Latent Class Approach Research Articles

Longitudinal Analysis of the Transition Between Multimorbidity and Mortality Patterns from a Syndemic Perspective.

Longitudinal analysis of multimorbidity (MM) from a holistic approach to the aging population is necessary to guide medical care. To estimate the probability of transition between different MM patterns and the "death" stage and to characterize the socioeconomic, health, behavioral, and functional dimensions of MM patterns from a syndemic perspective. A cohort of 7,120 people ≥50 from the Mexican Health and Aging Study and mortality data at 11 and 17 years of follow-up were analyzed using the latent class approach. MM was defined as the coexistence of two or more chronic diseases (CDs). Socioeconomic, health, behavioral, and functional indicators were registered. Descriptive analysis, latent class analysis, and Markov chain models were carried out. The model included three latent classes: "healthy" (low probability of CD); "multi-organic" (high probability of hypertension, arthritis, diabetes mellitus, lung disease, and myocardial infarction); and "sensitive-emotional, mental, neurological" (high probability of chronic pain, depression, and cognitive impairment). At the first follow-up, people in the "multi-organic" class were more likely to move on to death, and at the second follow-up, they were in the "sensitive-emotional, mental, neurological" class. In addition, the characteristic profiles of the indicators for each latent class were identified. The probabilities of transition between the three classes of MM and death, as well as the socioeconomic, behavioral, and functional dimensions in those over 50 years of age, revealed, from a syndemic perspective, the variability in the likelihood of progressing toward an unfavorable outcome.

The use of multiple systems estimation to estimate the number of unattributed paintings by Modigliani

Abstract The number of unattributed paintings by Amedeo Modigliani is estimated, using the method of multiple systems estimation (MSE). Most major artists’ works are listed in one catalogue raisonné, but there are five catalogues purporting to list Modigliani paintings. These can be treated as list sources from which MSE can be applied. We obtain estimates by following the classical MSE approach using log-linear models, and compare these with estimates obtained via a Bayesian non-parametric latent class approach. We also consider the impact of fake paintings through sensitivity analyses. Our estimates point to there being around 20–120 unattributed Modigliani paintings.

Diagnostic Accuracy of Three Methods of Body Temperature Measurement in Children: A Latent Class Approach

Diagnostic Accuracy of Three Methods of Body Temperature Measurement in Children: A Latent Class Approach

Assessing the Victim-Perpetrator Overlap in Adolescent Dating Violence in China: A Latent Class Analysis.

There is a limited understanding of the pathways that lead to victim-perpetrator overlap in adolescent dating violence (ADV) particularly in developing countries such as China. Applying a latent class approach, the present study focuses on whether the overlap can be explained by theoretical constructs grounded in social learning, general strain, and social bonding theories, and whether these constructs relate to underexplored mediating mechanisms or are direct precursors. The study sample consisted of 1,787 dating adolescents (39.8% male; mean age = 17.82 ± 0.92 years) from a population of 5,820 adolescents in 32 high schools in Chinese cities and rural counties. We found a significant victim-perpetrator overlap in controlling, psychological, and physical ADV, with the proclivity increasing with the severity of ADV. Membership of groups engaged in general delinquency and those engaged in dating violence increased the odds of victim-perpetrator overlap. Peer bullying victimization was the most consistent direct predictor of victim-perpetrator overlap in family/peer/community settings; victimization resulting from interparental violence during childhood, peer bullying, and community violence was the most consistent indirect predictor. Neighborhood bonds had a greater impact than social bonds with family and school in terms of direct and indirect protective effects against victim-perpetrator overlap. Patterns of exposure to violent victimization in family/peer/community settings and patterns of social bonds cast new light on the nature of deviant peer (general delinquency vs. ADV) networks, which were found to operate as major mediating mechanisms in victim-perpetrator overlap.

A Bayesian latent class approach to causal inference with longitudinal data.

Bayesian methods are becoming increasingly in demand in clinical and public health comparative effectiveness research. Limited literature has explored parametric Bayesian causal approaches to handle time-dependent treatment and time-dependent covariates. In this article, building on to the work on Bayesian g-computation, we propose a fully Bayesian causal approach, implemented using latent confounder classes which represent the patient's disease and health status. Our setting is suitable when the latent class represents a true disease state that the physician is able to infer without misclassification based on manifest variables. We consider a causal effect that is confounded by the visit-specific latent class in a longitudinal setting and formulate the joint likelihood of the treatment, outcome and latent class models conditionally on the class indicators. The proposed causal structure with latent classes features dimension reduction of time-dependent confounders. We examine the performance of the proposed method using simulation studies and compare the proposed method to other causal methods for longitudinal data with time-dependent treatment and time-dependent confounding. Our approach is illustrated through a study of the effectiveness of intravenous immunoglobulin in treating newly diagnosed juvenile dermatomyositis.

