Finite Mixtures of Latent Trait Analyzers With Concomitant Variables for Bipartite Networks: An Analysis of COVID-19 Data

Networks consist of interconnected units, known as nodes, and allow to formally describe interactions within a system. Specifically, bipartite networks depict relationships between two distinct sets of nodes, designated as sending and receiving nodes. An integral aspect of bipartite network analysis often involves identifying clusters of nodes with similar behaviors. The computational complexity of models for large bipartite networks poses a challenge. To mitigate this challenge, we employ a Mixture of Latent Trait Analyzers (MLTA) for node clustering. Our approach extends the MLTA to include covariates and introduces a double EM algorithm for estimation. Applying our method to COVID-19 data, with sending nodes representing patients and receiving nodes representing preventive measures, enables dimensionality reduction and the identification of meaningful groups. We present simulation results demonstrating the accuracy of the proposed method.

Model-based two-way clustering of second-level units in ordinal multilevel latent Markov models

In this paper, an ordinal multilevel latent Markov model based on separate random effects is proposed. In detail, two distinct second-level discrete effects are considered in the model, one affecting the initial probability vector and the other affecting the transition probability matrix of the first-level ordinal latent Markov process. To model these separate effects, we consider a bi-dimensional mixture specification that allows to avoid unverifiable assumptions on the random effect distribution and to derive a two-way clustering of second-level units. Starting from a general model where the two random effects are dependent, we also obtain the independence model as a special case. The proposal is applied to data on the physical health status of a sample of elderly residents grouped into nursing homes. A simulation study assessing the performance of the proposal is also included.

Multiple imputation and selection of ordinal level 2 predictors in multilevel models: An analysis of the relationship between student ratings and teacher practices and attitudes

The article is motivated by the analysis of the relationship between university student ratings and teacher practices and attitudes, which are measured via a set of binary and ordinal items collected by an innovative survey. The analysis is conducted through a two-level random intercept model, where student ratings are nested within teachers. The analysis must face two issues about the items measuring teacher practices and attitudes, which are level 2 predictors: (a) the items are severely affected by missingness due to teacher non-response and (b) there is redundancy in both the number of items and the number of categories of their measurement scale. We tackle the missing data issue by considering a multiple imputation strategy exploiting information at both student and teacher levels. For the redundancy issue, we rely on regularization techniques for ordinal predictors, also accounting for the multilevel data structure. The proposed solution addresses the problem at hand in an original way, and it can be applied whenever it is required to select level 2 predictors affected by missing values. The results obtained with the final model indicate that ratings on teacher ability to motivate students are related to certain teacher practices and attitudes.

Finite Mixtures of Hidden Markov Models for Longitudinal Responses Subject to Drop out

Drop out is a typical issue in longitudinal studies. When the missingness is non-ignorable, inference based on the observed data only may be biased. This paper is motivated by the Leiden 85+ study, a longitudinal study conducted to analyze the dynamics of cognitive functioning in the elderly. We account for dependence between longitudinal responses from the same subject using time-varying random effects associated with a heterogeneous hidden Markov chain. As several participants in the study drop out prematurely, we introduce a further random effect model to describe the missing data mechanism. The potential dependence between the random effects in the two equations (and, therefore, between the two processes) is introduced through a joint distribution specified via a latent structure approach. The application of the proposal to data from the Leiden 85+ study shows its effectiveness in modeling heterogeneous longitudinal patterns, possibly influenced by the missing data process. Results from a sensitivity analysis show the robustness of the estimates with respect to misspecification of the missing data mechanism. A simulation study provides evidence for the reliability of the inferential conclusions drawn from the analysis of the Leiden 85+ data.

