scholarly journals Specifying Location–Scale Models for Heterogeneous Variances as Multilevel SEMs

2020 ◽  
Author(s):  
Daniel McNeish
Keyword(s):  
Random Effects ◽  
Repeated Measures ◽  
Multilevel Models ◽  
Scale Model ◽  
General Version ◽  
Cross Sectional ◽  
Mixed Effect ◽  
Recent Developments ◽  
Multilevel Sem ◽  
The Mean

Standard multilevel models focus on variables that predict the mean while the within-group variability is largely treated as a nuisance. Recent work has shown the advantage of including predictors for both the mean (the location submodel) and the variability (the scale submodel) within a single model. Constrained versions of the model can be fit in standard mixed effect model software, but the most general version with random effects in each of the location and scale submodels has been noted for being difficult to fit and estimate in software. However, the latest release of Mplus includes new capabilities that facilitate fitting the general version of the model as a multilevel SEM. This paper introduces the general form of the model that includes location and scale random effects (called the location-scale model) and notes how it can be envisioned as a multilevel SEM. We provide a tutorial with example analyses and Mplus code for the model with two-level cross-sectional data and three-level repeated measures data and discuss how such a model has potential to extend recent developments in organizational science.

scholarly journals Specifying Location-Scale Models for Heterogeneous Variances as Multilevel SEMs

2020 ◽  
pp. 109442812091308 ◽  
Author(s):  
Daniel McNeish
Keyword(s):  
Random Effects ◽  
Repeated Measures ◽  
Structural Equation ◽  
Multilevel Models ◽  
Equation Model ◽  
Scale Model ◽  
General Version ◽  
Cross Sectional ◽  
Mixed Effect ◽  
The Mean

Standard multilevel models focus on variables that predict the mean while the within-group variability is largely treated as a nuisance. Recent work has shown the advantage of including predictors for both the mean (the location submodel) and the variability (the scale submodel) within a single model. Constrained versions of the model can be fit in standard mixed effect model software, but the most general version with random effects in each of the location and scale submodels has been noted for being difficult to fit and estimate in software. However, the latest release of Mplus includes new capabilities that facilitate fitting the general version of the model as a multilevel structural equation model (SEM). This article introduces the general form of the model that includes location and scale random effects (called the location-scale model) and notes how it can be envisioned as a multilevel SEM. We provide a tutorial with example analyses and Mplus code for the model with two-level cross-sectional data and three-level repeated measures data and discuss how such a model has potential to extend recent developments in organizational science.

scholarly journals Multilevel growth curve models that incorporate a random coefficient model for the level 1 variance function

2017 ◽  
Vol 27 (11) ◽  
pp. 3478-3491 ◽  
Author(s):  
Harvey Goldstein ◽  
George Leckie ◽  
Christopher Charlton ◽  
Kate Tilling ◽  
William J Browne
Keyword(s):  
Random Effects ◽  
Repeated Measures ◽  
Random Effect ◽  
Growth Function ◽  
Individual Variability ◽  
Growth Patterns ◽  
Variance Function ◽  
Longitudinal Growth ◽  
Growth Data ◽  
The Mean

Aim To present a flexible model for repeated measures longitudinal growth data within individuals that allows trends over time to incorporate individual-specific random effects. These may reflect the timing of growth events and characterise within-individual variability which can be modelled as a function of age. Subjects and methods A Bayesian model is developed that includes random effects for the mean growth function, an individual age-alignment random effect and random effects for the within-individual variance function. This model is applied to data on boys’ heights from the Edinburgh longitudinal growth study and to repeated weight measurements of a sample of pregnant women in the Avon Longitudinal Study of Parents and Children cohort. Results The mean age at which the growth curves for individual boys are aligned is 11.4 years, corresponding to the mean ‘take off’ age for pubertal growth. The within-individual variance (standard deviation) is found to decrease from 0.24 cm2 (0.50 cm) at 9 years for the ‘average’ boy to 0.07 cm2 (0.25 cm) at 16 years. Change in weight during pregnancy can be characterised by regression splines with random effects that include a large woman-specific random effect for the within-individual variation, which is also correlated with overall weight and weight gain. Conclusions The proposed model provides a useful extension to existing approaches, allowing considerable flexibility in describing within- and between-individual differences in growth patterns.

