The influence of attributes on community preferences regarding antibiotic treatment: evidence from a discrete choice model

Antibiotic resistance (AR) rates in Vietnam are among the highest in Asia, and recent infections due to multi-drug resistance in the country have caused thousands of deaths each year. This study investigated a Vietnamese community's preferences for antibiotic treatment and its knowledge and attitudes regarding antibiotics. A discrete choice experiment-based survey was developed and administered to the population of interest. The respondents were given sociodemographic-, knowledge- and attitude-related items and 17 pairs of choice tasks. Two hypothetical options were included in each choice task. Latent class analysis was conducted to determine the differences among the respondents' preferences. Among 1,014 respondents, 805 (79.4%) gave valid questionnaires. A three-latent-class model with four covariates (age, healthcare-related education or career, occupation, and attitude classifications) was used in the analysis. All five attributes significantly influenced the respondents' decisions. The majority, including young employed respondents with non-healthcare-related work or education, found treatment failure more important. Older respondents who had healthcare-related education/careers and/or appropriate antibiotic use- and antibiotics resistance-related attitudes, regarded contribution to antibiotic resistance as an important attribute in selecting antibiotic treatments. Unemployed individuals with correct knowledge identified the cost of antibiotic treatment as the most essential decision-making factor. Findings suggest minimal antibiotic impact on resistance; only 7.83% view it as amajor concern. The respondents exhibited substantial preference heterogeneity, and the general Vietnamese public had poor knowledge of and attitudes toward antibiotic use and antibiotic resistance. This study emphasizes the need for individual responsibility for antibiotic resistance and appropriate antibiotic use.

Clustering of health behaviors and their associations with cardiometabolic risk factors among adults at high risk for type 2 diabetes in India: A latent class analysis

BACKGROUND

We aimed to identify clusters of health behaviors and study their associations with cardiometabolic risk factors in adults at high risk for type 2 diabetes in India.

METHODS

Baseline data from the Kerala Diabetes Prevention Program (n = 1000; age 30-60 years) were used for this study. Information on physical activity (PA), sedentary behavior, fruit and vegetable intake, sleep, and alcohol and tobacco use was collected using questionnaires. Blood pressure, waist circumference, 2-h plasma glucose, high-density lipoprotein and low-density lipoprotein cholesterol, and triglycerides were measured using standardized protocols. Latent class analysis was used to identify clusters of health behaviors, and multilevel mixed-effects linear regression was employed to examine their associations with cardiometabolic risk factors.

RESULTS

Two classes were identified, with 87.4% of participants in class 1 and 12.6% in class 2. Participants in both classes had a high probability of not engaging in leisure-time PA (0.80 for class 1; 0.73 for class 2) and consuming <5 servings of fruit and vegetables per day (0.70 for class 1; 0.63 for class 2). However, participants in class 1 had a lower probability of sitting for >=3 h per day (0.26 vs 0.42), tobacco use (0.10 vs 0.75), and alcohol use (0.08 vs 1.00) compared to those in class 2. Class 1 had a significantly lower mean systolic blood pressure (β = -3.70 mm Hg, 95% confidence interval [CI] -7.05, -0.36), diastolic blood pressure (β = -2.45 mm Hg, 95% CI -4.74, -0.16), and triglycerides (β = -0.81 mg/dL, 95% CI -0.75, -0.89).

CONCLUSION

Implementing intervention strategies, tailored to cluster-specific health behaviors, is required for the effective prevention of cardiometabolic disorders among high-risk adults for type 2 diabetes.

Structural and Cognitive Social Capital of Older Korean Adults and Their Relationship with Depression Trajectories: Latent Class and Growth Curve Analyses

This study examined whether the identified latent classes of structural and cognitive social capital are differently associated with depression trajectories in older Korean adults. From the five waves (2006, 2009, 2012, 2015, and 2018) of the Korean Welfare Panel Study, 3,606 participants aged ≥65 were analyzed. The latent class analysis identifies structural and cognitive social capital subgroups. Latent growth curve analysis examined the latent classes' effect on depression trajectories. Three classes were identified: medium-structural and high-cognitive (Class 1), high-structural and cognitive (Class 2), and low-structural and cognitive (Class 3). Classes 1 and 2 showed lower depression at baseline; however, the trajectory change rate was opposite than Class 3. Compared to Classes 1 and 2, depression was highest at baseline but with a slower change rate in Class 3. Therefore, it is important to identify older adults' structural and cognitive social capital classes to depression trajectories.

Opioid therapy trajectories of patients with chronic non-cancer pain over 1 year of follow-up after initiation of short-acting opioid formulations

OBJECTIVE

This study compared opioid utilization trajectories of persons initiating tramadol, short-acting hydrocodone, or short-acting oxycodone, and it characterized opioid dose trajectories and type of opioid in persistent opioid therapy subsamples.

METHODS

A retrospective cohort study of adults with chronic non-cancer pain who were initiating opioid therapy was conducted with the IQVIA PharMetrics® Plus for Academics data (2008-2018). Continuous enrollment was required for 6 months before ("baseline") and 12 months after ("follow-up") the first opioid prescription ("index date"). Opioid therapy measures were assessed every 7 days over follow-up. Group-based trajectory modeling (GBTM) was used to identify trajectories for any opioid and total morphine milligram equivalent measures, and longitudinal latent class analysis was used for opioid therapy type.

RESULTS

A total of 40 276 tramadol, 141 023 hydrocodone, and 45 221 oxycodone initiators were included. GBTM on any opioid therapy identified 3 latent trajectories: early discontinuers (tramadol 39.0%, hydrocodone 54.1%, oxycodone 61.4%), late discontinuers (tramadol 37.9%, hydrocodone 39.4%, oxycodone 33.3%), and persistent therapy (tramadol 6.7%, hydrocodone 6.5%, oxycodone 5.3%). An additional fourth trajectory, intermittent therapy (tramadol 16.4%), was identified for tramadol initiators. Of those on persistent therapy, 2687 individuals were on persistent therapy with tramadol, 9169 with hydrocodone, and 2377 with oxycodone. GBTM on opioid dose resulted in 6 similar trajectory groups in each persistent therapy group. Longitudinal latent class analysis on opioid therapy type identified 6 latent classes for tramadol and oxycodone and 7 classes for hydrocodone.

CONCLUSION

Opioid therapy patterns meaningfully differed by the initial opioid prescribed, notably the presence of intermittent therapy among tramadol initiators and higher morphine milligram equivalents and prescribing of long-acting opioids among oxycodone initiators.

Health behavior profiles and association with mental health status among US active-duty service members

Introduction

This study investigated the clustering of health behaviors among US active duty servicemembers (ADSM) into risk profiles and explored the association between these profiles with ADSM sociodemographic characteristics and mental health status.

Methods

This study utilized secondary data from the 2018 Health Related Behaviors Survey (HRBS), a Department of Defense (DoD) self-administered online survey. Health behaviors included physical activity, screen use, sleep habits, tobacco/substance use, alcohol drinking, preventive health care seeking and condom use at last sex/having multiple sexual partners. Past-year mental health status was measured using the Kessler Screening Scale for Psychological Distress (K6). Latent class analysis (LCA) on health behaviors was used to cluster ADSMs into risk profiles. Multivariable logistic model was used to examine whether ADSM characteristics and mental health status were associated with ADSMs' risk profiles.

Results

The LCA identified a four-class model that clustered ADSMs into the following sub-groups: (1) Risk Inclined (14.4%), (2) High Screen Users (51.1%), (3) Poor Sleepers (23.9%) and (4) Risk Averse (10.6). Over a tenth (16.4%) of ADSMs were categorized as having serious psychological distress. Being male, younger, less educated, in the Army, Marine Corps or Navy were associated with higher odds of being Risk Inclined (AOR ranging from 1.26 to 2.42). Compared to the reference group of Risk Adverse ADSMs, those categorized as Risk Inclined (AOR: 8.30; 95% CI: 5.16-13.36), High Screen Users (AOR: 2.44; 95% CI: 1.56-3.82) and Poor Sleepers (AOR: 5.26; 95% CI: 3.38-8.19) had significantly higher odds of having serious psychological distress.

Discussion

Study findings suggest opportunities to tailor behavioral and health promotion interventions for each of the distinct risk profiles. For example, ADSM described as Risk Inclined may benefit from preventive mental health services. Solutions for ADSM described as Poor Sleepers may include education on sleep hygiene; instituting duty schedules; and shifting military cultural norms to promote sleep hygiene as a pathway to optimal performance and thus military readiness. ADSM with low-risk behavior profiles such as those described as Risk Averse may prove beneficial in the roll-out of interventions as they act as peer-educators or mentors.

