Joint Modeling Approach Research Articles

Bridging Models of Biometric and Psychometric Assessment: A Three-Way Joint Modeling Approach of Item Responses, Response Times, and Gaze Fixation Counts.

Recently, joint models of item response data and response times have been proposed to better assess and understand test takers' learning processes. This article demonstrates how biometric information such as gaze fixation counts obtained from an eye-tracking machine can be integrated into the measurement model. The proposed joint modeling framework accommodates the relations among a test taker's latent ability, working speed and test engagement level via a person-side variance-covariance structure, while simultaneously permitting the modeling of item difficulty, time-intensity, and the engagement intensity through an item-side variance-covariance structure. A Bayesian estimation scheme is used to fit the proposed model to data. Posterior predictive model checking based on three discrepancy measures corresponding to various model components are introduced to assess model-data fit. Findings from a Monte Carlo simulation and results from analyzing experimental data demonstrate the utility of the model.

Comparison of Joint Modelling and Landmarking Approaches for Dynamic Prediction Using Bootstrap Simulation

Prediction models for clinical outcomes can greatly help clinicians with early diagnosis, cost-effective management and primary prevention of many medical conditions. In conventional prediction models, predictors are typically measured at a fixed time point, either at baseline or at other time point of interest such as biomarker values measured at the most recent follow-up. Dynamic prediction has emerged as a more appealing prediction technique that takes account of longitudinal history of biomarkers for making predictions. We compared prediction performance of two well-known approaches for dynamic prediction, namely joint modelling and landmarking, using bootstrap simulation based on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) data with repeat Mini-Mental State Examination (MMSE) scores as the longitudinal biomarker and time-to-Alzheimer’s disease (AD) as the survival outcome. We assessed the performance of both approaches in terms of extended definitions of discrimination and calibration, namely dynamic area under the receiver operating characteristic curve (dynAUC) and expected prediction error (PE). We focused on real data-based bootstrap simulation in an attempt to be as impartial as possible to both methods as landmarking is a pragmatic approach which does not specify a statistical model for the longitudinal markers, and therefore any comparison based on model based data simulation may potentially be more advantageous to joint modelling approach. The dynAUC and PE were compared at landmarks t_{s}=1.0, 1.5, 2.0,text{ and },2.5 years and within a 2-year window from the landmark time points. The optimism corrected estimates of dynAUC for joint modelling were slightly higher (1.26, 3.22, 2.76 and 0.12% higher at the four landmark time points) than that of landmarking approach. Apart from the final landmark point (at 2.5 years), dynamic prediction based on joint models has also performed slightly better in terms of calibration. The expected prediction errors (PE) for joint models were 0.70, 2.56 and 2.04% lower at the first three landmark time points, respectively, compared to the landmarking approach. In general, joint modelling approach has performed better than the landmarking approach in terms of both discrimination (dynAUC) and calibration (PE), although the margin of gain in performance by using joint models over landmarking was relatively small indicating that landmarking approach was close enough, despite not having a precise statistical model characterising the evolution of the longitudinal markers. Future comparative studies should consider extended versions of joint modelling and landmarking approaches which may overcome some of the limitations of the standard methods.

Effect of Nonlinear Modeling of Beam-Column Joint on Pushover Analysis

The present paper is concerned with the seismic risk assessment of buildings in the Kingdom of Saudi Arabia. A critical review of the existing literature is presented to identify the shortcomings of extant studies. None of the extant studies considered nonlinear action of the beam-column joint (BCJ) but rather they dealt with BCJ as a rigid element for simplicity and the only plastic hinging has been considered in beams and/or columns. Hence the main focus of this paper is to demonstrate the significant effects of the nonlinear action of BCJ in the pushover analysis and in turn the inadequacy of all previous studies which overlooked such effect. In this study, nonlinear static pushover analysis is performed on two-dimensional RC frames of existing buildings in Jeddah city, with and without using macro node elements and pushover curves are compared. The beam-column joint modelling approach adopted in this study is through macro node element which accounts for failure due to shear collapse of the joint, concrete crushing, flexural and/or shear plastic damage of the beams or columns connected and bond-slip failure. The results clearly indicate that the RC frame in which the beam-column joints were modeled using a macro node element, tends to have lesser base shear values and higher displacement capacity when compared to the RC frame modeled without using the macro node. Furthermore, the status of plastic hinges developed in building frames modeled without using macro node element was found to be within the Immediate occupancy (IO) performance level, but this hinge status drastically changed to Collapse prevention (CP) performance level when BCJ was modeled using macro node. Hence, the results highlight that the nonlinear action of beam-column joint has a significant effect on the nonlinear response of a structure.

