Effects Of Individual Covariates Research Articles

Forward stability and model path selection

Most scientific publications follow the familiar recipe of (i) obtain data, (ii) fit a model, and (iii) comment on the scientific relevance of the effects of particular covariates in that model. This approach, however, ignores the fact that there may exist a multitude of similarly-accurate models in which the implied effects of individual covariates may be vastly different. This problem of finding an entire collection of plausible models has also received relatively little attention in the statistics community, with nearly all of the proposed methodologies being narrowly tailored to a particular model class and/or requiring an exhaustive search over all possible models, making them largely infeasible in the current big data era. This work develops the idea of forward stability and proposes a novel, computationally-efficient approach to finding collections of accurate models we refer to as model path selection (MPS). MPS builds up a plausible model collection via a forward selection approach and is entirely agnostic to the model class and loss function employed. The resulting model collection can be displayed in a simple and intuitive graphical fashion, easily allowing practitioners to visualize whether some covariates can be swapped for others with minimal loss.

Incidence and predictors of hepatocellular carcinoma in patients with autoimmune hepatitis

Autoimmune hepatitis (AIH) is a rare chronic liver disease of unknown aetiology; the risk of hepatocellular carcinoma (HCC) remains unclear and risk factors are not well-defined. We aimed to investigate the risk of HCC across a multicentre AIH cohort and to identify predictive factors. We performed a retrospective, observational, multicentric study of patients included in the International Autoimmune Hepatitis Group Retrospective Registry. The assessed clinical outcomes were HCC development, liver transplantation, and death. Fine and Gray regression analysis stratified by centre was applied to determine the effects of individual covariates; the cumulative incidence of HCC was estimated using the competing risk method with death as a competing risk. A total of 1,428 patients diagnosed with AIH from 1980 to 2020 from 22 eligible centres across Europe and Canada were included, with a median follow-up of 11.1 years (interquartile range 5.2-15.9). Two hundred and ninety-three (20.5%) patients had cirrhosis at diagnosis. During follow-up, 24 patients developed HCC (1.7%), an incidence rate of 1.44 cases/1,000 patient-years; the cumulative incidence of HCC increased over time (0.6% at 5 years, 0.9% at 10 years, 2.7% at 20 years, and 6.6% at 30 years of follow-up). Patients who developed cirrhosis during follow-up had a significantly higher incidence of HCC. The cumulative incidence of HCC was 2.6%, 4.6%, 5.6% and 6.6% at 5, 10, 15, and 20 years after the development of cirrhosis, respectively. Obesity (hazard ratio [HR] 2.94, p=0.04), cirrhosis (HR 3.17, p=0.01), and AIH/PSC variant syndrome (HR 5.18, p=0.007) at baseline were independent risk factors for HCC development. HCC incidence in AIH is low even after cirrhosis development and is associated with risk factors including obesity, cirrhosis, and AIH/PSC variant syndrome. The risk of developing hepatocellular carcinoma (HCC) in individuals with autoimmune hepatitis (AIH) seems to be lower than for other aetiologies of chronic liver disease. Yet, solid data for this specific patient group remain elusive, given that most of the existing evidence comes from small, single-centre studies. In our study, we found that HCC incidence in patients with AIH is low even after the onset of cirrhosis. Additionally, factors such as advanced age, obesity, cirrhosis, alcohol consumption, and the presence of the AIH/PSC variant syndrome at the time of AIH diagnosis are linked to a higher risk of HCC. Based on these findings, there seems to be merit in adopting a specialized HCC monitoring programme for patients with AIH based on their individual risk factors.

A pharmacokinetic model optimized by covariates for propofol target-controlled infusion in dogs