Joint modeling approaches for censored predictors due to detection limits with applications to metabolites data

Measures of substance concentration in urine, serum or other biological matrices often have an assay limit of detection. When concentration levels fall below the limit, exact measures cannot be obtained, and thus are left censored. The problem becomes more challenging when the censored data come from heterogeneous populations consisting of exposed and non-exposed subjects. If the censored data come from non-exposed subjects, their measures are always zero and hence censored, forming a latent class governed by a distinct censoring mechanism compared with the exposed subjects. The exposed group's censored measurements are always greater than zero, but less than the detection limit. It is very often that the exposed and non-exposed subjects may have different disease traits or different relationships with outcomes of interest, so we need to disentangle the two different populations for valid inference. In this article, we aim to fill the methodological gaps in the literature by developing a novel joint modeling approach to not only address the censoring issue in predictors, but also untangle different relationships of exposed and non-exposed subjects with the outcome. Simulation studies are performed to assess the numerical performance of our proposed approach when the sample size is small to moderate. The joint modeling approach is also applied to examine associations between plasma metabolites and blood pressure in Bogalusa Heart Study, and identify new metabolites that are highly associated with blood pressure.

Compliance with World Health Organization COVID-19 preventive behaviors in rural counties in Western Kenya: a cross-sectional study

Introduction

the World Health Organization (WHO) recommended various measures to tackle COVID-19, and were adopted by many governments, targeting behavior change among citizens to lower the transmission. There was a paucity of data on the patterns of compliance with different measures within individuals and whether people adhere to all recommended measures or cautiously prefer few but not others. Understanding compliance behaviors and associated factors is important for developing interventions to increase compliance.

Methods

cross-sectional study was conducted among adults in the western region of Kenya. A sample of 806 participants was selected using a stratified sampling method. A structured questionnaire was used to gather data from the participants. Compliance was assessed with six behaviors: hand sanitation, proper hygiene, no handshaking, social distancing, and other guidelines. Latent analysis was used to identify behavioral patterns. Descriptive statistics were used to assess demographic characteristics, in terms of frequency distribution, and percentages. Multinomial logistic regression was used to assess the association between demographic characteristics and compliance level.

Results

compliance was highest for masking (85.3%), and was lowest for social distancing (60.2%). The majority of participants were found to be full compliers (class 1: 40.5%), there was an increased probability of full compliance among those aged between 18-30 years (OR= 1.042; 95% CI: 0.307-13.052, p < 0.040) compared to those aged ≥70.

Conclusion

using facemasks had the highest rate of compliance, followed by hand sanitization and proper hygiene. However, overall, the findings showed that while compliance with some protocol behaviors is high, individuals comply consistently across recommended compliance behaviors.

Testing latent classes in gut microbiome data using generalized Poisson regression models

Human microbiome research has gained increasing importance due to its critical roles in comprehending human health and disease. Within the realm of microbiome research, the data generated often involves operational taxonomic unit counts, which can frequently present challenges such as over-dispersion and zero-inflation. To address dispersion-related concerns, the generalized Poisson model offers a flexible solution, effectively handling data characterized by over-dispersion, equi-dispersion, and under-dispersion. Furthermore, the realm of zero-inflated generalized Poisson models provides a strategic avenue to simultaneously tackle both over-dispersion and zero-inflation. The phenomenon of zero-inflation frequently stems from the heterogeneous nature of study populations. It emerges when specific microbial taxa fail to thrive in the microbial community of certain subjects, consequently resulting in a consistent count of zeros for these individuals. This subset of subjects represents a latent class, where their zeros originate from the genuine absence of the microbial taxa. In this paper, we introduce a novel testing methodology designed to uncover such latent classes within generalized Poisson regression models. We establish a closed-form test statistic and deduce its asymptotic distribution based on estimating equations. To assess its efficacy, we conduct an extensive array of simulation studies, and further apply the test to detect latent classes in human gut microbiome data from the Bogalusa Heart Study.

Subphenotypes of SARS-CoV-2-Associated ARDS Overlap Each Other: A Retrospective Analysis

Background  Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus-associated pneumonia and acute respiratory distress syndrome (ARDS) were often associated with hyperinflammation and elevation of several serum inflammatory markers but usually less than what is observed in non-coronavirus disease (COVID) ARDS. Elevated inflammatory markers such as C-reactive protein, interleukin (IL)-6, etc., are associated with severe infection. This study identified subphenotypes of COVID-19 ARDS patients by latent profile analysis in a cohort of Indian patients. Methods  Data of n  = 233 adult Indian patients with laboratory-confirmed SARS-CoV-2 infection admitted to a tertiary care teaching hospital were analyzed in this retrospective study. Only patients with acute respiratory failure (defined by partial pressure of oxygen/fraction of inspired oxygen ratio < 200 mm Hg) and chest X-ray showing bilateral infiltrates were included. Results  The patients' mean (standard deviation) age was 53.3 (14.9) years, and 62% were male. A two subphenotypic model was formulated based on the lowest Bayesian information criterion. Neutrophil-to-lymphocyte ratio and serum IL-6 were latent variables in that model (entropy 0.91). The second phenotype (hyperinflammatory) had lower platelet count ( p  = 0.02), higher serum creatinine ( p  = 0.004), higher C-reactive protein ( p  = 0.001), higher ferritin ( p  < 0.001), and serum lactate dehydrogenase ( p  = 0.009). Age-adjusted hospital mortality ( p  = 0.007), duration of hospital stay ( p  < 0.001), and duration of intensive care unit stay ( p  < 0.001) were significantly higher in the second subphenotype. Conclusion  Two distinct but overlapping subphenotypes were identified in SARS-CoV-2-associated respiratory failure. Hyperinflammatory subphenotype was associated with significantly poor short-term outcomes.

Procedures for Analyzing Multidimensional Mixture Data

The multidimensional mixture data structure exists in many test (or inventory) conditions. Heterogeneity also relatively exists in populations. Still, some researchers are interested in deciding to which subpopulation a participant belongs according to the participant's factor pattern. Thus, in this study, we proposed three analysis procedures based on the factor mixture model to analyze data in the multidimensional mixture context. Simulations were manipulated with different levels of factor numbers, factor correlations, numbers of latent classes, and class separation. Issues with regard to model selection were discussed at first. The results showed that in the two-class situations the procedures of "factor structure first then class number" (Procedure 1) and "factor structure and class number considered simultaneously" (Procedure 3) performed better than the "class number first then factor structure" (Procedure 2) and yielded precise parameter estimation and classification accuracy. It would be appropriate to choose Procedures 1 and 3 when strong measurement invariance is assumed while using an information criterion, but Procedure 1 saved more time than Procedure 3. In the three-class situations, the performance of all three procedures was limited. Implementations and suggestions have been addressed in this research.

A Bootstrap Approach for Evaluating Uncertainty in the Number of Groups Identified by Latent Class Growth Models

The use of longitudinal finite mixture models such as group-based trajectory modeling has seen a sharp increase during the last few decades in the medical literature. However, these methods have been criticized, especially because of the data-driven modeling process, which involves statistical decision-making. In this paper, we propose an approach that uses the bootstrap to sample observations with replacement from the original data to validate the number of groups identified and to quantify the uncertainty in the number of groups. The method allows investigation of the statistical validity and uncertainty of the groups identified in the original data by checking to see whether the same solution is also found across the bootstrap samples. In a simulation study, we examined whether the bootstrap-estimated variability in the number of groups reflected the replicationwise variability. We evaluated the ability of 3 commonly used adequacy criteria (average posterior probability, odds of correct classification, and relative entropy) to identify uncertainty in the number of groups. Finally, we illustrate the proposed approach using data from the Quebec Integrated Chronic Disease Surveillance System to identify longitudinal medication patterns between 2015 and 2018 in older adults with diabetes.

[Latent classes of health risk behaviors and their relationship with self-control among rural secondary school students in Guizhou Province]

OBJECTIVE

To understand the potential categories of health risk Guizhou Province.

