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

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.

Download Full-text

The impact of random-effect misspecification on percentile estimation for longitudinal growth data

Statistics in Medicine ◽

10.1002/sim.5437 ◽

2012 ◽

Vol 31 (28) ◽

pp. 3708-3718 ◽

Cited By ~ 5

Author(s):

Kyeongmi Cheon ◽

Paul S. Albert ◽

Zhiwei Zhang

Keyword(s):

Random Effect ◽

Longitudinal Growth ◽

Growth Data ◽

Percentile Estimation ◽

The Impact

Download Full-text

Specifying Location–Scale Models for Heterogeneous Variances as Multilevel SEMs

10.31234/osf.io/8ex32 ◽

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.

Download Full-text

Two-year tree growth patterns investigated from monthly girth records using dendrometer bands in a wet evergreen forest in India

Journal of Tropical Ecology ◽

10.1017/s0266467400001498 ◽

2000 ◽

Vol 16 (3) ◽

pp. 429-446 ◽

Cited By ~ 24

Author(s):

Raphaël Pélissier ◽

Jean-pierre Pascal

Keyword(s):

Growth Rate ◽

Tree Growth ◽

Growth Patterns ◽

Annual Growth ◽

Evergreen Forest ◽

Annual Growth Rate ◽

Growth Data ◽

Growth Profiles ◽

Dendrometer Bands ◽

The Mean

With the aim of characterizing tree growth patterns, this paper re-examines the growth data of 100 selected trees belonging to 24 species that were recorded monthly in a 0.2-ha plot of a wet evergreen forest in the Western Ghats of India during the period 1980–82 using dendrometer bands. The mean growth profile, combining all of the selected trees, showed: (a) a significantly lower annual growth rate during the second year of survey which seemed to be negatively related to monsoon precipitation; (b) significant intra-annual growth variation clearly related to the regular alternation between a period of heavy rain and a quite long dry season of the monsoon climatic regime. Analysis of the variability of the individual smoothed growth profiles representing the 2-y trend of the growth data showed that: (a) the mean growth rate depended on a combination of an intrinsic endogenous variable (the structural class grouping species according to their maximum size), a tree size variable (tree diameter at breast height, dbh) and a neighbourhood variable (the number of taller neighbours in a 10-m radius); (b) the sudden change in growth rate from one year to the other was not predictable using these variables. The amplitude of the seasonal variations, investigated from the detrended growth profiles, appeared to be dependent on a combination of tree dbh and the number of taller neighbours in a 10-m radius. A co-inertia analysis of the smoothed and the detrended growth profiles indicated that the trees with fast growth also exhibited high seasonal variation. It is suggested that fast growing trees are those with favourable crown positions, which are consequently subject to high transpiration rates due to radiation and wind exposure.

Download Full-text

Specifying Location-Scale Models for Heterogeneous Variances as Multilevel SEMs

Organizational Research Methods ◽

10.1177/1094428120913083 ◽

2020 ◽

pp. 109442812091308 ◽

Cited By ~ 1

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.

Download Full-text

A method of presenting longitudinal growth data

British Journal Of Nutrition ◽

10.1079/bjn19790011 ◽

1979 ◽

Vol 41 (1) ◽

pp. 53-56

Author(s):

A. E. Dugdale

Keyword(s):

Early Life ◽

Growth Function ◽

Statistical Properties ◽

Statistical Analyses ◽

Longitudinal Growth ◽

Growth Data ◽

Different Ages ◽

Growth Profiles

1. Longitudinal growth profiles contain much information but are difficult to incorporate into mathematical and statistical analyses.2. A growth function, which is a weighed average of growth achievement at different ages, is proposed.3. This function is a non-dimensional number with defined statistical properties, and emphasizes growth achievement in early life. It can be used to compare the growth of individuals and populations.