ObjectiveTo develop a population pharmacokinetic model for propofol target-controlled infusion (TCI) in dogs and to evaluate its performance for use in the clinical setting. Study designProspective clinical study. AnimalsA group of 40 client-owned dogs undergoing general anaesthesia for magnetic resonance imaging. MethodsPropofol was administered to 26 premedicated dogs and arterial blood samples were collected during the infusion and over 240 minutes after terminating the infusion. Propofol concentrations were measured by high-performance liquid chromatography. A population pharmacokinetic analysis was performed using a nonlinear mixed-effects modelling approach, allowing inter- and intra-individual variability estimation and quantitative evaluation of the influence of the following covariates: weight, body condition score, age, size-related age (Age_size), sex, premedication type, size and contrast agent administration. A final model was obtained using a stepwise approach in which individual covariate effects on each pharmacokinetic variable were incorporated. The performance of the developed TCI model was subsequently evaluated while inducing and maintaining anaesthesia in 14 premedicated dogs and assessed by comparing predicted and measured concentrations at specific time points. ResultsPropofol pharmacokinetics was best described by a three-compartment model. Weight, Age_size, premedication and sex showed significant pharmacokinetic effects. Addition of the significant covariate/variable associations to the final model resulted in a reduction of the objective function value from 285.53 to –22.34. The median values of prediction error and absolute performance error were 3.1% and 28.4%, respectively. Induction targets between 4.0 and 6.5 μg mL–1 allowed intubation within 5.0 ± 0.9 minutes. Anaesthesia was achieved with targets between 3.0 and 6.5 μg mL–1. Mean time to extubation was 9.7 ± 2.6 minutes. All dogs recovered smoothly and without complications. Conclusions and clinical relevanceOverall predictive performance of the pharmacokinetic model-driven infusion developed was clinically acceptable for administering propofol to dogs in routine anaesthesia.

An additive Gaussian process regression model for interpretable non-parametric analysis of longitudinal data

Biomedical research typically involves longitudinal study designs where samples from individuals are measured repeatedly over time and the goal is to identify risk factors (covariates) that are associated with an outcome value. General linear mixed effect models are the standard workhorse for statistical analysis of longitudinal data. However, analysis of longitudinal data can be complicated for reasons such as difficulties in modelling correlated outcome values, functional (time-varying) covariates, nonlinear and non-stationary effects, and model inference. We present LonGP, an additive Gaussian process regression model that is specifically designed for statistical analysis of longitudinal data, which solves these commonly faced challenges. LonGP can model time-varying random effects and non-stationary signals, incorporate multiple kernel learning, and provide interpretable results for the effects of individual covariates and their interactions. We demonstrate LonGP’s performance and accuracy by analysing various simulated and real longitudinal -omics datasets.

Subgroup analysis of zero-inflated Poisson regression model with applications to insurance data

Customized personal rate offering is of growing importance in the insurance industry. To achieve this, an important step is to identify subgroups of insureds from the corresponding heterogeneous claim frequency data. In this paper, a penalized Poisson regression approach for subgroup analysis in claim frequency data is proposed. Subjects are assumed to follow a zero-inflated Poisson regression model with group-specific intercepts, which capture group characteristics of claim frequency. A penalized likelihood function is derived and optimized to identify the group-specific intercepts and effects of individual covariates. To handle the challenges arising from the optimization of the penalized likelihood function, an alternating direction method of multipliers algorithm is developed and its convergence is established. Simulation studies and real applications are provided for illustrations.

Population Pharmacokinetics of Apixaban in Subjects With Nonvalvular Atrial Fibrillation.

This analysis describes the population pharmacokinetics (PPK) of apixaban in nonvalvular atrial fibrillation (NVAF) subjects, and quantifies the impact of intrinsic and extrinsic factors on exposure. The PPK model was developed using data from phase I–III studies. Apixaban exposure was characterized by a two‐compartment PPK model with first‐order absorption and elimination. Predictive covariates on apparent clearance included age, sex, Asian race, renal function, and concomitant strong/moderate cytochrome P450 (CYP)3A4/P‐glycoprotein (P‐gp) inhibitors. Individual covariate effects generally resulted in < 25% change in apixaban exposure vs. the reference NVAF subject (non‐Asian, male, aged 65 years, weighing 70 kg without concomitant CYP3A4/P‐gp inhibitors), except for severe renal impairment, which resulted in 55% higher exposure than the reference subject. The dose‐reduction algorithm resulted in a ~27% lower median exposure, with a large overlap between the 2.5‐mg and 5‐mg groups. The impact of Asian race on apixaban exposure was < 15% and not considered clinically significant.

