Bradley Carlin - Academia.edu (original) (raw)

Papers by Bradley Carlin

Research paper thumbnail of Combining nonexchangeable functional or survival data sources in oncology using generalized mixture commensurate priors

The Annals of Applied Statistics, 2015

Conventional approaches to statistical inference preclude structures that facilitate incorporatio... more Conventional approaches to statistical inference preclude structures that facilitate incorporation of supplemental information acquired from similar circumstances. For example, the analysis of data obtained using perfusion computed tomography to characterize functional imaging biomarkers in cancerous regions of the liver can benefit from partially informative data collected concurrently in non-cancerous regions. This paper presents a hierarchical model structure that leverages all available information about a curve, using penalized splines, while accommodating important between-source features. Our proposed methods flexibly borrow strength from the supplemental data to a degree that reflects the commensurability of the supplemental curve with the primary curve. We investigate our method's properties for nonparametric regression via simulation, and apply it to a set of liver cancer data. We also apply our method for a semiparametric hazard model to data from a clinical trial that compares time to disease progression for three colorectal cancer treatments, while supplementing inference with information from a previous trial that tested the current standard of care.

Research paper thumbnail of A Bayesian missing data framework for generalized multiple outcome mixed treatment comparisons

Research synthesis methods, Jan 4, 2015

Bayesian statistical approaches to mixed treatment comparisons (MTCs) are becoming more popular b... more Bayesian statistical approaches to mixed treatment comparisons (MTCs) are becoming more popular because of their flexibility and interpretability. Many randomized clinical trials report multiple outcomes with possible inherent correlations. Moreover, MTC data are typically sparse (although richer than standard meta-analysis, comparing only two treatments), and researchers often choose study arms based upon which treatments emerge as superior in previous trials. In this paper, we summarize existing hierarchical Bayesian methods for MTCs with a single outcome and introduce novel Bayesian approaches for multiple outcomes simultaneously, rather than in separate MTC analyses. We do this by incorporating partially observed data and its correlation structure between outcomes through contrast-based and arm-based parameterizations that consider any unobserved treatment arms as missing data to be imputed. We also extend the model to apply to all types of generalized linear model outcomes, suc...

Research paper thumbnail of Rejoinder to the discussion of "a Bayesian missing data framework for generalized multiple outcome mixed treatment comparisons," by S. Dias and A. E. Ades

Research synthesis methods, Jan 13, 2015

Research paper thumbnail of Flexible Cure Rate Modelling Under Latent Activation Schemes

With rapid improvements in medical treatment and health care, many data sets dealing with time to... more With rapid improvements in medical treatment and health care, many data sets dealing with time to relapse or death now reveal a substantial portion of patients who are cured (that is, who never experience the event). Extended survival models called cure rate models account for the probability of a subject being cured and can be broadly classified into the classical mixture models of Berkson and Gage (1952; "BG type") or the stochastic tumor models pioneered by Yakovlev (1996) and extended to a hierarchical framework by Chen, Ibrahim and Sinha (1999; "YCIS type"). Recent developments in Bayesian hierarchical cure models have evoked significant interest regarding relationships and preferences between these two classes of models. Our present work proposes a unifying class of cure rate models that facilitates flexible hierarchical model-building while including both existing cure model classes as special cases. This unifying class enables robust modelling by accounting for uncertainty in underlying mechanisms leading to cure. Issues such as regressing on the cure fraction and propriety of the associated posterior distributions under different modelling assumptions are also discussed. Finally, we offer a simulation study and also illustrate with two data sets (one on melanoma and the other on breast cancer) that reveal our framework's ability to distinguish among underlying mechanisms that lead to relapse and cure.