Modelling associations between animal social structure and demography

Successfully navigating a complex social environment can influence fitness (i.e. survival and reproductive success), which can cascade to outcomes in demography and population dynamics. Extensive research has focused on the mechanisms and drivers of fitness variation that result from differences in social structure and interactions, and separately, the effects of demography on population change. However, there have been few attempts to address the effects of social structure on population dynamics through demography. Here, we first review the effects of social structure and individual social position on fitness. We then address knowledge gaps, including the relationship between social structure and population dynamics, the carryover effects of social conditions, and the differential effects of social variables on specific demographic rates. We also review statistical tools and data requirements for the analysis of social networks and demography. We then propose that knowledge gaps could be filled by using joint modelling approaches. We used a simulation study to highlight the potential use of social networks to inform survival, a key demographic parameter. We developed a model that combines social network and survival (Cormack–Jolly–Seber) analyses and evaluated its performance to place inferences under different group structures and sampling scenarios using simulated data. Our results show that valid inferences on social and survival parameters and their connections can be achieved with realistic sample sizes, but precision is improved with more complete information (i.e. fewer missing individuals). Based on our review and simulation study, we suggest that further development of integrative modelling approaches can yield greater understanding and improved power to make predictions about the effects of social environments on populations and the feedback of population dynamics on social structure. • Social structure influences demography through fitness. • We conducted a simulation combining social network analysis and survival modelling. • Models showed minimal bias and good precision but incomplete data reduced precision. • We suggest further work to overcome limitations associated with data integration.

Automated Bot Detection Using Bayesian Latent Class Models in Online Surveys.

Behavioral scientists have become increasingly reliant on online survey platforms such as Amazon's Mechanical Turk (Mturk). These platforms have many advantages, for example it provides ease of access to difficult to sample populations, a large pool of participants, and an easy to use implementation. A major drawback is the existence of bots that are used to complete online surveys for financial gain. These bots contaminate data and need to be identified in order to draw valid conclusions from data obtained with these platforms. In this article, we will provide a Bayesian latent class joint modeling approach that can be routinely applied to identify bots and simultaneously estimate a model of interest. This method can be used to separate the bots' response patterns from real human responses that were provided in line with the item content. The model has the advantage that it is very flexible and is based on plausible assumptions that are met in most empirical settings. We will provide a simulation study that investigates the performance of the model under several relevant scenarios including sample size, proportion of bots, and model complexity. We will show that ignoring bots will lead to severe parameter bias whereas the Bayesian latent class model results in unbiased estimates and thus controls this source of bias. We will illustrate the model and its capabilities with data from an empirical political ideation survey with known bots. We will discuss the implications of the findings with regard to future data collection via online platforms.

Bayesian Joint Modeling of Multivariate Longitudinal and Survival Data With an Application to Diabetes Study.

Joint models of longitudinal and time-to-event data have received a lot of attention in epidemiological and clinical research under a linear mixed-effects model with the normal assumption for a single longitudinal outcome and Cox proportional hazards model. However, those model-based analyses may not provide robust inference when longitudinal measurements exhibit skewness and/or heavy tails. In addition, the data collected are often featured by multivariate longitudinal outcomes which are significantly correlated, and ignoring their correlation may lead to biased estimation. Under the umbrella of Bayesian inference, this article introduces multivariate joint (MVJ) models with a skewed distribution for multiple longitudinal exposures in an attempt to cope with correlated multiple longitudinal outcomes, adjust departures from normality, and tailor linkage in specifying a time-to-event process. We develop a Bayesian joint modeling approach to MVJ models that couples a multivariate linear mixed-effects (MLME) model with the skew-normal (SN) distribution and a Cox proportional hazards model. Our proposed models and method are evaluated by simulation studies and are applied to a real example from a diabetes study.