METHODS

From November to December 2021, 4452 rural students in middle school students with average age of(13.5±1.6) years were selected from Guizhou Province by multi-stage stratified random cluster sampling method.1505(33.8%) students in the first grade, 1958(44.0%) students in the second grade and 989(22.2%) students in the third grade. There were 2295 boys(51.5%) and 2157 girls(48.5%). Basic information questionnaire, health risk behavior questionnaire and self-control scale were used for questionnaire survey. Latent category analysis was used to explore the potential categories of health risk behaviors, and disordered multiple classification logistic regression analysis was used to explore the relationship between potential categories and self-control.

RESULTS

The health risk behaviors of rural middle school students in Guizhou Province could be divided into four potential categories: low risk group(71.4%), medium risk group(11.6%), sub-high risk group(5.2%) and high risk group(10.7%). There were statistically significant differences in the distribution characteristics of potential categories of junior middle school students with different gender, grade, nationality, only child, accommodation, stay-behind, academic performance, academic pressure, peer relationship, parent-child relationship, teacher-student relationship and domestic violence(P<0.05 or P<0.01). Taking the low-risk group as the reference group, the highest self-control scores were in the medium risk group(OR=1.049, 95%CI 1.040-1.058), the sub-high risk group(OR=1.098, 95%CI 1.083-1.113), and the high risk group(OR=1.077, 95%CI 1.066-1.087).

CONCLUSION

The latent characteristics of health risk behavior of rural junior middle school students in Guizhou Province are obvious. Improving self-control ability can reduce the occurrence of medium risk group, sub-high risk group and high risk group.

Healthcare center clustering for Cox's proportional hazards model by fusion penalty

There has been growing research interest in developing methodology to evaluate healthcare centers' performance with respect to patient outcomes. Conventional assessments can be conducted using fixed or random effects models, as seen in provider profiling. We propose a new method, using fusion penalty to cluster healthcare centers with respect to a survival outcome. Without any prior knowledge of the grouping information, the new method provides a desirable data-driven approach for automatically clustering healthcare centers into distinct groups based on their performance. An efficient alternating direction method of multipliers algorithm is developed to implement the proposed method. The validity of our approach is demonstrated through simulation studies, and its practical application is illustrated by analyzing data from the national kidney transplant registry.

Latent heterogeneity of muscle-invasive bladder cancer in patient characteristics and survival: A population-based nation-wide study in the Bladder Cancer Data Base Sweden (BladderBaSe)

BACKGROUND

Patients with muscle-invasive bladder cancer (MIBC) constitute a heterogenous group in terms of patient and tumour characteristics ('case-mix') and prognosis. The aim of the current study was to investigate whether differences in survival could be used to separate MIBC patients into separate classes using a recently developed latent class regression method for survival analysis with competing risks.

METHODS

We selected all participants diagnosed with MIBC in the Bladder Cancer Data Base Sweden (BladderBase) and analysed inter-patient heterogeneity in risk of death from bladder cancer and other causes.

RESULTS

Using data from 9653 MIBC patients, we detected heterogeneity with six distinct latent classes in the studied population. The largest, and most frail class included 50% of the study population and was characterised by a somewhat larger proportion of women, higher age at diagnosis, more advanced disease and lower probability of curative treatment. Despite this, patients in this class treated with curative intent by radical cystectomy or radiotherapy had a lower association to risk of death. The second largest class included 23% and was substantially less frail as compared to the largest class. The third and fourth class included each around 9%-10%, whereas the fifth and sixth class included each 3%-4% of the population.

CONCLUSIONS

Results from the current study are compatible with previous research and the method can be used to adjust comparisons in prognosis between MIBC populations for influential differences in the distribution of sub-classes.

Patient preferences do matter: a discrete choice experiment conducted with breast cancer patients in six European countries, with latent class analysis

OBJECTIVES

The evolution of breast cancer (BC) treatments has resulted in tailored therapies for the different types and stages of BC. Each treatment has a profile of benefits and adverse effects which are taken into consideration when planning a treatment pathway. This study examines whether patients' preferences are in line with what is considered important from decision makers viewpoint.

METHODS

An online discrete choice experiment was conducted in six European countries (France, Germany, Ireland, Poland, Spain, UK) with BC patients. Six attributes were included: overall survival (OS), hyperglycemia, rash, pain, functional well-being (FWB), and out-of-pocket payment (OOP). Sixteen choice sets with two hypothetical treatments and a "No treatment" option were presented. Data were analyzed with the use of heteroscedastic conditional, mixed logistic, and latent class models. Marginal rate of substitution (MRS) were estimated for OOP versus the rest of attributes to establish the ranking of preferences for each attribute.

RESULTS

Two hundred and forty-seven patients with advanced or metastatic BC and 314 with early-stage BC responded. Forty-nine percent of patients were < 44 years old and 65 percent had completed university education. The MRS of the analysis demonstrated that "severe pain" is the highest dis-preferred attribute level, followed by "severe impairment in FWB" and OS. Four classes of patients as "decision makers" were identified.

CONCLUSIONS

This study suggests that there is heterogeneity in treatment preferences of BC patients depending on their sociodemographic and disease-related characteristics. In combination with clinical guidelines, patient preferences can support the selection and tailoring of treatment options.

Inferring Latent Structure in Polytomous Data with a Higher-Order Diagnostic Model

Researchers continue to develop and advance models for diagnostic research in the social and behavioral sciences. These diagnostic models (DMs) provide researchers with a framework for providing a fine-grained classification of respondents into substantively meaningful latent classes as defined by a multivariate collection of binary attributes. A central concern for DMs is advancing exploratory methods for uncovering the latent structure, which corresponds with the relationship between unobserved binary attributes and observed polytomous items with two or more response options. Multivariate behavioral polytomous data are often collected within a higher-order design where general factors underlying first-order latent variables. This study advances existing exploratory DMs for polytomous data by proposing a new method for inferring the latent structure underlying polytomous response data using a higher-order model to describe dependence among the discrete latent attributes. We report a novel Bayesian formulation that uses variable selection techniques for inferring the latent structure along with a higher-order factor model for attributes. We report evidence of accurate parameter recovery in a Monte Carlo simulation study and present results from an application to the 2012 Programme for International Student Assessment (PISA) problem-solving vignettes to demonstrate the method.

A latent class model to multiply impute missing treatment indicators in observational studies when inferences of the treatment effect are made using propensity score matching

Analysts often estimate treatment effects in observational studies using propensity score matching techniques. When there are missing covariate values, analysts can multiply impute the missing data to create m completed data sets. Analysts can then estimate propensity scores on each of the completed data sets, and use these to estimate treatment effects. However, there has been relatively little attention on developing imputation models to deal with the additional problem of missing treatment indicators, perhaps due to the consequences of generating implausible imputations. However, simply ignoring the missing treatment values, akin to a complete case analysis, could also lead to problems when estimating treatment effects. We propose a latent class model to multiply impute missing treatment indicators. We illustrate its performance through simulations and with data taken from a study on determinants of children's cognitive development. This approach is seen to obtain treatment effect estimates closer to the true treatment effect than when employing conventional imputation procedures as well as compared to a complete case analysis.

Racial/Ethnic Differences in Chronic Disease Predictors Among American High School Students

BACKGROUND

Few studies have attempted to define clusters of chronic disease predictors with additional focus on racial/ethnic differences. The purpose of this study was to highlight differences in predictors of chronic diseases among American high school students by identifying subgroups using latent class analysis (LCA).

METHODS

The chronic disease predictor variable used in the analysis was created from 5 modified items in the 2019 Youth Risk Behavior Surveillance that were identified to be critical to healthy lifestyles in Healthy People 2020. Descriptive, bivariate, multinomial logistic regression and LCA were performed using SAS 9.4 and Mplus in 9th to 12th grade students, using data from the Youth Risk Behavior Survey (N = 13,677).

RESULTS

Three distinct classes emerged for US high school students and were characterized as high, moderate, and low risk of chronic disease (38%, 33%, and 29%, respectively). Black and Asian students had a higher chance of being in the high-risk class of chronic diseases.

IMPLICATIONS FOR SCHOOL HEALTH POLICY, PRACTICE, AND EQUITY

Emphasis should be placed on sociocultural and socio-environmentally structured prevention programs for at risk/students, ensuring that policy formation reflects the language, identity, and needs of the populations at risk.

CONCLUSIONS

The behavioral similarities of the classes identified highlight the need for continued research, novel interventions, and culturally sensitive strategies and policies in US high schools.

Preference Variation: Where Does Health Risk Attitude Come Into the Equation?

OBJECTIVES

Decisions about health often involve risk, and different decision makers interpret and value risk information differently. Furthermore, an individual's attitude toward health-specific risks can contribute to variation in health preferences and behavior. This study aimed to determine whether and how health-risk attitude and heterogeneity of health preferences are related.