Download Full-text

Efektivitas Comprehensive Breastfeeding Education terhadap Keberhasilan Pemberian Air Susu Ibu Postpartum

Jurnal Keperawatan Padjadjaran ◽

10.24198/jkp.v1i2.56 ◽

1970 ◽

Vol 1 (2) ◽

Author(s):

Irma Nurbaeti ◽

Kustati Budi Lestari

Keyword(s):

Repeated Measures ◽

Breastfeeding Support ◽

Breastfeeding Practices ◽

Breastfeeding Education ◽

Repeated Measures Analysis ◽

Post Test ◽

Quasi Experimental ◽

The Mean ◽

Convenience Sampling ◽

Postpartum Mothers

Pemberian Air Susu Ibu (ASI) masih merupakan masalah bagi pemenuhan kebutuhan nutrisi bayi baru lahir. Dukungan agar ibu menyusui bayi merupakan hal penting dalam menginisiasi dan mempertahankan pemberian ASI. Strategi dibutuhkan untuk mendukung keberhasilan menyusui. Tujuan penelitian adalah menganalisis efektivitas comprehensive breastfeeding education terhadap keberhasilan pemberian (ASI) pada periode postpartum. Jenis penelitian ini menggunakan kuasi eksperimen one group pre post test repeated measured design. Jumlah sampel sebanyak 22 ibu dengan menggunakan teknik accidental sampling. Pengumpulan data dilaksanakan pada bulan September–Oktober 2013 di Puskesmas wilayah Kota Tangerang Selatan. Intervensi dilakukan selama 30 menit. Pengumpulan data dilakukan sebelum intervensi, 3 hari setelah intervensi (post1), dan 10 hari setelah intervensi (post 2). Pengumpulan data menggunakan kuesioner dan observasi. Keberhasilan pemberian ASI berdasar pada parameter pengetahuan, langkah menyusui, perlekatan bayi, dan kecukupan ASI. Analisis data menggunakan general linear model repeated measureANOVA. Hasil penelitian menunjukkan adanya signifikansi comprehensive breastfeeding education (p=0.001). Rata-rata keberhasilan pemberian ASI sebelum dan setelah intervensi meningkat. Sebesar 93,9% intervensi memengaruhi tingkat keberhasilan. Rata-rata sebelum intervensi 56,74 (SD 5,92), post 1 sebesar 60,83 (SD 6,38) dan post2 sebesar 74,55 (SD 5,32). Subvariabel yang memiliki efek secara signifikan setelah intervensi adalah pengetahuan (p=0.001) dan langkah menyusui (p=0.001), sedangkan subvariabel perlekatan bayi (p=0.061) dan kecukupan ASI (p=0.162) tidak secara signifikan berbeda antara sebelum dan setelah intervensi. Pelaksanaanbreastfeeding education disarankan pada ibu agar dapat melakukan posisi perlekatan bayi yang benar sehingga dapat mengurangi masalah-masalah berkaitan dengan perlekatan yang tidak sesuai seperti puting perih, lecet atau berdarah, dan bayi kurang puas dalam menyusu yang bisa mengakibatkan gagalnya program ASI ekslusif.Kata kunci:Menyusui, pendidikan, perlekatan, postpartum AbstractBreastfeeding have still been problem for adequate newborn nutrition. Adequate breastfeeding support is essential for mothers to initiate and maintain optimal breastfeeding practices. A strategic needed to support successful breastfeeding. The purpose of research is to analyze the effectiveness comprehensive breastfeeding education on successful breastfeeding at postpartum periods. A quasi-experimental one group pretest, post test, repeated mesaured was used. This study was conducted at public health in Tangerang Selatan municipality in September–October 2013 among 22 postpartum mothers, convenience sampling methods. Intervention was done 30 minute. Data were collected before intervention (pretest), third day after intervention (post 1) and tenth day after intervention (repeated/post 2) using four parameter, that are knowledge, breastfeeding steps, proper lacth-on and adequate breastmilk. Using repeated measures analysis of variance there was a significant increase (p=0.001) in the overall Successful breastfeeding mean. Around 93,9% the effectiveness of intervention influence on successful. The mean before intervention is 56,74 (SD 5,92), increased at post 1:60,83 (SD 6,38) and post 2:74,55 (SD 5,32). Subvariable which has effect significantly after intervention is knowledge (p=0.001) and breastfeeding steps (p=0.001), in contrary, proper latch-on (p=0,061) and adequate breastmilk (p=0.162) have no significant effect after intervention. Suggestion to support breastfeeding education and counselling proper latch-on adequately that can decrease the problem such as painful, creaks or bloody putting.Key words: Breastfeeding, education, latch-on, postpartum

Download Full-text

Gingival Health of Porcelain Laminate Veneered Teeth: A Retrospective Assessment

Operative Dentistry ◽

10.2341/18-088-c ◽

2019 ◽

Vol 44 (5) ◽

pp. 452-458 ◽

Cited By ~ 1

Author(s):

R Arif ◽

JB Dennison ◽

D Garcia ◽

P Yaman

Keyword(s):

