Latent Trait Model Research Articles

Using the Mini-Mental State Examination to Screen for Delirium in Elderly Patients with Hip Fracture

Background: The main aim of this paper was to examine the usefulness of the Mini-Mental State Examination (MMSE) for screening delirium in elderly patients with hip fracture. Methods: The sample included 364 elderly patients with hip fracture admitted to two hospitals in Oslo. Delirium was assessed by the Confusion Assessment Method (CAM) as an approximation of the gold standard, DSM-IV. To evaluate the psychometric properties of the MMSE scale, we used the Mokken nonparametric latent trait model for unidimensional scaling. Results: In total, 76 (21%) patients were diagnosed with delirium based on the CAM, and 141 (43%) had preexisting cognitive impairment. As a screening tool, the recommended MMSE cut-point of 24 showed an acceptable sensitivity of 88% with a specificity of 54%. The prediction of delirium, based on logistic regression on the MMSE total score and on the 5 items selected by the stepwise logistic regression procedure, gave clearly less acceptable results. Conclusions: Our study indicates that the MMSE may be useful in screening for delirium, but the high percentage of false positives shows that it does not have diagnostic potential in patients with hip fracture.

A single latent trait model for multiple binary outcomes in a cluster randomized clinical trial

We formulate a latent variable model with Bayesian estimation to assess a cluster-level intervention effect on multiple binary outcomes from three-level hierarchical data. This approach incorporates the correlation structure into one latent trait, and simultaneously regresses the latent trait on observed covariates. Random effects are included to model the hierarchical structure. We illustrate use of this single-latent trait model in a 32-site cluster randomized clinical trial of a three-arm intervention to improve the quality of pneumonia care. Simulation studies verify the accuracy of the estimation. This latent variable model provides a comprehensive way to analyze multivariate hierarchical data, by estimating intervention effects with respect to multiple binary outcomes, quantifying relationships among outcomes, identifying those outcomes that are most informative regarding the assumed latent trait, and providing a summary measure of the latent trait (e.g., “quality of care”) at each site.

A latent trait model for response times on tests employing the proportional hazards model

For computer-administered tests, response times can be recorded conjointly with the corresponding responses. This broadens the scope of potential modelling approaches because response times can be analysed in addition to analysing the responses themselves. For this purpose, we present a new latent trait model for response times on tests. This model is based on the Cox proportional hazards model. According to this model, latent variables alter a baseline hazard function. Two different approaches to item parameter estimation are described: the first approach uses a variant of the Cox model for discrete time, whereas the second approach is based on a profile likelihood function. Properties of each estimator will be compared in a simulation study. Compared to the estimator for discrete time, the profile likelihood estimator is more efficient, that is, has smaller variance. Additionally, we show how the fit of the model can be evaluated and how the latent traits can be estimated. Finally, the applicability of the model to an empirical data set is demonstrated.

A flexible latent trait model for response times in tests

Latent trait models for response times in tests have become popular recently. One challenge for response time modeling is the fact that the distribution of response times can differ considerably even in similar tests. In order to reduce the need for tailor-made models, a model is proposed that unifies two popular approaches to response time modeling: Proportional hazard models and the accelerated failure time model with log–normally distributed response times. This is accomplished by resorting to discrete time. The categorization of response time allows the formulation of a response time model within the framework of generalized linear models by using a flexible link function. Item parameters of the proposed model can be estimated with marginal maximum likelihood estimation. Applicability of the proposed approach is demonstrated with a simulation study and an empirical application. Additionally, means for the evaluation of model fit are suggested.

Is the continuity of externalizing psychopathology the same in adolescents and middle-aged adults? A test of the externalizing spectrum's developmental coherence.

Externalizing psychopathology (EXT) is a framework for understanding diagnostic comorbidity and etiology of antisocial and substance-use behaviors. EXT indicates continuity in adulthood but the structure of adolescent EXT is less clear. This report examines whether adolescent EXT is trait-like, as has been found with adults, or categorical. We use tests of measurement invariance to determine how diagnostic indicators of EXT differ in adolescents compared to adults. The EXT measures employed were DSM-IIIR diagnoses of adult antisocial behavior, conduct disorder, and alcohol, marijuana, and drug dependence. Latent trait, latent class, and hybrid models were fit to two separate data sets: 2,769 seventeen-year-old adolescents and 2,619 adults from the Minnesota Twin Family Study. The best model in both samples was a single-trait LT model. Parameters from the adolescent and adult models were equivalent for all disorders except alcohol dependence. It appears that EXT in adolescence can be accurately represented by a single-trait model, and the measurement properties of EXT are similar during these time periods with the exception of alcohol dependence.