Patient Age, Sex, and Inflammatory Bowel Disease Phenotype Associate With Course of Primary Sclerosing Cholangitis

Primary sclerosing cholangitis (PSC) is an orphan hepatobiliary disorder associated with inflammatory bowel disease (IBD). We aimed to estimate the risk of disease progression based on distinct clinical phenotypes in a large international cohort of patients with PSC. We performed a retrospective outcome analysis of patients diagnosed with PSC from 1980 through 2010 at 37 centers in Europe, North America, and Australia. For each patient, we collected data on sex, clinician-reported age at and date of PSC and IBD diagnoses, phenotypes of IBD and PSC, and date and indication of IBD-related surgeries. The primary and secondary endpoints were liver transplantation or death (LTD) and hepatopancreatobiliary malignancy, respectively. Cox proportional hazards models were applied to determine the effects of individual covariates on rates of clinical events, with time-to-event analysis ascertained through Kaplan-Meier estimates. Of the 7121 patients in the cohort, 2616 met the primary endpoint (median time to event of 14.5 years) and 721 developed hepatopancreatobiliary malignancy. The most common malignancy was cholangiocarcinoma (n= 594);patients of advanced age at diagnosis had an increased incidence compared with younger patients (incidence rate: 1.2 per 100 patient-years for patients younger than 20 years old, 6.0 per 100 patient-years for patients 21-30 years old, 9.0 per 100patient-years for patients 31-40 years old, 14.0 per 100 patient-years for patients 41-50 years old, 15.2 per 100 patient-years for patients 51-60 years old, and 21.0 per 100patient-years for patients older than 60 years). Of all patients with PSC studied, 65.5% were men, 89.8% had classical or large-duct disease, and 70.0% developed IBD at some point. Assessing the development of IBD as a time-dependent covariate, Crohn's disease and no IBD (both vs ulcerative colitis) were associated with a lower risk of LTD (unadjusted hazard ratio [HR], 0.62; P < .001 and HR, 0.90; P= .03, respectively) and malignancy (HR, 0.68; P= .008 and HR, 0.77; P= .004, respectively). Small-duct PSC was associated with a lower risk of LTD or malignancy compared with classic PSC (HR, 0.30 and HR, 0.15, respectively; both P < .001). Female sex was also associated with a lower risk of LTD or malignancy (HR, 0.88; P= .002 and HR, 0.68; P < .001, respectively). In multivariable analyses assessing the primary endpoint, small-duct PSC characterized a low-risk phenotype in both sexes (adjusted HR formen, 0.23; P < .001 and adjusted HR for women, 0.48; P=.003). Conversely, patients with ulcerative colitis had an increased risk of liver disease progression compared with patients with Crohn's disease (HR, 1.56; P < .001) or no IBD (HR, 1.15; P= .002). In an analysis of data from individual patients with PSC worldwide, we found significant variation in clinical course associated with age at diagnosis, sex, and ductal and IBD subtypes. The survival estimates provided might be used to estimate risk levels for patients with PSC and select patients for clinical trials.

Likelihood-based inference for discretely observed birth-death-shift processes, with applications to evolution of mobile genetic elements.

Continuous-time birth-death-shift (BDS) processes are frequently used in stochastic modeling, with many applications in ecology and epidemiology. In particular, such processes can model evolutionary dynamics of transposable elements-important genetic markers in molecular epidemiology. Estimation of the effects of individual covariates on the birth, death, and shift rates of the process can be accomplished by analyzing patient data, but inferring these rates in a discretely and unevenly observed setting presents computational challenges. We propose a multi-type branching process approximation to BDS processes and develop a corresponding expectation maximization algorithm, where we use spectral techniques to reduce calculation of expected sufficient statistics to low-dimensional integration. These techniques yield an efficient and robust optimization routine for inferring the rates of the BDS process, and apply broadly to multi-type branching processes whose rates can depend on many covariates. After rigorously testing our methodology in simulation studies, we apply our method to study intrapatient time evolution of IS6110 transposable element, a genetic marker frequently used during estimation of epidemiological clusters of Mycobacterium tuberculosis infections.

