Search

The preclinical stage of many neurodegenerative diseases can span decades before symptoms become apparent. Understanding the sequence of preclinical biomarker changes provides a critical opportunity for early diagnosis and effective intervention prior to significant loss of patients' brain functions. The main challenge to early detection lies in the absence of direct observation of the disease state and the considerable variability in both biomarkers and disease dynamics among individuals. Recent research hypothesized the existence of subgroups with distinct biomarker patterns due to co-morbidities and degrees of brain resilience. Our ability to early diagnose and intervene during the preclinical stage of neurodegenerative diseases will be enhanced by further insights into heterogeneity in the biomarker-disease relationship. In this paper, we focus on Alzheimer's disease (AD) and attempt to identify the systematic patterns within the heterogeneous AD biomarker-disease cascade. Specifically, we quantify the disease progression using a dynamic latent variable whose mixture distribution represents patient subgroups. Model estimation uses Hamiltonian Monte Carlo with the number of clusters determined by the Bayesian Information Criterion (BIC). We report simulation studies that investigate the performance of the proposed model in finite sample settings that are similar to our motivating application. We apply the proposed model to the BIOCARD data, a longitudinal study that was conducted over two decades among individuals who were initially cognitively normal. Our application yields evidence consistent with the hypothetical model of biomarker dynamics presented in Jack et al. (2013). In addition, our analysis identified two subgroups with distinct disease-onset patterns. Finally, we develop a dynamic prediction approach to improve the precision of prognoses.

Verbal autopsy (VA) algorithms are routinely used to determine individual-level causes of death (COD) in many low-and-middle-income countries, which are then aggregated to derive population-level cause-specific mortality fractions (CSMF), essential to informing public health policies. However, VA algorithms frequently misclassify COD and introduce bias in CSMF estimates. A recent method, VA-calibration, can correct for this bias using a VA misclassification matrix estimated from paired data on COD from both VA and minimally invasive tissue sampling (MITS) from the Child Health and Mortality Prevention Surveillance (CHAMPS) Network. Due to the limited sample size, CHAMPS data are pooled across all countries, implicitly assuming that the misclassification rates are homogeneous. In this research, we show that the VA misclassification matrices are substantially heterogeneous across countries, thereby biasing the VA-calibration. We develop a coherent framework for modeling country-specific misclassification matrices in data-scarce settings. We first introduce a novel base model based on two latent mechanisms: intrinsic accuracy and systematic preference to parsimoniously characterize misclassifications. We prove that they are identifiable from the data and manifest as a form of invariance in certain misclassification odds, a pattern evident in the CHAMPS data. Then we expand from this base model, adding higher complexity and country-specific heterogeneity via interpretable effect sizes. Shrinkage priors balance the bias-variance tradeoff by adaptively favoring simpler models. We publish uncertainty-quantified estimates of VA misclassification rates for 6 countries. This effort broadens VA-calibration's future applicability and strengthens ongoing efforts of using VA for mortality surveillance.

Dynamic prediction of causal effects under different treatment regimes conditional on an individual's characteristics and longitudinal history is an essential problem in precision medicine. This is challenging in practice because outcomes and treatment assignment mechanisms are unknown in observational studies, an individual's treatment efficacy is a counterfactual, and the existence of selection bias is often unavoidable. We propose a Bayesian framework for identifying subgroup counterfactual benefits of dynamic treatment regimes by adapting Bayesian g-computation algorithm (J. Robins, 1986; Zhou, Elliott, & Little, 2019) to incorporate multivariate generalized linear mixed-effects models. Unmeasured time-invariant factors are identified as subject-specific random effects in the assumed joint distribution of outcomes, time-varying confounders, and treatment assignments. Existing methods mostly assume no unmeasured confounding and focus on balancing the observed confounder distributions between different treatments, while our method allows the presence of time-invariant unmeasured confounding. We propose a sequential ignorability assumption based on treatment assignment heterogeneity, which is analogous to balancing the latent tendency toward each treatment due to unmeasured time-invariant factors beyond the observables. We use simulation studies to assess the sensitivity of the proposed method's performance to various model assumptions. The method is applied to observational clinical data to investigate the efficacy of continuously using mycophenolate in different subgroups of scleroderma patients who were treated with the drug., Comment: 19 pages, 5 figures

