Statistics in Medicine最新文献

Intensive longitudinal biomarker data are increasingly common in scientific studies that seek temporally granular understanding of the role of behavioral and physiological factors in relation to outcomes of interest. Intensive longitudinal biomarker data, such as those obtained from wearable devices, are often obtained at a high frequency typically resulting in several hundred to thousand observations per individual measured over minutes, hours, or days. Often in longitudinal studies, the primary focus is on relating the means of biomarker trajectories to an outcome, and the variances are treated as nuisance parameters, although they may also be informative for the outcomes. In this paper, we propose a Bayesian hierarchical model to jointly model a cross-sectional outcome and the intensive longitudinal biomarkers. To model the variability of biomarkers and deal with the high intensity of data, we develop subject-level cubic B-splines and allow the sharing of information across individuals for both the residual variability and the random effects variability. Then different levels of variability are extracted and incorporated into an outcome submodel for inferential and predictive purposes. We demonstrate the utility of the proposed model via an application involving bio-monitoring of hertz-level heart rate information from a study on social stress.

To complement the conventional area under the ROC curve (AUC) which cannot fully describe the diagnostic accuracy of some non-standard biomarkers, we introduce a transformed ROC curve and its associated transformed AUC (TAUC) in this article, and show that TAUC can relate the original improper biomarker to a proper biomarker after a non-monotone transformation. We then provide nonparametric estimation of the non-monotone transformation and TAUC, and establish their consistency and asymptotic normality. We conduct extensive simulation studies to assess the performance of the proposed TAUC method and compare with the traditional methods. Case studies on real biomedical data are provided to illustrate the proposed TAUC method. We are able to identify more important biomarkers that tend to escape the traditional screening method.

As a crucial tool in neuroscience, mediation analysis has been developed and widely adopted to elucidate the role of intermediary variables derived from neuroimaging data. Typically, structural equation models (SEMs) are employed to investigate the influences of exposures on outcomes, with model coefficients being interpreted as causal effects. While existing SEMs have proven to be effective tools for mediation analysis involving various neuroimaging-related mediators, limited research has explored scenarios where these mediators are derived from the shape space. In addition, the linear relationship assumption adopted in existing SEMs may lead to substantial efficiency losses and decreased predictive accuracy in real-world applications. To address these challenges, we introduce a novel framework for shape mediation analysis, designed to explore the causal relationships between genetic exposures and clinical outcomes, whether mediated or unmediated by shape-related factors while accounting for potential confounding variables. Within our framework, we apply the square-root velocity function to extract elastic shape representations, which reside within the linear Hilbert space of square-integrable functions. Subsequently, we introduce a two-layer shape regression model to characterize the relationships among neurocognitive outcomes, elastic shape mediators, genetic exposures, and clinical confounders. Both estimation and inference procedures are established for unknown parameters along with the corresponding causal estimands. The asymptotic properties of estimated quantities are investigated as well. Both simulated studies and real-data analyses demonstrate the superior performance of our proposed method in terms of estimation accuracy and robustness when compared to existing approaches for estimating causal estimands.

High-dimensional multinomial regression models are very useful in practice but have received less research attention than logistic regression models, especially from the perspective of statistical inference. In this work, we analyze the estimation and prediction error of the contrast-based ${\ell }_1$ -penalized multinomial regression model and extend the debiasing method to the multinomial case, providing a valid confidence interval for each coefficient and $p$ value of the individual hypothesis test. We also examine cases of model misspecification and non-identically distributed data to demonstrate the robustness of our method when some assumptions are violated. We apply the debiasing method to identify important predictors in the progression into dementia of different subtypes. Results from extensive simulations show the superiority of the debiasing method compared to other inference methods.

In many trials and experiments, subjects are not only observed once but multiple times, resulting in a cluster of possibly correlated observations (e.g., brain regions per patient). Observations often do not fulfill model assumptions of mixed models and require the use of nonparametric methods. In this article, we develop and present a purely nonparametric rank-based procedure that flexibly allows the unbiased and consistent estimation of the Wilcoxon-Mann-Whitney effect in clustered data designs. Compared with existing methods, we allow flexible weights to be used in effect estimation. Additionally, we develop global and multiple contrast test procedures to test null hypotheses formulated regarding the generalized Mann-Whitney effects and for the computation of range-preserving simultaneous confidence intervals in a unified way. Extensive simulation studies show that these methods control the type-I error rate well and have reasonable power to detect alternatives in various situations.

