Multivariate reference and tolerance regions based on conditional transformation models: Application to glycemic markers.

The reference interval is the most widely used medical decision-making, constituting a central tool in determining whether an individual is healthy or not. When the results of several continuous diagnostic tests are available for the same patient, their clinical interpretation is more reliable if a multivariate reference region (MVR) is available rather than multiple univariate reference intervals. MVRs, defined as regions containing 95% of the results of healthy subjects, extend the concept of the reference interval to the multivariate setting. However, they are rarely used in clinical practice owing to difficulties associated with their interpretability and the restrictions inherent to the assumption of a Gaussian distribution. Further statistical research is thus needed to make MVRs more applicable and easier for physicians to interpret. Since the joint distribution of diagnostic test results may well change with patient characteristics independent of disease status, MVRs adjusted for covariates are desirable. The present work introduces a novel formulation for MVRs based on multivariate conditional transformation models (MCTMs). Additionally, we take into account the estimation uncertainty of such MVRs by means of tolerance regions. These conditional MVRs imply no parametric restriction on the response, and potentially nonlinear continuous covariate effects can be estimated. MCTMs allow the estimation of the effects of covariates on the joint distribution of multivariate response variables and on these variables' marginal distributions, via the use of most likely transformation estimation. Our contributions proved reliable when tested with simulated data and for a real data application with two glycemic markers.

A cross-country study on the impact of governmental responses to the COVID-19 pandemic on perinatal mental health.

This study aimed to analyse the role of governmental responses to the coronavirus disease 2019 (COVID-19) outbreak, measured by the Containment and Health Index (CHI), on symptoms of anxiety and depression during pregnancy and postpartum, while considering the countries' Inequality-adjusted Human Development Index (IHDI) and individual factors such as age, gravidity, and exposure to COVID-19. A cross-sectional study using baseline data from the Riseup-PPD-COVID-19 observational prospective international study (ClinicalTrials.gov: NCT04595123) was carried out between June and October 2020 in 12 countries (Albania, Brazil, Bulgaria, Chile, Cyprus, Greece, Israel, Malta, Portugal, Spain, Turkey, and the United Kingdom). Participants were 7645 pregnant women or mothers in the postpartum period-with an infant aged up to 6 months-who completed the Edinburgh Postnatal Depression Scale (EPDS) or the Generalised Anxiety Disorder Assessment (GAD-7) during pregnancy or the postpartum period. The overall prevalence of clinically significant depression symptoms (EPDS ≥ 13) was 30%, ranging from 20,5% in Cyprus to 44,3% in Brazil. The prevalence of clinically significant anxiety symptoms (GAD-7 ≥ 10) was 23,6% (ranging from 14,2% in Israel and Turkey to 39,5% in Brazil). Higher symptoms of anxiety or depression were observed in multigravida exposed to COVID-19 or living in countries with a higher number of deaths due to COVID-19. Furthermore, multigravida from countries with lower IHDI or CHI had higher symptoms of anxiety and depression. Perinatal mental health is context-dependent, with women from more disadvantaged countries at higher risk for poor mental health. Implementing more restrictive measures seems to be a protective factor for mental health, at least in the initial phase of the COVID-19.

Longitudinal relationship of liver injury with inflammation biomarkers in COVID-19 hospitalized patients using a joint modeling approach.

The mechanisms underlying liver disease in patients with COVID-19 are not entirely known. The aim is to investigate, by means of novel statistical techniques, the changes over time in the relationship between inflammation markers and liver damage markers in relation to survival in COVID-19. The study included 221 consecutive patients admitted to the hospital during the first COVID-19 wave in Spain. Generalized additive mixed models were used to investigate the influence of time and inflammation markers on liver damage markers in relation to survival. Joint modeling regression was used to evaluate the temporal correlations between inflammation markers (serum C-reactive protein [CRP], interleukin-6, plasma D-dimer, and blood lymphocyte count) and liver damage markers, after adjusting for age, sex, and therapy. The patients who died showed a significant elevation in serum aspartate transaminase (AST) and alkaline phosphatase levels over time. Conversely, a decrease in serum AST levels was observed in the survivors, who showed a negative correlation between inflammation markers and liver damage markers (CRP with serum AST, alanine transaminase [ALT], and gamma-glutamyl transferase [GGT]; and D-dimer with AST and ALT) after a week of hospitalization. Conversely, most correlations were positive in the patients who died, except lymphocyte count, which was negatively correlated with AST, GGT, and alkaline phosphatase. These correlations were attenuated with age. The patients who died during COVID-19 infection displayed a significant elevation of liver damage markers, which is correlated with inflammation markers over time. These results are consistent with the role of systemic inflammation in liver damage during COVID-19.