Regressive Class Modelling for Predicting Trajectories of COVID-19 Fatalities Using Statistical and Machine Learning Models.

The COVID-19 (SARS-CoV-2 virus) pandemic has led to a substantial loss of human life worldwide by providing an unparalleled challenge to the public health system. The economic, psychological, and social disarray generated by the COVID-19 pandemic is devastating. Public health experts and epidemiologists worldwide are struggling to formulate policies on how to control this pandemic as there is no effective vaccine or treatment available which provide long-term immunity against different variants of COVID-19 and to eradicate this virus completely. As the new cases and fatalities are recorded daily or weekly, the responses are likely to be repeated or longitudinally correlated. Thus, studying the impact of available covariates and new cases on deaths from COVID-19 repeatedly would provide significant insights into this pandemic’s dynamics. For a better understanding of the dynamics of spread, in this paper, we study the impact of various risk factors on the new cases and deaths over time. To do that, we propose a marginal-conditional based joint modelling approach to predict trajectories, which is crucial to the health policy planners for taking necessary measures. The conditional model is a natural choice to study the underlying property of dependence in consecutive new cases and deaths. Using this model, one can examine the relationship between outcomes and predictors, and it is possible to calculate risks of the sequence of events repeatedly. The advantage of repeated measures is that one can see how individual responses change over time. The predictive accuracy of the proposed model is also compared with various machine learning techniques. The machine learning algorithms used in this paper are extended to accommodate repeated responses. The performance of the proposed model is illustrated using COVID-19 data collected from the Texas Health and Human Services.

Joint analysis of multivariate failure time data with latent variables.

We propose a joint modeling approach to investigate the observed and latent risk factors of the multivariate failure times of interest. The proposed model comprises two parts. The first part is a distribution-free confirmatory factor analysis model that characterizes the latent factors by correlated multiple observed variables. The second part is a multivariate additive hazards model that assesses the observed and latent risk factors of the failure times. A hybrid procedure that combines the borrow-strength estimation approach and the asymptotically distribution-free generalized least square method is developed to estimate the model parameters. The asymptotic properties of the proposed estimators are derived. Simulation studies demonstrate that the proposed method performs well for practical settings. An application to a study concerning the risk factors of multiple diabetic complications is provided.

Without Commitment to an Ontology, There Could Be No Causal Inference.

Without Commitment to an Ontology, There Could Be No Causal Inference.

Longitudinal relationship of liver injury with inflammation biomarkers in COVID-19 hospitalized patients using a joint modeling approach

The mechanisms underlying liver disease in patients with COVID-19 are not entirely known. The aim is to investigate, by means of novel statistical techniques, the changes over time in the relationship between inflammation markers and liver damage markers in relation to survival in COVID-19. The study included 221 consecutive patients admitted to the hospital during the first COVID-19 wave in Spain. Generalized additive mixed models were used to investigate the influence of time and inflammation markers on liver damage markers in relation to survival. Joint modeling regression was used to evaluate the temporal correlations between inflammation markers (serum C-reactive protein [CRP], interleukin-6, plasma D-dimer, and blood lymphocyte count) and liver damage markers, after adjusting for age, sex, and therapy. The patients who died showed a significant elevation in serum aspartate transaminase (AST) and alkaline phosphatase levels over time. Conversely, a decrease in serum AST levels was observed in the survivors, who showed a negative correlation between inflammation markers and liver damage markers (CRP with serum AST, alanine transaminase [ALT], and gamma-glutamyl transferase [GGT]; and D-dimer with AST and ALT) after a week of hospitalization. Conversely, most correlations were positive in the patients who died, except lymphocyte count, which was negatively correlated with AST, GGT, and alkaline phosphatase. These correlations were attenuated with age. The patients who died during COVID-19 infection displayed a significant elevation of liver damage markers, which is correlated with inflammation markers over time. These results are consistent with the role of systemic inflammation in liver damage during COVID-19.