Exploring Adolescents’ Social Media Connection and Disconnection: A Latent Class Approach

In industrialized societies characterized by ubiquitous connectivity, many individuals disconnect from their phones or social media to break patterns of habitual use, reduce information overload, alleviate stress, and avoid distractions. Although research has predominantly focused on (young) adults, information on digital disconnection among adolescents remains limited. In response, in the present study, we analyzed survey data from 956 Flemish adolescents in Belgium (<em>M</em><sub>age</sub><em> </em>= 15.10, <em>SD </em>= 1.61, and 64.8% girls) and, using latent class analysis, identified two subgroups: Class 1 labeled as “low adoption of social media apps and disconnection tools,” and Class 2 labeled as “high adoption of social media apps and disconnection tools.” Adolescents in Class 2 were more likely to use social media, specifically social networking, instant messaging, and entertainment apps, and to adopt disconnection tools (e.g., iOS Screentime or the Forest app). Next, drawing on the media practice model, we investigated how sociodemographic and sociopsychological factors contribute to those usage patterns and found that girls, older adolescents, and adolescents with higher scores for depressive symptoms and flourishing were more likely to belong to Class 2. Those findings offer new insights into adolescents’ social media connection and disconnection that can inform interventions to promote healthier smartphone use among adolescents.

Latent class approach to classify e-scooter non-users: A comparative study of Helsinki and Tokyo

Recent years have seen a rapid deployment of shared electric scooters (e-scooters), across hundreds of cities worldwide. Despite the global diffusion, most urban residents are still not using e-scooters, and they often bear the negative consequences of e- scooter deployment. This comparative study investigates the attitudes and characteristics of the e-scooter non-users in Helsinki, Finland and Tokyo, Japan using latent class clustering analysis. Our findings reveal five classes of e-scooter non-users, ranging from very negative perspectives to the least negative perspectives towards e-scooter deployment. Non-users also have a diverse set of socio-demographics background, including variation across age, gender, and income. In addition, non-users mostly do not feel the necessity of using an e-scooter as well as lack competences and have safety concerns. Safety concerns are dominant among those with more negative perceptions, while classes with the least negative perceptions and low income tend to not use e-scooters due to high rental costs. Understanding the different categories and perspectives of non-users, alongside with understanding users, has clear smart mobility policy and management implications. On the one hand, it should help with understanding the effectiveness of different interventions stemming from e-scooter operators and public policy actors. On the other hand, this study’s findings contribute to understanding public acceptability of e-scooter technology and associated policy and management actions. The study also opens further pathways for both comparative studies and for understanding different groups of users and non-users of smart mobility technologies.

Ideal Family Size and Reproductive Orientations: An Exploration of Change Over Time in the United States

Drawing on more than 30 years of nationally representative microdata from the General Social Survey, this article comprehensively updates recent trends in ideal family size in the United States. It first documents stability in ideal family sizes between the mid-1980s and 2018, even in the face of a recent fertility decline. Next, the study adopts a latent class approach that identifies typologies of “reproductive orientations,” defined as multidimensional mental models of reproduction encompassing ideal family size, attitudes toward reproductive labor, and views on reproduction contexts. The findings indicate three distinct classes of reproductive orientations: Progressive Familialists, Conservative Familialists, and Blended Egalitarians. Further analyses suggest that the prevalence of these classes has changed over time and that class membership is associated with distinct patterns of childbearing and marriage. These findings deepen contemporary understandings of ideal family size in the United States and have broader implications for how demographers conceptualize and measure fertility preferences across diverse contexts.

Instrumental variable estimation with observed and unobserved heterogeneity of the treatment and instrument effect: a latent class approach

Instrumental variable estimation with observed and unobserved heterogeneity of the treatment and instrument effect: a latent class approach

Certified Community Behavioral Health Clinic Services for Clients With Co-occurring Disorders: A Latent Class Approach.

