Latent class profile analysis: an application to stage-sequential process in early-onset drinking behaviours.

The Latent Class Structure of Exercise and Eating Behavior

The aims of this study were to use latent class analysis (LCA) to identify subgroups of adults in Taiwan based on their reasons for seeking health information and to explore predictors of subgroup membership. A questionnaire survey of 752 adults from 25 communities in Taiwan was conducted. LCA was used to identify distinct classes of participants; latent class regression was performed to identify factors predicting latent class membership. Three classes emerged through LCA. The Health-Improving Group (50.40%) reported high probabilities of reasons relevant to improving their or someone else’s health but low probabilities of reasons relevant to patient–provider interaction. The Active Group (32.98%) showed high probabilities of almost all of the reasons for seeking health information. The Passive Group (16.62%) showed low probabilities across all of the reasons. Compared to the Health-Improving Group, the Active Group was significantly more likely to have higher education and perceive higher information-seeking self-efficacy. The individuals in the Passive Group were significantly more likely to be male, be younger, have lower health literacy, and have fewer years of education than those in the Health-Improving Group. This LCA approach can provide important information on how communication strategies should be applied to different population subgroups.

Early onset drinking is associated with later heavy drinking and related consequences. Early drinking onset and binge drinking are also independently associated with blackouts, which are periods of amnesia for events during a drinking episode. The objective of this study was to examine how early onset drinking relates to changes in the frequency of experiencing blackouts across 3years controlling for year-specific binge drinking. Participants (N=1,145; 67.9% female) from a 6-year, longitudinal study are included in these analyses. Measures of self-reported age at drinking onset included ages at first drink, first high, and first drunk, which were used to create a latent early onset drinking factor. Frequency of binge drinking and blackouts were assessed annually during Years 4 to 6. Overall, 69.2% of participants reported experiencing blackouts. After controlling for year-specific binge drinking, a growth curve model indicated that early onset drinkers reported more frequent blackouts at Year 4. There were, however, no significant effects of acceleration or deceleration in the frequency of blackouts across the 3years. Early onset drinkers continued to experience more frequent blackouts compared with those who initiated alcohol use later, despite decreases in binge drinking over time. Early onset drinkers reported more frequent blackouts across all 3years, indicating that early alcohol initiation predisposes those individuals to continue to experience more frequent blackouts, despite a decrease in their binge drinking. This may be due to various factors, such as altered hippocampal development and functioning resulting from early alcohol exposure.

Implicit and explicit drinking identity predict latent classes that differ on the basis of college students’ drinking behaviors

This study applied a three-step latent class analysis (LCA) approach to explore latent classes of learning strategy use and their moderation effects on the relationships between motivation and mathematics performance. The data of 15-year-old students from five East Asian educational systems related to Chinese culture in the Programme of International Student Assessment (PISA) in 2012 were analyzed. The findings indicated that Shanghai, Singapore, Taiwan, and Macau showed three latent classes of learning strategies, whereas Hong Kong had two latent classes. Most students in the five educational systems reported to use the control strategy, some students reported the use of combined learning strategies, and few students reported the use of memorization except for students in Shanghai. Furthermore, we found the moderation effects of learning strategy use on mathematics performance depended on the types of motivation and educational systems. This study provides insights into the advantages of a three-step LCA approach in educational research.

Despite the high drinking rates and the complexity of drinking behaviors in adolescents, insufficient attention has been paid to their drinking patterns. Therefore, we aimed to identify patterns of adolescent drinking behaviors and factors predicting the distinct subgroups of adolescent drinking behaviors. We analyzed nationally representative secondary data obtained in 2017. Our final sample included 24,417 Korean adolescents who had consumed at least one glass of alcohol in their lifetime. To investigate patterns of drinking behaviors, we conducted a latent class analysis using nine alcohol-related characteristics, including alcohol consumption levels, solitary drinking, timing of drinking initiation, and negative consequences of drinking. Furthermore, we investigated differences in demographics, mental health status, and characteristics of substance use across the latent classes identified in our study. To do so, we used the PROC LCA with COVARIATES statement in the SAS software. We identified three latent classes of drinking behaviors: current non-drinkers (CND), binge drinkers (BD), and problem drinkers (PD). Compared to the CND class, both BD and PD classes were strongly associated with higher academic year, lower academic performance, higher levels of stress, suicidal ideation, lifetime conventional or electronic cigarette use, and lifetime use of other drugs. Health professionals should develop and implement intervention strategies targeting individual subgroups of drinking behaviors to obtain better outcomes. In particular, health professionals should consider different characteristics across subgroups of adolescent drinking behaviors when developing the interventions, such as poor mental health status and other substance use among binge and problem drinkers.