Longitudinal Joint Modelling of Binary and Continuous Outcomes: A Comparison of Bridge and Normal Distributions


 Background: Longitudinal joint models consider the variation caused by repeated measurements over time as well as the association among the response variables. In the case of combining binary and continuous response variables using generalized linear mixed models, integrating over a normally distributed random intercept in the binary logistic regression sub-model does not yield to a closed form. In this paper, we assessed the impact of assuming a Bridge distribution for the random intercept in the binary logistic regression submodel and compared the results to that of normal distribution.
 Method: The response variables are combined through correlated random intercepts. The random intercept in the continuous outcome submodel follows a normal distribution. The random intercept in the binary outcome submodel follows a normal or Bridge distribution. The estimations were carried out using a likelihood-based approach in direct and conditional joint modeling approaches. To illustrate the performance of the models, a simulation study was conducted
 Results: Based on the simulation results and regardless of the joint modeling approach, the models with a Bridge distribution for the random intercept of the binary outcome resulted in a slightly more accurate estimations and better performance.
 Conclusion: In addition to the fact that assuming a bridge distribution for the random intercept in binary logistic regression yields to the same interpretation of parameter estimates in marginal and conditional forms, our study revealed that even if the random intercept of binary logistic regression is normally distributed, assuming a Bridge distribution in the model will result in more accurate results. 

Associations between Longitudinal Gestational Weight Gain and Scalar Infant Birth Weight: A Bayesian Joint Modeling Approach.

Despite the importance of maternal gestational weight gain, it is not yet conclusively understood how weight gain during different stages of pregnancy influences health outcomes for either mother or child. We partially attribute this to differences in and the validity of statistical methods for the analysis of longitudinal and scalar outcome data. In this paper, we propose a Bayesian joint regression model that estimates and uses trajectory parameters as predictors of a scalar response. Our model remedies notable issues with traditional linear regression approaches found in the clinical literature. In particular, our methodology accommodates nonprospective designs by correcting for bias in self-reported prestudy measures; truly accommodates sparse longitudinal observations and short-term variation without data aggregation or precomputation; and is more robust to the choice of model changepoints. We demonstrate these advantages through a real-world application to the Alberta Pregnancy Outcomes and Nutrition (APrON) dataset and a comparison to a linear regression approach from the clinical literature. Our methods extend naturally to other maternal and infant outcomes as well as to areas of research that employ similarly structured data.

Jointly modeling marine species to inform the effects of environmental change on an ecological community in the Northwest Atlantic

Single species distribution models (SSDMs) are typically used to understand and predict the distribution and abundance of marine fish by fitting distribution models for each species independently to a combination of abiotic environmental variables. However, species abundances and distributions are influenced by abiotic environmental preferences as well as biotic dependencies such as interspecific competition and predation. When species interact, a joint species distribution model (JSDM) will allow for valid inference of environmental effects. We built a joint species distribution model of marine fish and invertebrates of the Northeast US Continental Shelf, providing evidence on species relationships with the environment as well as the likelihood of species to covary. Predictive performance is similar to SSDMs but the Bayesian joint modeling approach provides two main advantages over single species modeling: (1) the JSDM directly estimates the significance of environmental effects; and (2) predicted species richness accounts for species dependencies. An additional value of JSDMs is that the conditional prediction of species distributions can use not only the environmental associations of species, but also the presence and abundance of other species when forecasting future climatic associations.

145. Comparing Antibiotic Use Across Inpatient Facilities with Different Antibiotic Stewardship Typologies using Machine Learning and Joint Modeling Approach

BackgroundHospital antibiotic stewardship programs (ASP) aim to promote the appropriate use of antimicrobials (including antibiotics) and play a critical role in controlling antibiotic costs and antibiotic-resistant bacterial infection risk, and improving patient outcomes. However, unlike other health care quality improvement intervention programs, the ASP implementation strategies vary among healthcare facilities, and little is known about whether different types of ASP implementation will lead to the shifting of antibiotic drug use from one class to another. MethodsWe proposed an analytical framework using unsupervised machine learning and joint model approach to 1) develop a typology of ASP strategies in facilities from the Veterans Health Administration, America’s largest integrated health care system; and 2) simultaneously evaluate the impacts of different ASP types on the annual antibiotic use rates across multiple drug classes. The unsupervised machine learning method was used to leverage the structural components in the surveys conducted by the Veteran Affair (VA) Healthcare Analysis and Information group and the Consolidated Framework for Implementation Research experts from Boston University, and reveal the underlying ASP patterns in the VA facilities in 2016. ResultsWe identified 4 groups in the VA facilities in terms of enthusiasm and implementation level of antibiotic control in our ASP typology. We found the facilities with high implementation level and high enthusiasm in ASP and those with high implementation level but low enthusiasm had statistically significant 30% (p-value=0.002) and 22% (p-value=0.031) lower antibiotic use rates in broad-spectrum agents used for community infections, respectively than those with low implementation level and low enthusiasm. However, the facilities with high implementation and high enthusiasm also marginally increased antibiotic use rates in beta-lactam antibiotics (p-value=0.096). ConclusionThe developed analytical framework in the study provided an approach to the granular assessment of the impact of the healthcare intervention programs and might be informative for future health service policy development.Disclosures Matthew B. Goetz, MD, Nothing to disclose