Research paper thumbnail of Bayesian hierarchical models for network meta-analysis incorporating nonignorable missingness

Statistical methods in medical research, Jan 28, 2015

Network meta-analysis expands the scope of a conventional pairwise meta-analysis to simultaneousl... more Network meta-analysis expands the scope of a conventional pairwise meta-analysis to simultaneously compare multiple treatments, synthesizing both direct and indirect information and thus strengthening inference. Since most of trials only compare two treatments, a typical data set in a network meta-analysis managed as a trial-by-treatment matrix is extremely sparse, like an incomplete block structure with significant missing data. Zhang et al. proposed an arm-based method accounting for correlations among different treatments within the same trial and assuming that absent arms are missing at random. However, in randomized controlled trials, nonignorable missingness or missingness not at random may occur due to deliberate choices of treatments at the design stage. In addition, those undertaking a network meta-analysis may selectively choose treatments to include in the analysis, which may also lead to missingness not at random. In this paper, we extend our previous work to incorporate...

Research paper thumbnail of Incorporation of individual-patient data in network meta-analysis for multiple continuous endpoints, with application to diabetes treatment

Statistics in medicine, Jan 30, 2015

Availability of individual patient-level data (IPD) broadens the scope of network meta-analysis (... more Availability of individual patient-level data (IPD) broadens the scope of network meta-analysis (NMA) and enables us to incorporate patient-level information. Although IPD is a potential gold mine in biomedical areas, methodological development has been slow owing to limited access to such data. In this paper, we propose a Bayesian IPD NMA modeling framework for multiple continuous outcomes under both contrast-based and arm-based parameterizations. We incorporate individual covariate-by-treatment interactions to facilitate personalized decision making. Furthermore, we can find subpopulations performing well with a certain drug in terms of predictive outcomes. We also impute missing individual covariates via an MCMC algorithm. We illustrate this approach using diabetes data that include continuous bivariate efficacy outcomes and three baseline covariates and show its practical implications. Finally, we close with a discussion of our results, a review of computational challenges, and ...

Research paper thumbnail of Bayesian modeling and analysis for gradients in spatiotemporal processes

Biometrics, Jan 20, 2015

Stochastic process models are widely employed for analyzing spatiotemporal datasets in various sc... more Stochastic process models are widely employed for analyzing spatiotemporal datasets in various scientific disciplines including, but not limited to, environmental monitoring, ecological systems, forestry, hydrology, meteorology, and public health. After inferring on a spatiotemporal process for a given dataset, inferential interest may turn to estimating rates of change, or gradients, over space and time. This manuscript develops fully model-based inference on spatiotemporal gradients under continuous space, continuous time settings. Our contribution is to offer, within a flexible spatiotemporal process model setting, a framework to estimate arbitrary directional gradients over space at any given timepoint, temporal derivatives at any given spatial location and, finally, mixed spatiotemporal gradients that reflect rapid change in spatial gradients over time and vice-versa. We achieve such inference without compromising on rich and flexible spatiotemporal process models and use nonse...

Research paper thumbnail of Detecting outlying trials in network meta-analysis

Statistics in medicine, Jan 8, 2015

Network meta-analysis (NMA) expands the scope of a conventional pairwise meta-analysis to simulta... more Network meta-analysis (NMA) expands the scope of a conventional pairwise meta-analysis to simultaneously handle multiple treatment comparisons. However, some trials may appear to deviate markedly from the others and thus be inappropriate to be synthesized in the NMA. In addition, the inclusion of these trials in evidence synthesis may lead to bias in estimation. We call such trials trial-level outliers. To the best of our knowledge, while heterogeneity and inconsistency in NMA have been extensively discussed and well addressed, few previous papers have considered the proper detection and handling of trial-level outliers. In this paper, we propose several Bayesian outlier detection measures, which are then applied to a diabetes data set. Simulation studies comparing our approaches in both arm-based and contrast-based model settings are provided in two supporting appendices. Copyright © 2015 John Wiley & Sons, Ltd.