Abstract 316: Using Reference Effect Measures to Identify Sources of Variation in 30-day Readmissions for Percutaneous Coronary Intervention

Background: Multilevel models for non-normal outcomes are widely used in outcomes research to estimate effects of covariates on outcomes, e.g. hospital readmissions following percutaneous coronary intervention (PCI). We refine Reference Effect Measures (REM) to compare effects of individual covariates, sets of covariates and random effects on the same scale (e.g. odds ratio, OR) and present a novel approach for displaying these effects. We illustrate this method by studying these sources of variation in 30-day readmission rates for Veterans Administration (VA) patients undergoing PCI. Methods: We used mixed effects logistic regression with 13 patient and 3 hospital covariates to study 30-day readmission rates in a national cohort of 45,521 VA Clinical Assessment, Reporting and Tracking (CART) patients who received a PCI during 10/2007-9/2012 at 49 VA hospitals. OR was used as a REM to compare percentiles of hospital or patient risk with median hospitals or patients to assess levels of variation. Results: Overall 30-day readmission rate was 11.5% ranging from 6.8% to 17.3% across the 49 sites. The figure below shows effects of individual patient and hospital covariates, combined effects of patient and hospital characteristics (shaded bars), and hospital random variation (shaded curve) for 30-day readmissions. The OR for comparing a 97.5 th percentile hospital with a median hospital, all covariates being equal, was 1.43, which was substantially less than the 2.88 OR found when comparing patients at the same percentile of the combined patient risk distribution. The largest patient covariate effect (congestive heart failure) had an OR of 1.58, equivalent to comparing a 99 th percentile hospital with a median hospital, or an 81 st percentile risk patient to a median risk patient. Combining all hospital characteristics, the OR for a 97.5 th percentile hospital versus a median hospital was 1.16. Conclusions: REMs are simple to compute, interpret and graph, and provide direct comparison on the same scale across random effects and covariates. The methods apply generally to linear, logistic, log, count, and time to event outcomes. These methods showed that patient risk drives 30-day readmission rates for VA PCI patients and the figure provided a new way of visualizing variation across these effects.

Varying Index Coefficient Models

It has been a long history of using interactions in regression analysis to investigate alterations in covariate-effects on response variables. In this article, we aim to address two kinds of new challenges arising from the inclusion of such high-order effects in the regression model for complex data. The first kind concerns a situation where interaction effects of individual covariates are weak but those of combined covariates are strong, and the other kind pertains to the presence of nonlinear interactive effects directed by low-effect covariates. We propose a new class of semiparametric models with varying index coefficients, which enables us to model and assess nonlinear interaction effects between grouped covariates on the response variable. As a result, most of the existing semiparametric regression models are special cases of our proposed models. We develop a numerically stable and computationally fast estimation procedure using both profile least squares method and local fitting. We establish both estimation consistency and asymptotic normality for the proposed estimators of index coefficients as well as the oracle property for the nonparametric function estimator. In addition, a generalized likelihood ratio test is provided to test for the existence of interaction effects or the existence of nonlinear interaction effects. Our models and estimation methods are illustrated by simulation studies, and by an analysis of child growth data to evaluate alterations in growth rates incurred by mother’s exposures to endocrine disrupting compounds during pregnancy. Supplementary materials for this article are available online.

Population pharmacokinetics of afatinib, an irreversible ErbB family blocker, in patients with various solid tumors

This population pharmacokinetic model was developed to characterize the pharmacokinetics of the oral irreversible ErbB family blocker afatinib in patients with solid tumors and to investigate the impact of selected intrinsic and extrinsic factors. Data from 927 patients (4,460 plasma concentrations) with advanced solid tumors in 7 Phase II or III studies were analyzed. Afatinib was administered orally in continuous 3 or 4 week cycles (starting dose 20, 40 or 50 mg once-daily). Plasma concentration-time data for up to 7 months dosing were analyzed using nonlinear mixed-effects modeling. The pharmacokinetic profile of afatinib was best described by a two-compartment disposition model with first-order absorption and linear elimination. There was a slightly more than proportional increase in exposure with increasing dose, accounted for by a dose-dependent relative bioavailability. For the therapeutic dose of 40 mg, the estimated apparent total clearance and distribution volume at steady state were 734 mL/min and 2,370 L, respectively. While food intake, body weight, gender, Eastern Cooperative Oncology Group performance score, renal function, and the level of alkaline phosphatase, lactate dehydrogenase or total protein were statistically significant covariates influencing afatinib exposure, none resulted in a proportional change in exposure of more than 27.8 % in a typical patient at model extremes (2.5th and 97.5th percentiles of baseline values for continuous covariates). In simulations of the individual covariate effects, none caused a change in the typical profile exceeding the observed variability range (90 % prediction interval) of afatinib. This population pharmacokinetic model adequately described the pharmacokinetics of afatinib in different cancer patient populations and therefore can be used for simulations exploring covariate effects and possible dose adaptations. The effect size for each of the individual covariates is not considered clinically relevant.