Evaluating the Mode of Presentation to Hospital and Time to Death/Discharge in Patients with COVID-19 in Southwest Iran: A Joint Modelling Approach

Evaluating the Mode of Presentation to Hospital and Time to Death/Discharge in Patients with COVID-19 in Southwest Iran: A Joint Modelling Approach

Shared parameter and copula models for analysis of semicontinuous longitudinal data with nonrandom dropout and informative censoring.

Analysis of longitudinal semicontinuous data characterized by subjects' attrition triggered by nonrandom dropout is complex and requires accounting for the within-subject correlation, and modeling of the dropout process. While methods that address the within-subject correlation and missing data are available, approaches that incorporate the nonrandom dropout, also referred to informative right censoring, in the modeling step are scarce due to the computational intensity and possible intractable integration needed for its implementation. Appreciating the complexity of this problem and the need for a new methodology that is feasible for implementation, we propose to extend a framework of likelihood-based marginalized two-part models to account for informative right censoring. The censoring process is modeled using two approaches: (1) Poisson censoring for the count of visits before dropout and (2) survival time to dropout. Novel consideration was given to the proposed joint modeling approaches for the semicontinuous and censoring components of the likelihood function which included (1) shared parameter, and (2) Clayton copula. The cross-part and within-part correlations were accounted for through a complex random effect structure that models correlated random intercepts and slopes. Feasibility of implementation, and accuracy of these approaches were investigated using extensive simulation studies and clinical application.

Weak correlations in health services research: Weak relationships or common error?

To examine whether the correlation between a provider's effect on one population of patients and the same provider's effect on another population is underestimated if the effects for each population are estimated separately as opposed to being jointly modeled as random effects, and to characterize how the impact of the estimation procedure varies with sample size. Medicare claims and enrollment data on emergency department (ED) visits, including patient characteristics, the patient's hospitalization status, and identification of the doctor responsible for the decision to hospitalize the patient. We used a three-pronged investigation consisting of analytical derivation, simulation experiments, and analysis of administrative data to demonstrate the fallibility of stratified estimation. Under each investigation method, results are compared between the joint modeling approach to those based on stratified analyses. We used data on ED visits from administrative claims from traditional (fee-for-service) Medicare from January 2012 through September 2015. The simulation analysis demonstrates that the joint modeling approach is generally close to unbiased, whereas the stratified approach can be severely biased in small samples, a consequence of joint modeling benefitting from bivariate shrinkage and the stratified approach being compromised by measurement error. In the administrative data analyses, the estimated correlation of doctor admission tendencies between female and male patients was estimated to be 0.98 under the joint model but only 0.38 using stratified estimation. The analogous correlations for White and non-White patients are 0.99 and 0.28 and for Medicaid dual-eligible and non-dual-eligible patients are 0.99 and 0.31, respectively. These results are consistent with the analytical derivations. Joint modeling targets the parameter of primary interest. In the case of population correlations, it yields estimates that are substantially less biased and higher in magnitude than naive estimators that post-process the estimates obtained from stratified models.