scholarly journals ehealth literacy and health literacy among immigrants and their descendants compared with women of Danish origin: a cross-sectional study using a multidimensional approach among pregnant women

BMJ Open ◽  
2020 ◽  
Vol 10 (5) ◽  
pp. e037076
Author(s):  
Sarah Fredsted Villadsen ◽  
Hajer Hadi ◽  
Israa Ismail ◽  
Richard H Osborne ◽  
Claus Thorn Ekstrøm ◽  
...  
Keyword(s):  
Health Literacy ◽  
Health System ◽  
Pregnant Women ◽  
Immigrant Women ◽  
Healthcare Providers ◽  
Cross Sectional ◽  
Mixed Effect ◽  
Ehealth Literacy ◽  
Digital Services ◽  
The Mean

ObjectiveTo explore ehealth literacy, ability to actively engage with healthcare providers and health system navigation among pregnant immigrant women and their descendants compared with women of Danish origin.Design and settingA cross-sectional survey at antenatal clinics in 2016, Denmark.ParticipantsPregnant women attending antenatal care (n=405).Outcome measuresThe eHealth Literacy Questionnaire (eHLQ) and two domains from the Health Literacy Questionnaire (HLQ): ability to actively engage with healthcare providers and health system navigation. Range of response options for eHLQ (1–4) and HLQ (1–5). With mixed-effect linear regressions, eHLQ and HLQ among immigrants and their descendants compared with women of Danish origin were assessed.ResultsThe response rate was 75%. The overall trend was lower ehealth literacy and HLQ domains among immigrants and their descendants compared with women of Danish origin. For ehealth literacy, the results suggest that challenges related more to digital abilities than motivation, trust and access to technology. The mean ability to engage with digital services was 3.20 (SD 0.44) for women of Danish origin. Non-Western descendants (−0.14, 95% CI −0.31 to 0.02), non-Western (−0.20, 95% CI −0.34 to −0.06) and Western (−0.22, 95% CI −0.39 to −0.06) immigrants had lower adjusted means of this outcome. No differences in motivation to engage with digital services were found for descendants (−0.00, 95% CI −0.17 to 0.17), non-Western (0.03, 95% CI −0.11 to 0.18) or Western (−0.06, 95% CI −0.23 to 0.10) immigrants compared with the mean of the reference (2.85, SD 0.45). Lower ability to engage with healthcare providers was found for non-Western born immigrants (−0.15, CI 95% −0.30 to −0.01) compared with the mean of women with Danish origin (4.15, SD 0.47).ConclusionGenerally, descendant and immigrant women had lower levels of ehealth literacy and health literacy than women of Danish origin. These differences are potentially antecedents of adverse birth outcomes and could inform structural efforts to mitigate health inequalities.

scholarly journals Multilevel modeling for longitudinal data: concepts and applications

2019 ◽  
Vol 54 (4) ◽  
pp. 459-489 ◽  
Author(s):  
Joseph F. Hair Jr. ◽  
Luiz Paulo Fávero
Keyword(s):  
Longitudinal Data ◽  
Multilevel Modeling ◽  
Random Effects ◽  
Repeated Measures ◽  
Design Methodology ◽  
Multilevel Models ◽  
Covariance Structures ◽  
Content Type ◽  
Intraclass Correlations ◽  
Fixed And Random Effects

Purpose This paper aims to discuss multilevel modeling for longitudinal data, clarifying the circumstances in which they can be used. Design/methodology/approach The authors estimate three-level models with repeated measures, offering conditions for their correct interpretation. Findings From the concepts and techniques presented, the authors can propose models, in which it is possible to identify the fixed and random effects on the dependent variable, understand the variance decomposition of multilevel random effects, test alternative covariance structures to account for heteroskedasticity and calculate and interpret the intraclass correlations of each analysis level. Originality/value Understanding how nested data structures and data with repeated measures work enables researchers and managers to define several types of constructs from which multilevel models can be used.