Certified community behavioral health clinics (CCBHCs) are designed to provide comprehensive care for individuals with co-occurring mental and substance use disorders. The authors classified outpatient mental health treatment facilities on the basis of provision of services for clients with co-occurring disorders and assessed whether CCBHCs differed from other outpatient mental health facilities in services provided. The authors used latent class analysis to identify distinct services for clients with co-occurring disorders in 5,692 outpatient mental health facilities in the 2021 National Substance Use and Mental Health Services Survey. Nine indicators were included: treatment for clients with substance or alcohol use disorder co-occurring with serious mental illness or serious emotional disturbance, specialized programs or groups for such clients, medication-assisted treatment (MAT) for alcohol use disorder, MAT for opioid use disorder, detoxification, individual counseling, group counseling, case management, and 12-step groups. A multinomial logistic regression was used to estimate whether CCBHCs were associated with any identified classes after analyses controlled for facility characteristics. A four-class solution provided a model with the best fit, comprising comprehensive services (23.4%), case management services (17.7%), counseling and self-help services (58.6%), and professional services (4.3%). Regressing class membership on facility type and covariates, the authors found that compared with community mental health clinics (CMHCs), CCBHCs were more likely to belong to the comprehensive services class than to the case management services, counseling and self-help services, and professional services classes. CCBHCs were more likely than other outpatient programs to offer comprehensive care, and CCBHC status of a CMHC facilitated enhanced service provisions.

Exploring preferences for contractual terms in a scenario of ecological transition for the agri-food sector: a latent class approach

Governance mechanisms along the agri-food supply chains are increasingly important in a scenario of ecological transition. Under the conceptual and analytical lens of the Transaction Cost Economics, we explored farmers’ preferences towards a variety of clauses usually adopted in production contracts. To this purpose, a discrete choice experiment among 190 durum wheat producers in Italy was conducted. Results from a latent class model showed that producers were mainly interested in fixed prices formula and to join shared rules of production but revealed little or no interest for compelling sustainable cultivation techniques and the provision of technical assistance. However, these preferences are heterogeneous across farmers and vary depending on their level of education and previous use of contractual arrangements, with relevant implications for contract design and management. Governance mechanisms along the agri-food supply chains are increasingly important in a scenario of ecological transition. Under the conceptual and analytical lens of the Neo Institutional Economics, we explored farmers’ preferences towards a variety of clauses usually adopted in production contracts. To this purpose, a discrete choice experiment among 190 durum wheat producers in Italy was conducted. Results from a latent class model showed that that producers were mainly interested in fixed prices formula and to join shared rules of production but revealed little or no interest for compelling sustainable cultivation techniques and the provision of technical assistance. However, these preferences are heterogenous across farmers and vary depending on their level of education and previous use of contractual arrangements, with relevant implications for contract design and management.

Predictors of substance use disorder treatment and mutual support group participation among Black women across the criminal legal spectrum: A latent class approach

BackgroundThe purpose of this study includes: 1) identifying classes of substance-related needs among Black women, and 2) examining the effect of substance-related need classes and culturally-relevant factors on Black women’s use of substance use disorder (SUD) treatment and mutual support groups. MethodsAs part of a longitudinal cohort study, Black women were recruited in prison nearing release, on probation, and in the community without involvement in the criminal legal system (CLS, n=565) and followed-up at 18-months. We conducted a baseline latent class analysis of substance-related needs among Black women. Logistic regression models adjusted for culturally-relevant factors to predict the use of treatment and frequency of mutual support group participation over 18-months among Black women who use drugs. ResultsFour classes by level of needs were found: low, daily marijuana use, high mental health, and high comorbidity. During the 18-month follow-up, women characterized by the high comorbidity need class and with higher scores of religious well-being were more likely to frequently participate in mutual support groups. Non-CLS-involved women were less likely to engage with both treatment and mutual support groups than women from the prison sample at 18-months. ConclusionsThis study highlights four distinct classes of substance-related needs among Black women, highlighting the complex patterns of behavior and within-racial group differences among Black women. Black women with high comorbidity needs were more likely to participate in mutual support groups, but the latent classes did not predict SUD treatment indicating other non-medical and social contextual need factors may be at play.

Trajectories of relationship and sexual satisfaction over 2 years in the Covid-19 pandemic: A latent class analysis.

Previous research on the impact of the Covid-19 pandemic on romantic relationships has mainly concentrated on short-term effects and average trends of change. This study aimed to explore different trajectories of relationship and sexual satisfaction from April 2020 to March 2022. Including a cross-national sample of 2859 individuals, a latent class approach was applied to identify subgroups of sexual and relationship satisfaction trajectories. Participants' satisfaction levels, attachment style, psychological symptoms, life satisfaction, and living arrangements were taken into account. Three latent classes were identified for both relationship and sexual satisfaction: (1) a group with high satisfaction levels but a slightly declining trajectory, (2) a group with low satisfaction levels but an ascending trajectory, and (3) a fluctuating group. Living apart, having children, psychological symptoms, low life satisfaction, and avoidant attachment characterized the two groups with low or fluctuating trajectories. The findings suggest that there were different trajectories of relationship and sexual satisfaction during the pandemic, which can be predicted by individual dispositions. Even 2 years after the pandemic, its effects had not disappeared.