Research paper thumbnail of Exploration of the use of Bayesian modeling of gradients for censored spatiotemporal data from the Deepwater Horizon oil spill

Spatial Statistics, 2014

This paper develops a hierarchical framework for identifying spatiotemporal patterns in data with... more This paper develops a hierarchical framework for identifying spatiotemporal patterns in data with a high degree of censoring using the gradient process. To do this, we impute censored values using a sampling-based inverse CDF method within our Markov chain Monte Carlo algorithm, thereby avoiding burdensome integration and facilitating efficient estimation of other model parameters. We illustrate use of our methodology using a simulated data example, and uncover the danger of simply substituting a space- and time-constant function of the level of detection for all missing values. We then fit our model to area measurement data of volatile organic compounds (VOC) air concentrations collected on vessels supporting the response and clean-up efforts of the Deepwater Horizon oil release that occurred starting April 20, 2010. These data contained a high percentage of observations below the detectable limits of the measuring instrument. Despite this, we were still able to make some interesting discoveries, including elevated levels of VOC near the site of the oil well on June 26th. Using the results from this preliminary analysis, we hope to inform future research on the Deepwater Horizon study, including the use of gradient methods for assigning workers to exposure categories.

Research paper thumbnail of Generalized Linear Models for Small-Area Estimation

Journal of The American Statistical Association, 1998

Bayesian methods have been used quite extensively in recent years for solving small-area estimati... more Bayesian methods have been used quite extensively in recent years for solving small-area estimation problems. Particularly effective in this regard has been the hierarchical or empirical Bayes approach, which is especially suitable for a systematic connection of local areas through models. However, the development to date has mainly concentrated on continuous-valued variates. Often the survey data are discrete or categorical,

Research paper thumbnail of Commensurate priors for incorporating historical information in clinical trials using general and generalized linear models

Assessing between-study variability in the context of conventional random-effects meta-analysis i... more Assessing between-study variability in the context of conventional random-effects meta-analysis is notoriously difficult when incorporating data from only a small number of historical studies. In order to borrow strength, historical and current data are often assumed to be fully homogeneous, but this can have drastic consequences for power and Type I error if the historical information is biased. In this paper, we propose empirical and fully Bayesian modifications of the commensurate prior model ) extending Pocock (1976, and evaluate their frequentist and Bayesian properties for incorporating patient-level historical data using general and generalized linear mixed regression models. Our proposed commensurate prior models lead to preposterior admissible estimators that facilitate alternative bias-variance trade-offs than those offered by pre-existing methodologies for incorporating historical data from a small number of historical studies. We also provide a sample analysis of a colon cancer trial comparing timeto-disease progression using a Weibull regression model.

Research paper thumbnail of Welcome to PubAlerts, a free service of the UNC Institute on Aging Information Center. Our goal is

Research paper thumbnail of Web Appendices: Hierarchical and Joint Site-Edge Methods for Medicare Hospice Service Region Boundary Analysis

Research paper thumbnail of Probability matching priors for linear calibration

Test, 1995

Summary In the linear calibration problem, a model is fit to paired observations arising from two... more Summary In the linear calibration problem, a model is fit to paired observations arising from two measurement techniques, one known to be far more accurate (but also more expensive) than the other. The fitted model is then used with univariate observations from the less ...

Research paper thumbnail of Effect of Dissemination of Evidence in Reducing Injuries from Falls

New England Journal of Medicine, 2008

Falling is a common and morbid condition among elderly persons. Effective strategies to prevent f... more Falling is a common and morbid condition among elderly persons. Effective strategies to prevent falls have been identified but are underutilized.

Research paper thumbnail of Bayesian measures of model complexity and fit

Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2002

We consider the problem of comparing complex hierarchical models in which the number of parameter... more We consider the problem of comparing complex hierarchical models in which the number of parameters is not clearly defined. Using an information theoretic argument we derive a measure p D for the effective number of parameters in a model as the difference between the posterior mean of the deviance and the deviance at the posterior means of the parameters of interest. In general p D approximately corresponds to the trace of the product of Fisher's information and the posterior covariance, which in normal models is the trace of the 'hat' matrix projecting observations onto fitted values. Its properties in exponential families are explored. The posterior mean deviance is suggested as a Bayesian measure of fit or adequacy, and the contributions of individual observations to the fit and complexity can give rise to a diagnostic plot of deviance residuals against leverages. Adding p D to the posterior mean deviance gives a deviance information criterion for comparing models, which is related to other information criteria and has an approximate decision theoretic justification. The procedure is illustrated in some examples, and comparisons are drawn with alternative Bayesian and classical proposals. Throughout it is emphasized that the quantities required are trivial to compute in a Markov chain Monte Carlo analysis.