METHODS

To study the association between health-risk attitude and preference heterogeneity, we selected 3 discrete choice experiment case studies in the health domain that included risk attributes and accounted for preference heterogeneity. Health-risk attitude was measured using the 13-item Health-Risk Attitude Scale (HRAS-13). We analyzed 2 types of heterogeneity via panel latent class analyses, namely, how health-risk attitude relates to (1) stochastic class allocation and (2) systematic preference heterogeneity.

RESULTS

Our study did not find evidence that health-risk attitude as measured by the HRAS-13 distinguishes people between classes. Nevertheless, we did find evidence that the HRAS-13 can distinguish people's preferences for risk attributes within classes. This phenomenon was more pronounced in the patient samples than in the general population sample. Moreover, we found that numeracy and health literacy did distinguish people between classes.

CONCLUSIONS

Modeling health-risk attitude as an individual characteristic underlying preference heterogeneity has the potential to improve model fit and model interpretations. Nevertheless, the results of this study highlight the need for further research into the association between health-risk attitude and preference heterogeneity beyond class membership, a different measure of health-risk attitude, and the communication of risks.

Assessing Consumer Willingness to Pay for Nutritional Information Using a Dietary App

A healthy society is the foundation of development in every country, and one way to achieve a healthy society is to promote healthy nutrition. An unbalanced diet is one of the leading causes of noncommunicable diseases globally. If food was correctly selected and correctly consumed, both the problems of overeating and lack of nutrition could be largely solved while also decreasing public health costs. Interventions such as presenting necessary information and warning labels would help consumers make better food choices. Hence, providing nutritional information to consumers becomes essential. The present study investigates the importance of nutrition information labels on consumers' preferences by estimating their willingness to pay for features and information provided by a dietary software program (app). An application can easily display the information to the consumers and help them make informed food choices. A discrete choice experiment investigated consumers' preferences and willingness to pay to receive nutritional information. Mixed multinomial logit and latent class analysis were applied. The results showed the existence of heterogeneity in consumer preferences for different nutritional information provided by the application. Consumers are willing to pay more for salt and fat alerts. The results of this study allow for the analysis of consumers' interest in nutritional information. Such results are essential for the industry for future investments in similar applications that potentially could help consumers make better informed choices.

Three-year trajectories of alcohol use among at-risk and among low-risk drinkers in a general population sample of adults: A latent class growth analysis of a brief intervention trial

Background

Few studies have assessed trajectories of alcohol use in the general population, and even fewer studies have assessed the impact of brief intervention on the trajectories. Especially for low-risk drinkers, it is unclear what trajectories occur, whether they benefit from intervention, and if so, when and how long. The aims were first, to identify alcohol use trajectories among at-risk and among low-risk drinkers, second, to explore potential effects of brief alcohol intervention and, third, to identify predictors of trajectories.

Methods

Adults aged 18-64 years were screened for alcohol use at a municipal registration office. Those with alcohol use in the past 12 months (N = 1646; participation rate: 67%) were randomized to assessment plus computer-generated individualized feedback letters or assessment only. Outcome was drinks/week assessed at months 3, 6, 12, and 36. Alcohol risk group (at-risk/low-risk) was determined using the Alcohol Use Disorders Identification Test-Consumption. Latent class growth models were estimated to identify alcohol use trajectories among each alcohol risk group. Sex, age, school education, employment status, self-reported health, and smoking status were tested as predictors.

Results

For at-risk drinkers, a light-stable class (46%), a medium-stable class (46%), and a high-decreasing class (8%) emerged. The light-stable class tended to benefit from intervention after 3 years (Incidence Rate Ratio, IRR=1.96; 95% Confidence Interval, CI: 1.14-3.37). Male sex, higher age, more years of school, and current smoking decreased the probability of belonging to the light-stable class (p-values<0.05). For low-risk drinkers, a very light-slightly increasing class (72%) and a light-increasing class (28%) emerged. The very light-slightly increasing class tended to benefit from intervention after 6 months (IRR=1.60; 95% CI: 1.12-2.28). Male sex and more years of school increased the probability of belonging to the light-increasing class (p-value < 0.05).

Conclusion

Most at-risk drinkers did not change, whereas the majority of low-risk drinkers increased alcohol use. There may be effects of alcohol feedback, with greater long-term benefits among persons with low drinking amounts. Our findings may help to identify refinements in the development of individualized interventions to reduce alcohol use.

Differential Associations Between Changes in Food Environment and Changes in BMI Among Adults Living in Urban, Low-Income Communities

BACKGROUND

Food environments can contribute to excess weight gain among adults, but the evidence is mixed.

OBJECTIVES

This longitudinal study investigated the associations between changes in the food environment and changes in BMI in adults and whether changes in the food environment differentially impact various subgroups.

METHODS

At 2 time points, BMI was calculated using self-reported height and weight data from 517 adults (mean age, 41 years) living in 4 New Jersey cities. The counts of different types of food outlets within 0.4, 0.8, and 1.6 km of respondents' residences were collected at baseline and tracked until follow-up. A binary measure of social standing (social-advantage group, n = 219; social-disadvantage group, n = 298) was created through a latent class analysis using social, economic, and demographic variables. Multivariable linear regression modeled the associations between changes in BMI with measures of the food environment; additionally, interaction terms between the measures of food environment and social standing were examined.

RESULTS

Overall, over 18 months, an increase in the number of small grocery stores within 0.4 km of a respondent's residence was associated with a decrease in BMI (β = -1.0; 95% CI: -1.9, -0.1; P = 0.024), while an increase in the number of fast-food restaurants within 1.6 km was associated with an increase in BMI (β = 0.1; 95% CI: 0.01, 0.2; P = 0.027). These overall findings, however, masked some group-specific associations. Interaction analyses suggested that associations between changes in the food environment and changes in BMI varied by social standing. For instance, the association between changes in fast-food restaurants and changes in BMI was only observed in the social-disadvantage group (β = 0.1; 95% CI: 0.02, 0.2; P = 0.021).

CONCLUSIONS

In a sample of adults living in New Jersey, changes in the food environment had differential effects on individuals' BMIs, based on their social standing.

Meta-analysis of dichotomous and ordinal tests with an imperfect gold standard

Standard methods for the meta-analysis of medical tests, without assuming a gold standard, are limited to dichotomous data. Multivariate probit models are used to analyse correlated dichotomous data, and can be extended to model ordinal data. Within the context of an imperfect gold standard, they have previously been used for the analysis of dichotomous and ordinal test data from a single study, and for the meta-analysis of dichotomous tests. However, they have not previously been used for the meta-analysis of ordinal tests. In this article, we developed a Bayesian multivariate probit latent class model for the simultaneous meta-analysis of ordinal and dichotomous tests without assuming a gold standard, which also allows one to obtain summary estimates of joint test accuracy. We fitted the models using the software Stan, which uses a state-of-the-art Hamiltonian Monte Carlo algorithm, and we applied the models to a dataset in which studies evaluated the accuracy of tests, and test combinations, for deep vein thrombosis. We demonstrate the issues with dichotomising ordinal test accuracy data in the presence of an imperfect gold standard, before applying and comparing several variations of our proposed model which do not require the data to be dichotomised. The models proposed will allow researchers to more appropriately meta-analyse ordinal and dichotomous tests without a gold standard, potentially leading to less biased estimates of test accuracy. This may lead to a better understanding of which tests, and test combinations, should be used for any given medical condition.

Classification of older adults who underwent lumbar-related surgery using pre-operative biopsychosocial predictors and relationships with surgical recovery: An observational study conducted in the United States

Lumbar surgery is a commonly prescribed intervention for low back pain but poses higher risks and worse outcomes for older adults. Identifying clinical phenotypes based on biopsychosocial factors may help identify older adults who are at greatest risk for poor postoperative recovery. This study aimed to (a) classify older adults who underwent lumbar surgery based on preoperative biopsychosocial factors, and (b) quantify the association between preoperative biopsychosocial classifications and 3 and 12 months postoperative improvement outcomes. Latent class analysis was used to identify biopsychosocial classifications in 10,283 individuals aged ≥60 from the Quality Outcomes Database (the United States, 2021-2018). Logistic regression models measured the association between biopsychosocial classifications and 3 and 12 months postoperative outcomes (back/leg pain intensity, disability and quality of life), adjusting for covariates. Three classes were identified based on 19 a priori biopsychosocial factors and were characterised as 'high-risk' (15%), 'physical-/social health-risk' (44%) and 'low-risk' (41%). The high-risk class demonstrated increased odds of failing to recover post-operatively compared to the other classes. Similarly, the physical-/social-risk class demonstrated increased odds of failing to recover in all outcomes and time points compared to the low-risk class. Biopsychosocial factors with higher prevalence in the high versus low-risk class were depression (92.5% vs. 10.6%), multiple morbidities (55.3% vs. 25.7%) and obesity (59.5% vs. 37.2%). This study introduces novel non-recovery phenotypes for older adults undergoing lumbar surgery and may lead to the development of tailored interventions to improve clinical care and outcomes for this population.