A bidimensional finite mixture model for longitudinal data subject to dropout

In longitudinal studies, subjects may be lost to follow up and, thus, present incomplete response sequences. When the mechanism underlying the dropout is nonignorable, we need to account for dependence between the longitudinal and the dropout process. We propose to model such a dependence through discrete latent effects, which are outcome-specific and account for heterogeneity in the univariate profiles. Dependence between profiles is introduced by using a probability matrix to describe the corresponding joint distribution. In this way, we separately model dependence within each outcome and dependence between outcomes. The major feature of this proposal, when compared with standard finite mixture models, is that it allows the nonignorable dropout model to properly nest its ignorable counterpart. We also discuss the use of an index of (local) sensitivity to nonignorability to investigate the effects that assumptions about the dropout process may have on model parameter estimates. The proposal is illustrated via the analysis of data from a longitudinal study on the dynamics of cognitive functioning in the elderly.

Mixed hidden Markov quantile regression models for longitudinal data with possibly incomplete sequences

Quantile regression provides a detailed and robust picture of the distribution of a response variable, conditional on a set of observed covariates. Recently, it has be been extended to the analysis of longitudinal continuous outcomes using either time-constant or time-varying random parameters. However, in real-life data, we frequently observe both temporal shocks in the overall trend and individual-specific heterogeneity in model parameters. A benchmark dataset on HIV progression gives a clear example. Here, the evolution of the CD4 log counts exhibits both sudden temporal changes in the overall trend and heterogeneity in the effect of the time since seroconversion on the response dynamics. To accommodate such situations, we propose a quantile regression model, where time-varying and time-constant random coefficients are jointly considered. Since observed data may be incomplete due to early drop-out, we also extend the proposed model in a pattern mixture perspective. We assess the performance of the proposals via a large-scale simulation study and the analysis of the CD4 count data.

Relationship between EDNOS and its subtypes and borderline personality disorder

OBJECTIVE

The purpose of this study was to assess the prevalence of eating disorder not otherwise specified (EDNOS) and four well-defined subtypes of this disorder found in a sample of female borderline patients.

METHOD

The lifetime prevalence of EDNOS and its various subtypes among 233 female borderline patients and 46 female Axis II comparison subjects was assessed using the Structured Clinical Interview for DSM-III-R Axis I disorders.

RESULTS

Thirty-three percent of female borderline patients met DSM-III-R criteria for EDNOS at some point in their lives. Of these 76 women, 20% reported a pattern of restricting without low weight, 37% reported a pattern of binging without purging, 37% reported a pattern of purging without binging, and 33% reported a pattern of low weight without loss of menses. However, less than 25% of these 76 borderline women had ever met criteria for anorexia nervosa or bulimia nervosa.

CONCLUSIONS

The results of this study suggest that EDNOS is a separate cluster of eating disorders among borderline women, rather than a prodromal or residual form of a more clear-cut case of anorexia or bulimia nervosa.

Biparental failure in the childhood experiences of borderline patients

The purpose of this study was to assess the role of biparental abuse and neglect in the development of borderline personality disorder (BPD). A semistructured research interview was used to blindly assess the childhood experiences of biparental abuse and neglect reported by 358 borderline inpatients and 109 axis II controls. Eighty-four percent of borderline patients reported having experienced some type of biparental abuse or neglect before the age of 18; 55% reported a childhood history of biparental abuse; 77% reported a childhood history of biparental neglect. These experiences were also reported by a substantial percentage of Axis II controls (biparental abuse or neglect [61%], biparental abuse [31%], and biparental neglect [55%]). However, borderline patients were significantly more likely than axis II controls to report having been verbally, emotionally, and physically but not sexually abused by caretakers of both sexes. They were also significantly more likely than controls to report having caretakers of both sexes deny the validity of their thoughts and feelings, fail to provide them with needed protection, neglect their physical care, withdraw from them emotionally, and treat them inconsistently. It was also found that female borderlines who reported a previous history of neglect by a female caretaker and abuse by a male caretaker were at significantly higher risk for having been sexually abused by a noncaretaker. Taken together, the results of this study suggest that biparental failure may be a significant factor in the etiology of BPD. They also suggest that biparental failure may significantly increase a preborderline girl's risk of being sexually abused by someone other than her parents.

