Comments about Joint Modeling of Cluster Size and Binary and Continuous Subunit-Specific Outcomes

In longitudinal studies and in clustered situations often binary and continuous response variables are observed and need to be modeled together. In a recent publication Dunson, Chen, and Harry (2003, Biometrics 59, 521-530) (DCH) propose a Bayesian approach for joint modeling of cluster size and binary and continuous subunit-specific outcomes and illustrate this approach with a developmental toxicity data example. In this note we demonstrate how standard software (PROC NLMIXED in SAS) can be used to obtain maximum likelihood estimates in an alternative parameterization of the model with a single cluster-level factor considered by DCH for that example. We also suggest that a more general model with additional cluster-level random effects provides a better fit to the data set. An apparent discrepancy between the estimates obtained by DCH and the estimates obtained earlier by Catalano and Ryan (1992, Journal of the American Statistical Association 87, 651-658) is also resolved. The issue of bias in inferences concerning the dose effect when cluster size is ignored is discussed. The maximum-likelihood approach considered herein is applicable to general situations with multiple clustered or longitudinally measured outcomes of different type and does not require prior specification and extensive programming.     

A semiparametric joint model for cluster size and subunit-specific interval-censored outcomes

Clustered data frequently arise in biomedical studies, where observations, or subunits, measured within a cluster are associated. The cluster size is said to be informative, if the outcome variable is associated with the number of subunits in a cluster. In most existing work, the informative cluster size issue is handled by marginal approaches based on within-cluster resampling, or cluster-weighted generalized estimating equations. Although these approaches yield consistent estimation of the marginal models, they do not allow estimation of within-cluster associations and are generally inefficient. In this paper, we propose a semiparametric joint model for clustered interval-censored event time data with informative cluster size. We use a random effect to account for the association among event times of the same cluster as well as the association between event times and the cluster size. For estimation, we propose a sieve maximum likelihood approach and devise a computationally-efficient expectation-maximization algorithm for implementation. The estimators are shown to be strongly consistent, with the Euclidean components being asymptotically normal and achieving semiparametric efficiency. Extensive simulation studies are conducted to evaluate the finite-sample performance, efficiency and robustness of the proposed method. We also illustrate our method via application to a motivating periodontal disease dataset.     

Association Models for Clustered Data with Binary and Continuous Responses

Summary .  We consider analysis of clustered data with mixed bivariate responses, i.e., where each member of the cluster has a binary and a continuous outcome. We propose a new bivariate random effects model that induces associations among the binary outcomes within a cluster, among the continuous outcomes within a cluster, between a binary outcome and a continuous outcome from different subjects within a cluster, as well as the direct association between the binary and continuous outcomes within the same subject. For the ease of interpretations of the regression effects, the marginal model of the binary response probability integrated over the random effects preserves the logistic form and the marginal expectation of the continuous response preserves the linear form. We implement maximum likelihood estimation of our model parameters using standard software such as PROC NLMIXED of SAS . Our simulation study demonstrates the robustness of our method with respect to the misspecification of the regression model as well as the random effects model. We illustrate our methodology by analyzing a developmental toxicity study of ethylene glycol in mice.     

Factor analytic models of clustered multivariate data with informative censoring

This article describes a general class of factor analytic models for the analysis of clustered multivariate data in the presence of informative missingness. We assume that there are distinct sets of cluster-level latent variables related to the primary outcomes and to the censoring process, and we account for dependency between these latent variables through a hierarchical model. A linear model is used to relate covariates and latent variables to the primary outcomes for each subunit. A generalized linear model accounts for covariate and latent variable effects on the probability of censoring for subunits within each cluster. The model accounts for correlation within clusters and within subunits through a flexible factor analytic framework that allows multiple latent variables and covariate effects on the latent variables. The structure of the model facilitates implementation of Markov chain Monte Carlo methods for posterior estimation. Data from a spermatotoxicity study are analyzed to illustrate the proposed approach.     

Sample Size Estimation in Cluster Randomized Studies with Varying Cluster Size

Cluster randomized studies are common in community trials. The standard method for estimating sample size for cluster randomized studies assumes a common cluster size. However often in cluster randomized studies, size of the clusters vary. In this paper, we derive sample size estimation for continuous outcomes for cluster randomized studies while accounting for the variability due to cluster size. It is shown that the proposed formula for estimating total cluster size can be obtained by adding a correction term to the traditional formula which uses the average cluster size. Application of these results to the design of a health promotion educational intervention study is discussed.     