Effects of chronic intermittent hypobaric hypoxia on prostate-specific antigen (PSA) in Chilean miners

2021 ◽  
pp. oemed-2020-107232
Author(s):  
Diana Elizabeth Alcantara-Zapata ◽  
Shrikant I Bangdiwala ◽  
Daniel Jiménez ◽  
Manolis Kogevinas ◽  
Nella Marchetti ◽  
...  
Keyword(s):  
Prostate Specific Antigen ◽  
Hypobaric Hypoxia ◽  
Multilevel Models ◽  
Specific Antigen ◽  
Mining Operation ◽  
Cross Sectional Study ◽  
Cross Sectional ◽  
Mixed Effect ◽  
Intermittent Hypobaric Hypoxia ◽  
Different Altitudes

ObjectiveThe aim was to determine the effects of chronic intermittent hypobaric hypoxia (CIHH) on prostate-specific antigen (PSA) levels in Chilean miners who work at different altitudes.MethodsA cross-sectional study was conducted between April and July 2019. Miners from five mines (N=338) at different altitudes were evaluated. We recorded sociodemographic, working and altitude information. Haemoglobin oxygen saturation (SaO2) and haemoglobin (Hb) were measured in situ, while PSA and testosterone were analysed at a low level. Linear mixed-effect models were used to evaluate the association between PSA level and two CIHH exposures: composite CIHH (with four descriptors) and ChileStd-CIHH (CIHH Chilean standard; based on the Chilean technical guide for occupational exposure to CIHH). All models were adjusted by age, body mass index and day of the work the samples were taken.ResultsHighest and lowest PSA levels were found in mines ≥3000 m above sea level (mine 3: median=0.75, IQR=−0.45; mine 4: median=0.46, IQR=−0.35). In the multilevel models, the wider altitude difference between mining operation and camp showed lower PSA levels (model D: βPSA=−0.93 ng/mL, βlogPSA=−0.07, p<0001), adjusted for other CIHH descriptors, SaO2, Hb and testosterone. The descriptors of composite CIHH explained better PSA variations than ChileStd-CIHH (model D: marginal R2=0.090 vs model A: marginal R2=0.016).ConclusionsOccupational health regulations and high altitude medicine should consider these results as initial evidence on the inclusion of new descriptors for CIHH and the possible effect of this exposure on PSA levels in this male-dominated occupational sector.

scholarly journals The role of interleukin-6 in the early diagnosis of sepsis in premature infants

2017 ◽  
Vol 9 (3) ◽  
Author(s):  
Mehrdad Mirzarahimi ◽  
Manouchehr Barak ◽  
Abdolkarim Eslami ◽  
Afsaneh Enteshari-Moghaddam
Keyword(s):  
Early Diagnosis ◽  
Interleukin 6 ◽  
Repeated Measures ◽  
Complete Blood Count ◽  
Cross Sectional Study ◽  
Cross Sectional ◽  
Normal Value ◽  
Mortality And Morbidity ◽  
The Mean

Septicemia is one of the major causes of mortality and morbidity in newborns. The aim of this study was to investigate the role of interleukin-6 (IL-6) in the early diagnosis of Sepsis in premature newborns. This cross-sectional study was conducted on preterm infants admitted to NICU of Hospital Buali in Ardabil city, Iran. C-reactive protein (CRP), IL-6 and complete blood count tests have been done in baseline, third and seventh day. Collected data analyzed by one-sample t-test, repeated measures and ANOVA in SPSS.21. The mean of IL-6 in the first and third day after hospitalization was significantly more than normal value. The mean of CRP in the third and seventh day after hospitalization significantly more than normal value. We found that increasing in IL-6 level can occur earlier than CRP and it can be used as a good index in early sepsis diagnosis compare than CRP.

