Central venous catheterization is a standard procedure in intensive care therapy. In developing countries, this intervention is frequently performed by physicians in training and without the availability of ultrasound guidance. Purpose of this study was to determine the incidence and potential risk factors for mechanical complications during central venous catheterization in an intensive care setting performed by a mixed group of practitioners without the use of adjunct ultrasound.
To introduce a method to optimize structural retinal nerve fiber layer (RNFL) models based on glaucomatous visual field data and to show how such an optimized model can be used to reduce noise in visual fields while probably preserving clinically important features.
Nowadays, treatment regimens for cancer often involve a combination of drugs. The determination of the doses of each of the combined drugs in phase I dose escalation studies poses methodological challenges. The most common phase I design, the classic '3+3' design, has been criticized for poorly estimating the maximum tolerated dose (MTD) and for treating too many subjects at doses below the MTD. In addition, the classic '3+3' is not able to address the challenges posed by combinations of drugs. Here, we assume that a control drug (commonly used and well-studied) is administered at a fixed dose in combination with a new agent (the experimental drug) of which the appropriate dose has to be determined. We propose a randomized design in which subjects are assigned to the control or to the combination of the control and experimental. The MTD is determined using a model-based Bayesian technique based on the difference of probability of dose limiting toxicities (DLT) between the control and the combination arm. We show, through a simulation study, that this approach provides better and more accurate estimates of the MTD. We argue that this approach may differentiate between an extreme high probability of DLT observed from the control and a high probability of DLT of the combination. We also report on a fictive (simulation) analysis based on published data of a phase I trial of ifosfamide combined with sunitinib.
Austerity measures and health-system redesign to minimise hospital expenditures risk adversely affecting patient outcomes. The RN4CAST study was designed to inform decision making about nursing, one of the largest components of hospital operating expenses. We aimed to assess whether differences in patient to nurse ratios and nurses' educational qualifications in nine of the 12 RN4CAST countries with similar patient discharge data were associated with variation in hospital mortality after common surgical procedures.
A systematic review and meta-analyses were performed to identify the risk factors associated with carbapenem-resistant Pseudomonas aeruginosa and to identify sources and reservoirs for the pathogen. A systematic search of PubMed and Embase databases from 1 January 1987 until 27 January 2012 identified 1,662 articles, 53 of which were included in a systematic review and 38 in a random-effects meta-analysis study. The use of carbapenem, use of fluoroquinolones, use of vancomycin, use of other antibiotics, having medical devices, intensive care unit (ICU) admission, having underlying diseases, patient characteristics, and length of hospital stay were significant risk factors in multivariate analyses. The meta-analyses showed that carbapenem use (odds ratio [OR] = 7.09; 95% confidence interval [CI] = 5.43 to 9.25) and medical devices (OR = 5.11; 95% CI = 3.55 to 7.37) generated the highest pooled estimates. Cumulative meta-analyses showed that the pooled estimate of carbapenem use was stable and that the pooled estimate of the risk factor "having medical devices" increased with time. We conclude that our results highlight the importance of antibiotic stewardship and the thoughtful use of medical devices in helping prevent outbreaks of carbapenem-resistant P. aeruginosa.
Aortic gradient and aortic regurgitation are echocardiographic markers of aortic valve function. Both are biomarkers repeatedly measured in patients with valve abnormalities, and thus, it is expected that they are biologically interrelated. Loss of follow-up could be caused by multiple reasons, including valve progression related, such as an intervention or even the death of the patient. In that case, it would be of interest and appropriate to analyze these outcomes jointly. Joint models have recently received much attention because they cover a wide range of clinical applications and have promising results. We propose a joint model consisting of two longitudinal outcomes, one continuous (aortic gradient) and one ordinal (aortic regurgitation), and two time-to-events (death and reoperation). Moreover, we allow for more flexibility for the average evolution and the subject-specific profiles of the continuous repeated outcome by using B-splines. A disadvantage, however, is that when adopting a non-linear structure for the model, we may have difficulties when interpreting the results. To overcome this problem, we propose a graphical approach. In this paper, we apply the proposed joint models under the Bayesian framework, using a data set including serial echocardiographic measurements of aortic gradient and aortic regurgitation and measurements of the occurrence of death and reoperation in patients who received a human tissue valve in the aortic position. The interpretation of the results will be discussed.