Assessing Toxicities in a Clinical Trial: Bayesian Inference for Ordinal Data Nested within Categories

Summary This article addresses modeling and inference for ordinal outcomes nested within categorical responses. We propose a mixture of normal distributions for latent variables associated with the ordinal data. This mixture model allows us to fix without loss of generality the cutpoint parameters that link the latent variable with the observed ordinal outcome. Moreover, the mixture model is shown to be more flexible in estimating cell probabilities when compared to the traditional Bayesian ordinal probit regression model with random cutpoint parameters. We extend our model to take into account possible dependence among the outcomes in different categories. We apply the model to a randomized phase III study to compare treatments on the basis of toxicities recorded by type of toxicity and grade within type. The data include the different (categorical) toxicity types exhibited in each patient. Each type of toxicity has an (ordinal) grade associated to it. The dependence among the different types of toxicity exhibited by the same patient is modeled by introducing patient‐specific random effects.     

Estimation and testing with overdispersed proportions using the beta-logistic regression model of Heckman and Willis

Methods are presented for modeling dose-related effects in proportion data when extra-binomial variability is a concern. Motivation is taken from experiments in developmental toxicology, where similarity among conceptuses within a litter leads to intralitter correlations and to overdispersion in the observed proportions. Appeal is made to the well-known beta-binomial distribution to represent the overdispersion. From this, an exponential function of the linear predictor is used to model the dose-response relationship. The specification was introduced previously for econometric applications by Heckman and Willis; it induces a form of logistic regression for the mean response, together with a reciprocal biexponential model for the intralitter correlation. Large-sample, likelihood-based methods for estimating and testing the joint proportion-correlation response are studied. A developmental toxicity data set illustrates the methods.     

Multivariate methods for clustered binary data with multiple subclasses, with application to binary longitudinal data.

Clustered binary data occur frequently in biostatistical work. Several approaches have been proposed for the analysis of clustered binary data. In Rosner (1984, Biometrics 40, 1025-1035), a polychotomous logistic regression model was proposed that is a generalization of the beta-binomial distribution and allows for unit- and subunit-specific covariates, while controlling for clustering effects. One assumption of this model is that all pairs of subunits within a cluster are equally correlated. This is appropriate for ophthalmologic work where clusters are generally of size 2, but may be inappropriate for larger cluster sizes. A beta-binomial mixture model is introduced to allow for multiple subclasses within a cluster and to estimate odds ratios relating outcomes for pairs of subunits within a subclass as well as in different subclasses. To include covariates, an extension of the polychotomous logistic regression model is proposed, which allows one to estimate effects of unit-, class-, and subunit-specific covariates, while controlling for clustering using the beta-binomial mixture model. This model is applied to the analysis of respiratory symptom data in children collected over a 14-year period in East Boston, Massachusetts, in relation to maternal and child smoking, where the unit is the child and symptom history is divided into early-adolescent and late-adolescent symptom experience.     

Cryptic individual scaling relationships and the evolution of morphological scaling

Morphological scaling relationships between organ and body size—also known as allometries—describe the shape of a species, and the evolution of such scaling relationships is central to the generation of morphological diversity. Despite extensive modeling and empirical tests, however, the modes of selection that generate changes in scaling remain largely unknown. Here, we mathematically model the evolution of the group‐level scaling as an emergent property of individual‐level variation in the developmental mechanisms that regulate trait and body size. We show that these mechanisms generate a “cryptic individual scaling relationship” unique to each genotype in a population, which determines body and trait size expressed by each individual, depending on developmental nutrition. We find that populations may have identical population‐level allometries but very different underlying patterns of cryptic individual scaling relationships. Consequently, two populations with apparently the same morphological scaling relationship may respond very differently to the same form of selection. By focusing on the developmental mechanisms that regulate trait size and the patterns of cryptic individual scaling relationships they produce, our approach reveals the forms of selection that should be most effective in altering morphological scaling, and directs researcher attention on the actual, hitherto overlooked, targets of selection.     

Random effects modeling of multiple binomial responses using the multivariate binomial logit-normal distribution

The multivariate binomial logit-normal distribution is a mixture distribution for which, (i) conditional on a set of success probabilities and sample size indices, a vector of counts is independent binomial variates, and (ii) the vector of logits of the parameters has a multivariate normal distribution. We use this distribution to model multivariate binomial-type responses using a vector of random effects. The vector of logits of parameters has a mean that is a linear function of explanatory variables and has an unspecified or partly specified covariance matrix. The model generalizes and provides greater flexibility than the univariate model that uses a normal random effect to account for positive correlations in clustered data. The multivariate model is useful when different elements of the response vector refer to different characteristics, each of which may naturally have its own random effect. It is also useful for repeated binary measurement of a single response when there is a nonexchangeable association structure, such as one often expects with longitudinal data or when negative association exists for at least one pair of responses. We apply the model to an influenza study with repeated responses in which some pairs are negatively associated and to a developmental toxicity study with continuation-ratio logits applied to an ordinal response with clustered observations.     