The Association Between Adverse Childhood Experiences and Preterm Delivery: A Latent Class Approach.

Background: Racial and socioeconomic status (SES) disparities in preterm delivery (PTD) have existed in the United States for decades. Disproportionate maternal exposures to adverse childhood experiences (ACEs) may increase the risk for adverse birth outcomes. Moreover, racial and SES disparities exist in the prevalence of ACEs, underscoring the need for research that examines whether ACEs contribute to racial and SES disparities in PTD. Methods: We examined the relationship between ACEs and PTD in a longitudinal sample of N = 3,884 women from the National Longitudinal Study of Adolescent to Adult Health (1994-2018). We applied latent class analysis to (1) identify subgroups of women characterized by patterns of ACE occurrence; (2) estimate the association between latent class membership (LCM) and PTD, and (3) examine whether race and SES influence LCM or the association between LCM and PTD. Results: Two latent classes were identified, with women in the high ACEs class characterized by a higher probability of emotional abuse, physical abuse, sexual abuse, and foster care placement compared with the low ACEs class, but neither class was associated with PTD. Race and SES did not predict LCM. Conclusions: Our findings suggest that ACEs may not impact PTD risk in previously hypothesized ways. Future research should assess the impact of ACEs on the probability of having live birth pregnancies as well as the role of potential protective factors in mitigating the impact of ACEs on PTD.

Profiles of Death Penalty Attitudes: A Latent Class Approach

ABSTRACT Objectives This study examines the heterogeneity in the demographic, normative, and experiential features of those who support, oppose, and are undecided on capital punishment to determine the complexity of characteristics that exist beyond the “averages.” This study also evaluates macro-level changes in death penalty supporter, opposer, and undecided sub-types in the U.S. over time. Methods Drawing upon national trend data from the General Social Survey (1972–2016), a series of Latent Class Analyses (LCA) are employed to identify the number, size, and compositional profile of emergent subtypes of supporters, opposers, and undecideds on the death penalty. We also examine variation in the size and relative distribution of these latent classes over five decades. Results Substantial heterogeneity among death penalty supporters and opposers was found, yielding nine and seven distinct sub-types of each, respectively. Two unique sub-types of undecideds were also uncovered. Conclusion Variation in the composition and size of these latent classes over time provides important insights into the variation in death penalty support over the past 50 years. This is the first study of which we are aware to employ an LCA to national trend data of public opinion toward the death penalty across multiple decades in the United States.

Comparing youth and adult homicide victimization in Finland 2002–2018: A latent class approach

Comparing youth and adult homicide victimization in Finland 2002–2018: A latent class approach

Abstract C012: Integration of social variables resolves subclasses of tumor microenvironment

Abstract Intro: Recent work has identified subclasses of the tumor microenvironment, with most work emphasizing a robust adaptive immune response associated with ER-negative and/or basal-like tumors. Since adaptive immune responses may influence response to chemotherapy and may be targetable by immunotherapy, understanding population-level heterogeneity in immune response is important. Methods: We leveraged the Carolina Breast Cancer Study (CBCS), a large population-based study that oversampled young (≤50 years) and Black women with invasive breast cancer (BC). We performed NanoString gene expression profiling on tissue from 1,329 BC patients, including 704 (53%) Black and 625 (47%) non-Black women, and assayed genes representative of at least 10 individual immune cell types from both adaptive and innate arms of the immune system, homologous recombination deficiency signature and TP53 mutation status. Using a latent class approach that included variables for self-reported race, triple negative status, individual measures of socioeconomic status and community-level measures of advantage and disadvantage, we identified latent classes that integrated tumor microenvironment with tumor biology, individual and community level exposures under a Cells-to-Society conceptual model of breast cancer outcomes. Results: Latent Class Analysis identified 6 integrated classes of the tumor microenvironment, further subdividing a previously identified and validated adaptive immune class into two subgroups. The two adaptive classes were distinguished by different expression levels of B-cells, CD8+ T cells, cytotoxic cells and PD-L1, and differing frequency of basal-like subtype (Adaptive Class 1 RFD [95% CI]: 19.0 [11.9, 26.2]; Adaptive Class 2 RFD [95% CI]: 32.8 [23.3,42.3]) and presence of genomic instability gene expression subtypes (Adaptive Class-1 RFD [95% CI]: 22.1 [14.0, 30.1]; Adaptive Class-2 RFD [95% CI]: 37.5 [28.1,46.2]) relative to the immune-quiet tumors. Discussion: Social variables have complex effects on tumor biology and microenvironment, but latent class models and other model-based approaches may help to integrate biological data with other social factors to identify susceptible subgroups and to optimize precision medicine in diverse populations. Citation Format: Alina M. Hamilton, Kara McCormack, Marc A. Emerson, Melissa A. Troester, Terry Hyslop. Integration of social variables resolves subclasses of tumor microenvironment [abstract]. In: Proceedings of the 16th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2023 Sep 29-Oct 2;Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2023;32(12 Suppl):Abstract nr C012.