A New Econometric Approach for Modeling Several Count Variables: A Case Study of Crash Frequency Analysis by Crash Type and Severity

There is limited adoption of research modeling crash severity frequency considering different crash types due to the challenge associated with analyzing large number of dependent variables. The proposed research contributes to burgeoning econometric and safety literature by developing a joint modeling approach that can accommodate for several dependent variables within a parsimonious structure. By recasting the analysis levels for dependent variables, the proposed approach allows for flexible consideration of crashes by type and severity within a single framework. Specifically, we employ a Panel Mixed Negative Binomial- Generalized Ordered Probit Fractional Spilt (PMNB-GOPFS) model where the first component (NB) accommodates for crash frequency by crash type and the later component (GOPFS) studies the fraction of severity outcome for different crash types. The proposed model system increases interaction between dependent variables through observed variables thus reducing the dependency on unobserved interactions across dependent variables. Thus, the proposed approach allows for the estimation of parsimonious specifications reducing the need for computationally intensive simulation based estimation. The proposed system is also flexible to accommodate for common unobserved effects including: 1) common unobserved factors simultaneously affecting crash counts of different crash types; 2) common unobserved factors simultaneously affecting crash severity proportions of different crash types; and 3) common unobserved factors that simultaneously impact crash counts and severity proportions by different crash types. The model system performance is illustrated using a simulation study. The empirical analysis was conducted using zonal level crash count data for the year 2016 from Central Florida while considering a comprehensive set of exogenous variables including roadway, built environment, land-use, traffic and sociodemographic characteristics. To illustrate the applicability of our proposed system, we carried out a comparison exercise between our proposed joint PMNB-GOPFS and the traditional multivariate system for predicting crash counts across different crash severities. The resulting goodness of fit measures clearly highlight the superior/equivalent performance of the proposed PMNB-GOPFS model over the traditional RPMNB model with less than half the number of parameters. The proposed framework can predict several dimensions including total crash counts, total crash counts by crash types, crash counts for each severity level and finally, proportions and counts of crashes for each crash type by severity.

Biomarkers and Bioactivity in Drug Discovery using a Joint Modelling Approach

Biomarkers that are validated and robust are required for the enhancement of diagnosis, the observation of drug-related activity, therapeutic reactions, and as the blueprint for developing safer and more direct therapeutic efforts for a variety of chronic ailments. Various kinds of biomarkers have proven impactful when it comes to the discovery and development of drugs, but the procedure that involves identifying and verifying ailment-specific biomarkers has proven to be hassling. In recent times, there have been some advancements in multiple omics (also known as multi-omics) methods like transcriptomic, cytometry, genomics, proteomics, metabolomics and imaging. These advancements have made it possible for the discovery and development of distinct biomarkers for complicated chronic ailments to be accelerated expeditiously. In spite of the fact that numerous drawbacks still need to be looked into, ongoing efforts for the discovery and improvement of illness-associated biomarkers will go a long way in optimizing decision-making across the entire process of drug development and expand our comprehension of the infection processes. In addition, when the preclinical biomarkers are effectively translated into the clinic, the way will pave well to an equally effective implementation of personalized therapies throughout complicated illness environments to become beneficial to patients, healthcare service providers and the industry of bio-pharma.

What structures diurnal visitation rates to flowering trees in an Afrotropical lowland rainforest understory?

Abstract As environmental change and degradation accelerate, perturbing insect populations, we need to better understand the resource use dynamics of diverse wild pollinators. Most tropical trees are adapted for biotic pollination, yet we still know little about plant–pollinator interactions in African rainforests. We addressed this gap from a community perspective, identifying what floral traits – colour, scent, reward accessibility and visibility – structure visitation patterns among insect functional groups to tree species flowering in the understory of Korup National Park, Cameroon. To understand how visitor groups share resources, we used joint modelling that explicitly considers zero‐inflation in visitation rates and correlation among visitor groups. Most tree species had exposed floral rewards, and all were visited by multiple insect groups among which ants, bees, beetles, and flies were most abundant. Visitation rates varied more among tree species than among individual trees. Floral scent differences were important for structuring visits, particularly for flies, bees, and ants. Ant and bee visitation rates decreased while fly visits increased marginally throughout the season, correlated with the dry to wet season transition. Comparison with other lowland tropical understories suggests flies may be uniquely diverse and important to this system, and differences in seasonality and forest structure may be drivers of community differences. Floral scent is likely a key functional trait structuring flower–insect interactions in tropical forest environments and should be emphasised in future studies. Lastly, a joint modelling approach can elucidate community structure, particularly in communities with ecologically generalised interactions.