The COVID-19 infection fatality rate (IFR) is the proportion of individuals infected with SARS-CoV-2 who subsequently die. As COVID-19 disproportionately affects older individuals, age-specific IFR estimates are imperative to facilitate comparisons of the impact of COVID-19 between locations and prioritize distribution of scarce resources. However, there lacks a coherent method to synthesize available data to create estimates of IFR and seroprevalence that vary continuously with age and adequately reflect uncertainties inherent in the underlying data. In this article, we introduce a novel Bayesian hierarchical model to estimate IFR as a continuous function of age that acknowledges heterogeneity in population age structure across locations and accounts for uncertainty in the estimates due to seroprevalence sampling variability and the imperfect serology test assays. Our approach simultaneously models test assay characteristics, serology, and death data, where the serology and death data are often available only for binned age groups. Information is shared across locations through hierarchical modeling to improve estimation of the parameters with limited data. Modeling data from 26 developing country locations during the first year of the COVID-19 pandemic, we found seroprevalence did not change dramatically with age, and the IFR at age 60 was above the high-income country estimate for most locations.

Many individually randomized group treatment (IRGT) trials randomly assign individuals to study arms but deliver treatments via shared agents, such as therapists, surgeons, or trainers. Post-randomization interactions induce correlations in outcome measures between participants sharing the same agent. Agents can be nested in or crossed with trial arm, and participants may interact with a single agent or with multiple agents. These complications have led to ambiguity in choice of models but there have been no systematic efforts to identify appropriate analytic models for these study designs. To address this gap, we undertook a simulation study to examine the performance of candidate analytic models in the presence of complex clustering arising from multiple membership, single membership, and single agent settings, in both nested and crossed designs and for a continuous outcome. With nested designs, substantial type I error rate inflation was observed when analytic models did not account for multiple membership and when analytic model weights characterizing the association with multiple agents did not match the data generating mechanism. Conversely, analytic models for crossed designs generally maintained nominal type I error rates unless there was notable imbalance in the number of participants that interact with each agent.

The current high-dimensional linear factor models fail to account for the different types of variables, while high-dimensional nonlinear factor models often overlook the overdispersion present in mixed-type data. However, overdispersion is prevalent in practical applications, particularly in fields like biomedical and genomics studies. To address this practical demand, we propose an overdispersed generalized factor model (OverGFM) for performing high-dimensional nonlinear factor analysis on overdispersed mixed-type data. Our approach incorporates an additional error term to capture the overdispersion that cannot be accounted for by factors alone. However, this introduces significant computational challenges due to the involvement of two high-dimensional latent random matrices in the nonlinear model. To overcome these challenges, we propose a novel variational EM algorithm that integrates Laplace and Taylor approximations. This algorithm provides iterative explicit solutions for the complex variational parameters and is proven to possess excellent convergence properties. We also develop a criterion based on the singular value ratio to determine the optimal number of factors. Numerical results demonstrate the effectiveness of this criterion. Through comprehensive simulation studies, we show that OverGFM outperforms state-of-the-art methods in terms of estimation accuracy and computational efficiency. Furthermore, we demonstrate the practical merit of our method through its application to two datasets from genomics. To facilitate its usage, we have integrated the implementation of OverGFM into the R package GFM.

Sensor devices, such as accelerometers, are widely used for measuring physical activity (PA). These devices provide outputs at fine granularity (e.g., 10-100 Hz or minute-level), which while providing rich data on activity patterns, also pose computational challenges with multilevel densely sampled data, resulting in PA records that are measured continuously across multiple days and visits. On the other hand, a scalar health outcome (e.g., BMI) is usually observed only at the individual or visit level. This leads to a discrepancy in numbers of nested levels between the predictors (PA) and outcomes, raising analytic challenges. To address this issue, we proposed a multilevel longitudinal functional principal component analysis (mLFPCA) model to directly model multilevel functional PA inputs in a longitudinal study, and then implemented a longitudinal functional principal component regression (FPCR) to explore the association between PA and obesity-related health outcomes. Additionally, we conducted a comprehensive simulation study to examine the impact of imbalanced multilevel data on both mLFPCA and FPCR performance and offer guidelines for selecting optimal methods.

From early in the coronavirus disease 2019 (COVID-19) pandemic, there was interest in using machine learning methods to predict COVID-19 infection status based on vocal audio signals, for example, cough recordings. However, early studies had limitations in terms of data collection and of how the performances of the proposed predictive models were assessed. This article describes how these limitations have been overcome in a study carried out by the Turing-RSS Health Data Laboratory and the UK Health Security Agency. As part of the study, the UK Health Security Agency collected a dataset of acoustic recordings, SARS-CoV-2 infection status and extensive study participant meta-data. This allowed us to rigorously assess state-of-the-art machine learning techniques to predict SARS-CoV-2 infection status based on vocal audio signals. The lessons learned from this project should inform future studies on statistical evaluation methods to assess the performance of machine learning techniques for public health tasks.