Loss-Related Characteristics and Symptoms of Depression, Prolonged Grief, and Posttraumatic Stress Following Suicide Bereavement

(1) Background: The aim of the present study was to examine symptom classes of major depressive disorder (MDD), prolonged grief disorder (PGD), and posttraumatic stress disorder (PTSD) in a sample of suicide-bereaved individuals, while accounting for loss-related characteristics. (2) Methods: A latent class analysis was conducted to identify classes of the suicide bereaved, sharing symptom profiles, in a German suicide-bereaved sample (N = 159). (3) Results: Our analyses revealed three main classes: a resilient class (16%), a class with high endorsement probability for PGD symptoms (50%), and a class with high endorsement probability for combined PGD/PTSD symptoms (34%). Prolonged grief and intrusive symptoms emerged across all classes, while MDD showed low endorsement probability. Our results indicate an association between class membership and time passed since the loss; however, this applies only to the comparison between the PGD and the resilient class, and not for the PGD/PTSD class. (4) Conclusions: Our results may provide information about the predictability of symptom clusters following suicide bereavement. The findings also represent a significant step towards tailoring treatments based on the needs of relevant suicide-bereaved subgroups through a symptom-level approach. Time passed since loss might explain differences between symptom clusters.

How to Define Mild to Severe Hidradenitis Suppurativa? A Simple New Tool Based on Latent Class Analysis of EPIVER Data Study

Purpose

Hidradenitis suppurativa (HS) is an inflammatory skin disease characterized by recurrent or chronic painful and suppurating lesions in the apocrine gland-bearing regions. The lack of knowledge about HS and its extremely heterogeneous clinical presentation, in terms of both lesion appearance and sites of involvement, frequently delay its diagnosis for several years. Objectives: in this study, using the latent class analysis, it was demonstrated that severity of HS could be evaluated not only with clinical or surgical characteristics but also with gender specificities.

Patients and Methods

Clinical and sociodemographic data of HS patients were retrospectively analysed with the latent class method in order to create a classification tool of disease severity.

Results

From the study of 1428 HS patients (544 men and 884 women), two classification models, depending on gender, were developed. Each classification model was composed of three distinct latent classes clearly identified and defined from mild-to-severe cases of HS. These classification models of HS severity were not distorted by patient ages and were coherent with Hurley stages but were more clinically precise.

Conclusion

In this study, a convenient classification tool, useful for facilitating decision support in routine practice, has been developed. This tool could be used to define clinical subgroups within a study population.

Constructing Taxonomies: Identifying Distinctive Class of HIV Support and Risk Networks among People Who Use Drugs (PWID) and Their Network Members in the HPTN 037 Randomized Controlled Trial

Injection drug use is a significant mode of HIV transmission. Social networks are potential avenues for behavior change among high-risk populations. Increasing knowledge should include a classification or taxonomy system of networks’ attributes, risks, and needs. The current study employed 232 networks comprising 232 indexes, with 464 network members enrolled in Philadelphia. LCA revealed a three-class solution, Low-Risk, Paraphernalia Risk, and High Sex/Moderate Paraphernalia Risk class, among participants. The analysis found receiving money or drugs for sex and employment status increased the odds of belonging to PR and PSR classes. Homelessness and incarceration increased the odds of belonging to the PR class when compared to the LR class. Our findings suggest that classes of risk among PWID comprise clusters of information concerning their members. These findings add depth to our understanding while extending our knowledge of the contextual environment that nurtures or exacerbates the problem.

Differences in Circulating Extracellular Vesicle and Soluble Cytokines in Older Versus Younger Breast Cancer Patients With Distinct Symptom Profiles

Because extracellular vesicle (EV)-associated cytokines, both encapsulated and surface bound, have been associated with symptom severity, and may vary over the lifespan, they may be potential biomarkers to uncover underlying mechanisms of various conditions. This study evaluated the associations of soluble and EV-associated cytokine concentrations with distinct symptom profiles reported by 290 women with breast cancer prior to surgery. Patients were classified into older (≥60 years, n = 93) and younger (< 60 years, n = 197) cohorts within two previously identified distinct symptom severity profiles, that included pain, depressive symptoms, sleep disturbance, and fatigue (i.e., High Fatigue Low Pain and All Low). EVs were extracted using ExoQuick. Cytokine concentrations were determined using Luminex multiplex assay. Mann Whitney U test evaluated the differences in EV and soluble cytokine levels between symptom classes and between and within the older and younger cohorts adjusting for Karnofsky Performance Status (KPS) score, body mass index (BMI), and stage of disease. Partial correlation analyses were run between symptom severity scores and cytokine concentrations. Results of this study suggest that levels of cytokine concentrations differ between EV and soluble fractions. Several EV and soluble pro-inflammatory cytokines had positive associations with depressive symptoms and fatigue within both age cohorts and symptom profiles. In addition, in the older cohort with High Fatigue Low Pain symptom profile, EV GM-CSF concentrations were higher compared to the All Low symptom profile (p < 0.05). Albeit limited by a small sample size, these exploratory analyses provide new information on the association between cytokines and symptom profiles of older and younger cohorts. Of note, unique EV-associated cytokines were found in older patients and in specific symptom classes. These results suggest that EVs may be potential biomarker discovery tools. Understanding the mechanisms that underlie distinct symptom class profiles categorized by age may inform intervention trials and offer precision medicine approaches.

Modeling Wheezing Spells Identifies Phenotypes with Different Outcomes and Genetic Associates

Rationale: Longitudinal modeling of current wheezing identified similar phenotypes, but their characteristics often differ between studies. Objectives: We propose that a more comprehensive description of wheeze may better describe trajectories than binary information on the presence/absence of wheezing. Methods: We derived six multidimensional variables of wheezing spells from birth to adolescence (including duration, temporal sequencing, and the extent of persistence/recurrence). We applied partition-around-medoids clustering on these variables to derive phenotypes in five birth cohorts. We investigated within- and between-phenotype differences compared with binary latent class analysis models and ascertained associations of these phenotypes with asthma and lung function and with polymorphisms in asthma loci 17q12-21 and CDHR3 (cadherin-related family member 3). Measurements and Main Results: Analysis among 7,719 participants with complete data identified five spell-based wheeze phenotypes with a high degree of certainty: never (54.1%), early-transient (ETW) (23.7%), late-onset (LOW) (6.9%), persistent (PEW) (8.3%), and a novel phenotype, intermittent wheeze (INT) (6.9%). FEV₁/FVC was lower in PEW and INT compared with ETW and LOW and declined from age 8 years to adulthood in INT. 17q12-21 and CDHR3 polymorphisms were associated with higher odds of PEW and INT, but not ETW or LOW. Latent class analysis- and spell-based phenotypes appeared similar, but within-phenotype individual trajectories and phenotype allocation differed substantially. The spell-based approach was much more robust in dealing with missing data, and the derived clusters were more stable and internally homogeneous. Conclusions: Modeling of spell variables identified a novel intermittent wheeze phenotype associated with lung function decline to early adulthood. Using multidimensional spell variables may better capture wheeze development and provide a more robust input for phenotype derivation.