Latent state trait modeling of children's cortisol at two points of the diurnal cycle

One challenge in examining stable individual differences in basal activity of the HPA axis is controlling for internally or externally based situational factors that lead to day-to-day variation in ambulatory cortisol. Disturbed basal activity is of particular interest in studies with children, for whom a dysregulated HPA axis may play an etiologic role in emotional or health outcomes. The purpose of this study was to determine whether trait vs. situationally specific sources of variation can be identified at different points of the diurnal cycle in children and if so, whether state and trait components vary according to time of measurement. Early morning and late evening salivary cortisol was collected from 164 children aged 7 to 11 years. Samples were collected 30 min after wake-up and 30 min before bedtime on 3 weekdays. State, trait, and error components of cortisol levels were assessed using a latent state trait model. Possible influences of sampling day and outlier treatment on parameter estimates were examined. The results showed that a latent trait factor superimposed on state residuals and measurement error was identified for both early morning and late evening cortisol. Model fit was excellent and criteria for invariance tests were met. Trait factors accounted for 41% and 57% of the variance in morning and evening cortisol, respectively. These findings suggest cortisol attributed to trait factors can be identified and are of substantial magnitude at both the peak and nadir of the diurnal cycle. Latent state trait modeling is a potentially useful tool in understanding the role of stable individual differences in cortisol levels for development and health.

A questionnaire on treatment satisfaction and disease specific knowledge among patients with acute coronary syndrome. I: Are treatment satisfaction and disease specific knowledge continuous latent traits?

ObjectiveTo assess psychometric properties of a questionnaire covering treatment satisfaction and disease-specific knowledge among patients discharged after treatment for acute coronary syndrome. MethodsComparative scaling of the questionnaire using latent class analysis (LCA) and exploratory factor analysis (EFA) in a consecutive sample of 2015 patients. ResultsLCA revealed four qualitatively differing patterns of patients’ knowledge and five distinct patterns of treatment satisfaction. EFA for patients’ knowledge identified four uncorrelated “dimensions”. Patient satisfaction was scored in a two-factor solution despite proven heterogeneity of persons. ConclusionsLCA was helpful to identify classes of patients that cannot be scaled according to a latent trait model. Practice implicationsSpecific patterns of insufficient disease-specific knowledge and satisfaction requiring specific interventions became visible.

The Dimensional Nature of Externalizing Behaviors in Adolescence: Evidence from a Direct Comparison of Categorical, Dimensional, and Hybrid Models

Researchers have recognized the importance of developing an accurate classification system for externalizing disorders, though much of this work has been framed by a priori preferences for categorical vs. dimensional constructs. Newer statistical technologies now allow categorical and dimensional models of psychopathology to be compared empirically. In this study, we directly compared the fit of categorical and dimensional models of externalizing behaviors in a large and representative community sample of adolescents at two time points separated by nearly 2.5years (N = 2027; mean age at Time 1 = 11.09years; 50.8% female). Delinquent and aggressive behaviors were assessed with child and parent Child Behavior Checklist reports. Latent trait, latent class, and factor mixture models were fit to the data, and at both time points, the latent trait model provided the best fit to the data. The item parameters were inspected and interpreted, and it was determined that the items were differentially sensitive across all regions of the dimension. We conclude that classification models can be based on empirical evidence rather than a priori preferences, and while current classification systems conceptualize externalizing problems in terms of discrete groups, they can be better conceptualized as dimensions.

Using Testlet Response Theory to analyze data from a survey of attitude change among breast cancer survivors

In this paper we examine alternative measurement models for fitting data from health surveys. We show why a testlet-based latent trait model that includes covariate information, embedded within a fully Bayesian framework, can allow multiple simultaneous inferences and aid interpretation. We illustrate our approach with a survey of breast cancer survivors that reveals how the attitudes of those patients change after diagnosis toward a focus on appreciating the here-and-now, and away from consideration of longer-term goals. Using the covariate information, we also show the extent to which individual-level variables such as race, age and Tamoxifen treatment are related to a patient's change in attitude.The major contribution of this research is to demonstrate the use of a hierarchical Bayesian IRT model with covariates in this application area; hence a novel case study, and one that is certainly closely aligned with but distinct from the educational testing applications that have made IRT the dominant test scoring model.

Bayesian Semiparametric Structural Equation Models with Latent Variables

Structural equation models (SEMs) with latent variables are widely useful for sparse covariance structure modeling and for inferring relationships among latent variables. Bayesian SEMs are appealing in allowing for the incorporation of prior information and in providing exact posterior distributions of unknowns, including the latent variables. In this article, we propose a broad class of semiparametric Bayesian SEMs, which allow mixed categorical and continuous manifest variables while also allowing the latent variables to have unknown distributions. In order to include typical identifiability restrictions on the latent variable distributions, we rely on centered Dirichlet process (CDP) and CDP mixture (CDPM) models. The CDP will induce a latent class model with an unknown number of classes, while the CDPM will induce a latent trait model with unknown densities for the latent traits. A simple and efficient Markov chain Monte Carlo algorithm is developed for posterior computation, and the methods are illustrated using simulated examples, and several applications.