Fitting Growth Model Using Nonlinear Regression with Random Parameters

2011 ◽  
Vol 480-481 ◽  
pp. 1308-1312
Author(s):  
Yao Xiang Li ◽  
Li Chun Jiang
Keyword(s):  
Random Effects ◽  
Repeated Measures ◽  
Fixed Effects ◽  
Model Building ◽  
Mixed Effects ◽  
Mixed Effects Models ◽  
Ratio Test ◽  
Nonlinear Mixed Effects ◽  
Nonlinear Mixed Effects Models ◽  
Mixed Effect

Mixed Effect models are flexible models to analyze grouped data including longitudinal data, repeated measures data, and multivariate multilevel data. One of the most common applications is nonlinear growth data. The Chapman-Richards model was fitted using nonlinear mixed-effects modeling approach. Nonlinear mixed-effects models involve both fixed effects and random effects. The process of model building for nonlinear mixed-effects models is to determine which parameters should be random effects and which should be purely fixed effects, as well as procedures for determining random effects variance-covariance matrices (e.g. diagonal matrices) to reduce the number of the parameters in the model. Information criterion statistics (AIC, BIC and Likelihood ratio test) are used for comparing different structures of the random effects components. These methods are illustrated using the nonlinear mixed-effects methods in S-Plus software.

Use, costs, and predictors of psychiatric healthcare services following an autism spectrum diagnosis: Population-based cohort study

Autism ◽  
2019 ◽  
Vol 23 (8) ◽  
pp. 2020-2030
Author(s):  
Caroline Croteau ◽  
Laurent Mottron ◽  
Marc Dorais ◽  
Jean-Eric Tarride ◽  
Sylvie Perreault
Keyword(s):  
Drug Use ◽  
Repeated Measures ◽  
Healthcare Services ◽  
Autism Spectrum ◽  
Psychoactive Drug ◽  
Psychiatric Services ◽  
Administrative Database ◽  
Cross Sectional ◽  
Mixed Effect

A number of cross-sectional studies report extensive use of psychiatric services and high healthcare costs in autistic youths. However, little is known about how the use of these services evolves from the time of diagnosis, as children grow up. Our objectives were to investigate the use, costs, and predictors of psychiatric services following autism spectrum diagnosis. We built a cohort of 1227 newly diagnosed autism spectrum individuals identified in the Quebec (Canada) Régie de l’assurance maladie du Québec administrative database (January 1998 to December 2010). Mean number and cost per individual of psychiatric healthcare use (hospitalizations, medical visits, psychoactive drug use) were calculated yearly for 5 years following autism spectrum diagnosis. Mean number of psychiatric visits decreased over time by more than threefold (7.5 vs 2.1 visits) from year 1 to year 5, whereas psychoactive drug use increased from 16.0 to 25.2 claims. Psychiatric hospitalizations decreased during follow-up, but still represented the greatest costs per individual (CAD9820 for year 1; CAD4628 for year 5). Antipsychotics represented over 50% of drug costs. Mixed-effect model with repeated measures showed that previous psychoactive drug use was the strongest predictor of greater psychiatric healthcare cost during follow-up (odds ratio: 9.96; 95% confidence interval: 7.58–13.10). These trends contrast with guidelines advocating cautious prescribing of antipsychotics with periodical re-assessment of their benefit.