Fitting Birth-Death Processes to Panel Data with Applications to Bacterial DNA Fingerprinting.

Continuous-time linear birth-death-immigration (BDI) processes are frequently used in ecology and epidemiology to model stochastic dynamics of the population of interest. In clinical settings, multiple birth-death processes can describe disease trajectories of individual patients, allowing for estimation of the effects of individual covariates on the birth and death rates of the process. Such estimation is usually accomplished by analyzing patient data collected at unevenly spaced time points, referred to as panel data in the biostatistics literature. Fitting linear BDI processes to panel data is a nontrivial optimization problem because birth and death rates can be functions of many parameters related to the covariates of interest. We propose a novel expectation-maximization (EM) algorithm for fitting linear BDI models with covariates to panel data. We derive a closed-form expression for the joint generating function of some of the BDI process statistics and use this generating function to reduce the E-step of the EM algorithm, as well as calculation of the Fisher information, to one-dimensional integration. This analytical technique yields a computationally efficient and robust optimization algorithm that we implemented in an open-source R package. We apply our method to DNA fingerprinting of Mycobacterium tuberculosis, the causative agent of tuberculosis, to study intrapatient time evolution of IS6110 copy number, a genetic marker frequently used during estimation of epidemiological clusters of Mycobacterium tuberculosis infections. Our analysis reveals previously undocumented differences in IS6110 birth-death rates among three major lineages of Mycobacterium tuberculosis, which has important implications for epidemiologists that use IS6110 for DNA fingerprinting of Mycobacterium tuberculosis.

Population pharmacokinetics of palivizumab, a humanized anti-respiratory syncytial virus monoclonal antibody, in adults and children.

Although it has been on the market for over a decade, confusion remains regarding the pharmacokinetics (PK) and optimal dosing of palivizumab, a humanized IgG1κ monoclonal antibody indicated for the prevention of serious lower respiratory tract disease caused by respiratory syncytial virus (RSV) in pediatric patients at high risk of RSV disease. The objectives of this analysis were to characterize the population PK of palivizumab in adults and children using nonlinear mixed-effect modeling, quantify the effects of individual covariates on variability in palivizumab disposition, and compare palivizumab exposures for various dosing scenarios. Palivizumab PK data from 22 clinical studies were used for model development. The model was developed using a two-stage approach: (i) a 2-compartment model with first-order absorption after intramuscular administration was fitted to adult data, and (ii) the same structural model was fitted to the sparse pediatric data using the NONMEM $PRIOR subroutine, with informative priors obtained from the adult analysis. Body weight and an age descriptor that combines gestational age and postnatal age (PAGE) using an asymptotic-exponential model best described palivizumab clearance in pediatric patients. Palivizumab clearance increased slightly from 10.2 ml/day to 11.9 ml/day as a function of PAGE ranging from 7 to 18 months. Covariate analysis indicated a 20% higher clearance in children with chronic lung disease and in children with antidrug antibody titer values of ≥80. These covariates did not substantially explain interindividual variability. In the label-indicated pediatric population, body weight was the primary demographic factor affecting palivizumab PK. Body weight-based dosing of 15 mg/kg yields similar palivizumab concentrations in children of different gestational and postnatal ages. Simulations demonstrated that there was little difference in palivizumab PK between healthy term and premature infants. Simulations also demonstrated that the 5 monthly palivizumab doses of 15 mg/kg, consistent with the label and studied in two randomized, clinical trials, provided greater and more prolonged palivizumab exposure than did an abbreviated dosing regimen of 3 monthly doses.