This paper describes and illustrates the functionality of the baker R package. The package estimates a suite of nested partially-latent class models (NPLCM) for multivariate binary responses that are observed under a case-control design. The baker package allows researchers to flexibly estimate population-level class prevalences and posterior probabilities of class membership for individual cases. Estimation is accomplished by calling a cross-platform automatic Bayesian inference software JAGS through a wrapper R function that parses model specifications and data inputs. The baker package provides many useful features, including data ingestion, exploratory data analyses, model diagnostics, extensive plotting and visualization options, catalyzing communications between practitioners and domain scientists. Package features and workflows are illustrated using simulated and real data sets. Package URL: https://github.com/zhenkewu/baker, Comment: 30 pages

COVID-19 has challenged health systems to learn how to learn. This paper describes the context, methods and challenges for learning to improve COVID-19 care at one academic health center. Challenges to learning include: (1) choosing a right clinical target; (2) designing methods for accurate predictions by borrowing strength from prior patients' experiences; (3) communicating the methodology to clinicians so they understand and trust it; (4) communicating the predictions to the patient at the moment of clinical decision; and (5) continuously evaluating and revising the methods so they adapt to changing patients and clinical demands. To illustrate these challenges, this paper contrasts two statistical modeling approaches - prospective longitudinal models in common use and retrospective analogues complementary in the COVID-19 context - for predicting future biomarker trajectories and major clinical events. The methods are applied to and validated on a cohort of 1,678 patients who were hospitalized with COVID-19 during the early months of the pandemic. We emphasize graphical tools to promote physician learning and inform clinical decision making., Comment: 16 pages, 7 figures, 1 table1

Compositional data are common in many fields, both as outcomes and predictor variables. The inventory of models for the case when both the outcome and predictor variables are compositional is limited and the existing models are difficult to interpret, due to their use of complex log-ratio transformations. We develop a transformation-free linear regression model where the expected value of the compositional outcome is expressed as a single Markov transition from the compositional predictor. Our approach is based on generalized method of moments thereby not requiring complete specification of data likelihood and is robust to different data generating mechanism. Our model is simple to interpret, allows for 0s and 1s in both the compositional outcome and covariates, and subsumes several interesting subcases of interest. We also develop a permutation test for linear independence. Finally, we show that despite its simplicity, our model accurately captures the relationship between compositional data from education and medical research., Comment: 19 pages, 7 figures

Quantification learning is the task of prevalence estimation for a test population using predictions from a classifier trained on a different population. Quantification methods assume that the sensitivities and specificities of the classifier are either perfect or transportable from the training to the test population. These assumptions are inappropriate in the presence of dataset shift, when the misclassification rates in the training population are not representative of those for the test population. Quantification under dataset shift has been addressed only for single-class (categorical) predictions and assuming perfect knowledge of the true labels on a small subset of the test population. We propose generalized Bayes quantification learning (GBQL) that uses the entire compositional predictions from probabilistic classifiers and allows for uncertainty in true class labels for the limited labeled test data. Instead of positing a full model, we use a model-free Bayesian estimating equation approach to compositional data based only on a first-moment assumption. The idea will be useful in Bayesian compositional data analysis in general as it is robust to different generating mechanisms for compositional data and includes categorical outputs as a special case. We show how our method yields existing quantification approaches as special cases. Extension to an ensemble GBQL that uses predictions from multiple classifiers yielding inference robust to inclusion of a poor classifier is discussed. We outline a fast and efficient Gibbs sampler using a rounding and coarsening approximation to the loss functions. We also establish posterior consistency, asymptotic normality and valid coverage of interval estimates from GBQL, as well as finite sample posterior concentration rate. Empirical performance of GBQL is demonstrated through simulations and analysis of real data with evident dataset shift., Comment: 59 pages, 15 figures

Cookstove replacement trials have found mixed results on their impact on respiratory health. The limited range of concentrations and small sample sizes of individual studies are important factors that may be limiting their statistical power. We present a hierarchical approach to modeling exposure concentrations and pooling data from multiple studies in order to estimate a common exposure-response curve. The exposure concentration model accommodates temporally sparse, clustered longitudinal observations. The exposure-response curve model provides a flexible, semi-parametric estimate of the exposure-response relationship while accommodating heterogeneous clustered data. We apply this model to data from three studies of cookstoves and respiratory infections in children in Nepal, which represent three study types: crossover trial, parallel trial, and case-control study. We find evidence of increased odds of disease for particulate matter concentrations between 50 and 200 $\mu$g/m$^3$ and a flattening of the exposure-response curve for higher exposure concentrations. The model we present can incorporate additional studies and be applied to other settings., Comment: 22 pages, 8 figures