Some exact tests for manifest properties of latent trait models

Item response theory is one of the modern test theories with applications in educational and psychological testing. Recent developments made it possible to characterize some desired properties in terms of a collection of manifest ones, so that hypothesis tests on these traits can, in principle, be performed. But the existing test methodology is based on asymptotic approximation, which is impractical in most applications since the required sample sizes are often unrealistically huge. To overcome this problem, a class of tests is proposed for making exact statistical inference about four manifest properties: covariances given the sum are non-positive (CSN), manifest monotonicity (MM), conditional association (CA), and vanishing conditional dependence (VCD). One major advantage is that these exact tests do not require large sample sizes. As a result, tests for CSN and MM can be routinely performed in empirical studies. For testing CA and VCD, the exact methods are still impractical in most applications, due to the unusually large number of parameters to be tested. However, exact methods are still derived for them as an exploration toward practicality. Some numerical examples with applications of the exact tests for CSN and MM are provided.

Neuroticism and psychological distress: To what extent is their association due to person-environment correlation?

Objective To examine to what extent the association between neuroticism and psychological distress is related to individuals’ inherent vulnerability or their tendency to self-select high-risk environments or situations. Method Data was drawn from both waves (1984/1985 and 1991) of the Health and Lifestyle Survey. Psychological distress was evaluated using the 30-item General Health Questionnaire (GHQ-30) and neuroticism was assessed with the Eysenck Personality Inventory. A checklist of life events was completed in the second wave only. A Latent State Trait model was estimated to decompose psychological distress into its environmental-contextual and individual-specific components. Results Neuroticism accounted for 31 and 10% respectively of the variance of the environmental-contextual and individual-specific psychological distress components. Conclusion Our results favour the notion that individuals with high-levels of neuroticism tend to self-select situations likely to lead to adversity and distress.

Body image disturbance in 1000 male appearance and performance enhancing drug users

Body image disturbance (BID) among men has only recently become a phenomenon of clinical significance with noted heterogeneity in the behavioral consequences of these disturbances. The degree of heterogeneity among appearance and performance enhancing drug (APED) users is unknown and an empirically derived framework for studying BID is necessary. APED users ( N = 1000) were recruited via the Internet and they completed a comprehensive online assessment APED use patterns, motivations, consequences, and BID. Data were evaluated using latent trait, latent class, and factor mixture models. Model results were validated using a range of covariates including cycle characteristics, age, APED history, and APED risk. A 1-Factor, 4-Class model provided the best fit to the data with Class 1 scoring the highest on all measures of BID and Class 4 the lowest on all measures. Class 2 differed in their preference for being lean over muscular and Class 3 preferred adding mass and size. Each class was associated with unique risks, APED history, and training identity. Not all APED users suffer from significant BID and there are unique profiles for those with elevated BID. Future research on male BID should account for this structure in order to better define relevant diagnostic categories and evaluate the clinical significance of BID.

Dynamic latent trait models: An application to Latin American banking crises

Dynamic latent trait models combine information from a variety of manifest variables, possibly measured on different scales, that are presumed to be indicators of an unobserved latent phenomenon, while allowing appropriate consideration of the longitudinal character of time series. I use a Bayesian dynamic latent trait model of banking sector financial accounts measured at the country/quarter level to build an indicator of banking system robustness in Latin America. As a methodological innovation, I extend dynamic latent trait models to take into account country-specific effects of bank regulatory regimes through hierarchical modeling of factor loadings. I suggest how these models can be applied to other types of phenomena—for example to combine existing political regime indicators to build a more informative measure of democracy.

Evaluación de factores asociados a la comprensión del tratamiento en pacientes que inician la terapia antirretroviral

The aim of this study was to evaluate factors associated with patients' comprehension of antiretroviral therapy (ART). Cross-sectional analysis in which patients at 2 HIV/AIDS public referral centers (Belo Horizonte, Brazil) were interviewed after initiating ART. Information was recorded on variables related to the patient's characteristics, the treatment prescribed, and the healthcare professional involved. A score indicating the patients' level of comprehension regarding the medications prescribed was obtained using a latent trait model estimated by the item response theory. A total of 406 patients were interviewed. Mean (SD) age was 35 (10) years, 227 were men (56%), 302 of Afro-American ethnicity (77%), and 213 had <8 years of education (53%). The regression model determined that 52.25% of the variability of comprehension was explained by the individual's characteristics. Variables associated (P<0.05) with poorest understanding about ART were lower education (<8 years), lack of knowledge about treatment duration and clinical severity, inadequate information provided by physicians, inability to understand pharmaceutical information, daily number of tablets, and the ART regimen prescribed. Comprehension of information about the ART regimen prescribed varies considerably between individuals. Nonetheless, several factors were found to be associated with the level of understanding: characteristics of the patient (education, clinical severity), characteristics of treatment (daily number of tablets, ART regimen prescribed), and contribution of healthcare professionals (information from physicians and pharmacists). Strategies to reinforce information about ART should be a priority for patients with a low level of understanding.