Modeling conditional reference regions: Application to glycemic markers.

Many clinical decisions are taken based on the results of continuous diagnostic tests. Usually, only the results of one single test is taken into consideration, the interpretation of which requires a reference range for the healthy population. However, the use of two different tests, can be necessary in the diagnosis of certain diseases. This obliges a bivariate reference region be available for their interpretation. It should also be remembered that reference regions may depend on patient variables (eg, age and sex) independent of the suspected disease. However, few proposals have been made regarding the statistical modeling of such reference regions, and those put forward have always assumed a Gaussian distribution, which can be rather restrictive. The present work describes a new statistical method that allows such reference regions to be estimated with no insistence on the results being normally distributed. The proposed method is based on a bivariate location-scale model that provides probabilistic regions covering a specific percentage of the bivariate data, dependent on certain covariates. The reference region is estimated nonparametrically and the nonlinear effects of continuous covariates via polynomial kernel smoothers in additive models. The bivariate model is estimated using a backfitting algorithm, and the optimal smoothing parameters of the kernel smoothers selected by cross-validation. The model performed satisfactorily in simulation studies under the assumption of non-Gaussian conditions. Finally, the proposed methodology was found to be useful in estimating a reference region for two continuous diagnostic tests for diabetes (fasting plasma glucose and glycated hemoglobin), taking into account the age of the patient.

Impact of the Covid-19 pandemic on perinatal mental health (Riseup-PPD-COVID-19): protocol for an international prospective cohort study.

BACKGROUND: Corona Virus Disease 19 (COVID-19) is a new pandemic, declared a public health emergency by the World Health Organization, which could have negative consequences for pregnant and postpartum women. The scarce evidence published to date suggests that perinatal mental health has deteriorated since the COVID-19 outbreak. However, the few studies published so far have some limitations, such as a cross-sectional design and the omission of important factors for the understanding of perinatal mental health, including governmental restriction measures and healthcare practices implemented at the maternity hospitals. Within the Riseup-PPD COST Action, a study is underway to assess the impact of COVID-19 in perinatal mental health. The primary objectives are to (1) evaluate changes in perinatal mental health outcomes; and (2) determine the risk and protective factors for perinatal mental health during the COVID-19 pandemic. Additionally, we will compare the results between the countries participating in the study. METHODS: This is an international prospective cohort study, with a baseline and three follow-up assessments over a six-month period. It is being carried out in 11 European countries (Albania, Bulgaria, Cyprus, France, Greece, Israel, Malta, Portugal, Spain, Turkey, and the United Kingdom), Argentina, Brazil and Chile. The sample consists of adult pregnant and postpartum women (with infants up to 6 months of age). The assessment includes measures on COVID-19 epidemiology and public health measures (Oxford COVID-19 Government Response Tracker dataset), Coronavirus Perinatal Experiences (COPE questionnaires), psychological distress (BSI-18), depression (EPDS), anxiety (GAD-7) and post-traumatic stress symptoms (PTSD checklist for DSM-V). DISCUSSION: This study will provide important information for understanding the impact of the COVID-19 pandemic on perinatal mental health and well-being, including the identification of potential risk and protective factors by implementing predictive models using machine learning techniques. The findings will help policymakers develop suitable guidelines and prevention strategies for perinatal mental health and contribute to designing tailored mental health interventions. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04595123 .

Offering, participation and adherence to cardiac rehabilitation programmes in the elderly: a European comparison based on the EU-CaRE multicentre observational study.