Research paper thumbnail of Generalized Linear Models for Small-Area Estimation

Journal of the American Statistical Association, 1998

Bayesian methods have been used quite extensively in recent years for solving small-area estimati... more Bayesian methods have been used quite extensively in recent years for solving small-area estimation problems. Particularly effective in this regard has been the hierarchical or empirical Bayes approach, which is especially suitable for a systematic connection of local ...

Research paper thumbnail of Flexible Cure Rate Modeling Under Latent Activation Schemes

Journal of the American Statistical Association, 2007

With rapid improvements in medical treatment and health care, many datasets dealing with time to ... more With rapid improvements in medical treatment and health care, many datasets dealing with time to relapse or death now reveal a substantial portion of patients who are cured (i.e., who never experience the event). Extended survival models called cure rate models account for the probability of a subject being cured and can be broadly classified into the classical mixture models of Berkson and Gage (BG type) or the stochastic tumor models pioneered by Yakovlev and extended to a hierarchical framework by Chen, Ibrahim, and Sinha (YCIS type). Recent developments in Bayesian hierarchical cure models have evoked significant interest regarding relationships and preferences between these two classes of models. Our present work proposes a unifying class of cure rate models that facilitates flexible hierarchical model-building while including both existing cure model classes as special cases. This unifying class enables robust modeling by accounting for uncertainty in underlying mechanisms leading to cure. Issues such as regressing on the cure fraction and propriety of the associated posterior distributions under different modeling assumptions are also discussed. Finally, we offer a simulation study and also illustrate with two datasets (on melanoma and breast cancer) that reveal our framework's ability to distinguish among underlying mechanisms that lead to relapse and cure.

Research paper thumbnail of Hierarchical Spatio-Temporal Mapping of Disease Rates

Journal of the American Statistical Association, 1997

Maps of regional morbidity and mortality rates are useful tools in determining spatial patterns o... more Maps of regional morbidity and mortality rates are useful tools in determining spatial patterns of disease. Combined with socio-demographic census information, they also permit assessment of environmental justice, i.e., whether certain subgroups su er disproportionately from certain diseases or other adverse e ects of harmful environmental exposures. Bayes and empirical Bayes methods have proven useful in smoothing crude maps of disease risk, eliminating the instability of estimates in low-population areas while maintaining geographic resolution. In this paper we extend existing hierarchical spatial models to account for temporal e ects and spatio-temporal interactions. Fitting the resulting highly-parametrized models requires careful implementation of Markov chain Monte Carlo (MCMC) methods, as well as novel techniques for model evaluation and selection. We illustrate our approach using a dataset of county-speci c lung cancer rates in the state of Ohio during the period 1968{1988.

Research paper thumbnail of Access to Home-Based Hospice Care for Rural Populations: Identification of Areas Lacking Service

Journal of Palliative Medicine, 2006

Many persons dying of cancer enroll in home-based hospice prior to death. It is established in th... more Many persons dying of cancer enroll in home-based hospice prior to death. It is established in the literature that persons in rural settings are less likely to use hospice than persons living in urban areas. We examine whether this is due, in part, to a lack of hospice providers serving rural areas. The 100% Medicare enrollment and hospice files for 2000-2002 were the basis for this study. We used a Bayesian smoothing technique to estimate the ZIP-code-level service area for each Medicare-certified hospice in the United States. These service areas were combined to identify ZIP codes not served by any hospice. Overall, approximately 332,000 elders (7.5% of ZIP codes) reside in areas not served by home-based hospice. Each year over 15,000 deaths occur in these unserved areas. There was a strong association between lack of service and urban/rural gradient. One hundred percent of the ZIP codes in the most urban areas (>1,000,000 people) are served by hospice and only 2.8% of the ZIP codes in urban areas of less than 1,000,000 are unserved. In rural areas adjacent to urban areas, over 9% of ZIP codes are unserved and in rural areas not adjacent to an urban area almost 24% of ZIP codes are not served by hospice. While the majority of the elderly population of the US resides in areas currently served by Medicare-certified hospice, there is a geographically large area that lacks home-based hospice services. Current payment policies may need to be adjusted to facilitate hospice availability to these rural populations.