Joint modelling techniques have seen great advances in the recent years, with several types of joint models having been developed in literature that can handle a wide range of applications. This special issue of Statistical Methods in Medical Research presents some recent developments from this field. This introductory article contains some background material and highlights the contents of the contributions.
Although crime victimisation is as prevalent in psychiatric patients as crime perpetration (and possibly more so), few European figures for it are available. We therefore assessed its one-year prevalence and incident rates in Dutch severely mentally ill outpatients, and compared the results with victimisation rates in the general population.
Classic regression is based on certain assumptions that conflict with visual field (VF) data. We investigate and evaluate different regression models and their assumptions in order to determine point-wise VF progression in glaucoma and to better predict future field loss for personalised clinical glaucoma management.
In clinical trials, it is frequently of interest to estimate the time between the onset of two events (e.g. duration of response in oncology). Here, we consider the case where subjects are assessed at fixed visits but the initial event and the terminating event occur in between visits. This type of data, called doubly interval censored, is often analyzed with standard survival techniques, assuming either that the survival time (between initial and terminating event) is known exactly or is single interval censored. We introduce a motivating dataset in which the interest is to evaluate the impact of the treatment on the duration of response endpoint. We review the existing approaches and discuss their limitations with respect to the characteristics of our motivating dataset. Furthermore, we propose a stochastic EM algorithm that overcomes the problems in the existing approaches. We show by simulations the finite sample properties of our approach.
Genome-wide association studies have become very popular in identifying genetic contributions to phenotypes. Millions of SNPs are being tested for their association with diseases and traits using linear or logistic regression models. This conceptually simple strategy encounters the following computational issues: a large number of tests and very large genotype files (many Gigabytes) which cannot be directly loaded into the software memory. One of the solutions applied on a grand scale is cluster computing involving large-scale resources. We show how to speed up the computations using matrix operations in pure R code.
To optimize the planning of blood donations but also to continue motivating the volunteers it is important to streamline the practical organization of the timing of donations. While donors are asked to return for donation after a suitable period, still a relevant proportion of blood donors is deferred from donation each year due to a too low hemoglobin level. Rejection of donation may demotivate the candidate donor and implies an inefficient planning of the donation process. Hence, it is important to predict the future hemoglobin level to improve the planning of donors visits to the blood bank.
The analysis of growth curves of children can be done on either the original scale or in standard deviation scores. The first approach is found in many statistical textbooks, while the second approach is common in endocrinology, for instance in the evaluation of the effect of growth hormone in children that are born small for gestational age that remain small later in childhood. We illustrate here that the second approach may involve more complex modeling and hence a worse model fit.
Purpose: This controlled clinical trial aimed to compare the 3-year outcomes of glass fiber posts and composite cores with gold alloy-based posts and cores for the restoration of endodontically treated teeth. Materials and Methods: One hundred forty-four patients in need of 205 restorations on endodontically treated teeth were selected and followed for 7 to 37 months (mean: 21 ± 9 months). The teeth were primarily stratified based on the remaining tissue available to restore the tooth core with or without a post. Then, randomization allocated the teeth to either test group 1 (prefabricated glass fiber posts), test group 2 (custom-made glass fiber posts), or test group 3 (composite cores without posts). The control group consisted of gold alloy-based posts and cores. All posts/cores were covered with all-ceramic single crowns. Failures were either absolute, such as root fractures or irreparable fractures of the post/core, or relative, such as loss of post retention or reparable fractures of the core. Success and survival probability lifetime curves, corrected for clustering, were drawn for the entire data set. Results: The recall rate at 3 years was 97.1%. Absolute failures consisted of two root fractures and one endodontic failure, while relative failures included three instances of retention loss of the post/core and one post fracture. Because of the low number of events, no statistical tests were performed. The success and survival probabilities over all groups together at 3 years amounted to 91.7% and 97.2%, respectively. Conclusions: After being followed for up to 3 years, both cast gold and composite post and core systems performed well clinically. Longer follow-up times are needed to detect possible significant differences. Int J Prosthodont 2011;24:363-372.