Multiple treatment comparisons in a series of anti-malarial trials with an ordinal primary outcome and repeated treatment evaluations

BackgroundArtemisinin-based combination therapies (ACT) are widely used in African countries, including Cameroon. Between 2005 and 2007, five randomized studies comparing different treatment arms among artesunate-amodiaquine and other ACT were conducted in Cameroonian children aged two to 60 months who had uncomplicated Plasmodium falciparum malaria. In these studies, the categorical criterion proposed by the World Health Organization (WHO) to assess the relative effectiveness of anti-malarial drugs was repeatedly evaluated on Days 14, 21 and 28 after treatment initiation. The aim of the present study was to compare the effects of different treatments on this repeated ordinal outcome, hence using the fully available information.MethodsThe quantitative synthesis was based on individual patient data. Due to the incomplete block design concerning treatment arms between different trials, a mixed treatment comparison (MTC) meta-analysis approach was adopted. The repeated ordinal outcome was modelled through a latent variable, as a proportional odds mixed model with trial, period and treatment arms as covariates. The model was further complexified to account for the variance heterogeneity, and the individual log-residual variance was modelled as a linear mixed model, as well. The effects of individual covariates at inclusion, such as parasitaemia, fever, gender and weight, were also tested. Model parameters were estimated using a Bayesian approach via the WinBUGS software. After selecting the best model using Deviance Information Criterion (DIC), mixed treatment comparisons were based on the estimated treatment effects.ResultsModeling the residual variance improved the model ability to adjust the data. The results showed that, compared to artesunate-amodiaquine (ASAQ), dihydroartemisinin-piperaquine (DHPP) was significantly more efficacious. Artesunate-chlorproguanil-dapsone (ASCD) was less efficacious than artesunate-sulphadoxine-pyrimethamine (ASSP), artemether-lumefantrine (AMLM) and DHPP, the difference with the latter being significant. No difference in efficacy was found between ASAQ and AMLM.ConclusionsBayesian mixed treatment comparisons of a network of connected randomized trials with repeated measurements of the primary categorical outcome allowed to take into account both the individual- and between- studies sources of heterogeneity. The results of the present study complete the previous quantitative review based on a binary outcome at a fixed time point, suggesting that DHPP represents an alternative for the treatment of uncomplicated P. falciparum malaria in Cameroonian children.

To move or not to move to find a new job: spatial duration time model with dynamic covariate effects

The aim of this paper is to show the flexibility and capacity of penalized spline smoothing as estimation routine for modelling duration time data. We analyse the unemployment behaviour in Germany between 2000 and 2004 using a massive database from the German Federal Employment Agency. To investigate dynamic covariate effects and differences between competing job markets depending on the distance between former and recent working place, a functional duration time model with competing risks is used. It is build upon a competing hazard function where some of the smooth covariate effects are allowed to vary with unemployment duration. The focus of our analysis is on contrasting the spatial, economic and individual covariate effects of the competing job markets and on analysing their general influence on the unemployed's re-employment probabilities. As a result of our analyses, we reveal differences concerning gender, age and education. We also discover an effect between the newly formed and the old West German states. Moreover, the spatial pattern between the considered job markets differs.

Territory settlement and site fidelity in Reed Buntings Emberiza schoeniclus

Capsule Site fidelity and territory settlement is probably limited by the occupancy of suitable sites by males and the high mortality in females. Aim To investigate the relationship between territory settlement and site fidelity in Reed Buntings. Methods Two hundred and seventy Reed Buntings were colour‐ringed at five fishponds in South Bohemia, the Czech Republic, in 2000–2009. Model selection was used to test the effect of individual covariates (sex, age, weight and body measurements) on apparent survival and recapture probabilities. Results There is a correlation between territory settlement and site fidelity. Older males were more frequently re‐sighted in the study area during the year of ringing. Moreover, males were more frequently recovered in subsequent breeding seasons. An effect of body measurements as well as body weight did not appear evident. An effect of breeding performance in previous seasons on site fidelity was not found. Conclusion There is a connection between territory settlement and site fidelity in the early‐arriving Reed Buntings. Site fidelity and territory settlement is probably limited by the occupancy of suitable sites by males and the high mortality in females most likely caused by the extended breeding season in the study area. Moreover, territory settlement, and marginally, site fidelity is associated with the previous experience of males. Forming pre‐breeding flocks probably provides an addition of nutrition reserves after spring arrival. High competition between males and lack of suitable breeding habitats, owing to occupancy by older residents, indicated the crucial importance of small fragmented wetlands to this species in this study area.