Research paper thumbnail of Combining nonexchangeable functional or survival data sources in oncology using generalized mixture commensurate priors

The Annals of Applied Statistics, 2015

Conventional approaches to statistical inference preclude structures that facilitate incorporatio... more Conventional approaches to statistical inference preclude structures that facilitate incorporation of supplemental information acquired from similar circumstances. For example, the analysis of data obtained using perfusion computed tomography to characterize functional imaging biomarkers in cancerous regions of the liver can benefit from partially informative data collected concurrently in non-cancerous regions. This paper presents a hierarchical model structure that leverages all available information about a curve, using penalized splines, while accommodating important between-source features. Our proposed methods flexibly borrow strength from the supplemental data to a degree that reflects the commensurability of the supplemental curve with the primary curve. We investigate our method's properties for nonparametric regression via simulation, and apply it to a set of liver cancer data. We also apply our method for a semiparametric hazard model to data from a clinical trial that compares time to disease progression for three colorectal cancer treatments, while supplementing inference with information from a previous trial that tested the current standard of care.

Research paper thumbnail of A Bayesian missing data framework for generalized multiple outcome mixed treatment comparisons

Research synthesis methods, Jan 4, 2015

Bayesian statistical approaches to mixed treatment comparisons (MTCs) are becoming more popular b... more Bayesian statistical approaches to mixed treatment comparisons (MTCs) are becoming more popular because of their flexibility and interpretability. Many randomized clinical trials report multiple outcomes with possible inherent correlations. Moreover, MTC data are typically sparse (although richer than standard meta-analysis, comparing only two treatments), and researchers often choose study arms based upon which treatments emerge as superior in previous trials. In this paper, we summarize existing hierarchical Bayesian methods for MTCs with a single outcome and introduce novel Bayesian approaches for multiple outcomes simultaneously, rather than in separate MTC analyses. We do this by incorporating partially observed data and its correlation structure between outcomes through contrast-based and arm-based parameterizations that consider any unobserved treatment arms as missing data to be imputed. We also extend the model to apply to all types of generalized linear model outcomes, suc...

Research paper thumbnail of Rejoinder to the discussion of "a Bayesian missing data framework for generalized multiple outcome mixed treatment comparisons," by S. Dias and A. E. Ades

Research synthesis methods, Jan 13, 2015

Research paper thumbnail of Flexible Cure Rate Modelling Under Latent Activation Schemes

With rapid improvements in medical treatment and health care, many data sets dealing with time to... more With rapid improvements in medical treatment and health care, many data sets dealing with time to relapse or death now reveal a substantial portion of patients who are cured (that is, who never experience the event). Extended survival models called cure rate models account for the probability of a subject being cured and can be broadly classified into the classical mixture models of Berkson and Gage (1952; "BG type") or the stochastic tumor models pioneered by Yakovlev (1996) and extended to a hierarchical framework by Chen, Ibrahim and Sinha (1999; "YCIS type"). Recent developments in Bayesian hierarchical cure models have evoked significant interest regarding relationships and preferences between these two classes of models. Our present work proposes a unifying class of cure rate models that facilitates flexible hierarchical model-building while including both existing cure model classes as special cases. This unifying class enables robust modelling by accounting for uncertainty in underlying mechanisms leading to cure. Issues such as regressing on the cure fraction and propriety of the associated posterior distributions under different modelling assumptions are also discussed. Finally, we offer a simulation study and also illustrate with two data sets (one on melanoma and the other on breast cancer) that reveal our framework's ability to distinguish among underlying mechanisms that lead to relapse and cure.

Research paper thumbnail of Bayesian hierarchical models for network meta-analysis incorporating nonignorable missingness

Statistical methods in medical research, Jan 28, 2015

Network meta-analysis expands the scope of a conventional pairwise meta-analysis to simultaneousl... more Network meta-analysis expands the scope of a conventional pairwise meta-analysis to simultaneously compare multiple treatments, synthesizing both direct and indirect information and thus strengthening inference. Since most of trials only compare two treatments, a typical data set in a network meta-analysis managed as a trial-by-treatment matrix is extremely sparse, like an incomplete block structure with significant missing data. Zhang et al. proposed an arm-based method accounting for correlations among different treatments within the same trial and assuming that absent arms are missing at random. However, in randomized controlled trials, nonignorable missingness or missingness not at random may occur due to deliberate choices of treatments at the design stage. In addition, those undertaking a network meta-analysis may selectively choose treatments to include in the analysis, which may also lead to missingness not at random. In this paper, we extend our previous work to incorporate...