AIMS: Cardiac rehabilitation (CR) is strongly recommended but participation of elderly patients has not been well characterized. This study aims to analyse current rates and determinants of CR referral, participation, adherence, and compliance in a contemporary European cohort of elderly patients. METHODS AND RESULTS: The EU-CaRE observational study included data from consecutive patients aged ≥ 65 with acute coronary syndrome, revascularization, stable coronary artery disease, or heart valve replacement, recruited in eight European centres. Rates and factors determining offering, participation, and adherence to CR programmes and compliance with training sessions were studied across centres, under consideration of extensive-outpatient vs. intensive-inpatient programmes. Three thousand, four hundred, and seventy-one patients were included in the offering and participation analysis. Cardiac rehabilitation was offered to 80.8% of eligible patients, formal contraindications being the main reason for not offering CR. Mean participation was 68.0%, with perceived lack of usefulness and transport issues being principal barriers. Mean adherence to CR programmes of participants in the EU-CaRE study (n = 1663) was 90.3%, with hospitalization/physical impairment as principal causes of dropout. Mean compliance with training sessions was 86.1%. Older age was related to lower offering and participation, and comorbidity was associated with lower offering, participation, adherence, and compliance. Intensive-inpatient programmes displayed higher adherence (97.1% vs. 85.9%, P < 0.001) and compliance (full compliance: 66.0% vs. 38.8%, P < 0.001) than extensive-outpatient programmes. CONCLUSION: In this European cohort of elderly patients, older age and comorbidity tackled patients' referral and uptake of CR programmes. Intensive-inpatient CR programmes showed higher completion than extensive-outpatient CR programmes, suggesting this formula could suit some elderly patients.

Patterns of monogenean abundance in native and invasive populations of Planiliza haematocheila (Teleostei: Mugilidae): interactions between climate and host defence mechanisms explain parasite release.

One of the most intriguing questions in invasive biology is how an introduced species changes its population parameters in a new environment. Translocation of free-living species commonly results in co-introduction of their parasites. The current study focuses on the abundance pattern of the monogenean parasite Ligophorus llewellyni of the pacific so-iuy mullet, Planiliza haematocheila, across the native and introduced distribution ranges. We evaluated parasite release by the so-iuy mullet by comparing abundance patterns of L. llewellyni under effects of the host length, water temperature and month of the year in the Sea of Japan and the Sea of Azov. Generalised additive models applied to analysis of parasite abundance data showed that relationships between the mean number of L. llewellyni and the three tested independent variables were not linear. Our results suggest that the introduced host lost a large amount of parasite abundance due to the effect of warm climate in a new region, which is mediated by host defence mechanisms. The abundance of L. llewellyni rapidly rose in autumn, as fish activity and immune response decrease, reached the maximum in winter and began to fall in spring as a warm temperature facilitates the fish immune defence. The abundance of L. llewellyni showed an initial increase in response to fish growth and reached an asymptote. The response curves built for native and introduced regions reached an asymptote at different fish body lengths, reflecting the fish growth rate, which is higher in the introduced range of P. haematocheila. We found that the carried parasite species holds the same trend in relationships compared with its native area, between the mean number of monogeneans per host and independent variables increasing abundance with fish length, low temperature and cold months. Our results open new perspectives for future research on statistical modelling of parasite abundance across native and introduced distribution ranges in order to provide deeper insight into host-parasite interactions of invasive populations.

Volumetric Absorptive Microsampling (VAMS) for assaying immunosuppressants from venous whole blood by LC-MS/MS using a novel atmospheric pressure ionization probe (UniSpray™).

