Robust Inferences For Covariate Effects On Survival Time With Censored Linear Regression Models

Various inference procedures for linear regression models with censored failure times have been studied extensively. Recent developments on efficient algorithms to implement these procedures enhance the practical usage of such models in survival analysis. In this article, we present robust inferences for certain covariate effects on the failure time in the presence of "nuisance" confounders under a semiparametric, partial linear regression setting. Specifically, the estimation procedures for the regression coefficients of interest are derived from a working linear model and are valid even when the function of the confounders in the model is not correctly specified. The new proposals are illustrated with two examples and their validity for cases with practical sample sizes is demonstrated via a simulation study.

On Causal Mediation Analysis with a Survival Outcome

Suppose that having established a marginal total effect of a point exposure on a time-to-event outcome, an investigator wishes to decompose this effect into its direct and indirect pathways, also know as natural direct and indirect effects, mediated by a variable known to occur after the exposure and prior to the outcome. This paper proposes a theory of estimation of natural direct and indirect effects in two important semiparametric models for a failure time outcome. The underlying survival model for the marginal total effect and thus for the direct and indirect effects, can either be a marginal structural Cox proportional hazards model, or a marginal structural additive hazards model. The proposed theory delivers new estimators for mediation analysis in each of these models, with appealing robustness properties. Specifically, in order to guarantee ignorability with respect to the exposure and mediator variables, the approach, which is multiply robust, allows the investigator to use several flexible working models to adjust for confounding by a large number of pre-exposure variables. Multiple robustness is appealing because it only requires a subset of working models to be correct for consistency; furthermore, the analyst need not know which subset of working models is in fact correct to report valid inferences. Finally, a novel semiparametric sensitivity analysis technique is developed for each of these models, to assess the impact on inference, of a violation of the assumption of ignorability of the mediator.

Predicting the evolution of Kaposi sarcoma, in the highly active antiretroviral therapy era

BACKGROUND: The outcome of Kaposi sarcoma varies. While many patients do well on highly active antiretroviral therapy, others have progressive disease and need chemotherapy. In order to predict which patients are at risk of unfavorable evolution, we established a prognostic score. METHOD: The survival analysis (Kaplan-Meier method; Cox proportional hazards models) of 144 patients with Kaposi sarcoma prospectively included in the Swiss HIV Cohort Study, from January 1996 to December 2004, was conducted. OUTCOME ANALYZED: use of chemotherapy or death. VARIABLES ANALYZED: demographics, tumor staging [T0 or T1 (16)], CD4 cell counts and HIV-1 RNA concentration, human herpesvirus 8 (HHV8) DNA in plasma and serological titers to latent and lytic antigens. RESULTS: Of 144 patients, 54 needed chemotherapy or died. In the univariate analysis, tumor stage T1, CD4 cell count below 200 cells/microl, positive HHV8 DNA and absence of antibodies against the HHV8 lytic antigen at the time of diagnosis were significantly associated with a bad outcome.Using multivariate analysis, the following variables were associated with an increased risk of unfavorable outcome: T1 [hazard ratio (HR) 5.22; 95% confidence interval (CI) 2.97-9.18], CD4 cell count below 200 cells/microl (HR 2.33; 95% CI 1.22-4.45) and positive HHV8 DNA (HR 2.14; 95% CI 1.79-2.85).We created a score with these variables ranging from 0 to 4: T1 stage counted for two points, CD4 cell count below 200 cells/microl for one point, and positive HHV8 viral load for one point. Each point increase was associated with a HR of 2.26 (95% CI 1.79-2.85). CONCLUSION: In the multivariate analysis, staging (T1), CD4 cell count (<200 cells/microl), positive HHV8 DNA in plasma, at the time of diagnosis, predict evolution towards death or the need of chemotherapy.