Research paper thumbnail of Incorporation of individual-patient data in network meta-analysis for multiple continuous endpoints, with application to diabetes treatment

Statistics in medicine, Jan 30, 2015

Availability of individual patient-level data (IPD) broadens the scope of network meta-analysis (... more Availability of individual patient-level data (IPD) broadens the scope of network meta-analysis (NMA) and enables us to incorporate patient-level information. Although IPD is a potential gold mine in biomedical areas, methodological development has been slow owing to limited access to such data. In this paper, we propose a Bayesian IPD NMA modeling framework for multiple continuous outcomes under both contrast-based and arm-based parameterizations. We incorporate individual covariate-by-treatment interactions to facilitate personalized decision making. Furthermore, we can find subpopulations performing well with a certain drug in terms of predictive outcomes. We also impute missing individual covariates via an MCMC algorithm. We illustrate this approach using diabetes data that include continuous bivariate efficacy outcomes and three baseline covariates and show its practical implications. Finally, we close with a discussion of our results, a review of computational challenges, and ...

Research paper thumbnail of Bayesian modeling and analysis for gradients in spatiotemporal processes

Biometrics, Jan 20, 2015

Stochastic process models are widely employed for analyzing spatiotemporal datasets in various sc... more Stochastic process models are widely employed for analyzing spatiotemporal datasets in various scientific disciplines including, but not limited to, environmental monitoring, ecological systems, forestry, hydrology, meteorology, and public health. After inferring on a spatiotemporal process for a given dataset, inferential interest may turn to estimating rates of change, or gradients, over space and time. This manuscript develops fully model-based inference on spatiotemporal gradients under continuous space, continuous time settings. Our contribution is to offer, within a flexible spatiotemporal process model setting, a framework to estimate arbitrary directional gradients over space at any given timepoint, temporal derivatives at any given spatial location and, finally, mixed spatiotemporal gradients that reflect rapid change in spatial gradients over time and vice-versa. We achieve such inference without compromising on rich and flexible spatiotemporal process models and use nonse...

Research paper thumbnail of Detecting outlying trials in network meta-analysis

Statistics in medicine, Jan 8, 2015

Network meta-analysis (NMA) expands the scope of a conventional pairwise meta-analysis to simulta... more Network meta-analysis (NMA) expands the scope of a conventional pairwise meta-analysis to simultaneously handle multiple treatment comparisons. However, some trials may appear to deviate markedly from the others and thus be inappropriate to be synthesized in the NMA. In addition, the inclusion of these trials in evidence synthesis may lead to bias in estimation. We call such trials trial-level outliers. To the best of our knowledge, while heterogeneity and inconsistency in NMA have been extensively discussed and well addressed, few previous papers have considered the proper detection and handling of trial-level outliers. In this paper, we propose several Bayesian outlier detection measures, which are then applied to a diabetes data set. Simulation studies comparing our approaches in both arm-based and contrast-based model settings are provided in two supporting appendices. Copyright © 2015 John Wiley & Sons, Ltd.

Research paper thumbnail of Exploration of the use of Bayesian modeling of gradients for censored spatiotemporal data from the Deepwater Horizon oil spill

Spatial Statistics, 2014

This paper develops a hierarchical framework for identifying spatiotemporal patterns in data with... more This paper develops a hierarchical framework for identifying spatiotemporal patterns in data with a high degree of censoring using the gradient process. To do this, we impute censored values using a sampling-based inverse CDF method within our Markov chain Monte Carlo algorithm, thereby avoiding burdensome integration and facilitating efficient estimation of other model parameters. We illustrate use of our methodology using a simulated data example, and uncover the danger of simply substituting a space- and time-constant function of the level of detection for all missing values. We then fit our model to area measurement data of volatile organic compounds (VOC) air concentrations collected on vessels supporting the response and clean-up efforts of the Deepwater Horizon oil release that occurred starting April 20, 2010. These data contained a high percentage of observations below the detectable limits of the measuring instrument. Despite this, we were still able to make some interesting discoveries, including elevated levels of VOC near the site of the oil well on June 26th. Using the results from this preliminary analysis, we hope to inform future research on the Deepwater Horizon study, including the use of gradient methods for assigning workers to exposure categories.