Therapeutic drug monitoring (TDM) of immunosuppressants (IMS) is crucial to prevent rejection or toxicity after solid organ transplantation. Microsampling techniques (sampling <50 µL of blood) can be a good alternative to conventional venous sampling for TDM, due to their numerous advantages, including its easy and low-invasive sampling, enabling self-collection, and cost-saving shipment and storage. Furthermore, volumetric absorptive microsampling (VAMS) enables the collection of precise and accurate blood volumes, overcoming the hematocrit (Hct) effect related to dried blood spots, while offering the same benefits. In this work, an LC-MS/MS method for the determination of the 5 most common IMS (mycophenolic acid -MPA-, tacrolimus -TAC-, sirolimus -SIR-, everolimus -EVE- and cyclosporin A -CsA-) in venous blood collected with Mitra™ VAMS devices was developed and validated, employing a novel LC-MS/MS interface, Unispray™. The method was fully validated including linearity, limits of detection (LOD) and quantification (LLOQ), accuracy, precision, selectivity, carry-over, matrix effect, recovery, impact of Hct on recovery and autosampler and short-/long-term stability, satisfying acceptance criteria in all cases. LLOQs were 0.5 ng/mL for TAC, SIR and EVE, 20 ng/mL for CsA and 75 ng/mL for MPA. No impact of the Hct (range: 0.2 to 0.62 L/L) on recovery was found for any analyte. All compounds were stable in VAMS for at least 8 months at -20 °C. In addition, as part of the VAMS analytical method validation, we performed for the first time a broad statistical study to compare liquid venous blood concentrations from patients under TAC (n = 53) and MPA (n = 20) treatment to those observed when the same specimens were absorbed into VAMS. Our results showed that venous blood VAMS concentrations were correlated to those found in the original liquid venous blood, proving that the VAMS material itself will not bias blood drug concentrations. Therefore, the present method could be applied to evaluate possible correlations between venous blood and capillary blood collected with VAMS.

Improvement in VO2peak predicts readmissions for cardiovascular disease and mortality in patients undergoing cardiac rehabilitation.

BACKGROUND: Improvement in exercise capacity is a main goal of cardiac rehabilitation but the effects are often lost at long-term follow-up and thus also the benefits on prognosis. We assessed whether improvement in VO2peak during a cardiac rehabilitation programme predicts long-term prognosis. METHODS AND RESULTS: We performed a retrospective analysis of 1561 cardiac patients completing cardiac rehabilitation in 2011-2017 in Copenhagen. Mean age was 63.6 (11) years, 74% were male and 84% had coronary artery disease, 6% chronic heart failure and 10% heart valve replacement. The association between baseline VO2peak and improvement after cardiac rehabilitation and being readmitted for cardiovascular disease and/or all-cause mortality was assessed with three different analyses: Cox regression for the combined outcome, for all-cause mortality and a multi-state model. During a median follow-up of 2.3 years, 167 readmissions for cardiovascular disease and 77 deaths occurred. In adjusted Cox regression there was a non-linear decreasing risk of the combined outcome with higher baseline VO2peak and with improvement of VO2peak after cardiac rehabilitation. A similar linear association was seen for all-cause mortality. Applying the multi-state model, baseline VO2peak and change in VO2peak were associated with risk of a cardiovascular disease readmission and with all-cause mortality but not with mortality in those having an intermediate readmission for cardiovascular disease. CONCLUSION: VO2peak as well as change in VO2peak were highly predictive of future risk of readmissions for cardiovascular disease and all-cause mortality. The predictive value did not extend beyond the next admission for a cardiovascular event.

Liraglutide in a real-world setting: Joint modeling of metabolic response, prediction of efficacy, and cardiovascular risk / Liraglutida en vida real: modelización conjunta de la respuesta metabólica, predicción de eficacia y riesgo cardiovascular

Introduction and objectives: The worldwide prevalence of type 2 diabetes mellitus increases in parallel to that of obesity. Liraglutide (LRG), a glucagon-like peptide-1 receptor agonist, can reduce body weight. This study assessed the metabolic efficacy of LRG in real-world clinical practice. Methods: An observational, retrospective cohort study including patients treated with LRG for at least one year (187 patients). Anthropometric and metabolic variables, a composite endpoint, factors predicting response to LRG, and cardiovascular risk over time were assessed. A linear mixed-effects model with a bivariate structure was constructed to investigate the time-dependent relationship between weight and HbA1c values. Results: HbA1c levels and weight significantly decreased in the first 12 weeks, and the decrease persisted at 12 and 24 months in all subgroups studied. Mean weight and HbA1c decreases after 24 months were 8.5kg and 1.7% respectively. HbA1c values <7% were achieved by 42% of patients at 12 months and by 40% at 24 months. Treatment with LRG allowed for reduction in insulin dose. No serious adverse events were noted. Cardiovascular risk decreased from high to moderate-low. Conclusions: Under standard clinical practice conditions, LRG achieved a better metabolic response than seen in clinical trials. Efficacy at 12 weeks of treatment is a good predictor of response. LRG allows for delaying or reducing insulin dose by improving both weight and glucose control. Cardiovascular risk improved