We conducted this observational study to investigate tissue oxygen saturation during a vascular occlusion test in relationship with the condition of peripheral circulation and outcome in critically ill patients.
Psychotic conditions and especially schizophrenia, have been associated with increased morbidity and mortality. Many studies are performed in specialized settings with a strong focus on schizophrenia. Somatic comorbidity after psychosis is studied, using a general practice comorbidity registration network.
Logistic random effects models are a popular tool to analyze multilevel also called hierarchical data with a binary or ordinal outcome. Here, we aim to compare different statistical software implementations of these models.
Lives saved predictions are used to quantify the impact of certain remedial measures in nurse staffing and patient safety research, giving an indication of the potential gain in patient safety. Data collected in nurse staffing and patient safety are often multilevel in structure, requiring statistical techniques to account for clustering in the data.
The objective of this study was to verify how valid misclassification measurements obtained from a pre-survey calibration exercise are by comparing them to validation scores obtained in field conditions. Validation data were collected from the Smile for Life project, an oral health intervention study in Flemish children. A calibration exercise was organized under pre-survey conditions (32 age-matched children examined by eight examiners and the benchmark scorer). In addition, using a pre-determined sampling scheme blinded to the examiners, the benchmark scorer re-examined between six and 11 children screened by each of the dentists during the survey. Factors influencing sensitivity and specificity for scoring caries experience (CE) were investigated, including examiner, tooth type, surface type, tooth position (upper/lower jaw, right/left side) and validation setting (pre-survey versus field). In order to account for the clustering effect in the data, a generalized estimating equations approach was applied. Sensitivity scores were influenced not only by the calibration setting (lower sensitivity in field conditions, p?0.01), but also by examiner, tooth type (lower sensitivity in molar teeth, p?0.01) and tooth position (lower sensitivity in the lower jaw, p?0.01). Factors influencing specificity were examiner, tooth type (lower specificity in molar teeth, p?0.01) and surface type (the occlusal surface with a lower specificity than other surfaces) but not the validation setting. Misclassification measurements for scoring CE are influenced by several factors. In this study, the validation setting influenced sensitivity, with lower scores obtained when measuring data validity in field conditions. Results obtained in a pre-survey calibration setting need to be interpreted with caution and do not (always) reflect the actual performance of examiners during the field work.
Count data often exhibit overdispersion. One type of overdispersion arises when there is an excess of zeros in comparison with the standard Poisson distribution. Zero-inflated Poisson and hurdle models have been proposed to perform a valid likelihood-based analysis to account for the surplus of zeros. Further, data often arise in clustered, longitudinal or multiple-membership settings. The proper analysis needs to reflect the design of a study. Typically random effects are used to account for dependencies in the data. We examine the h-likelihood estimation and inference framework for hurdle models with random effects for complex designs. We extend the h-likelihood procedures to fit hurdle models, thereby extending h-likelihood to truncated distributions. Two applications of the methodology are presented.
We have developed a method to longitudinally classify subjects into two or more prognostic groups using longitudinally observed values of markers related to the prognosis. We assume the availability of a training data set where the subjects allocation into the prognostic group is known. The proposed method proceeds in two steps as described earlier in the literature. First, multivariate linear mixed models are fitted in each prognostic group from the training data set to model the dependence of markers on time and on possibly other covariates. Second, fitted mixed models are used to develop a discrimination rule for future subjects. Our method improves upon existing approaches by relaxing the normality assumption of random effects in the underlying mixed models. Namely, we assume a heteroscedastic multivariate normal mixture for random effects. Inference is performed in the Bayesian framework using the Markov chain Monte Carlo methodology. Software has been written for the proposed method and it is freely available. The methodology is applied to data from the Dutch Primary Biliary Cirrhosis Study.