Several studies have used latent class analyses to explore obesogenic behaviors and substance use in adolescents independently. We explored a variety of health risks jointly to identify distinct patterns of risk behaviors among adolescents. Latent class models were estimated using Youth Risk Behavior Surveillance System (YRBSS) 2011 data. Behaviors in the models included substance use, physical fights, physical activity (PA), sedentary activity, sleep, fruit and vegetables (F&V) consumption, and weight-related variables. Models were estimated separately by sex. Constrained and unconstrained models were computed for race. Age and perceived and calculated weight status were compared among emergent latent classes using latent multinomial logistic regressions. Four and 5 latent class solutions emerged for boys and girls, respectively. Healthy, sedentary, and physically active (PA) groups were common in both boys and girls. The sample of boys also included a health risk behavior (HRB)/PA group and the sample of girls included HRB and HRB/F&V groups. The results confirm that obesogenic and HRBs should be conceptualized as combined patterns of risk. Health education programming in schools should adopt a multiple health behavior approach to target the continuum of risks adolescents engage in simultaneously.

This study investigated the relationship between age of first drink (AFD) and a broad range of negative alcohol-related outcomes among college students exhibiting unhealthy alcohol use. We conducted an anonymous on-line survey to collect self-report data from first-year college students at a large northeastern university. Among 1,792 respondents who reported ever drinking, 14% reported an AFD before age 14. These early onset drinkers were more likely than later onset drinkers to report frequent drinking, heavy drinking, and other unhealthy alcohol use behaviors. Among the subset of drinkers with unhealthy alcohol use (36%), early drinkers were more likely than later onset drinkers to report experiencing five out of 13 alcohol-related consequences, including driving while intoxicated, missing work or school due to drinking, getting into trouble at work or school due to drinking, receiving lower grades than they should have due to drinking, and developing a tolerance to alcohol.

A method for incorporating unobserved heterogeneity into aggregate count data frameworks is presented and used to control for endogenous spatial sorting in zonal recreation models. The method is based on latent class analysis, which has become a popular tool for analyzing heterogeneous preferences with individual data but has not yet been applied to aggregate count data. The method is tested using data on backcountry hikers for a southern California study site and performs well for relatively small numbers of classes. The latent class model produces substantially smaller welfare estimates compared to a constrained version that assumes homogeneity throughout the population. (JEL Q51)

Background: Over the last two decades, latent class analysis (LCA) has been extensively used to statistically identify phenotypes of childhood wheeze; however, less has been discussed on the external factors affecting the outcomes of LCA models in phenotype identification research. Objective: To understand the potential cause of variability in the number/types of wheeze phenotypes identified by means of LCA and to assess the performance of these models under different conditions. Methods: Here, we examined the effects of several data characteristics (sample size, data collection age and intervals), and model dimensionality (number of latent classes) on the classification of wheeze phenotypes by means of latent class and transition analysis. We also identified critical and redundant data collection points in discriminating phenotypes of wheeze using two different variable selection methods, stepwise-backward elimination, and genetic algorithm. Results: Small sample sizes suggested fewer wheeze sub-groups as they were underpowered to detect less frequently observed wheeze phenotypes. There was a strong interplay between sample size, number of data collection points and the optimal number of wheeze phenotypes extracted. A variable selection algorithm applied to LCA resulted in the exclusion of 6 redundant/non-informative time points (Month 6, Month 30, Month 69, Month 91, Month 140 and Month 198) as they contained very little discriminative and/or extra information about the wheeze classes. Conclusions: Input data characteristics and model complexity were found to be very influential on the classification of wheeze phenotypes and prevalence of each class in the population.