Introducción y objetivos: La prevalencia mundial de diabetes mellitus tipo 2 aumenta junto a la de la obesidad. Liraglutida (LRG), un agonista del receptor del péptido similar al GLP1, es un fármaco antidiabético capaz de reducir peso. Evaluamos en práctica clínica de vida real su eficacia metabólica. Método: Estudio de cohorte observacional retrospectivo. Se incluyeron los pacientes tratados al menos durante un año con LRG (187 pacientes). Evaluamos variables antropométricas, metabólicas, objetivos combinados, factores predictivos de respuesta y evolución del riesgo cardiovascular. Se construyó un modelo de efectos mixtos lineales de estructura bivariante para investigar la relación tiempo-dependiente entre el peso y los valores de HbA1c. Resultados: Descenso significativo de los valores de HbA1c y peso en las primeras 12 semanas de tratamiento, mantenido a los 12 y 24 meses, en todos los subgrupos estudiados. Reducción media de peso y HbA1c tras 24 meses de tratamiento de 8,5 kg y 1,7%. El valor de HbA1c fue <7% en 42% de pacientes a los 12 meses, 40% a los 24 meses. El tratamiento con LRG permitió reducir la dosis de insulina. No registramos eventos adversos graves. El riesgo cardiovascular mejoró. Conclusiones: Bajo condiciones de práctica clínica habitual la respuesta metabólica a LRG resultó mejor que la observada en ensayos clínicos. La eficacia a las 12 semanas de tratamiento es un buen predictor de respuesta. LRG permite retrasar o reducir la insulinoterapia. Los pacientes mejoraron su riesgo cardiovascular

Liraglutide in a real-world setting: Joint modeling of metabolic response, prediction of efficacy, and cardiovascular risk.

INTRODUCTION AND OBJECTIVES: The worldwide prevalence of type 2 diabetes mellitus increases in parallel to that of obesity. Liraglutide (LRG), a glucagon-like peptide-1 receptor agonist, can reduce body weight. This study assessed the metabolic efficacy of LRG in real-world clinical practice. METHODS: An observational, retrospective cohort study including patients treated with LRG for at least one year (187 patients). Anthropometric and metabolic variables, a composite endpoint, factors predicting response to LRG, and cardiovascular risk over time were assessed. A linear mixed-effects model with a bivariate structure was constructed to investigate the time-dependent relationship between weight and HbA1c values. RESULTS: HbA1c levels and weight significantly decreased in the first 12 weeks, and the decrease persisted at 12 and 24 months in all subgroups studied. Mean weight and HbA1c decreases after 24 months were 8.5kg and 1.7% respectively. HbA1c values <7% were achieved by 42% of patients at 12 months and by 40% at 24 months. Treatment with LRG allowed for reduction in insulin dose. No serious adverse events were noted. Cardiovascular risk decreased from high to moderate-low. CONCLUSIONS: Under standard clinical practice conditions, LRG achieved a better metabolic response than seen in clinical trials. Efficacy at 12 weeks of treatment is a good predictor of response. LRG allows for delaying or reducing insulin dose by improving both weight and glucose control. Cardiovascular risk improved.

Corneal Biomechanics in Unilateral Keratoconus and Fellow Eyes with a Scheimpflug-based Tonometer.