The Role of School Social Environment on Dental Caries Experience in 8- to 12-Year-Old Brazilian Children: A Multilevel Analysis

2015 ◽  
Vol 49 (5) ◽  
pp. 548-556 ◽  
Author(s):  
María Raquel Fernández ◽  
Marília L. Goettems ◽  
Thiago M. Ardenghi ◽  
Flávio F. Demarco ◽  
Marcos Britto Correa
Keyword(s):  
Public Schools ◽  
Dental Caries ◽  
Social Environment ◽  
School Environment ◽  
Dental Plaque ◽  
Multilevel Models ◽  
Caries Experience ◽  
Caries Prevalence ◽  
Cross Sectional ◽  
The Mean

Aim: Although children spend most of their time involved in activities related to school, few studies have focused on the association between school social environment and oral health. This cross-sectional study assessed individual and school-related social environment correlates of dental caries in Brazilian schoolchildren aged 8-12 years. Methods: A sample of children from 20 private and public schools (n = 1,211) was selected. Socio-economic data were collected from parents, and data regarding children characteristics were collected from children using a questionnaire. Dental examinations were performed to assess the presence of dental plaque: dental caries experience (DMFT ≥1) and dental caries severity (mean dmf-t/DMF-T). The social school environment was assessed by a questionnaire administered to school coordinators. Multilevel Poisson regression was used to investigate the association between school social environment and dental caries prevalence and experience. Results: The dental caries prevalence was 32.4% (95% confidence interval: 29.7-35.2) and the mean dmf-t/DMF-T was 1.84 (standard deviation: 2.2). Multilevel models showed that the mean dmf-t/DMF-T and DMFT ≥1 were associated with lower maternal schooling and higher levels of dental plaque. For contextual variables, schools offering after-hours sports activities were associated with a lower prevalence of dental caries and a lower mean of dmf-t/DMF-T, while the occurrence of violence and theft episodes was positively associated with dental caries. Conclusions: The school social environment has an influence on dental caries in children. The results suggest that strategies focused on the promotion of healthier environments should be stimulated to reduce inequalities in dental caries.

Serum ionic dysequilibria in clinical opioid dependence: Cross-sectional and longitudinal studies

2017 ◽  
Vol 36 (8) ◽  
pp. 776-784
Author(s):  
AS Reece ◽  
A Norman ◽  
GK Hulse
Keyword(s):  
Linear Regression ◽  
Repeated Measures ◽  
Opioid Dependence ◽  
Clinical Pathology ◽  
Antibody Test ◽  
Cross Sectional ◽  
Electrophysiological Properties ◽  
Medical Control ◽  
Excitable Membranes ◽  
The Mean

Introduction: Despite an increasing awareness that the activity of excitable membranes is determined by the underlying ionic gradients across them, and their importance in drug dependency, we were not able to identify any reports of comparing the electrolyte composition of opioid-dependent and non-addicted controls. Methods: Linear regression was used to compare clinical pathology blood results taken from 2699 opioid-dependent patients (ODP) and 5307 medical control (MC) patients on a total of 21,734 occasions for the period 1995–2015. The presence of a hepatitis C antibody test was used to separate OPD and MC patients. Results: The mean age among ODP and MC was 33.51 ± 0.16 and 37.99 ± 0.23 years, respectively ( p < 0.0001). The groups were 71.5% and 54.2% male ( p < 0.0001). Drug use in this cohort has been reported previously. Analysis of sodium, haemoglobin and albumin were used to exclude marked effects of haemodilution/haemoconcentration. Repeated measures linear regression against age and time showed depressed levels of bicarbonate ( p < 0.0001) and potassium ( p < 0.05) and elevated levels of chloride ( p < 0.025) and anions ( p < 0.01) in ODP in both sexes. Multiple regression in mixed-effects models showed that these effects were all worse in females ( p = 0.0001). Conclusion: This data shows that opioid dependence is associated with significant changes in chloride, potassium, bicarbonate and anions in both sexes, and worse in females. This likely has implications for the electrophysiological properties of excitable membranes. It is consistent with the reported impairment of potassium-chloride exchangers in opioid dependence. Explication of the mechanisms responsible must await further studies.

Sign in / Sign up

Export Citation Format

Share Document