Compreensão da terapia anti-retroviral: uma aplicação de modelo de traço latente

The aim of this study was to develop a score to determine the level of understanding regarding information on antiretroviral therapy (ART) among patients initiating treatment. This was a cross-sectional analysis based on interviews with HIV patients in outpatient public referral centers (Belo Horizonte, Minas Gerais State, Brazil). The score for patients' understanding of their medicines was obtained using a latent trait model, estimated by the Item Response Theory, based on the concordance between each patient answer and the written prescription. Hierarchical linear regression was used to assess patients' global understanding of ART, considering each class of drugs (level 1) and the individual (level 2). Among 406 patients, 37.9% failed to reach a minimum level of understanding of their treatment. The item with the highest level of difficulty was "precaution in use". The item "dosage" showed the most varied understanding of ART. A high proportion of patients displayed minimal understanding of ART, indicating a high potential risk for non-adherence to therapy. It is thus necessary to identify factors associated with insufficient understanding of ART.

Using a factor mixture modeling approach in alcohol dependence in a general population sample

Alcohol dependence (AD) is a complex and heterogeneous disorder. The identification of more homogeneous subgroups of individuals with drinking problems and the refinement of the diagnostic criteria are inter-related research goals. They have the potential to improve our knowledge of etiology and treatment effects, and to assist in the identification of risk factors or specific genetic factors. Mixture modeling has advantages over traditional modeling that focuses on either the dimensional or categorical latent structure. The mixture modeling combines both latent class and latent trait models, but has not been widely applied in substance use research. The goal of the present study is to assess whether the AD criteria in the population could be better characterized by a continuous dimension, a few discrete subgroups, or a combination of the two. More than seven thousand participants were recruited from the population-based Virginia Twin Registry, and were interviewed to obtain DSM-IV (Diagnostic and Statistical Manual of Mental Disorder, version IV) symptoms and diagnosis of AD. We applied factor analysis, latent class analysis, and factor mixture models for symptom items based on the DSM-IV criteria. Our results showed that a mixture model with 1 factor and 3 classes for both genders fit well. The 3 classes were a non-problem drinking group and severe and moderate drinking problem groups. By contrast, models constrained to conform to DSM-IV diagnostic criteria were rejected by model fitting indices providing empirical evidence for heterogeneity in the AD diagnosis. Classification analysis showed different characteristics across subgroups, including alcohol-caused behavioral problems, comorbid disorders, age at onset for alcohol-related milestones, and personality. Clinically, the expanded classification of AD may aid in identifying suitable treatments, interventions and additional sources of comorbidity based on these more homogenous subgroups of alcohol use problems.

Skills Diagnosis Using IRT‐Based Continuous Latent Trait Models

This article summarizes the continuous latent trait IRT approach to skills diagnosis as particularized by a representative variety of continuous latent trait models using item response functions (IRFs). First, several basic IRT‐based continuous latent trait approaches are presented in some detail. Then a brief summary of estimation, model checking, and assessment scoring aspects are discussed. Finally, the University of California at Berkeley multidimensional Rasch‐model‐grounded SEPUP middle school science‐focused embedded assessment project is briefly described as one significant illustrative application.

Bayesian methods for latent trait modelling of longitudinal data

Latent trait models have long been used in the social science literature for studying variables that can only be measured indirectly through multiple items. However, such models are also very useful in accounting for correlation in multivariate and longitudinal data, particularly when outcomes have mixed measurement scales. Bayesian methods implemented with Markov chain Monte Carlo provide a flexible framework for routine fitting of a broad class of latent variable (LV) models, including very general structural equation models. However, in considering LV models, a number of challenging issues arise, including identifiability, confounding between the mean and variance, uncertainty in different aspects of the model, and difficulty in computation. Motivated by the problem of modelling multidimensional longitudinal data, this article reviews the recent literature, provides some recommendations and highlights areas in need of additional research, focusing on methods for model uncertainty.

A Multidimensional Unfolding Latent Trait Model for Binary Data

We introduce a multidimensional latent trait model for binary data with non-monotone item response functions. We assume that the conditional probability of endorsing an item is a normal probability density function, and that the latent traits are normally distributed. The model yields closed form expressions for the moments of the multivariate Bernoulli (MVB) distribution. As a result, cell probabilities can be computed also in closed form, regardless of the dimensionality of the latent traits. The model is an ideal point model in the sense that a respondent - precisely at the ideal point (the mode of the item response function) - endorses the item with probability one.