Physical Activity Patterns among Individuals with Prediabetes or Type 2 Diabetes across Two Years-A Longitudinal Latent Class Analysis

Background: This study aimed to identify distinct profiles of physical activity (PA) patterns among individuals with prediabetes or type 2 diabetes participating in a two-year PA trial and to investigate predictors of the profiles. Methods: Data (n = 168, collected 2013–2020) from the cohort of a randomized trial aimed at increasing PA in individuals with prediabetes and type 2 diabetes were used. PA and sedentary behaviours were assessed by waist-worn ActiGraph GT1M accelerometers at baseline and at 6, 12, 18 and 24 months. Fifteen PA and sedentary variables were entered into a latent class mixed model for multivariate longitudinal outcomes. Multinominal regression analysis modelled profile membership based on baseline activity level, age, gender, BMI, disease status and group randomisation. Results: Two profiles of PA patterns were identified: “Increased activity” (n = 37, 22%) included participants increasing time in PA and decreasing sedentary time. “No change in activity” (n = 131, 78%) included participants with no or minor changes. “Increased activity” were younger (p = 0.003) and more active at baseline (p = 0.011), compared to “No change in activity”. No other predictor was associated with profile membership. Conclusions: A majority of participants maintained PA and sedentary patterns over two years despite being part of a PA intervention. Individuals improving PA patterns were younger and more active at baseline.

Bayesian approaches to variable selection in mixture models with application to disease clustering

In biomedical research, cluster analysis is often performed to identify patient subgroups based on patients' characteristics or traits. In the model-based clustering for identifying patient subgroups, mixture models have played a fundamental role in modeling. While there is an increasing interest in using mixture modeling for identifying patient subgroups, little work has been done in selecting the predictors that are associated with the class assignment. In this study, we develop and compare two approaches to perform variable selection in the context of a mixture model to identify important predictors that are associated with the class assignment. These two approaches are the one-step approach and the stepwise approach. The former refers to an approach in which clustering and variable selection are performed simultaneously in one overall model, whereas the latter refers to an approach in which clustering and variable selection are performed in two sequential steps. We considered both shrinkage prior and spike-and-slab prior to select the importance of variables. Markov chain Monte Carlo algorithms are developed to estimate the posterior distribution of the model parameters. Practical applications and simulation studies are carried out to evaluate the clustering and variable selection performance of the proposed models.

Couple Religiosity and Well-Being Among Older Adults in the United States

Objectives: This article evaluates whether couples' religious similarity is consequential for the health of older married men and women. Alternatively, we examine whether women's religiosity alone is health-protective to their husbands. Methods: Using dyadic data from the US National Social Life, Health, and Aging Project, a representative sample of 913 individuals ages 62-91 plus their marital partners, we perform latent-class analysis to separate older couples into classes based on religious characteristics. Ordered logistic regression models are then used to assess whether different combinations of religious (dis)similarity are associated with married men and women's well-being. Results: We find that older women in highly religious, homogamous marriages report better mental and physical health relative to women in heterogamous and secular (non-religious) marriages. No significant associations were observed for men. Discussion: Our results emphasize that religiosity is not only an individual trait-dis/similarities within a couple have important implications for older women's well-being.

Combined Approach to Multi-Informant Data Using Latent Factors and Latent Classes: Trifactor Mixture Model

Although collecting data from multiple informants is highly recommended, methods to model the congruence and incongruence between informants are limited. Bauer and colleagues suggested the trifactor model that decomposes the variances into common factor, informant perspective factors, and item-specific factors. This study extends their work to the trifactor mixture model that combines the trifactor model and the mixture model. This combined approach allows researchers to investigate the common and unique perspectives of multiple informants on targets using latent factors and simultaneously take into account potential heterogeneity of targets using latent classes. We demonstrate this model using student self-rated and teacher-rated academic behaviors (N = 24,094). Model specification and testing procedures are explicated in detail. Methodological and practical issues in conducting the trifactor mixture analysis are discussed.

Developmental trajectories of sleep during childhood and adolescence are related to health in young adulthood

AIM

Sleep behaviour is correlated and causally related to physical and mental health. Limited longitudinal data exist on the associations of poor sleep behaviour in childhood and adolescence with adult health. Parent-reported sleep behaviours from 1993 participants of the Raine Study (at ages 5, 8, 10, 14, 17) were used to determine sleep trajectories (using latent class growth analysis).

METHODS

Measures of physical and mental health were compared between sleep trajectories using generalised linear models (at age 20).

RESULTS

Three sleep trajectories were identified as follows: 43% of participants belonged to a trajectory with 'consistently minimal' sleep problems, 49% showed some 'declining' in reporting of sleep problems incidence and 8% had 'persistent' sleep problems. Participants in the 'consistently minimal' trajectory had better physical and mental health outcomes at age 20 compared to those in the 'declining' and 'persistent' trajectories. For example, 'consistently minimal' participants had significantly lower body fat percentage (mean difference: -3.89% (95% CI: -7.41 to -0.38)) and a higher (better) SF-12 mental component score (mean difference: 4.78 (95% CI: 2.35-7.21)) compared to participants in the 'persistent' trajectory.

CONCLUSION

Poor sleep behaviour across childhood and adolescent years is related to poorer physical and mental health in young adulthood.

A Brief Report on Older Working Caregivers: Developing a Typology of Work Environments

OBJECTIVES

A growing proportion of the US labor force juggles paid work with family caregiving of older adults. However, no research has examined caregivers' work environments. The purpose of this brief report is to develop typologies of the work environments of family caregivers.

METHODS

This study used data drawn from the 2008-2012 waves of the Health and Retirement Study. Our sample includes employed individuals who also provided regular help with daily activities to a parent or spouse (n=976). We used latent class analysis to develop caregiver work environment typologies.

RESULTS

Our analyses revealed four typologies among caregivers: a) high quality work environments (n=340; 35%); b) average work environments with high job lock (n=293; 30%); c) low-quality work environments (n=203; 21%), and d) high personal interference in supportive work environments (n=140; 14%). Although only 21% of working caregivers were in a low-quality work environment (Type C), descriptive results suggest that these workers were most likely to be minorities who needed to work for financial reasons, reporting the highest number of health problems, and the most work hours.

DISCUSSION

Our findings provide insights into the types of environments that caregivers work in, and the characteristics of individuals in those environments. We discuss implications of our findings for future research and work-based policy development.

Ranking populations in terms of inequality of health opportunity: A flexible latent type approach

We offer a flexible latent type approach to rank populations according to unequal health opportunities. Building upon the latent-class method, an approch increasingly adopted to estimate health inequalities, our contribution is to let the number of socioeconomic groups considered vary to obtain an opportunity-inequality curve for a population that gives how the between-type inequality varies with the number of types. A population A is said to have less inequality of opportunity than population B if its curve is statistically below that of population B. This version of the latent class approach allows for a robust ranking of 31 European countries regarding inequality of opportunity in health.

Childhood wheeze trajectory-specific risk factors: A systematic review and meta-analysis

BACKGROUND

There is growing interest in the use of latent trajectory methodology to identify wheeze patterns in heterogeneous populations of children. This study systematically reviewed and meta-analyzed childhood wheeze trajectory studies to identify childhood wheeze trajectory group-specific risk factors among children from birth through to adolescence.

METHODS

We included studies that used latent trajectory methodology to identify wheeze trajectories and associated risk factors. We searched PubMed, EMBASE, and Google Scholar from 2000 through September 30, 2019, for relevant studies. The study was conducted according to the PRISMA recommendations.

RESULTS

Thirteen cohort studies conducted in eleven high-income countries were included in our meta-analysis with the length of follow-up ranging from 3 to 18 years. Five distinct latent wheeze trajectory groups were identified: Never/Infrequent, Early-Transient, Early-Persistent, Intermediate-Onset, and Late-Onset. We found moderate-to-strong evidence that family history of asthma predicted persistent childhood wheezing among male children but with lower risk among first-born children. There was weak-to-moderate evidence for childhood atopy, male sex, short duration of breastfeeding, tobacco exposure, daycare attendance, and having siblings as risk factors for Early-Transient wheezing; except for breastfeeding, these factors were also associated with intermediate and Late-Onset wheezing with varying effect sizes in high-risk vs general population cohorts.

CONCLUSIONS

Our findings confirm the consistency of wheeze trajectory groups defined by onset, peak prevalence, and duration; we also suggest a common nomenclature for future trajectory studies. With the exception of the relationship between a family history of asthma and persistent childhood wheezing, commonly suspected wheeze risk factors (childhood atopy, male sex, short duration of breastfeeding, tobacco exposure, daycare attendance, and having siblings) are not trajectory-specific and have varying effects in high-risk vs general population cohorts. Delineation of time-varying risk factor effects may be critical to the specificity of wheeze trajectory group prediction to better inform prognosis and targeted early preventive intervention among at-risk children.