Statistical power and sample size requirements for three level hierarchical cluster randomized trials

SUMMARY: Cluster randomized clinical trials (cluster-RCT), where the community entities serve as clusters, often yield data with three hierarchy levels. For example, interventions are randomly assigned to the clusters (level three unit). Health care professionals (level two unit) within the same cluster are trained with the randomly assigned intervention to provide care to subjects (level one unit). In this study, we derived a closed form power function and formulae for sample size determination required to detect an intervention effect on outcomes at the subject's level. In doing so, we used a test statistic based on maximum likelihood estimates from a mixed-effects linear regression model for three level data. A simulation study follows and verifies that theoretical power estimates based on the derived formulae are nearly identical to empirical estimates based on simulated data. Recommendations at the design stage of a cluster-RCT are discussed.     

A Bayesian approach to jointly modeling toxicity and biomarker expression in a phase I/II dose-finding trial

In this article, we propose a Bayesian approach to phase I/II dose-finding oncology trials by jointly modeling a binary toxicity outcome and a continuous biomarker expression outcome. We apply our method to a clinical trial of a new gene therapy for bladder cancer patients. In this trial, the biomarker expression indicates biological activity of the new therapy. For ethical reasons, the trial is conducted sequentially, with the dose for each successive patient chosen using both toxicity and activity data from patients previously treated in the trial. The modeling framework that we use naturally incorporates correlation between the binary toxicity and continuous activity outcome via a latent Gaussian variable. The dose-escalation/de-escalation decision rules are based on the posterior distributions of both toxicity and activity. A flexible state-space model is used to relate the activity outcome and dose. Extensive simulation studies show that the design reliably chooses the preferred dose using both toxicity and expression outcomes under various clinical scenarios.     

A potential outcomes approach to developmental toxicity analyses

Estimating the effects of a toxin on fetal development in animal models such as mice can be problematic, because the number of pups that develop and survive until birth may simultaneously affect developmental outcomes such as birth weight and be affected by the introduction of a toxin into the fetal environment. Also, comparing pups that survived until birth at a high dose of the toxin with pups that survived at low doses may underestimate the effect of the toxin, because the lower dose means include the less healthy pups that would not survive if exposed to a higher level of toxin. We consider this problem in a potential outcomes framework that defines the effect of the dose on the outcome as the difference between what the outcome would have been for a pup had the dam in which the pup develops been exposed to dose level Z=z* rather than dose level Z=z. To disentangle the direct effect of dose from the effect of litter size, we focus on effects defined within principal strata that are a function of the survival status of the pups at each of the possible dose levels. A unique contribution to the potential outcomes literature is that we allow the outcome for a subject to be dependent on the principal stratum to which other subjects within a cluster belong.     

Developmental plasticity and maternal effects of reproductive characteristics in the frog,Bombina orientalis

Summary Life history theory suggests that reproductive characteristics such as ovum size and clutch size should be well buffered against vararies of the environment. However, studies which demonstrate environmental sensitivity of reproductive characteristics are increasing in number, as are studies which find that maternal effects are responsible for much of the variation in developmental and growth rates in embryonic and larval fish and amphibians. The data reported here demonstrate that the environment, in terms of temperature and food availability that a specific individual encounters during vitellogenesis, exerts a strong influence on both egg size and number. Warmer temperatures and less food decrease ovum size, while colder temperatures and less food decrease clutch size. The variation in ovum size that is induced by the environment can exert a strong influence on variation in offspring development and growth and serve as an excellent model for studies on the evolution of developmental plasticity.     

Isolating cost drivers in interstitial lung disease treatment using nonparametric Bayesian methods

Mixture modeling is a popular approach to accommodate overdispersion, skewness, and multimodality features that are very common for health care utilization data. However, mixture modeling tends to rely on subjective judgment regarding the appropriate number of mixture components or some hypothesis about how to cluster the data. In this work, we adopt a nonparametric, variational Bayesian approach to allow the model to select the number of components while estimating their parameters. Our model allows for a probabilistic classification of observations into clusters and simultaneous estimation of a Gaussian regression model within each cluster. When we apply this approach to data on patients with interstitial lung disease, we find distinct subgroups of patients with differences in means and variances of health care costs, health and treatment covariates, and relationships between covariates and costs. The subgroups identified are readily interpretable, suggesting that this nonparametric variational approach to inference can discover valid insights into the factors driving treatment costs. Moreover, the learning algorithm we employed is very fast and scalable, which should make the technique accessible for a broad range of applications.     