Research paper thumbnail of Generalized Linear Models for Small-Area Estimation

Journal of The American Statistical Association, 1998

Bayesian methods have been used quite extensively in recent years for solving small-area estimati... more Bayesian methods have been used quite extensively in recent years for solving small-area estimation problems. Particularly effective in this regard has been the hierarchical or empirical Bayes approach, which is especially suitable for a systematic connection of local areas through models. However, the development to date has mainly concentrated on continuous-valued variates. Often the survey data are discrete or categorical,

Research paper thumbnail of Commensurate priors for incorporating historical information in clinical trials using general and generalized linear models

Assessing between-study variability in the context of conventional random-effects meta-analysis i... more Assessing between-study variability in the context of conventional random-effects meta-analysis is notoriously difficult when incorporating data from only a small number of historical studies. In order to borrow strength, historical and current data are often assumed to be fully homogeneous, but this can have drastic consequences for power and Type I error if the historical information is biased. In this paper, we propose empirical and fully Bayesian modifications of the commensurate prior model ) extending Pocock (1976, and evaluate their frequentist and Bayesian properties for incorporating patient-level historical data using general and generalized linear mixed regression models. Our proposed commensurate prior models lead to preposterior admissible estimators that facilitate alternative bias-variance trade-offs than those offered by pre-existing methodologies for incorporating historical data from a small number of historical studies. We also provide a sample analysis of a colon cancer trial comparing timeto-disease progression using a Weibull regression model.

Research paper thumbnail of Welcome to PubAlerts, a free service of the UNC Institute on Aging Information Center. Our goal is

Research paper thumbnail of Web Appendices: Hierarchical and Joint Site-Edge Methods for Medicare Hospice Service Region Boundary Analysis

Research paper thumbnail of Probability matching priors for linear calibration

Test, 1995

Summary In the linear calibration problem, a model is fit to paired observations arising from two... more Summary In the linear calibration problem, a model is fit to paired observations arising from two measurement techniques, one known to be far more accurate (but also more expensive) than the other. The fitted model is then used with univariate observations from the less ...

Research paper thumbnail of Effect of Dissemination of Evidence in Reducing Injuries from Falls

New England Journal of Medicine, 2008

Falling is a common and morbid condition among elderly persons. Effective strategies to prevent f... more Falling is a common and morbid condition among elderly persons. Effective strategies to prevent falls have been identified but are underutilized.

Research paper thumbnail of Bayesian measures of model complexity and fit

Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2002

We consider the problem of comparing complex hierarchical models in which the number of parameter... more We consider the problem of comparing complex hierarchical models in which the number of parameters is not clearly defined. Using an information theoretic argument we derive a measure p D for the effective number of parameters in a model as the difference between the posterior mean of the deviance and the deviance at the posterior means of the parameters of interest. In general p D approximately corresponds to the trace of the product of Fisher's information and the posterior covariance, which in normal models is the trace of the 'hat' matrix projecting observations onto fitted values. Its properties in exponential families are explored. The posterior mean deviance is suggested as a Bayesian measure of fit or adequacy, and the contributions of individual observations to the fit and complexity can give rise to a diagnostic plot of deviance residuals against leverages. Adding p D to the posterior mean deviance gives a deviance information criterion for comparing models, which is related to other information criteria and has an approximate decision theoretic justification. The procedure is illustrated in some examples, and comparisons are drawn with alternative Bayesian and classical proposals. Throughout it is emphasized that the quantities required are trivial to compute in a Markov chain Monte Carlo analysis.