Designing combination drug phase I trials has become increasingly complex, due to the increasing diversity in classes of agents, mechanisms of action, safety profiles and drug-administration schedules. With approximately 850 agents currently in development for cancer treatment, it is evident that combination development must be prioritised, as based on a specific hypothesis, as well as a projected development path for the involved combination. In this manuscript the most relevant issues and pitfalls for combination drug phase I trial design are discussed. Several phase I study designs that incorporate controls to circumvent bias due to imbalances in observed background toxicity are discussed.
Progression-related endpoints (such as time to progression or progression-free survival) and time to death are common endpoints in cancer clinical trials. It is of interest to study the link between progression-related endpoints and time to death (e.g. to evaluate the degree of surrogacy). However, current methods ignore some aspects of the definitions of progression-related endpoints. We review those definitions and investigate their impact on modeling the joint distribution. Further, we propose a multi-state model in which the association between the endpoints is modeled through a frailty term. We also argue that interval-censoring needs to be taken into account to more closely match the latent disease evolution. The joint distribution and an expression for Kendalls tau are derived. The model is applied to data from a clinical trial in advanced metastatic ovarian cancer.
This paper is a report of a cost-effectiveness analysis from a hospital perspective of increased nurse staffing levels (to the level of the 75th percentile) in Belgian general cardiac postoperative nursing units.
Peginterferon (PEG-IFN) results in HBeAg loss combined with virologic response in only a minority of patients with HBeAg positive chronic hepatitis B. Baseline predictors of response to PEG-IFN include HBV-genotype, pre-treatment HBV DNA levels, and ALT. The aims of this study were to develop a model, which improves the baseline prediction of response to PEG-IFN for individual patients by including early HBV DNA measurements during treatment and to establish an early indication for cessation of treatment. One hundred thirty-six patients treated with PEG-IFN were included in the study. Response was defined as loss of HBeAg and HBV DNA <10,000 copies/ml at 26 weeks post-treatment. Logistic regression analysis techniques were used to develop a dynamic prediction model with HBV DNA during the first 32 weeks of therapy. An early clinically useful rule for dis(continuation) of treatment was identified with a grid of cut-off values of HBV DNA decline during treatment. Adding HBV DNA decline to baseline prediction increased c-statistics from 0.846 to 0.857, 0.855 to 0.866 at weeks 4, 12, and 24. A HBV DNA decline of at least 2 log(10) within 24 weeks was strongly associated with response when added to the baseline prediction model: OR 5.7 (95% CI: 1.70-20.0; P = 0.004). A dynamic model including HBV DNA decline during treatment provides more accurate predictions of response to PEG-IFN. The model strongly supports individual decision making on treatment (dis)continuation in patients with HBeAg positive chronic hepatitis B. It is recommended that PEG-IFN treatment is stopped by 24 weeks if HBV DNA declined <2 log(10).
Important long-term health problems have been described after severe paediatric trauma. The International Classification of Functioning (ICF) was developed as a universal framework to describe that health. We evaluated outcome in children after severe trauma (defined as: hospitalised >48 h) by means of a questionnaire based on this ICF construct (IROS). Questionnaires were sent to children; one year after this trauma and to control children without any previous severe trauma. We created propensity score-matched pairs (n = 133) and evaluated differences in health perception. IROS characteristics were investigated by means of Item Response Theory models. We then estimated the health state of each individual based on his/her response pattern (factor score z01) and investigated the effect of selected covariates with simple linear regression. Significant odds ratios for differences between matched groups (p < 0.05) were observed for among others emotional problems, mobility, societal life and family burden, but not for chronic pain. Children in the trauma group showed, e.g. significant more physician (estimated relative risk R 1.7) and psychologist (R 3.5) visits. IROS primarily provides information from medium to high health burden levels and factor scores ranged from 0.41 (lowest) to 0.967 (highest burden). A significant impact on health burden could only be proven for the state at discharge (p = 0.015), although there was a tendency towards worse factor scores for children that were older, had a higher Injury Severity Score or after traffic injury. In conclusion, we showed that the burden of health problems for children and families after severe trauma is still high and physical, as well as psychosocial in nature. The health state at discharge seems to predict long-term outcome, which might be of importance in view of, e.g. trajectory assistance. IROS may provide an improved scoring system to evaluate outcome after (paediatric) injury or critical illness.