Violence in the lives of adult borderline patients

The purpose of this study was to assess the experiences of adult violence reported by a sample of criteria-defined borderline patients and axis II controls. The experiences of having had a physically abusive partner and/or having been raped reported by 362 personality-disordered inpatients were assessed blind to diagnostic status using a semistructured research interview. Forty-six percent of borderline patients reported having been a victim of violence since the age of 18. Borderline patients (N = 290) were significantly more likely than axis II controls (N = 72) to report having had a physically abusive partner, having been raped, having been raped multiple times, having been raped by a known perpetrator, and having been both physically assaulted by a partner and raped. Female borderline patients were significantly more likely than male borderline patients to have been physically and/or sexually assaulted as adults (50% vs. 26%). However, a significantly higher percentage of borderline patients of both genders reported experiences of adult violence than controls of the same gender. Four risk factors were found to significantly predict whether borderline patients had an adult history of being a victim of violence: female gender, a substance use disorder that began before the age of 18, childhood sexual abuse, and emotional withdrawal by a caretaker. The results of this study suggest that both male and female borderline patients are at substantial risk for being physically and/or sexually victimized as adults.

Reported pathological childhood experiences associated with the development of borderline personality disorder

OBJECTIVE

The purpose of this study was to assess a full range of pathological childhood experiences reported by patients with criteria-defined borderline personality disorder and comparison patients with other personality disorders.

METHOD

The pathological childhood experiences reported by 467 inpatients with personality disorders were assessed by interviewers who used a semistructured research interview and were blind to clinical diagnosis.

RESULTS

Of the 358 patients with borderline personality disorder, 91% reported having been abused, and 92% reported having been neglected, before the age of 18. The borderline patients were significantly more likely than the 109 patients with other personality disorders to report having been emotionally and physically abused by a caretaker and sexually abused by a noncaretaker. They were also significantly more likely to report having a caretaker withdraw from them emotionally, treat them inconsistently, deny their thoughts and feelings, place them in the role of a parent, and fail to provide them with needed protection. The borderline patients with a childhood history of sexual abuse were significantly more likely than those without such a history to report having experienced all but one of the types of abuse and neglect studied. When all significant risk factors were considered together, four were found to be significant predictors of a borderline diagnosis: female gender, sexual abuse by a male noncaretaker, emotional denial by a male caretaker, and inconsistent treatment by a female caretaker.

CONCLUSIONS

The results suggest that sexual abuse is neither necessary nor sufficient for the development of borderline personality disorder and that other childhood experiences, particularly neglect by caretakers of both genders, represent significant risk factors.

Childhood experiences of borderline patients

The childhood histories of 50 outpatients meeting both Diagnostic Interview for Borderlines (DIB) and DSM-III criteria for Borderline Personality Disorder, 29 outpatients meeting DSM-III criteria for Antisocial Personality Disorder, and 26 outpatients meeting DSM-III for Dysthymic Disorder as well as DSM-III criteria for some other type of Axis II disorder were assessed, blind to proband diagnosis, using a semistructured interview. Borderlines were significantly more likely than those in either control group to report a history of abuse, particularly verbal and sexual abuse. They were also significantly more likely than antisocial controls to report a history of neglect, particularly emotional withdrawal, and significantly more likely than dysthymic other personality disorder controls to report a history of early separation experiences. The authors conclude that the development of Borderline Personality Disorder is more strongly associated with (1) exposure to chronically disturbed caretakers than prolonged separations from these same adults and (2) a history of abuse than a history of neglect.

Interacting sex and familial sinistrality characteristics influence both language lateralization and spatial ability in right handers

In 1981 McKeever and Hoff found a pattern of sex-familial sinistrality (FS)-visual field interaction on an Object Naming Latency Task. The FS - females and FS + males performed more symmetrically. Also in 1981, Hécaen, DeAgostini, and Monzon-Montes found a substantially lower incidence of aphasia following left hemisphere lesions in FS - females and FS + males. Both findings suggest a less strict left hemisphere dominance in FS - females and FS + males. We report a replication of the McKeever and Hoff findings and also evidence of a pattern of sex-FS interaction in spatial visualization ability wherein FS - females and FS + males perform better than FS + females and FS - males.