Computer-coded verbal autopsy (CCVA) algorithms predict cause of death from high-dimensional family questionnaire data (verbal autopsies) of a deceased individual. CCVA algorithms are typically trained on non-local data, then used to generate national and regional estimates of cause-specific mortality fractions. These estimates may be inaccurate if the non-local training data is different from the local population of interest. This problem is a special case of transfer learning. However, most transfer learning classification approaches are concerned with individual (e.g. a person's) classification within a target domain (e.g. a particular population) with training performed in data from a source domain. Epidemiologists are often more interested in estimating population-level etiological distributions, using datasets much smaller than those used in common transfer learning applications. We present a parsimonious hierarchical Bayesian transfer learning framework to directly estimate population-level class probabilities in a target domain. To address small sample sizes, we introduce a novel shrinkage prior for the transfer error rates guaranteeing that, in absence of any labeled target domain data or when the baseline classifier has zero transfer error, the calibrated estimate of class probabilities coincides with the naive estimates from the baseline classifier, thereby subsuming the default practice as a special case. A novel Gibbs sampler using data-augmentation enables fast implementation. We extend our approach to use not one, but an ensemble of baseline classifiers. Theoretical and empirical results demonstrate how the ensemble model favors the most accurate baseline classifier. We present extensions allowing class probabilities to vary with covariates, and an EM-algorithm-based MAP estimation. An R-package implementing this method is developed.

In this paper, we propose a general framework for combining evidence of varying quality to estimate underlying binary latent variables in the presence of restrictions imposed to respect the scientific context. The resulting algorithms cluster the multivariate binary data in a manner partly guided by prior knowledge. The primary model assumptions are that 1) subjects belong to classes defined by unobserved binary states, such as the true presence or absence of pathogens in epidemiology, or of antibodies in medicine, or the "ability" to correctly answer test questions in psychology, 2) a binary design matrix $\Gamma$ specifies relevant features in each class, and 3) measurements are independent given the latent class but can have different error rates. Conditions ensuring parameter identifiability from the likelihood function are discussed and inform the design of a novel posterior inference algorithm that simultaneously estimates the number of clusters, design matrix $\Gamma$, and model parameters. In finite samples and dimensions, we propose prior assumptions so that the posterior distribution of the number of clusters and the patterns of latent states tend to concentrate on smaller values and sparser patterns, respectively. The model readily extends to studies where some subjects' latent classes are known or important prior knowledge about differential measurement accuracy is available from external sources. The methods are illustrated with an analysis of protein data to detect clusters representing auto-antibody classes among scleroderma patients., Comment: 37 pages and three figures in Main Paper, 24 pages in Supplementary Materials

Autoimmune diseases are characterized by highly specific immune responses against molecules in self-tissues. Different autoimmune diseases are characterized by distinct immune responses, making autoantibodies useful for diagnosis and prediction. In many diseases, the targets of autoantibodies are incompletely defined. Although the technologies for autoantibody discovery have advanced dramatically over the past decade, each of these techniques generates hundreds of possibilities, which are onerous and expensive to validate. We set out to establish a method to greatly simplify autoantibody discovery, using a pre-filtering step to define subgroups with similar specificities based on migration of radiolabeled, immunoprecipitated proteins on sodium dodecyl sulfate (SDS) gels and autoradiography [$\textbf{G}$el $\textbf{E}$lectrophoresis and band detection on $\textbf{A}$utoradiograms (GEA)]. Human recognition of patterns is not optimal when the patterns are complex or scattered across many samples. Multiple sources of errors - including irrelevant intensity differences and warping of gels - have challenged automation of pattern discovery from autoradiograms. In this paper, we address these limitations using a Bayesian hierarchical model with shrinkage priors for pattern alignment and spatial dewarping. The Bayesian model combines information from multiple gel sets and corrects spatial warping for coherent estimation of autoantibody signatures defined by presence or absence of a grid of landmark proteins. We show the pre-processing creates more clearly separated clusters and improves the accuracy of autoantibody subset detection via hierarchical clustering. Finally, we demonstrate the utility of the proposed methods with GEA data from scleroderma patients., Comment: Appendix containing algorithmic details and extra figures can be found here: http://zhenkewu.com/assets/pdfs/papers/wu-2017-gel-preprocessing_supp.pdf