Canine Olfactory Detection of a Non-Systemic Phytobacterial Citrus Pathogen of International Quarantine Significance

For millennia humans have benefitted from application of the acute canine sense of smell to hunt, track and find targets of importance. In this report, canines were evaluated for their ability to detect the severe exotic phytobacterial arboreal pathogen Xanthomonas citri pv. citri (Xcc), which is the causal agent of Asiatic citrus canker (Acc). Since Xcc causes only local lesions, infections are non-systemic, limiting the use of serological and molecular diagnostic tools for field-level detection. This necessitates reliance on human visual surveys for Acc symptoms, which is highly inefficient at low disease incidence, and thus for early detection. In simulated orchards the overall combined performance metrics for a pair of canines were 0.9856, 0.9974, 0.9257 and 0.9970, for sensitivity, specificity, precision, and accuracy, respectively, with 1–2 s/tree detection time. Detection of trace Xcc infections on commercial packinghouse fruit resulted in 0.7313, 0.9947, 0.8750, and 0.9821 for the same performance metrics across a range of cartons with 0–10% Xcc-infected fruit despite the noisy, hot and potentially distracting environment. In orchards, the sensitivity of canines increased with lesion incidence, whereas the specificity and overall accuracy was >0.99 across all incidence levels; i.e., false positive rates were uniformly low. Canines also alerted to a range of 1–12-week-old infections with equal accuracy. When trained to either Xcc-infected trees or Xcc axenic cultures, canines inherently detected the homologous and heterologous targets, suggesting they can detect Xcc directly rather than only volatiles produced by the host following infection. Canines were able to detect the Xcc scent signature at very low concentrations (10,000× less than 1 bacterial cell per sample), which implies that the scent signature is composed of bacterial cell volatile organic compound constituents or exudates that occur at concentrations many fold that of the bacterial cells. The results imply that canines can be trained as viable early detectors of Xcc and deployed across citrus orchards, packinghouses, and nurseries.

A new statistical test for latent class in censored data due to detection limit

Biomarkers of interest in urine, serum, or other biological matrices often have an assay limit of detection. When concentration levels of the biomarkers for some subjects fall below the limit, the measures for those subjects are censored. Censored data due to detection limits are very common in public health and medical research. If censored data from a single exposure group follow a normal distribution or follow a normal distribution after some transformations, Tobit regression models can be applied. Given a Tobit regression model and a detection limit, the proportion of censored data can be determined. However, in practice, it is common that the data can exhibit excessive censored observations beyond what would be expected under a Tobit regression model. One common cause is heterogeneity of the study population, that is, there exists a subpopulation who lack such biomarkers and their values are always under the detection limit, and hence are censored. In this article, we develop a new test for testing such latent class under a Tobit regression model by directly comparing the amount of observed censored data with what would be expected under the Tobit regression model. A closed form of the test statistic as well as its asymptotic properties are derived based on estimating equations. Simulation studies are conducted to investigate the performance of the new test and compare the new one with the existing ones including the Wald test, likelihood ratio test, and score test. Two real data examples are also included for illustrative purpose.

Empirically Derived Dietary Patterns Using Robust Profile Clustering in the Hispanic Community Health Study/Study of Latinos

BACKGROUND

Latent class models (LCMs) have been used in exploring dietary behaviors over a wide set of foods and beverages in a given population, but are prone to overgeneralize these habits in the presence of variation by subpopulations.

OBJECTIVES

This study aimed to highlight unique dietary consumption differences by both study site and ethnic background of Hispanic/Latino populations in the United States, that otherwise might be missed in a traditional LCM of the overall population. This was achieved using a new model-based clustering method, referred to as robust profile clustering (RPC).

METHODS

A total of 11,320 individuals aged 18-74 y from the Hispanic Community Health Study/Study of Latinos (2008-2011) with complete diet data were classified into 9 subpopulations, defined by study site (Bronx, Chicago, Miami, San Diego) and ethnic background. At baseline, dietary intake was ascertained using a food propensity questionnaire. Dietary patterns were derived from 132 food groups using the RPC method to identify patterns of the general Hispanic/Latino population and those specific to an identified subpopulation. Dietary patterns derived from the RPC were compared to those identified from an LCM.

RESULTS

The LCM identified 48 shared consumption behaviors of foods and beverages across the entire cohort, whereas significant consumption differences in subpopulations were identified in the RPC model for these same foods. Several foods were common within study site (e.g., chicken, orange juice, milk), ethnic background (e.g., papayas, plantain, coffee), or both (e.g., rice, tomatoes, seafood). Post hoc testing revealed an improved model fit in the RPC model [Deviance Information Criterion DICRPC = 2.3 × 104, DICLCM  = 9.5 × 106].

CONCLUSIONS

Dietary pattern behaviors of Hispanics/Latinos in the United States tend to align by ethnic background for some foods and by location for other foods. Consideration of both factors is imperative to better understand their contributions to population health and developing targeted nutrition intervention studies.

Substance Use and Suicide Attempts Among Adolescent Males Who Are Members of a Sexual Minority: A Comparison of Synthesized Substance-Use Measures

Adolescent sexual minority males (ASMM) are among the highest risk groups for suicide in the United States, with substance use as a significant risk factor. We tested for an association between substance use and suicidality among ASMM from the 2015 and 2017 Youth Risk Behavior Survey (n = 849). We compared several approaches to synthesizing measures of marijuana, cocaine, heroin, ecstasy, methamphetamines, synthetic marijuana, and prescription drug abuse, including several categorized and continuous indices, latent class analysis based on any use of each substance, and latent profile analysis based on use frequency. Using all approaches, substance use was positively associated with suicide attempts independent of covariates. A continuous cumulative index was the best fit to our data (quasi-information criterion = 853.9969) and detected the largest association, with the highest prevalence of suicide attempts among ASMM who used all substances compared with those who used none (adjusted prevalence ratio = 3.35, 95% confidence interval: 2.41, 4.66). A 3-latent-class model had the second best fit to the data (quasi-information criterion = 878.4464), with the highest prevalence of suicide attempts (adjusted prevalence ratio = 2.54, 95% confidence interval: 1.80, 3.57) among the high-substance-use class compared with the low-use class. Substance use is an especially important focal point for targeted interventions reducing suicidality among ASMM.

Statistical tests for latent class in censored data due to detection limit

Measures of substance concentration in urine, serum or other biological matrices often have an assay limit of detection. When concentration levels fall below the limit, the exact measures cannot be obtained. Instead, the measures are censored as only partial information that the levels are under the limit is known. Assuming the concentration levels are from a single population with a normal distribution or follow a normal distribution after some transformation, Tobit regression models, or censored normal regression models, are the standard approach for analyzing such data. However, in practice, it is often the case that the data can exhibit more censored observations than what would be expected under the Tobit regression models. One common cause is the heterogeneity of the study population, caused by the existence of a latent group of subjects who lack the substance measured. For such subjects, the measurements will always be under the limit. If a censored normal regression model is appropriate for modeling the subjects with the substance, the whole population follows a mixture of a censored normal regression model and a degenerate distribution of the latent class. While there are some studies on such mixture models, a fundamental question about testing whether such mixture modeling is necessary, i.e. whether such a latent class exists, has not been studied yet. In this paper, three tests including Wald test, likelihood ratio test and score test are developed for testing the existence of such latent class. Simulation studies are conducted to evaluate the performance of the tests, and two real data examples are employed to illustrate the tests.

Distinguishing Wheezing Phenotypes from Infancy to Adolescence. A Pooled Analysis of Five Birth Cohorts

Rationale: Pooling data from multiple cohorts and extending the time frame across childhood should minimize study-specific effects, enabling better characterization of childhood wheezing.

Objectives: To analyze wheezing patterns from early childhood to adolescence using combined data from five birth cohorts.

Methods: We used latent class analysis to derive wheeze phenotypes among 7,719 participants from five birth cohorts with complete report of wheeze at five time periods. We tested the associations of derived phenotypes with late asthma outcomes and lung function, and investigated the uncertainty in phenotype assignment.

Results: We identified five phenotypes: never/infrequent wheeze (52.1%), early onset preschool remitting (23.9%), early onset midchildhood remitting (9%), persistent (7.9%), and late-onset wheeze (7.1%). Compared with the never/infrequent wheeze, all phenotypes had higher odds of asthma and lower forced expiratory volume in 1 second and forced expiratory volume in 1 second/forced vital capacity in adolescence. The association with asthma was strongest for persistent wheeze (adjusted odds ratio, 56.54; 95% confidence interval, 43.75–73.06). We observed considerable within-class heterogeneity at the individual level, with 913 (12%) children having low membership probability (<0.60) of any phenotype. Class membership certainty was highest in persistent and never/infrequent, and lowest in late-onset wheeze (with 51% of participants having membership probabilities <0.80). Individual wheezing patterns were particularly heterogeneous in late-onset wheeze, whereas many children assigned to early onset preschool remitting class reported wheezing at later time points.