Modeling package delivery acceptance in Crowdshipping systems by Public Transportation Passengers: A latent class approach

The subject of this research is to model the probability of accepting package delivery tasks in Crowd-Shipping (CS) systems focusing on shipping packages by Public Transportation (PT) passengers. Real-world implementation of such a PT-based CS initiative requires a proper grasp of the behavior of crowd shippers, package receivers, and senders, among which crowdshippers are at the center of attention. Given survey data collected from 2,208PT passengers in the Sydney metropolitan area, this study constructs a model for package delivery task acceptance among PT passengers who express a willingness to participate as crowdshippers. A Latent Class (LC) choice model is developed and estimated to explore different classes of participant characteristics and preferences for accepting CS delivery tasks, under different levels of offered incentive, package weight, and required detour distance at the destination. As a key finding, the model identifies three classes, namely “leisurely”, “avid”, and “skeptical” users with class sizes of 19%, 53%, and 28%, respectively. Leisurely users tend to be majorly employed with relatively higher ages and lower PT trip frequency per month. This class shows a potential desire to take CS tasks as long as they are not logistically burdensome. Avid users, forming the largest class of the population, are mostly younger PT passengers who are less sensitive to task difficulty and might accept even CS tasks with minimum incentives (around AU$3). Skeptical users, displaying high PT trip frequency per month and high income, are unwilling to accept the CS tasks unless get compensated well.

Estimating Costs Associated with Disease Model States Using Generalized Linear Models: A Tutorial.

Estimates of costs associated with disease states are required to inform decision analytic disease models to evaluate interventions that modify disease trajectory. Increasingly, decision analytic models are developed using patient-level data with a focus on heterogeneity between patients, and there is a demand for costs informing such models to reflect individual patient costs. Statistical models of health care costs need to recognize the specific features of costs data which typically include a large number of zero observations for non-users, and a skewed and heavy right-hand tailed distribution due to a small number of heavy healthcare users. Different methods are available for modelling costs, such as generalized linear models (GLMs), extended estimating equations and latent class approaches. While there are tutorials addressing approaches to decision modelling, there is no practical guidance on the cost estimation to inform such models. Therefore, this tutorial aims to provide a general guidance on estimating healthcare costs associated with disease states in decision analytic models. Specifically, we present a step-by-step guide to how individual participant data can be used to estimate costs over discrete periods for participants with particular characteristics, based on the GLM framework. We focus on the practical aspects of cost modelling from the conceptualization of the research question to the derivation of costs for an individual in particular disease states. We provide a practical example with step-by-step R code illustrating the process of modelling the hospital costs associated with disease states for a cardiovascular disease model.

Do bipolar disorder symptom profiles matter for suicide risk? A latent class approach to investigating differences in suicidal desire and acquired capability.

Depressive and mixed symptoms in bipolar disorder (BD) have been linked to higher suicide risk. Based on Klonsky and May's three-step theory and Joiner's Interpersonal Psychological Theory of Suicide, we hypothesized that patients diagnosed with BD who reported severe levels of depressive symptoms and mixed depressive and manic symptoms would also report higher levels of suicidal desire and acquired capability of suicide, as well as suicidal thoughts and behaviors. The sample included 177 outpatients diagnosed with BD. Latent class analysis was conducted to replicate the identified groups of a previously conducted study using a smaller but overlapping dataset. Between-class and pairwise analyses with measures of suicidal desire and acquired capability were conducted. As expected, the classes characterized by severe depressive symptoms and mixed symptoms reported higher levels of suicidal desire. However, the results regarding acquired capability were less consistent. Given the overall elevated suicide risk of BD and the consistent relationship between depressive symptoms and other strong correlates of suicide, clinicians who work with patients diagnosed with BD should closely monitor changes in their depressive symptoms.