Early changes in biochemical markers of bone formation during teriparatide therapy correlate with improvements in vertebral strength in men with glucocorticoid-induced osteoporosis

Summary Changes of the bone formation marker PINP correlated positively with improvements in vertebral strength in men with glucocorticoid-induced osteoporosis (GIO) who received 18-month treatment with teriparatide, but not with risedronate. These results support the use of PINP as a surrogate marker of bone strength in GIO patients treated with teriparatide. Introduction To investigate the correlations between biochemical markers of bone turnover and vertebral strength estimated by finite element analysis (FEA) in men with GIO. Methods A total of 92 men with GIO were included in an 18-month, randomized, open-label trial of teriparatide (20 μg/day, n = 45) and risedronate (35 mg/week, n = 47). High-resolution quantitative computed tomography images of the 12th thoracic vertebra obtained at baseline, 6 and 18 months were converted into digital nonlinear FE models and subjected to anterior bending, axial compression and torsion. Stiffness and strength were computed for each model and loading mode. Serum biochemical markers of bone formation (amino-terminal-propeptide of type I collagen [PINP]) and bone resorption (type I collagen cross-linked C-telopeptide degradation fragments [CTx]) were measured at baseline, 3 months, 6 months and 18 months. A mixed-model of repeated measures analysed changes from baseline and between-group differences. Spearman correlations assessed the relationship between changes from baseline of bone markers with FEA variables. Results PINP and CTx levels increased in the teriparatide group and decreased in the risedronate group. FEA-derived parameters increased in both groups, but were significantly higher at 18 months in the teriparatide group. Significant positive correlations were found between changes from baseline of PINP at 3, 6 and 18 months with changes in FE strength in the teriparatide-treated group, but not in the risedronate group. Conclusions Positive correlations between changes in a biochemical marker of bone formation and improvement of biomechanical properties support the use of PINP as a surrogate marker of bone strength in teriparatide-treated GIO patients.

THE EFFECTS OF CURCUMIN ON BODY COMPOSITION OF PATIENTS WITH ADVANCED PANCREATIC CANCER

Cachexia is very common among patients with advanced pancreatic cancer and is a marker of poor prognosis. Weight loss in cachexia is due to both adipose and muscle compartments, and sarcopenia (severe muscle depletion) is associated with worse outcomes. Curcumin has shown a myriad of biological effects, including anti-cancer and anti-inflammatory. The ability of curcumin to attenuate cachexia and muscle loss has been tested in animal models, with conflicting results so far. The hypothesis of this study was that patients with advanced pancreatic cancer treated with curcumin for two months have less fat and muscle loss as compared to matched controls not treated with this compound. A matched 1:2 case-control retrospective study was conducted with 22 patients with pancreatic cancer who were treated with curcumin on a previous protocol and 44 untreated controls with the same diagnosis matched by age, gender, time from advanced cancer, body mass index, and number of prior therapies. Data was collected regarding oncologic treatment, medication use, weights, heights, and survival. Body composition was determined by computerized tomography analyses at two timepoints separated by 60±20 days. For treated patients, the first image was at the beginning of treatment and for controls it was determined by the matching time from advanced cancer. The evolution of body composition over time was quantitatively analyzed comparing both groups. All patients lost weight both due to fat and muscle losses, and patients treated with curcumin presented greater losses both in lean adipose body mass. Use of medications, chemotherapy, age, time from advanced cancer, baseline albumin, performance status, and number of prior therapies were not independently correlated with changes in body composition variables. Patients treated with curcumin had borderline shorter survival when compared with untreated patients. Sarcopenic treated patients had significantly shorter survival than non-sarcopenic counterparts, and sarcopenia status was not associated with survival among the controls. Treated patients with shorter survival showed a tendency to lose more lean and especially fat body mass as compared to untreated patients, maybe suggesting an effect of curcumin on shifting weight loss towards the end of life by impacting its mechanisms.