Conclusions: All wheeze phenotypes had significantly diminished lung function in school-age children, suggesting that the notion that early life episodic wheeze has a benign prognosis may not be true for a proportion of transient wheezers. We observed considerable within-phenotype heterogeneity in individual wheezing patterns.

A Note on the Presence of Spurious Pseudo-Guessing Parameters for Three-Parameter Logistic Models in Heterogeneous Populations

This note raises caution that a finding of a marked pseudo-guessing parameter for an item within a three-parameter item response model could be spurious in a population with substantial unobserved heterogeneity. A numerical example is presented wherein each of two classes the two-parameter logistic model is used to generate the data on a multi-item measuring instrument, while the three-parameter logistic model is found to be associated with a considerable pseudo-guessing parameter estimate on an item. The implications of the reported results for empirical educational research are subsequently discussed.

A typology of nonsuicidal self-injury in a clinical sample: A latent class analysis

Nonsuicidal self‐injury(NSSI) is a behavioural concern and can present in diverse ways, varying by method, frequency, severity, function and so forth. The possible combinations of these features of NSSI produce an array of profiles that makes evaluation and management of this behaviour challenging. The aim of this study was to build upon previous work that reduces the heterogeneity of NSSI patterns by using latent class analysis (LCA) to identify a typology of NSSI. Participants consisted of 235 outpatients aged 14–35 years attending a tertiary psychiatric hospital in Singapore who had reported at least one NSSI behaviour within the last year. Eight indicators captured using the Functional Assessment of Self‐Mutilation were used in the LCA: frequency of NSSI, length of contemplation before engaging in NSSI, usage of more than three NSSI methods, suicidal ideation and four psychological functions of NSSI, that is, social‐positive, social‐negative, automatic‐positive and automatic‐negative. The LCA revealed three distinct groups: Class 1—Experimental/Mild NSSI, Class 2—Multiple functions NSSI/Low Suicide Ideation and Class 3—Multiplefunctions NSSI/Possible Suicide Ideation. Multinomial logistic regression analyses were conducted to examine the associations between class membership and sociodemographic variables as well as measures of emotion dysregulation, childhood trauma, depression and quality of life. Females were overrepresented in Class 3. In general, Class 3 had the poorest scores followed by Class 2. Our analyses suggest that different NSSI subtypes require different treatment indications. Profiling patterns of NSSI may be a potentially useful step in guiding treatment plans and strategies.

Young Adults With Higher Motives and Expectancies of Regular Cannabis Use Show Poorer Psychosocial Functioning

Background: Young adults regularly using cannabis represent a uniquely vulnerable yet heterogeneous cohort. Few studies have examined user profiles using cannabis use motives and expectations. The association between user profiles and psychosocial functioning among only regular users remains unexplored. This exploration is important to improve public education efforts and design tailor treatment approaches. Methods: Regular cannabis users (at least weekly; n = 329) completed an online survey via Amazon Mechanical Turk. The survey measured levels of cannabis use, other substance use, motives and expectations of cannabis use, symptoms of psychosis, depression, anxiety and stress, and reckless behavior such as getting high before work or driving under the influence of cannabis. Latent class analysis was performed using motives and expectations to identify data driven patterns of regular cannabis use. Classes were then used to investigate mental health and behavioral correlates of differences in motives and expectations. Results: A 2-class solution provided the best fit to the data; Class 1: Low Motives and Expectancies (n = 158) characterized by lower endorsement across all motivation and expectation variables, and Class 2: High Motives and Expectancies (n = 171) characterized by endorsing multiple motivations, and higher positive and negative expectations of cannabis use. Classes differed in a range of cannabis use variables; e.g., greater proportion of peer use in Class 2. The High Motives and Expectancies users reported higher symptoms of psychosis (positive and negative symptoms), depression, anxiety, and stress. A higher proportion met the criteria for a cannabis use disorder compared with Low Motives and Expectancies users. High Motives and Expectancies users reported higher mean problems with nicotine dependence and illicit drug use other than cannabis and were more likely to get high before work and drive under the influence of cannabis. Conclusions: There is heterogeneity among young regular cannabis users in their motivations and expectancies of use and associated psychosocial functioning. Understanding motives and expectancies can help segregate which users are at higher risk of worse functioning. These findings are timely when designing targeted assessment and treatment strategies, particularly as cannabis is further decriminalized and accessibility increases.

Consumer preferences for alternatives to chick culling in Germany

The use of highly specialized breeds in poultry production has led to the situation in which fattening layer-type males are not economically viable, when competing with the conventional broiler meat. The vast majority of male layer chicks are therefore culled soon after hatching. Ethical concerns about this practice have led to a public debate in a number of countries, and its tenor seems unambiguous: the practice should end. Political and industrial representatives have also promoted putting an end to chick culling. Two alternatives that are already available or soon to be in the market in a number of countries are dual-use poultry production and in ovo gender determination. However, the alternatives are also not free from controversy. The presented study analyzes consumer attitudes towards these 2 alternatives. A discrete choice experiment on eggs with different production attributes was conducted among a sample of 400 German citizens. Results from a latent class model show that there is considerable heterogeneity in preferences, which can be depicted in 5 consumer segments. Consumer segments differ significantly in socioeconomic characteristics and attitudes towards chick-culling alternatives. One segment decides mainly based on product price. However, 28% of the sample show no price sensitivity but choose based on other product attributes such as the preferred chick-culling alternative or egg type. We find wide approval for in ovo gender determination with no segment disapproving of the technology. When it comes to dual-use poultry, the type of husbandry of cockerels is crucial for the approval of this production scheme. Rearing male chicks in free-range husbandry is the preferred alternative for one segment representing 27% of the sample. Results provide empirical evidence for a diversified egg demand, indicating diverse expectations for poultry production in the future.

An Exploratory Diagnostic Model for Ordinal Responses with Binary Attributes: Identifiability and Estimation

Diagnostic models (DMs) provide researchers and practitioners with tools to classify respondents into substantively relevant classes. DMs are widely applied to binary response data; however, binary response models are not applicable to the wealth of ordinal data collected by educational, psychological, and behavioral researchers. Prior research developed confirmatory ordinal DMs that require expert knowledge to specify the underlying structure. This paper introduces an exploratory DM for ordinal data. In particular, we present an exploratory ordinal DM, which uses a cumulative probit link along with Bayesian variable selection techniques to uncover the latent structure. Furthermore, we discuss new identifiability conditions for structured multinomial mixture models with binary attributes. We provide evidence of accurate parameter recovery in a Monte Carlo simulation study across moderate to large sample sizes. We apply the model to twelve items from the public-use, Early Childhood Longitudinal Study, Kindergarten Class of 1998-1999 approaches to learning and self-description questionnaire and report evidence to support a three-attribute solution with eight classes to describe the latent structure underlying the teacher and parent ratings. In short, the developed methodology contributes to the development of ordinal DMs and broadens their applicability to address theoretical and substantive issues more generally across the social sciences.

A latent class based imputation method under Bayesian quantile regression framework using asymmetric Laplace distribution for longitudinal medication usage data with intermittent missing values

Evaluating the association between diseases and the longitudinal pattern of pharmacological therapy has become increasingly important. However, in many longitudinal studies, self-reported medication usage data collected at patients' follow-up visits could be missing for various reasons. These pieces of missing or inaccurate/untenable information complicate determining the trajectory of medication use and its complete effects for patients. Although longitudinal models can deal with specific types of missing data, inappropriate handling of this issue can lead to a biased estimation of regression parameters especially when missing data mechanisms are complex and depend upon multiple sources of variation. We propose a latent class-based multiple imputation (MI) approach using a Bayesian quantile regression (BQR) that incorporates cluster of unobserved heterogeneity for medication usage data with intermittent missing values. Findings from our simulation study indicate that the proposed method performs better than traditional MI methods under certain scenarios of data distribution. We also demonstrate applications of the proposed method to data from the Prospective Study of Outcomes in Ankylosing Spondylitis (AS) cohort when assessing an association between longitudinal nonsteroidal anti-inflammatory drugs (NSAIDs) usage and radiographic damage in AS, while the longitudinal NSAID index data are intermittently missing.