Repeated Measures ◽

Gingival Crevicular Fluid ◽

Statistical Significance ◽

Fluid Collection ◽

Gingival Recession ◽

Gingival Health ◽

Long Term Effect ◽

The Mean ◽

Crevicular Fluid

SUMMARY Statement of Problem: The long-term effect of the presence of porcelain laminate veneers (PLVs) on the health of the surrounding gingival issues is not available in the restorative literature. Purpose: To assess the long-term effect of PLVs on the health of the surrounding gingival tissues. A secondary aim was to correlate gingival crevicular fluid (GCF) scores with clinical parameters used for gingival health assessment in teeth treated with PLVs. Methods and Materials: Patients who received PLVs placed at the Graduate Restorative Clinic within a seven- to 14-year period were recalled for clinical evaluations. Periodontal measurements including gingival index (GI), periodontal pocket depth (PPD), gingival recession (GR), and clinical attachment level (CAL) were measured using a standard probe and indices. Gingival Crevicular Fluid (GCF) was measured with a Periotron machine (Periotron 8000, Oraflow Inc), using Periopaper (Periopaper Gingival Fluid Collection Strip, Oraflow Inc.) for fluid collection. Photographs of any observed clinical defect were taken. Data were tabulated using Excel 2010 (Microsoft Corp). Statistical analysis for all descriptive statistics was performed using SPSS 21 (SPSS Software, IBM Corp.) and Stata SE 13 (Stata Software, StataCorp). Repeated-measures analysis of variance (ANOVA) was done to test for statistical significance of the mean pocket depths between the restored and unrestored surfaces of the veneered teeth. The significance level for all tests was p<0.05. Pearson's correlation coefficient was performed for testing statistical significance between GCF and GI and between GCF and PPD. Results: The frequency distribution of the GI included 47 PLVs (43%) with normal gingiva, 16 (15%) with mild inflammation, and 46 (42%) with moderate inflammation and bleeding on probing. The average PPD on the facial surface of the maxillary and mandibular PLVs was 2.17 mm and 2.16 mm, respectively. On the lingual surface, the average PPD was 2.10 mm for maxillary and 2.22 mm for mandibular PLVs. Gingival recession was seen in 27% of the evaluated PLVs. The repeated-measures ANOVA revealed p≥0.136, showing no statistical difference in the mean pocket depths between restored facial and unrestored lingual surfaces of the veneered teeth. A moderate correlation (r=0.407) was found between GCF and GI, which was significant at p<0.001. No correlation (r=0.124) was found between GCF and PPD, which was not significant at p=0.197. Conclusions: Gingival response to the evaluated PLVs was in the satisfactory range, with overall GI scores ranging between normal and moderate inflammation, pocket depths ranging from 1 to 2 mm, and recession present in 27% of the evaluated PLVs. No statistically significant difference was found between the mean pocket depths of the restored and unrestored surfaces of veneered teeth (p≥0.136). A moderate correlation was found between GCF and GI.

Download Full-text

A random effects population dynamics model based on proportions-at-age and removal data for estimating total mortality

Canadian Journal of Fisheries and Aquatic Sciences ◽

10.1139/f2012-103 ◽

2012 ◽

Vol 69 (11) ◽

pp. 1881-1893 ◽

Cited By ~ 6

Author(s):

Verena M. Trenkel ◽

Mark V. Bravington ◽

Pascal Lorance

Keyword(s):

Population Dynamics ◽

Random Effects ◽

Fixed Effects ◽

Random Effect ◽

Total Mortality ◽

Estimation Bias ◽

Effective Sample Size ◽

Dynamics Model ◽

Population Dynamics Model ◽

Study Parameter

Catch curves are widely used to estimate total mortality for exploited marine populations. The usual population dynamics model assumes constant recruitment across years and constant total mortality. We extend this to include annual recruitment and annual total mortality. Recruitment is treated as an uncorrelated random effect, while total mortality is modelled by a random walk. Data requirements are minimal as only proportions-at-age and total catches are needed. We obtain the effective sample size for aggregated proportion-at-age data based on fitting Dirichlet-multinomial distributions to the raw sampling data. Parameter estimation is carried out by approximate likelihood. We use simulations to study parameter estimability and estimation bias of four model versions, including models treating mortality as fixed effects and misspecified models. All model versions were, in general, estimable, though for certain parameter values or replicate runs they were not. Relative estimation bias of final year total mortalities and depletion rates were lower for the proposed random effects model compared with the fixed effects version for total mortality. The model is demonstrated for the case of blue ling (Molva dypterygia) to the west of the British Isles for the period 1988 to 2011.