Recurrent stroke predictors differ in medically treated patients with pathogenic vs other PFOs

OBJECTIVE: To examine predictors of stroke recurrence in patients with a high vs a low likelihood of having an incidental patent foramen ovale (PFO) as defined by the Risk of Paradoxical Embolism (RoPE) score.METHODS: Patients in the RoPE database with cryptogenic stroke (CS) and PFO were classified as having a probable PFO-related stroke (RoPE score of >6, n = 647) and others (RoPE score of variables for associations with stroke recurrence using Cox survival models with component database as a stratification factor. An interaction with RoPE score was checked for the variables that were significant.RESULTS: Follow-up was available for 92%, 79%, and 57% at 1, 2, and 3 years. Overall, a higher recurrence risk was associated with an index TIA. For all other predictors, effects were significantly different in the 2 RoPE score categories. For the low RoPE score group, but not the high RoPE score group, older age and antiplatelet (vs warfarin) treatment predicted recurrence. Conversely, echocardiographic features (septal hypermobility and a small shunt) and a prior (clinical) stroke/TIA were significant predictors in the high but not low RoPE score group.CONCLUSION: Predictors of recurrence differ when PFO relatedness is classified by the RoPE score, suggesting that patients with CS and PFO form a heterogeneous group with different stroke mechanisms. Echocardiographic features were only associated with recurrence in the high RoPE score group.

Multi-channel search for squarks and gluinos in sqrts=7 TeV pp collisions with the ATLAS detector

A search for supersymmetric particles in final states with zero, one, and two leptons, with and without jets identified as originating from b-quarks, in 4.7 fb(-1) of root s = 7 TeV pp collisions produced by the Large Hadron Collider and recorded by the ATLAS detector is presented. The search uses a set of variables carrying information on the event kinematics transverse and parallel to the beam line that are sensitive to several topologies expected in supersymmetry. Mutually exclusive final states are defined, allowing a combination of all channels to increase the search sensitivity. No deviation from the Standard Model expectation is observed. Upper limits at 95 % confidence level on visible cross-sections for the production of new particles are extracted. Results are interpreted in the context of the constrained minimal supersymmetric extension to the Standard Model and in supersymmetry-inspired models with diverse, high-multiplicity final states.

Elucidating the Functional Relationship Between Working Memory Capacity and Psychometric Intelligence: A Fixed-Links Modeling Approach for Experimental Repeated-Measures Designs

Numerous studies reported a strong link between working memory capacity (WMC) and fluid intelligence (Gf), although views differ in respect to how close these two constructs are related to each other. In the present study, we used a WMC task with five levels of task demands to assess the relationship between WMC and Gf by means of a new methodological approach referred to as fixed-links modeling. Fixed-links models belong to the family of confirmatory factor analysis (CFA) and are of particular interest for experimental, repeated-measures designs. With this technique, processes systematically varying across task conditions can be disentangled from processes unaffected by the experimental manipulation. Proceeding from the assumption that experimental manipulation in a WMC task leads to increasing demands on WMC, the processes systematically varying across task conditions can be assumed to be WMC-specific. Processes not varying across task conditions, on the other hand, are probably independent of WMC. Fixed-links models allow for representing these two kinds of processes by two independent latent variables. In contrast to traditional CFA where a common latent variable is derived from the different task conditions, fixed-links models facilitate a more precise or purified representation of the WMC-related processes of interest. By using fixed-links modeling to analyze data of 200 participants, we identified a non-experimental latent variable, representing processes that remained constant irrespective of the WMC task conditions, and an experimental latent variable which reflected processes that varied as a function of experimental manipulation. This latter variable represents the increasing demands on WMC and, hence, was considered a purified measure of WMC controlled for the constant processes. Fixed-links modeling showed that both the purified measure of WMC (β = .48) as well as the constant processes involved in the task (β = .45) were related to Gf. Taken together, these two latent variables explained the same portion of variance of Gf as a single latent variable obtained by traditional CFA (β = .65) indicating that traditional CFA causes an overestimation of the effective relationship between WMC and Gf. Thus, fixed-links modeling provides a feasible method for a more valid investigation of the functional relationship between specific constructs.