Surgeons realize that safe and efficient care processes for total joint replacement requires more than just well-performed operations. Orthopaedic teams are reorganizing care process to improve efficacy and shorten length of stay. Little is known on the impact of organizational changes on patient outcome. This paper studies the relation between the organization of care processes and patient outcomes in hip and knee. Clinical pathways are used as one of the methods to structure the care process. Although evidence is available on the effect of pathways in total joint replacement, their impact with the organization of the care process has not been studied previously.
Matrix metalloproteinases (MMP)-13 can initiate bone resorption and activate proMMP-9 in vitro, and both these MMPs have been widely implicated in tissue destruction associated with chronic periodontitis. We studied whether MMP-13 activity and TIMP-1 levels in gingival crevicular fluid (GCF) associated with progression of chronic periodontitis assessed clinically and by measuring carboxy-terminal telopeptide of collagen I (ICTP) levels. We additionally addressed whether MMP-13 could potentiate gelatinase activation in diseased gingival tissue.
Traditional approaches to the analysis of dmfs/DMFS count data pose analytical challenges, considering the increasing proportion of zeroes in the distribution. The aim of this paper was to predict the probability of caries-free subjects and the dependence of dmfs index on the influence of childhood sociodemographic factors, through the application of regression models.
The split-mouth design is a popular design in oral health research. In the most common split-mouth study, each of two treatments are randomly assigned to either the right or left halves of the dentition. The attractiveness of the design is that it removes a lot of inter-individual variability from the estimates of the treatment effect. However, already about 20 years ago the pitfalls of the design have been reported in the oral health literature. Yet, many clinicians are not aware of the potential problems with the split-mouth design. Further, it is our experience that most statisticians are not even aware of the existence of this design. Since most of the critical remarks appeared in the oral health literature, we argue that it is necessary to introduce the split-mouth design to a statistical audience, so that both clinicians and statisticians clearly understand the advantages, limitations, statistical considerations, and implications of its use in clinical trials and advise them on its use in practice.
Longitudinal studies often generate incomplete response patterns according to a missing not at random mechanism. Shared parameter models provide an appealing framework for the joint modelling of the measurement and missingness processes, especially in the nonmonotone missingness case, and assume a set of random effects to induce the interdependence. Parametric assumptions are typically made for the random effects distribution, violation of which leads to model misspecification with a potential effect on the parameter estimates and standard errors. In this article we avoid any parametric assumption for the random effects distribution and leave it completely unspecified. The estimation of the model is then made using a semi-parametric maximum likelihood method. Our proposal is illustrated on a randomized longitudinal study on patients with rheumatoid arthritis exhibiting nonmonotone missingness.
This study aimed to unravel the multidimensional profile of stroke outcomes by investigating the global correlation structure of motor, functional, and emotional problems of patients, as well as their caregivers strain, at 6 months after stroke. Potential differential associations based on patients level of functioning on admission to the rehabilitation center were analyzed.
A history of caries in the primary molars is associated with an advanced emergence of their permanent successors. Hence, caries in the primary molars may have an impact on the emergence order of the permanent teeth. The aim of the present study was to fully investigate the variability in permanent tooth emergence, taking into account the (caries) status of the primary molars.
In most multicenter studies that examine the relationship between nurse staffing and patient safety, nurse-staffing levels are measured per hospital. This can obscure relationships between staffing and outcomes at the unit level and lead to invalid inferences.
