An overview of mixture modelling for latent evolutions in longitudinal data: Modelling approaches, fit statistics and software

Nest, Gavin van der; Passos, Valéria Lima; Candel, Math J. J. M.; Breukelen, Gerard J. P. van

doi:10.1016/j.alcr.2019.100323

Cited by 162 publications

(157 citation statements)

References 89 publications

Supporting

Mentioning

157

Contrasting

Order By: Relevance

“…Development of psychological distress followed four distinct trajectories, as revealed by the latent class mixture models. The four-trajectory solution yielded the best model fit to the data according to the log-likelihood based statistics (Van der Nest et al ., 2020), i.e. the lowest value for the Bayes Information Criterion, as seen in Table 2.…”

Section: Resultsmentioning

confidence: 99%

“…Because we had no a priori expectations of specific distributions, we opted for an explorative approach to determine number and prevalence of trajectories, using the six repeated measurements of psychological distress. Mixture models for the clustering of longitudinal data series identify latent subpopulations that share similar trajectories (Van der Nest et al ., 2020). These trajectories, which are highly comparable within subpopulations, are deemed mutually exclusive between subpopulations.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Heterogeneous Mental Health Development During the COVID-19 Pandemic in the United Kingdom

Ellwardt

Praeg

2021

Preprint

View full text Add to dashboard Cite

Aim. The COVID-19 pandemic and the mitigation measures by governments have upended the economic and social lives of many, leading to widespread psychological distress. However, how distress developed during the pandemic and who was most affected is poorly understood. We explore heterogeneity in trajectories of psychological distress during the first six months of the pandemic in the United Kingdom and relate this heterogeneity to socio-demographic and health factors. Subjects and Methods. We analyze six waves of longitudinal, nationally representative survey data from the UK Household Longitudinal Study (N = 15,218), covering the first lockdown in 2020. First, latent class mixture modelling (LCCM) is used to identify trajectories of psychological distress. Second, associations of the trajectories with covariates are tested with multinomial logistic regressions. Results. We find four different trajectories of distress: continuously low, continuously moderate, temporarily elevated, and continuously elevated distress. One-fifth of the population experienced severely elevated risks of distress. Long-term exposure was highest among younger people, women, those who lost income, and those with previous health conditions or COVID-19 symptoms. Conclusion. Given the threat of persistent stress on health, policy measures should be sensitized to the unintended yet far-reaching consequences of non-pharmaceutical interventions.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Heterogeneous Mental Health Development During the COVID-19 Pandemic in the United Kingdom

Ellwardt

Praeg

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…A smaller number of aBIC indicated a better-fitting model [ 43 ], while a larger value of entropy represented a smaller likelihood of misclassification [ 43 , 44 ]. The adjusted Lo–Mendell–Rubin likelihood ratio test (aLMR-test) and bootstrapped likelihood ratio test (BLRT) were used to compare the n-class model versus the n−1 class model [ 45 , 46 ]. The significant p -value ( p < 0.05) suggested that the n-class model was well improved over the n-1-class model.…”

Section: Methodsmentioning

confidence: 99%

The Association between Blood Pressure Trajectories and Risk of Cardiovascular Diseases among Non-Hypertensive Chinese Population: A Population-Based Cohort Study

Lin

et al. 2021

IJERPH

View full text Add to dashboard Cite

Although previous studies have discussed the association between trajectories of blood pressure (BP) and risk of cardiovascular diseases (CVDs), the association among the non-hypertensive general population of youth and middle age has not been elucidated. We used the growth mixture model to explore the trajectories of BP among the non-hypertensive Chinese population and applied Cox regression to evaluate the association between trajectories of BP and the risk of stroke or myocardial infarction (MI). Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were categorized into three classes, respectively. Statistically significant associations were observed between SBP trajectories and stroke (range of adjusted hazard ratios (aHR): 1.369–3.837) or MI (rang of aHR = 6.047–13.017). Association between DBP trajectories and stroke (aHR: 3.685) or MI (range of aHR = 1.312–2.821) were also observed, although they did not reach statistical significance. Trajectories of SBP were more important risk factors than that of DBP in stroke and MI in our study population. BP management is important among pre-hypertensive adults to prevent stroke and MI when they age. Well-designed research with a larger sample size is required to confirm our findings and develop efficient methods to prevent CVDs.

show abstract

“…Statistical methods like latent class growth analysis (LCGA) can help to provide a more accurate picture of the heterogeneous course in psychosocial functioning that can be observed following FEP, as it allows considering different outcomes of the same characteristic simultaneously [ 12 , 13 ]. To our knowledge, only few studies so far have applied these statistical techniques to assess functional outcomes in FEP samples [ 14 , 15 , 16 ], and none of them has considered simultaneously sociodemographic variables, clinical features and an extensive set of cognitive domains, all of them previously related to poor functional outcomes [ 17 ].…”

Section: Introductionmentioning

confidence: 99%

Exploring Risk and Resilient Profiles for Functional Impairment and Baseline Predictors in a 2-Year Follow-Up First-Episode Psychosis Cohort Using Latent Class Growth Analysis

Salagre

Grande

Solé

et al. 2020

JCM

View full text Add to dashboard Cite

Being able to predict functional outcomes after First-Episode Psychosis (FEP) is a major goal in psychiatry. Thus, we aimed to identify trajectories of psychosocial functioning in a FEP cohort followed-up for 2 years in order to find premorbid/baseline predictors for each trajectory. Additionally, we explored diagnosis distribution within the different trajectories. A total of 261 adults with FEP were included. Latent class growth analysis identified four distinct trajectories: Mild impairment-Improving trajectory (Mi-I) (38.31% of the sample), Moderate impairment-Stable trajectory (Mo-S) (18.39%), Severe impairment-Improving trajectory (Se-I) (12.26%), and Severe impairment-Stable trajectory (Se-S) (31.03%). Participants in the Mi-I trajectory were more likely to have higher parental socioeconomic status, less severe baseline depressive and negative symptoms, and better premorbid adjustment than individuals in the Se-S trajectory. Participants in the Se-I trajectory were more likely to have better baseline verbal learning and memory and better premorbid adjustment than those in the Se-S trajectory. Lower baseline positive symptoms predicted a Mo-S trajectory vs. Se-S trajectory. Diagnoses of Bipolar disorder and Other psychoses were more prevalent among individuals falling into Mi-I trajectory. Our findings suggest four distinct trajectories of psychosocial functioning after FEP. We also identified social, clinical, and cognitive factors associated with more resilient trajectories, thus providing insights for early interventions targeting psychosocial functioning.

show abstract

An overview of mixture modelling for latent evolutions in longitudinal data: Modelling approaches, fit statistics and software

Cited by 162 publications

References 89 publications

Heterogeneous Mental Health Development During the COVID-19 Pandemic in the United Kingdom

Heterogeneous Mental Health Development During the COVID-19 Pandemic in the United Kingdom

The Association between Blood Pressure Trajectories and Risk of Cardiovascular Diseases among Non-Hypertensive Chinese Population: A Population-Based Cohort Study

Exploring Risk and Resilient Profiles for Functional Impairment and Baseline Predictors in a 2-Year Follow-Up First-Episode Psychosis Cohort Using Latent Class Growth Analysis

Contact Info

Product

Resources

About