SIGNIFICANCE: Before the appearance of evident keratoconus, corneal biomechanical changes may be detectable. Here, these properties are analyzed to detect any difference that could help in the early recognition of keratoconus to allow patients to benefit from early treatments and to avoid refractive procedures in these corneas. PURPOSE: The purpose of this study was to compare corneal biomechanical characteristics as determined by Corvis Scheimpflug Technology tonometry between normal eyes and asymmetric keratoconic eyes. METHODS: Retrospective data from normal eyes (n = 100), keratoconic eyes (n = 18), and their topographically normal fellow eyes (n = 18) were analyzed. Differences in the variables among the groups were determined. For the parameters that showed significant differences, the receiver operating characteristic curve and the area under the curve (AUC) were used to assess the diagnostic accuracy of each variable. The optimal cutoff points were determined when comparing normal and fellow eyes. Also, a new linear combination of variables was performed to obtain better discriminative values. RESULTS: The following variables differed significantly between normal and fellow eyes: length of the flattened cornea in the second applanation, peak distance, curvature radius at highest concavity, and central corneal thickness. When each variable was independently considered, AUCs, sensitivity, and specificity were insufficiently high for good discrimination between the two groups. However, using a linear combination of variables, an optimal cutoff point (0.157) was obtained with an AUC of 0.78, sensitivity of 0.84, and specificity of 0.69. CONCLUSIONS: A best predictive linear combination of corneal biomechanical variables was tested including diameter of the flattened cornea in the second applanation and central corneal thickness. This combination was considered as the best in terms of its prediction capacity, simplicity and clinical application. This formula may be useful in clinical practice to discriminate between normal eyes and incipient keratoconus.

Post-operative stress hyperglycemia is a predictor of mortality in liver transplantation.

BACKGROUND: A significant association is known between increased glycaemic variability and mortality in critical patients. To ascertain whether glycaemic profiles during the first week after liver transplantation might be associated with long-term mortality in these patients, by analysing whether diabetic status modified this relationship. METHOD: Observational long-term survival study includes 642 subjects undergoing liver transplantation from July 1994 to July 2011. Glucose profiles, units of insulin and all variables with influence on mortality are analysed using joint modelling techniques. RESULTS: Patients registered a survival rate of 85% at 1 year and 65% at 10 years, without differences in mortality between patients with and without diabetes. In glucose profiles, however, differences were observed between patients with and without diabetes: patients with diabetes registered lower baseline glucose values, which gradually rose until reaching a peak on days 2-3 and then subsequently declined, diabetic subjects started from higher values which gradually decreased across the first week. Patients with diabetes showed an association between mortality and age, Model for End-Stage Liver Disease score (MELD) score and hepatitis C virus; among non-diabetic patients, mortality was associated with age, body mass index, malignant aetiology, red blood cell requirements and parenteral nutrition. Glucose profiles were observed to be statistically associated with mortality among patients without diabetes (P = 0.022) but not among patients who presented with diabetes prior to transplantation (P = 0.689). CONCLUSIONS: Glucose profiles during the first week after liver transplantation are different in patients with and without diabetes. While glucose profiles are associated with long-term mortality in patients without diabetes, after adjusting for potential confounding variables such as age, cause of transplantation, MELD, nutrition, immunosuppressive drugs, and units of insulin administered, this does not occur among patients with diabetes.

25-Hydroxyvitamin D levels of children are inversely related to adiposity assessed by body mass index

Vitamin D deficiency is associated with wide range of pathologies. Some evidences have shown that low vitamin D circulating levels in children and adolescent are related to fat mass and obesity. The objectives of the present study were to characterize vitamin D status in children and adolescents and to determine if serum 25-hydroxyvitamin D (25(OH)D) concentration is related to adiposity assessed by body mass index (BMI). Serum 25(OH)D levels were measured by LIAISON method in 471 children and adolescents (2 to 18 years age) and analyzed according to gender, pubertal period, age, and BMI. An overall prevalence of 25(OH)D insufficiency and deficiency was present in the 67.1%. Lower 25(OH)D levels were found in females (25.56 ± 14.03 vs 29.71 ± 17.10 ng ml−1; P = 0.004) and pubertal children (25.52 ± 13.97 vs 29.21 ± 16.83 ng ml−1; P = 0.011). In addition, an inverse relation of BMI and age on 25(OH)D concentrations was observed in children. In conclusion, low vitamin D status was highly prevalent among children and adolescents. Of note, a non-lineal regression model showed that 39.6% of vitamin D levels variability was explained by BMI. These results indicate that adiposity assessed by BMI impacts vitamin D status

25-Hydroxyvitamin D levels of children are inversely related to adiposity assessed by body mass index.