The performance of a diagnostic test is often expressed in terms of sensitivity and specificity compared with the reference standard. Calculations of sensitivity and specificity commonly involve multiple observations per patient, which implies that the data are clustered. Whether analysis of sensitivity and specificity per patient or using multiple observations per patient is preferable depends on the clinical context and consequences. The purpose of this article was to discuss and illustrate the most common statistical methods that calculate sensitivity and specificity of clustered data, adjusting for the possible correlation between observations within each patient. This tutorial presents and illustrates the following methods: (a) analysis at different levels ignoring correlation, (b) variance adjustment, (c) logistic random-effects models, and (d) generalized estimating equations. The choice of method and the level of reporting should correspond with the clinical decision problem. If multiple observations per patient are relevant to the clinical decision problem, the potential correlation between observations should be explored and taken into account in the statistical analysis. Supplemental material: http://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.12120509/-/DC1.
(1) To describe the levels of implicit rationing of nursing care in Swiss acute care hospitals; (2) to explore the associations between nine selected potential rationing predictors and implicit rationing of nursing care.
Genome-wide association studies are characterized by a huge number of statistical tests performed to discover new disease-related genetic variants [in the form of single-nucleotide polymorphisms (SNPs)] in human DNA. Many SNPs have been identified for cross-sectionally measured phenotypes. However, there is a growing interest in genetic determinants of the evolution of traits over time. Dealing with correlated observations from the same individual, we need to apply advanced statistical techniques. The linear mixed model is popular but also much more computationally demanding than fitting a linear regression model to independent observations. We propose a conditional two-step approach as an approximate method to explore the longitudinal relationship between the trait and the SNP. In a simulation study, we compare several fast methods with respect to their accuracy and speed. The conditional two-step approach is applied to relate SNPs to longitudinal bone mineral density responses collected in the Rotterdam Study.
In a follow-up study, patients are monitored over time. Longitudinal and time-to-event studies are the two most important types of a follow-up study. In this paper, the focus is on longitudinal studies with a continuous response where patients are examined at several time points. While longitudinal studies provide a powerful tool for the evaluation of a treatment effect over time, a major problem is missing data caused, for example, by patients who drop out from the study. Many longitudinal studies in rheumatology use inappropriate statistical methodology because either they do not address correctly the correlated nature of the repeated measurements, or they treat the problem of missing data incorrectly. We will illustrate that there are interpretational and computational issues with the "classical" approaches. Further, we expand here on more appropriate statistical techniques to analyze longitudinal studies. To this end, we focus on randomized controlled trials (RCTs) and illustrate the approaches on data from a fictive randomized controlled trial in rheumatology.
Nurses work environments are associated with burnout experiences among nurses. The RN4CAST project provides data on these constructs within a four-level structure (nurse, nursing unit, hospital, and country), implying more complicated multilevel analysis strategies than have been used in previous efforts studying this relationship.
Several studies have concluded that the use of nurses time and energy is often not optimized. Given widespread migration of nurses from developing to developed countries, it is important for human resource planning to know whether nursing education in developing countries is associated with more exaggerated patterns of inefficiency.
Kappa-like agreement indexes are often used to assess the agreement among examiners on a categorical scale. They have the particularity of correcting the level of agreement for the effect of chance. In the present paper, we first define two agreement indexes belonging to this family in a hierarchical context. In particular, we consider the cases of a random and fixed set of examiners. Then, we develop a method to evaluate the influence of factors on these indexes. Agreement indexes are directly related to a set of covariates through a hierarchical model. We obtain the posterior distribution of the model parameters in a Bayesian framework. We apply the proposed approach on dental data and compare it with the generalized estimating equations approach.
An important target of many clinical studies is to identify biomarkers, including risk scores, with strong prognostic capabilities. While biomarker evaluations are commonly utilized to predict the progress of the disease at single time points, appropriate statistical tools to assess the prognostic value of serial biomarker evaluation are rarely used. The goal of this paper is to demonstrate flexible and appropriate statistical methodology to assess the predictive capability of serial echocardiographic measurements of allograft aortic valve function. Moreover, the concept of joint modeling of longitudinal and survival data to optimally utilize the relationship between repeated valve function measurements and time-to-death or time-to-reoperation, is introduced and illustrated. Optimal and suboptimal methods are illustrated using a prospective cohort of patients who survived aortic valve or root replacement with an allograft valve and who were followed clinically and echocardiographically over time.
