultaneously addresses the issues of covariate shift and missing covariates with a quantile-based optimal treatment objective. Specifically, we compare the outcome uncertainty across treatment arms that is due to missing covariates and use it to guide treatment selection to reduce the likelihood of worse outcomes. The performance of this method is evaluated in simulations and a sepsis data application.

For many rare diseases with no approved preventive interventions, promising interventions exist. However, it has proven difficult to conduct a pivotal phase 3 trial that could provide direct evidence demonstrating a beneficial effect of the intervention on the target disease outcome. When a promising putative surrogate endpoint(s) for the target outcome is available, surrogate-based provisional approval of an intervention may be pursued. Following the general Causal Roadmap rubric, we describe a surrogate endpoint-based provisional approval causal roadmap. Based on an observational study data set and a phase 3 randomized trial data set, this roadmap defines an approach to analyze the combined data set to draw a conservative inference about the treatment effect (TE) on the target outcome in the phase 3 study population. The observational study enrolls untreated individuals and collects baseline covariates, surrogate endpoints, and the target outcome, and is used to estimate the surrogate index-the regression of the target outcome on the surrogate endpoints and baseline covariates. The phase 3 trial randomizes participants to treated vs. untreated and collects the same data but is much smaller and hence very underpowered to directly assess TE, such that inference on TE is based on the surrogate index. This inference is made conservative by specifying 2 bias functions: one that expresses an imperfection of the surrogate index as a surrogate endpoint in the phase 3 study, and the other that expresses imperfect transport of the surrogate index in the untreated from the observational to the phase 3 study. Plug-in and nonparametric efficient one-step estimators of TE, with inferential procedures, are developed. The finite-sample performance of the estimators is evaluated in simulation studies. The causal roadmap is motivated by and illustrated with contemporary Group B Streptococcus vaccine development.

In nutritional epidemiology, self-reported dietary data are commonly used to investigate diet-disease relationships. However, the resulting association estimates are often subject to biases due to random and systematic measurement errors. Regression calibration has emerged as a crucial method for addressing these biases by refining self-reported nutrient intake with objective biomarkers, which differ from the true values only by a random "noise" component. This paper presents methodological tools for analyzing nutritional epidemiology cohort studies involving time-to-event data when a biomarker subsample is available alongside dietary assessments. We introduce novel regression calibration methods to tackle two common challenges in this field. First, a widely used approach assumes that the log hazard ratio (HR) follows a linear function of dietary exposure. However, assessing whether this assumption holds-or if a more flexible model is needed to capture potential deviations from linearity-is often necessary. Second, another prevalent analytical strategy involves estimating HRs based on categorized dietary exposure variables. New methods are critically needed to minimize bias in defining category boundaries and estimating hazard ratios within exposure categories, both of which can be distorted by measurement error. We apply these methods to reassess the relationship between sodium and potassium intake and cardiovascular disease risk using data from the Women's Health Initiative.

While methods for measuring and correcting differential performance in risk prediction models have proliferated in recent years, most existing techniques can only be used to assess fairness across relatively large subgroups. The purpose of algorithmic fairness efforts is often to redress discrimination against groups that are both marginalized and small, so this sample size limitation can prevent existing techniques from accomplishing their main aim. In clinical applications, this challenge combines with statistical issues that arise when models are used to guide treatment. We take a 3-step approach to addressing both of these challenges, building on the "counterfactual fairness" framework that accounts for confounding by treatment. First, we propose new estimands that leverage information across groups. Second, we estimate these quantities using a larger volume of data than existing techniques. Finally, we propose a novel data borrowing approach to incorporate "external data" that lacks outcomes and predictions but contains covariate and group membership information. We demonstrate application of our estimators to a risk prediction model used by a major Midwestern health system during the coronavirus disease 2019 (COVID-19) pandemic.