Analysis of Phenotypic Variation in Childhood Wheezing Disorders

Recurrent wheezing or asthma is a common problem in children that has increased considerably in prevalence in the past few decades. The causes and underlying mechanisms are poorly understood and it is thought that a numb er of distinct diseases causing similar symptoms are involved. Due to the lack of a biologically founded classification system, children are classified according to their observed disease related features (symptoms, signs, measurements) into phenotypes. The objectives of this PhD project were a) to develop tools for analysing phenotypic variation of a disease, and b) to examine phenotypic variability of wheezing among children by applying these tools to existing epidemiological data. A combination of graphical methods (multivariate co rrespondence analysis) and statistical models (latent variables models) was used. In a first phase, a model for discrete variability (latent class model) was applied to data on symptoms and measurements from an epidemiological study to identify distinct phenotypes of wheezing. In a second phase, the modelling framework was expanded to include continuous variability (e.g. along a severity gradient) and combinations of discrete and continuo us variability (factor models and factor mixture models). The third phase focused on validating the methods using simulation studies. The main body of this thesis consists of 5 articles (3 published, 1 submitted and 1 to be submitted) including applications, methodological contributions and a review. The main findings and contributions were: 1) The application of a latent class model to epidemiological data (symptoms and physiological measurements) yielded plausible pheno types of wheezing with distinguishing characteristics that have previously been used as phenotype defining characteristics. 2) A method was proposed for including responses to conditional questions (e.g. questions on severity or triggers of wheezing are asked only to children with wheeze) in multivariate modelling.ii 3) A panel of clinicians was set up to agree on a plausible model for wheezing diseases. The model can be used to generate datasets for testing the modelling approach. 4) A critical review of methods for defining and validating phenotypes of wheeze in children was conducted. 5) The simulation studies showed that a parsimonious parameterisation of the models is required to identify the true underlying structure of the data. The developed approach can deal with some challenges of real-life cohort data such as variables of mixed mode (continuous and categorical), missing data and conditional questions. If carefully applied, the approach can be used to identify whether the underlying phenotypic variation is discrete (classes), continuous (factors) or a combination of these. These methods could help improve precision of research into causes and mechanisms and contribute to the development of a new classification of wheezing disorders in children and other diseases which are difficult to classify.

A significant effect of the KIR ligand HLA-C on fibrosis progression in chronic C hepatitis with or without liver transplantation.

BACKGROUND & AIMS The interaction of KIR with their HLA ligands drives the activation and inhibition of natural killer (NK) cells. NK cells could be implicated in the development of liver fibrosis in chronic hepatitis C. METHODS We analysed 206 non-transplanted and 53 liver transplanted patients, selected according to their Metavir fibrosis stage. Several variables such as the number of activator KIR or the HLA ligands were considered in multinomial and logistic regression models. Possible confounding variables were also investigated. RESULTS The KIRs were not significant predictors of the fibrosis stage. Conversely, a significant reduction of the HLA-C1C2 genotype was observed in the most advanced fibrosis stage group (F4) in both cohorts. Furthermore, the progression rate of fibrosis was almost 10 times faster in the subgroup of patients after liver transplantation and HLA-C1C2 was significantly reduced in this cohort compared to non-transplanted patients. CONCLUSION This study suggests a possible role of KIR and their ligands in the development of liver damage. The absence of C1 and C2 ligands heterozygosity could lead to less inhibition of NK cells and a quicker progression to a high level of fibrosis in patients infected by HCV, especially following liver transplantation. This article is protected by copyright. All rights reserved.