Background: Risk taking behaviors have several negative consequences. This study aimed to identify the subgroups of students based on risk-taking behaviors and to assess the role of demographic characteristics, depression, anxiety, socioeconomic status (SES), physical inactivity and screen time on membership of specific subgroup.Study design: Cross-sectional study.Methods: This nationwide survey was conducted in 2011-2012 among 14880 students, aged 6-18 yr, selected by multistage, cluster-sampling method from 30 provinces of Iran. The students completed two sets of anonymous and validated questionnaires, obtained from the World Health Organization-Global School Health Survey questionnaires. Latent class analysis was performed to achieve the study objectives.Results: Overall, 13486 children and adolescents participated were enrolled (response rate 90.6%). They consisted of 50.8% boys, with a mean age of 12.47 ±3.36 year. The prevalence of physical fight, bullying, victimization, active smoking, passive hookah and passive cigarette smoking was 39.7%, 17.4%, 27.2%, 5.9%, 21.1 and 33.8%, respectively. Five latent classes were identified: (a) low risk (46.7%), (b) passive smoker (25.2%), (c violence and aggression taker with passive smoking (13.5%), (d) violence and aggression taker without passive smoking (10.8%) and (e) high risk (3.8%). Higher age (OR=1.41), being male (OR=5.21), depression (OR=4.58), anxiety (OR=3.38) and screen time (OR=3.11) were associated with high-risk class.Conclusion: The prevalence of some risk-taking behaviors among Iranian students is high. Our findings emphasize the importance of planning and evaluating preventive interventions by considering different high-risk behaviors simultaneously

Classifying substance use disorder treatment facilities with co-located mental health services: A latent class analysis approach

Dealing with multiple local modalities in latent class profile analysis

PurposeResearch on symptom clusters in oncology is progressing, but knowledge gaps remain. One question is whether the number and types of symptom subgroups (i.e., latent classes) differ based on cancer diagnosis. The purpose of this study was to: (1) identify and compare latent class subgroups based on four highly prevalent symptoms (pain, fatigue, sleep disturbance, and depression), and (2) examine the differences in sociodemographic and clinical factors in the identified latent classes across the seven cancer types (i.e., prostate, non-small cell lung, non-Hodgkin’s lymphoma, breast, uterine, cervical, and colorectal cancer).MethodsThis study is a cross-sectional secondary analysis of data obtained from the My-Health study in partnership with four Surveillance, Epidemiology, and End Results (SEER) cancer registries located in California (two), Louisiana, and New Jersey. The sample included 4,762 cancer survivors 6-13 months following diagnosis of one of the seven cancer types mentioned. Latent class profile analysis was used.ResultsSubjects were primarily young (59% age 21-64 years), Caucasian (41%), married/cohabitating (58%) and unemployed (55%). The number and types of symptom subgroups varied across these seven cancer populations: four-subgroups were the common in prostate, lung, non-Hodgkin’s lymphoma, and breast cancer survivors. Unmarried, low education, and unemployment status were associated with high risk of symptom burden across the cancer types.ConclusionIdentifying symptom subgroups by cancer diagnosis has the potential to develop innovative and effective targeted interventions in cancer survivors. Further research is needed to establish extensive knowledge in symptom clustering between treatment regimens, and short-term and long-term cancer survivors.

BackgroundHeterogeneity in patients with low back pain (LBP) is well recognised and different approaches to subgrouping have been proposed. Latent Class Analysis (LCA) is a statistical technique that is increasingly being used to identify subgroups based on patient characteristics. However, as LBP is a complex multi-domain condition, the optimal approach when using LCA is unknown. Therefore, this paper describes the exploration of two approaches to LCA that may help improve the identification of clinically relevant and interpretable LBP subgroups.MethodsFrom 928 LBP patients consulting a chiropractor, baseline data were used as input to the statistical subgrouping. In a single-stage LCA, all variables were modelled simultaneously to identify patient subgroups. In a two-stage LCA, we used the latent class membership from our previously published LCA within each of six domains of health (activity, contextual factors, pain, participation, physical impairment and psychology) (first stage) as the variables entered into the second stage of the two-stage LCA to identify patient subgroups. The description of the results of the single-stage and two-stage LCA was based on a combination of statistical performance measures, qualitative evaluation of clinical interpretability (face validity) and a subgroup membership comparison.ResultsFor the single-stage LCA, a model solution with seven patient subgroups was preferred, and for the two-stage LCA, a nine patient subgroup model. Both approaches identified similar, but not identical, patient subgroups characterised by (i) mild intermittent LBP, (ii) recent severe LBP and activity limitations, (iii) very recent severe LBP with both activity and participation limitations, (iv) work-related LBP, (v) LBP and several negative consequences and (vi) LBP with nerve root involvement.ConclusionsBoth approaches identified clinically interpretable patient subgroups. The potential importance of these subgroups needs to be investigated by exploring whether they can be identified in other cohorts and by examining their possible association with patient outcomes. This may inform the selection of a preferred LCA approach.