Research paper thumbnail of Generalized Linear Models for Small-Area Estimation

Journal of the American Statistical Association, 1998

Bayesian methods have been used quite extensively in recent years for solving small-area estimati... more Bayesian methods have been used quite extensively in recent years for solving small-area estimation problems. Particularly effective in this regard has been the hierarchical or empirical Bayes approach, which is especially suitable for a systematic connection of local ...

Research paper thumbnail of Flexible Cure Rate Modeling Under Latent Activation Schemes

Journal of the American Statistical Association, 2007

With rapid improvements in medical treatment and health care, many datasets dealing with time to ... more With rapid improvements in medical treatment and health care, many datasets dealing with time to relapse or death now reveal a substantial portion of patients who are cured (i.e., who never experience the event). Extended survival models called cure rate models account for the probability of a subject being cured and can be broadly classified into the classical mixture models of Berkson and Gage (BG type) or the stochastic tumor models pioneered by Yakovlev and extended to a hierarchical framework by Chen, Ibrahim, and Sinha (YCIS type). Recent developments in Bayesian hierarchical cure models have evoked significant interest regarding relationships and preferences between these two classes of models. Our present work proposes a unifying class of cure rate models that facilitates flexible hierarchical model-building while including both existing cure model classes as special cases. This unifying class enables robust modeling by accounting for uncertainty in underlying mechanisms leading to cure. Issues such as regressing on the cure fraction and propriety of the associated posterior distributions under different modeling assumptions are also discussed. Finally, we offer a simulation study and also illustrate with two datasets (on melanoma and breast cancer) that reveal our framework's ability to distinguish among underlying mechanisms that lead to relapse and cure.

Research paper thumbnail of Hierarchical Spatio-Temporal Mapping of Disease Rates

Journal of the American Statistical Association, 1997

Maps of regional morbidity and mortality rates are useful tools in determining spatial patterns o... more Maps of regional morbidity and mortality rates are useful tools in determining spatial patterns of disease. Combined with socio-demographic census information, they also permit assessment of environmental justice, i.e., whether certain subgroups su er disproportionately from certain diseases or other adverse e ects of harmful environmental exposures. Bayes and empirical Bayes methods have proven useful in smoothing crude maps of disease risk, eliminating the instability of estimates in low-population areas while maintaining geographic resolution. In this paper we extend existing hierarchical spatial models to account for temporal e ects and spatio-temporal interactions. Fitting the resulting highly-parametrized models requires careful implementation of Markov chain Monte Carlo (MCMC) methods, as well as novel techniques for model evaluation and selection. We illustrate our approach using a dataset of county-speci c lung cancer rates in the state of Ohio during the period 1968{1988.

Research paper thumbnail of Access to Home-Based Hospice Care for Rural Populations: Identification of Areas Lacking Service

Journal of Palliative Medicine, 2006

Many persons dying of cancer enroll in home-based hospice prior to death. It is established in th... more Many persons dying of cancer enroll in home-based hospice prior to death. It is established in the literature that persons in rural settings are less likely to use hospice than persons living in urban areas. We examine whether this is due, in part, to a lack of hospice providers serving rural areas. The 100% Medicare enrollment and hospice files for 2000-2002 were the basis for this study. We used a Bayesian smoothing technique to estimate the ZIP-code-level service area for each Medicare-certified hospice in the United States. These service areas were combined to identify ZIP codes not served by any hospice. Overall, approximately 332,000 elders (7.5% of ZIP codes) reside in areas not served by home-based hospice. Each year over 15,000 deaths occur in these unserved areas. There was a strong association between lack of service and urban/rural gradient. One hundred percent of the ZIP codes in the most urban areas (>1,000,000 people) are served by hospice and only 2.8% of the ZIP codes in urban areas of less than 1,000,000 are unserved. In rural areas adjacent to urban areas, over 9% of ZIP codes are unserved and in rural areas not adjacent to an urban area almost 24% of ZIP codes are not served by hospice. While the majority of the elderly population of the US resides in areas currently served by Medicare-certified hospice, there is a geographically large area that lacks home-based hospice services. Current payment policies may need to be adjusted to facilitate hospice availability to these rural populations.