The Pneumonia Etiology Research for Child Health (PERCH) study seeks to use modern measurement technology to infer the causes of pneumonia for which gold-standard evidence is unavailable. The paper describes a latent variable model designed to infer from case-control data the etiology distribution for the population of cases, and for an individual case given his or her measurements. We assume each observation is drawn from a mixture model for which each component represents one cause or disease class. The model addresses a major limitation of the traditional latent class approach by taking account of residual dependence among multivariate binary outcome given disease class, hence reduces estimation bias, retains efficiency and offers more valid inference. Such "local dependence" on a single subject is induced in the model by nesting latent subclasses within each disease class. Measurement precision and covariation can be estimated using the control sample for whom the class is known. In a Bayesian framework, we use stick-breaking priors on the subclass indicators for model-averaged inference across different numbers of subclasses. Assessment of model fit and individual diagnosis are done using posterior samples drawn by Gibbs sampling. We demonstrate the utility of the method on simulated and on the motivating PERCH data., Comment: 30 pages with 5 figures and 1 table; 1 appendix with 4 figures and 1 table

Hierarchical Bayesian models can be especially useful in precision medicine settings, where clinicians are interested in estimating the patient-level latent variables associated with an individual's current health state and its trajectory. Such models are often fit using batch Markov Chain Monte Carlo (MCMC). However, the slow speed of batch MCMC computation makes it difficult to implement in clinical settings, where immediate latent variable estimates are often desired in response to new patient data. In this report, we discuss how importance sampling (IS) can instead be used to obtain fast, in-clinic estimates of patient-level latent variables. We apply IS to the hierarchical model proposed in Coley et al (2015) for predicting an individual's underlying prostate cancer state. We find that latent variable estimates via IS can typically be obtained in 1-10 seconds per person and have high agreement with estimates coming from longer-running batch MCMC methods. Alternative options for out-of-sample fitting and online updating are also discussed., Comment: 9 pages, 2 figures

In this article, we present a Bayesian hierarchical model for predicting a latent health state from longitudinal clinical measurements. Model development is motivated by the need to integrate multiple sources of data to improve clinical decisions about whether to remove or irradiate a patient's prostate cancer. Existing modeling approaches are extended to accommodate measurement error in cancer state determinations based on biopsied tissue, clinical measurements possibly not missing at random, and informative partial observation of the true state. The proposed model enables estimation of whether an individual's underlying prostate cancer is aggressive, requiring surgery and/or radiation, or indolent, permitting continued surveillance. These individualized predictions can then be communicated to clinicians and patients to inform decision-making. We demonstrate the model with data from a cohort of low risk prostate cancer patients at Johns Hopkins University and assess predictive accuracy among a subset for whom true cancer state is observed. Simulation studies confirm model performance and explore the impact of adjusting for informative missingness on true state predictions. R code and simulated data available at https://github.com/rycoley/prediction-prostate-surveillance., Comment: 25 pages (double-spaced, with references), 6 figures for primary paper and 44 pages, 22 figures, 15 tables for appendix; Revision submittied for review 5/16; Changes from original version include changed title, additional simulations and figures, editing for clarity

In population studies on the etiology of disease, one goal is the estimation of the fraction of cases attributable to each of several causes. For example, pneumonia is a clinical diagnosis of lung infection that may be caused by viral, bacterial, fungal, or other pathogens. The study of pneumonia etiology is challenging because directly sampling from the lung to identify the etiologic pathogen is not standard clinical practice in most settings. Instead, measurements from multiple peripheral specimens are made. This paper introduces the statistical methodology designed for estimating the population etiology distribution and the individual etiology probabilities in the Pneumonia Etiology Research for Child Health (PERCH) study of 9; 500 children for 7 sites around the world. We formulate the scientific problem in statistical terms as estimating the mixing weights and latent class indicators under a partially-latent class model (pLCM) that combines heterogeneous measurements with different error rates obtained from a case-control study. We introduce the pLCM as an extension of the latent class model. We also introduce graphical displays of the population data and inferred latent-class frequencies. The methods are tested with simulated data, and then applied to PERCH data. The paper closes with a brief description of extensions of the pLCM to the regression setting and to the case where conditional independence among the measures is relaxed., Comment: 25 pages, 4 figures, 1 supplementary material