Likelihood models for clustered binary and continuous outcomes: application to developmental toxicology

In developmental toxicology, methods based on dose response modeling and quantitative risk assessment are being actively pursued. Among live fetuses, the presence of malformations and reduction in fetal weight are of primary interest, but ordinarily, the dose-response relationships are characterized in each of the outcomes separately while appropriately accounting for clustering within litters. Jointly modeling the outcomes, allowing different relationships with dose while incorporating the correlation between the fetuses and the outcomes, may be more appropriate. We propose a likelihood-based model that is an extension of a correlated probit model to incorporate continuous outcomes. Our model maintains a marginal dose-response interpretation for the individual outcomes while taking into account both the correlations between outcomes on an individual fetus and those due to clustering. The joint risk of malformation and low birth weight can then be estimated directly. This approach is particularly well suited to estimating safe dose levels as part of quantitative risk assessment.     

A Bayesian semiparametric joint hierarchical model for longitudinal and survival data

This article proposes a new semiparametric Bayesian hierarchical model for the joint modeling of longitudinal and survival data. We relax the distributional assumptions for the longitudinal model using Dirichlet process priors on the parameters defining the longitudinal model. The resulting posterior distribution of the longitudinal parameters is free of parametric constraints, resulting in more robust estimates. This type of approach is becoming increasingly essential in many applications, such as HIV and cancer vaccine trials, where patients' responses are highly diverse and may not be easily modeled with known distributions. An example will be presented from a clinical trial of a cancer vaccine where the survival outcome is time to recurrence of a tumor. Immunologic measures believed to be predictive of tumor recurrence were taken repeatedly during follow-up. We will present an analysis of this data using our new semiparametric Bayesian hierarchical joint modeling methodology to determine the association of these longitudinal immunologic measures with time to tumor recurrence.     

Time‐Varying Latent Effect Model for Longitudinal Data with Informative Observation Times

Summary In analysis of longitudinal data, it is not uncommon that observation times of repeated measurements are subject‐specific and correlated with underlying longitudinal outcomes. Taking account of the dependence between observation times and longitudinal outcomes is critical under these situations to assure the validity of statistical inference. In this article, we propose a flexible joint model for longitudinal data analysis in the presence of informative observation times. In particular, the new procedure considers the shared random‐effect model and assumes a time‐varying coefficient for the latent variable, allowing a flexible way of modeling longitudinal outcomes while adjusting their association with observation times. Estimating equations are developed for parameter estimation. We show that the resulting estimators are consistent and asymptotically normal, with variance–covariance matrix that has a closed form and can be consistently estimated by the usual plug‐in method. One additional advantage of the procedure is that it provides a unified framework to test whether the effect of the latent variable is zero, constant, or time‐varying. Simulation studies show that the proposed approach is appropriate for practical use. An application to a bladder cancer data is also given to illustrate the methodology.     

Assessing the application of a geographic presence-only model for land suitability mapping

Recent advances in ecological modeling have focused on novel methods for characterizing the environment that use presence-only data and machine-learning algorithms to predict the likelihood of species occurrence. These novel methods may have great potential for land suitability applications in the developing world where detailed land cover information is often unavailable or incomplete. This paper assesses the adaptation and application of the presence-only geographic species distribution model, MaxEnt, for agricultural crop suitability mapping in a rural Thailand where lowland paddy rice and upland field crops predominant. To assess this modeling approach, three independent crop presence datasets were used including a social-demographic survey of farm households, a remote sensing classification of land use/land cover, and ground control points, used for geodetic and thematic reference that vary in their geographic distribution and sample size. Disparate environmental data were integrated to characterize environmental settings across Nang Rong District, a region of approximately 1300 sq. km in size. Results indicate that the MaxEnt model is capable of modeling crop suitability for upland and lowland crops, including rice varieties, although model results varied between datasets due to the high sensitivity of the model to the distribution of observed crop locations in geographic and environmental space. Accuracy assessments indicate that model outcomes were influenced by the sample size and the distribution of sample points in geographic and environmental space. The need for further research into accuracy assessments of presence-only models lacking true absence data is discussed. We conclude that the MaxEnt model can provide good estimates of crop suitability, but many areas need to be carefully scrutinized including geographic distribution of input data and assessment methods to ensure realistic modeling results.     

Interpreting and Testing Interactions in Conditional Mixture Models

Mixture modeling applications in psychology often include covariates to explain class membership and aid in construct validation of the latent classification variable. These applications tend to use between-class models involving only main effects of predictors. However, a variety of developmental theories posit interactions among risk and protective variables in predicting membership in trajectory classes or behavioral symptom profiles. This article bridges this disconnect between substantive theory and methodological practice by presenting and comparing two approaches for testing interactive effects of predictors on class membership: product term (PT) and multiple group (MG) approaches. For each approach, we discuss alternative interpretation strategies involving predicted probabilities and odds ratios; we also discuss when the approaches provide equivalent inferences. Published longitudinal and cross-sectional mixture model applications that had originally allowed for only additive effects on class membership are re-analyzed to illustrate the testing and interpretation of interactive effects on class membership using both PT and MG approaches.