The Value of Waiting: Foreign Direct Investment with Uncertainty and Imperfect Local Knowledge

This paper examines the role of uncertainty and imperfect local knowledge in foreign direct investment. The main idea comes from the literature on investment under uncertainty, such as Pindyck (1991) and Dixit and Pindyck (1994). We empirically test .the value of waiting. with a dataset on foreign direct investment (FDI). Many factors (e.g., political and economic regulations) as well as uncertainty and the risks due to imperfect local knowledge, determine the attractiveness of FDI. The uncertainty and irreversibility of FDI links the time interval between permission and actual execution of such FDI with explanatory variables, including information on foreign (home) countries and domestic industries. Common factors, such as regulatory change and external shocks, may affect the uncertainty when foreign investors make irreversible FDI decisions. We derive testable hypotheses from models of investment under uncertainty to determine those possible factors that induce delays in FDI, using Korean data over 1962 to 2001.

Existence and Uniqueness of Equilibrium in Nonoptimal Unbounded Infinite Horizon Economies

In applied work in macroeconomics and finance, nonoptimal infinite horizon economies are often studied in the the state space is unbounded. Important examples of such economies are single vector growth models with production externalities, valued fiat money, monopolistic competition, and/or distortionary government taxation. Although sufficient conditions for existence and uniqueness of Markovian equilibrium are well known for the compact state space case, no similar sufficient conditions exist for unbounded growth. This paper provides such a set of sufficient conditions, and also present a computational algorithm that will prove asymptotically consistent when computing Markovian equilibrium.

Slope estimation in a parametric measurement error model: A simulation study

In regression analysis, covariate measurement error occurs in many applications. The error-prone covariates are often referred to as latent variables. In this proposed study, we extended the study of Chan et al. (2008) on recovering latent slope in a simple regression model to that in a multiple regression model. We presented an approach that applied the Monte Carlo method in the Bayesian framework to the parametric regression model with the measurement error in an explanatory variable. The proposed estimator applied the conditional expectation of latent slope given the observed outcome and surrogate variables in the multiple regression models. A simulation study was presented showing that the method produces estimator that is efficient in the multiple regression model, especially when the measurement error variance of surrogate variable is large.^

Mixture modeling for joint analysis of survival, discrete, and continuous data

Mixture modeling is commonly used to model categorical latent variables that represent subpopulations in which population membership is unknown but can be inferred from the data. In relatively recent years, the potential of finite mixture models has been applied in time-to-event data. However, the commonly used survival mixture model assumes that the effects of the covariates involved in failure times differ across latent classes, but the covariate distribution is homogeneous. The aim of this dissertation is to develop a method to examine time-to-event data in the presence of unobserved heterogeneity under a framework of mixture modeling. A joint model is developed to incorporate the latent survival trajectory along with the observed information for the joint analysis of a time-to-event variable, its discrete and continuous covariates, and a latent class variable. It is assumed that the effects of covariates on survival times and the distribution of covariates vary across different latent classes. The unobservable survival trajectories are identified through estimating the probability that a subject belongs to a particular class based on observed information. We applied this method to a Hodgkin lymphoma study with long-term follow-up and observed four distinct latent classes in terms of long-term survival and distributions of prognostic factors. Our results from simulation studies and from the Hodgkin lymphoma study demonstrated the superiority of our joint model compared with the conventional survival model. This flexible inference method provides more accurate estimation and accommodates unobservable heterogeneity among individuals while taking involved interactions between covariates into consideration.^