An incomplete miscarriage occurs when all the products of conception are not expelled through the cervix. Curettage or vacuum aspiration have been used to remove retained tissues. The anaesthetic techniques used to facilitate this procedure have not been systematically evaluated in order to determine which provide better outcomes to the patients.
Developmental adaptations due to early nutritional exposures may have permanent health consequences. Studies of diet and fetal size have mainly focused on individual nutrients despite evidence that the pattern of food consumption may be of significance. Hence, we evaluated the associations of dietary habits in early pregnancy (gestational age < 18 weeks) with fetal size, uteroplacental vascular resistance, placental weight and birth weight in a prospective observational study of 3207 Caucasian pregnant mothers in Rotterdam, the Netherlands. Participants completed a semiquantitative FFQ during early pregnancy. Logistic regression analysis was used to predict the occurrence of intra-uterine growth retardation at birth as a function of food intake. The derived solution was considered as the dietary pattern. As it was characterised by higher intakes of fruit, vegetables, vegetable oil, fish, pasta and rice, and lower intakes of meat, potatoes and fatty sauces, it was labelled the Mediterranean diet. The degree of adherence to the diet was positively associated with plasma folate and serum vitamin B12 concentrations and showed an inverse relationship with homocysteine and high-sensitivity C-reactive protein plasma concentrations (P <0·05). Important fetal size and placental parameters were associated with the degree of adherence to the diet, revealing a 72 g lower birth weight (95% CI -110·8, -33·3) and a 15 g lower placental weight (95% CI -29·8, -0·2) for women with low adherence to the diet. To conclude, low adherence to a Mediterranean diet in early pregnancy seems associated with decreased intra-uterine size with a lower placental and a lower birth weight.
The objective of finding a parsimonious representation of the observed data by a statistical model that is also capable of accurate prediction is commonplace in all domains of statistical applications. The parsimony of the solutions obtained by variable selection is usually counterbalanced by a limited prediction capacity. On the other hand, methodologies that assure high prediction accuracy usually lead to models that are neither simple nor easily interpretable. Regularization methodologies have proven to be useful in addressing both prediction and variable selection problems. The Bayesian approach to regularization constitutes a particularly attractive alternative as it is suitable for high-dimensional modeling, offers valid standard errors, and enables simultaneous estimation of regression coefficients and complexity parameters via computationally efficient MCMC techniques. Bayesian regularization falls within the versatile framework of Bayesian hierarchical models, which encompasses a variety of other approaches suited for variable selection such as spike and slab models and the MC(3) approach. In this article, we review these Bayesian developments and evaluate their variable selection performance in a simulation study for the classical small p large n setting. The majority of the existing Bayesian methodology for variable selection deals only with classical linear regression. Here, we present two applications in the contexts of binary and survival regression, where the Bayesian approach was applied to select markers prognostically relevant for the development of rheumatoid arthritis and for overall survival in acute myeloid leukemia patients.
Related JoVE Video
Journal of Visualized Experiments
What is Visualize?
JoVE Visualize is a tool created to match the last 5 years of PubMed publications to methods in JoVE's video library.
How does it work?
We use abstracts found on PubMed and match them to JoVE videos to create a list of 10 to 30 related methods videos.
Video X seems to be unrelated to Abstract Y...
In developing our video relationships, we compare around 5 million PubMed articles to our library of over 4,500 methods videos. In some cases the language used in the PubMed abstracts makes matching that content to a JoVE video difficult. In other cases, there happens not to be any content in our video library that is relevant to the topic of a given abstract. In these cases, our algorithms are trying their best to display videos with relevant content, which can sometimes result in matched videos with only a slight relation.