Download Full-text

Identifying longitudinal-growth patterns from infancy to childhood: a study comparing multiple clustering techniques

International Journal of Epidemiology ◽

10.1093/ije/dyab021 ◽

2021 ◽

Author(s):

Paraskevi Massara ◽

Charles D G Keown-Stoneman ◽

Lauren Erdman ◽

Eric O Ohuma ◽

Celine Bourdon ◽

...

Keyword(s):

Latent Class ◽

Growth Patterns ◽

Pattern Detection ◽

Distance Measures ◽

Longitudinal Growth ◽

Important Health ◽

Modelling Techniques ◽

Different Populations ◽

Growth Features ◽

Time Invariant

Abstract Background Most studies on children evaluate longitudinal growth as an important health indicator. Different methods have been used to detect growth patterns across childhood, but with no comparison between them to evaluate result consistency. We explored the variation in growth patterns as detected by different clustering and latent class modelling techniques. Moreover, we investigated how the characteristics/features (e.g. slope, tempo, velocity) of longitudinal growth influence pattern detection. Methods We studied 1134 children from The Applied Research Group for Kids cohort with longitudinal-growth measurements [height, weight, body mass index (BMI)] available from birth until 12 years of age. Growth patterns were identified by latent class mixed models (LCMM) and time-series clustering (TSC) using various algorithms and distance measures. Time-invariant features were extracted from all growth measures. A random forest classifier was used to predict the identified growth patterns for each growth measure using the extracted features. Results Overall, 72 TSC configurations were tested. For BMI, we identified three growth patterns by both TSC and LCMM. The clustering agreement was 58% between LCMM and TS clusters, whereas it varied between 30.8% and 93.3% within the TSC configurations. The extracted features (n = 67) predicted the identified patterns for each growth measure with accuracy of 82%–89%. Specific feature categories were identified as the most important predictors for patterns of all tested growth measures. Conclusion Growth-pattern detection is affected by the method employed. This can impact on comparisons across different populations or associations between growth patterns and health outcomes. Growth features can be reliably used as predictors of growth patterns.

Download Full-text

Dyslipidaemia and mortality in COVID-19 patients - a meta-analysis

QJM ◽

10.1093/qjmed/hcab071 ◽

2021 ◽

Author(s):

Marco Zuin ◽

Gianluca Rigatelli ◽

Claudio Bilato ◽

Carlo Cervellati ◽

Giovanni Zuliani ◽

...

Keyword(s):

Random Effects ◽

Mortality Risk ◽

Meta Analysis ◽

Random Effect ◽

Final Analysis ◽

Short Term ◽

Random Effects Models ◽

Pooled Prevalence ◽

Statistical Heterogeneity ◽

Trim And Fill

Abstract Objective The prevalence and prognostic implications of pre-existing dyslipidaemia in patients infected by the SARS-CoV-2 remain unclear. To perform a systematic review and meta-analysis of prevalence and mortality risk in COVID-19 patients with pre-existing dyslipidaemia. Methods Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed in abstracting data and assessing validity. We searched MEDLINE and Scopus to locate all the articles published up to January 31, 2021, reporting data on dyslipidaemia among COVID-19 survivors and non-survivors. The pooled prevalence of dyslipidaemia was calculated using a random effects model and presenting the related 95% confidence interval (CI), while the mortality risk was estimated using the Mantel-Haenszel random effects models with odds ratio (OR) and related 95% CI. Statistical heterogeneity was measured using the Higgins I2 statistic. Results Eighteen studies, enrolling 74.132 COVID-19 patients [mean age 70.6 years], met the inclusion criteria and were included in the final analysis. The pooled prevalence of dyslipidaemia was 17.5% of cases (95% CI: 12.3-24.3%, p < 0.0001), with high heterogeneity (I2=98.7%). Pre-existing dyslipidaemia was significantly associated with higher risk of short-term death (OR: 1.69, 95% CI: 1.19-2.41, p = 0.003), with high heterogeneity (I2=88.7%). Due to publication bias, according to the Trim-and-Fill method, the corrected random-effect ORs resulted 1.61, 95% CI 1.13-2.28, p < 0.0001 (one studies trimmed). Conclusions Dyslipidaemia represents a major comorbidity in about 18% of COVID-19 patients but it is associated with a 60% increase of short-term mortality risk.

Download Full-text