Extended Bayesian LASSO for multiple quantitative trait loci mapping and unobserved phenotype prediction.

The Bayesian LASSO (BL) has been pointed out to be an effective approach to sparse model representation and successfully applied to quantitative trait loci (QTL) mapping and genomic breeding value (GBV) estimation using genome-wide dense sets of markers. However, the BL relies on a single parameter known as the regularization parameter to simultaneously control the overall model sparsity and the shrinkage of individual covariate effects. This may be idealistic when dealing with a large number of predictors whose effect sizes may differ by orders of magnitude. Here we propose the extended Bayesian LASSO (EBL) for QTL mapping and unobserved phenotype prediction, which introduces an additional level to the hierarchical specification of the BL to explicitly separate out these two model features. Compared to the adaptiveness of the BL, the EBL is "doubly adaptive" and thus, more robust to tuning. In simulations, the EBL outperformed the BL in regard to the accuracy of both effect size estimates and phenotypic value predictions, with comparable computational time. Moreover, the EBL proved to be less sensitive to tuning than the related Bayesian adaptive LASSO (BAL), which introduces locus-specific regularization parameters as well, but involves no mechanism for distinguishing between model sparsity and parameter shrinkage. Consequently, the EBL seems to point to a new direction for QTL mapping, phenotype prediction, and GBV estimation.

Adult-type soft tissue sarcomas in paediatric age: A nomogram-based prognostic comparison with adult sarcoma

Introduction The study compares the outcome of paediatric patients with adult-type soft tissue sarcomas with that reported for adults, taking into account the effect of established prognostic factors. Methods The actual mortality of our series was compared with that predicted by the nomogram developed for adults at the Memorial Sloan Kettering Cancer Center. From a previously-published series of 182 patients <18 years, 112 cases fulfilling the criteria for the nomogram application were selected. Results Actual 10-year mortality for the series was 29%, compared with a 16% predicted mortality. The effect of individual covariates was qualitatively consistent with that of adults, but the unfavourable prognostic effect of tumour size was stronger in paediatric cases. Conclusion The variables known to have a prognostic role in adults are relevant also in children. The worse outcome observed in our series might be explained by the stronger adverse effect of tumour size in young patients.

Population pharmacokinetic analysis of mirtazapine.

Mirtazapine belongs to the new generation of antidepressants that is commonly used in clinical routine. Therefore, we feel it mandatory to control compliance in the context of non-response, adverse events or other clinical situations by means of plasma concentration measurements. While controlled clinical studies have evaluated the effect of individual covariates on the pharmacokinetics of mirtazapine, our analysis aims to identify covariates within a naturalistic clinical setting. We performed non-linear mixed-effects modelling with data from 65 depressed inpatients whose plasma concentrations were measured weekly during their stay in hospital. Each patient's age, height, weight, co-medication, alcohol, coffee and cigarette consumption, weekly serum creatinine concentrations, liver enzyme activity, blood pressure and pulse was noted. From 49 patients, the genotype of cytochrome P450 (CYP) isoenzymes 2D6, 2C9 and 2C19 was analysed. The clearance of CYP2D6 intermediate metabolisers was reduced by 26% compared with extensive metabolisers. No other factor significantly influenced the clearance of these patients. The variability of mirtazapine plasma concentrations in clinical routine is caused to a relevant degree by CYP2D6. This should be taken into account when therapeutic drug monitoring is carried out to check treatment adherence or when a special clinical situation, such as co-morbidity and add-on medication, demands careful dosing of this drug.

Estimation of population size based on additive hazards models for continuous-time recapture experiments.

We use the semiparametric additive hazards model to formulate the effects of individual covariates on the capture rates in the continuous-time capture-recapture experiment, and then construct a Horvitz-Thompson-type estimator for the unknown population size. The resulting estimator is consistent and asymptotically normal with an easily estimated variance. Simulation studies show that the asymptotic approximations are adequate for practical use when the average capture probabilities exceed .5. Ignoring covariates would underestimate the population size and the coverage probability is poor. A wildlife example is provided.