TIME SERIES ANALYSIS AS INPUT FOR PREDICTIVE MODELING: PREDICTING CARDIAC ARREST IN A PEDIATRIC INTENSIVE CARE UNIT

The first manuscript, entitled "Time-Series Analysis as Input for Clinical Predictive Modeling: Modeling Cardiac Arrest in a Pediatric ICU" lays out the theoretical background for the project. There are several core concepts presented in this paper. First, traditional multivariate models (where each variable is represented by only one value) provide single point-in-time snapshots of patient status: they are incapable of characterizing deterioration. Since deterioration is consistently identified as a precursor to cardiac arrests, we maintain that the traditional multivariate paradigm is insufficient for predicting arrests. We identify time series analysis as a method capable of characterizing deterioration in an objective, mathematical fashion, and describe how to build a general foundation for predictive modeling using time series analysis results as latent variables. Building a solid foundation for any given modeling task involves addressing a number of issues during the design phase. These include selecting the proper candidate features on which to base the model, and selecting the most appropriate tool to measure them. We also identified several unique design issues that are introduced when time series data elements are added to the set of candidate features. One such issue is in defining the duration and resolution of time series elements required to sufficiently characterize the time series phenomena being considered as candidate features for the predictive model. Once the duration and resolution are established, there must also be explicit mathematical or statistical operations that produce the time series analysis result to be used as a latent candidate feature. In synthesizing the comprehensive framework for building a predictive model based on time series data elements, we identified at least four classes of data that can be used in the model design. The first two classes are shared with traditional multivariate models: multivariate data and clinical latent features. Multivariate data is represented by the standard one value per variable paradigm and is widely employed in a host of clinical models and tools. These are often represented by a number present in a given cell of a table. Clinical latent features derived, rather than directly measured, data elements that more accurately represent a particular clinical phenomenon than any of the directly measured data elements in isolation. The second two classes are unique to the time series data elements. The first of these is the raw data elements. These are represented by multiple values per variable, and constitute the measured observations that are typically available to end users when they review time series data. These are often represented as dots on a graph. The final class of data results from performing time series analysis. This class of data represents the fundamental concept on which our hypothesis is based. The specific statistical or mathematical operations are up to the modeler to determine, but we generally recommend that a variety of analyses be performed in order to maximize the likelihood that a representation of the time series data elements is produced that is able to distinguish between two or more classes of outcomes. The second manuscript, entitled "Building Clinical Prediction Models Using Time Series Data: Modeling Cardiac Arrest in a Pediatric ICU" provides a detailed description, start to finish, of the methods required to prepare the data, build, and validate a predictive model that uses the time series data elements determined in the first paper. One of the fundamental tenets of the second paper is that manual implementations of time series based models are unfeasible due to the relatively large number of data elements and the complexity of preprocessing that must occur before data can be presented to the model. Each of the seventeen steps is analyzed from the perspective of how it may be automated, when necessary. We identify the general objectives and available strategies of each of the steps, and we present our rationale for choosing a specific strategy for each step in the case of predicting cardiac arrest in a pediatric intensive care unit. Another issue brought to light by the second paper is that the individual steps required to use time series data for predictive modeling are more numerous and more complex than those used for modeling with traditional multivariate data. Even after complexities attributable to the design phase (addressed in our first paper) have been accounted for, the management and manipulation of the time series elements (the preprocessing steps in particular) are issues that are not present in a traditional multivariate modeling paradigm. In our methods, we present the issues that arise from the time series data elements: defining a reference time; imputing and reducing time series data in order to conform to a predefined structure that was specified during the design phase; and normalizing variable families rather than individual variable instances. The final manuscript, entitled: "Using Time-Series Analysis to Predict Cardiac Arrest in a Pediatric Intensive Care Unit" presents the results that were obtained by applying the theoretical construct and its associated methods (detailed in the first two papers) to the case of cardiac arrest prediction in a pediatric intensive care unit. Our results showed that utilizing the trend analysis from the time series data elements reduced the number of classification errors by 73%. The area under the Receiver Operating Characteristic curve increased from a baseline of 87% to 98% by including the trend analysis. In addition to the performance measures, we were also able to demonstrate that adding raw time series data elements without their associated trend analyses improved classification accuracy as compared to the baseline multivariate model, but diminished classification accuracy as compared to when just the trend analysis features were added (ie, without adding the raw time series data elements). We believe this phenomenon was largely attributable to overfitting, which is known to increase as the ratio of candidate features to class examples rises. Furthermore, although we employed several feature reduction strategies to counteract the overfitting problem, they failed to improve the performance beyond that which was achieved by exclusion of the raw time series elements. Finally, our data demonstrated that pulse oximetry and systolic blood pressure readings tend to start diminishing about 10-20 minutes before an arrest, whereas heart rates tend to diminish rapidly less than 5 minutes before an arrest.