Vitamin D deficiency is associated with wide range of pathologies. Some evidences have shown that low vitamin D circulating levels in children and adolescent are related to fat mass and obesity. The objectives of the present study were to characterize vitamin D status in children and adolescents and to determine if serum 25-hydroxyvitamin D (25(OH)D) concentration is related to adiposity assessed by body mass index (BMI). Serum 25(OH)D levels were measured by LIAISON method in 471 children and adolescents (2 to 18 years age) and analyzed according to gender, pubertal period, age, and BMI. An overall prevalence of 25(OH)D insufficiency and deficiency was present in the 67.1%. Lower 25(OH)D levels were found in females (25.56 ± 14.03 vs 29.71 ± 17.10 ng ml-1; P = 0.004) and pubertal children (25.52 ± 13.97 vs 29.21 ± 16.83 ng ml-1; P = 0.011). In addition, an inverse relation of BMI and age on 25(OH)D concentrations was observed in children. In conclusion, low vitamin D status was highly prevalent among children and adolescents. Of note, a non-lineal regression model showed that 39.6% of vitamin D levels variability was explained by BMI. These results indicate that adiposity assessed by BMI impacts vitamin D status.

Bootstrap-based procedures for inference in nonparametric receiver-operating characteristic curve regression analysis.

Prior to using a diagnostic test in a routine clinical setting, the rigorous evaluation of its diagnostic accuracy is essential. The receiver-operating characteristic curve is the measure of accuracy most widely used for continuous diagnostic tests. However, the possible impact of extra information about the patient (or even the environment) on diagnostic accuracy also needs to be assessed. In this paper, we focus on an estimator for the covariate-specific receiver-operating characteristic curve based on direct regression modelling and nonparametric smoothing techniques. This approach defines the class of generalised additive models for the receiver-operating characteristic curve. The main aim of the paper is to offer new inferential procedures for testing the effect of covariates on the conditional receiver-operating characteristic curve within the above-mentioned class. Specifically, two different bootstrap-based tests are suggested to check (a) the possible effect of continuous covariates on the receiver-operating characteristic curve and (b) the presence of factor-by-curve interaction terms. The validity of the proposed bootstrap-based procedures is supported by simulations. To facilitate the application of these new procedures in practice, an R-package, known as npROCRegression, is provided and briefly described. Finally, data derived from a computer-aided diagnostic system for the automatic detection of tumour masses in breast cancer is analysed.

Reassessment of fluctuating dental asymmetry in Down syndrome.

Fluctuating dental asymmetry (FDA) is a tool to measure developmental stability that could be increased in gonosomal aneuploidies. The aim of this study was to quantify FDA in individuals with Down syndrome (DS). The study group comprised 40 individuals with DS, and a control group matched for age and sex was created. The target teeth were the maxillary central incisors (11,21), maxillary lateral incisors (12,22), maxillary canines (13,23), and maxillary first molars (16,26). Dental morphometric variables measured on CBCT images included tooth length, crown height, root length, mesio-distal diameter, crown-to-root ratio, vestibular-palatine diameter, mid mesio-distal diameter, mid buccal-palatal diameter, maximum buccal-palatal diameter, and cervical circumference. The FA2 fluctuating asymmetry index (Palmer and Strobeck, 1986) was applied. Some discrepancies in crown-to-root ratios and root length asymmetry were significantly lower in the DS individuals than in controls. Combining the crown-to-root ratio of tooth 11 versus 21, tooth 12 versus 22, and tooth 13 versus 23, we developed a predictive model with a discriminatory power between DS and controls of 0.983. Some dental morphometric variables may actually be more stable in DS individuals than in the general population. This offers a new perspective on the relationship between canalization, fluctuating asymmetry, and aneuploidy.