Search

Life history analysis has evolved in the last 50 years as a methodology for analyzing processes associated with human health, education, employment, and other areas. The complexity of many processes, the difficulty of obtaining complete and accurate data, and the increased use of observational data from registries and administrative sources have posed many recent challenges. We review the evolution of life history analysis, discuss some recent work, and consider three areas currently receiving much attention. A theme we stress is the use of expanded models that include selection and observation processes for studies in addition to the life history process of interest. Examples from health research are presented. [ABSTRACT FROM AUTHOR]

Tests for variance or scale effects due to covariates are used in many areas and recently, in genomic and genetic association studies. We study score tests based on location-scale models with arbitrary error distributions that allow incorporation of additional adjustment covariates. Tests based on Gaussian and Laplacian double generalized linear models are examined in some detail. Numerical properties of the tests under Gaussian and other error distributions are examined. Our results show that the use of model-based asymptotic distributions with score tests for scale effects does not control type 1 error well in many settings of practical relevance. We consider simple statistics based on permutation distribution approximations, which correspond to well-known statistics derived by another approach. They are shown to give good type 1 error control under different error distributions and under covariate distribution imbalance. The methods are illustrated through a differential gene expression analysis involving breast cancer tumor samples. [ABSTRACT FROM AUTHOR]

Multistate models provide a powerful framework for the analysis of life history processes when the goal is to characterize transition intensities, transition probabilities, state occupancy probabilities, and covariate effects thereon. Data on such processes are often only available at random visit times occurring over a finite period. We formulate a joint multistate model for the life history process, the recurrent visit process, and a random loss to follow-up time at which the visit process terminates. This joint model is helpful when discussing the independence conditions necessary to justify the use of standard likelihoods involving the life history model alone and provides a basis for analyses that accommodate dependence. We consider settings with disease-driven visits and routinely scheduled visits and develop likelihoods that accommodate partial information on the types of visits. Simulation studies suggest that suitably constructed joint models can yield consistent estimates of parameters of interest even under dependent visit processes, providing the models are correctly specified; identifiability and estimability issues are also discussed. An application is given to a cohort of individuals attending a rheumatology clinic where interest lies in progression of joint damage. [ABSTRACT FROM AUTHOR]

Two‐phase, response‐dependent sampling is often used in regression settings that involve expensive covariate measurements. Conditional maximum likelihood (CML) is an attractive approach in many cases as it avoids modelling of the covariate distribution, unlike full maximum likelihood. Scott & Wild (2011) introduced an augmented CML approach which is semi‐parametric efficient in certain settings with a discrete response variable. We consider general regression models and show the Scott–Wild estimator of covariate effects has the same asymptotic efficiency as two empirical likelihood estimators, and that these estimators dominate the CML estimator. We compare the efficiencies of various estimators in simulation studies and illustrate the methodology in a two‐phase genetics study. [ABSTRACT FROM AUTHOR]

In health research interest often lies in modeling a failure time process but in many cohort studies failure status is only determined at scheduled assessment times. While the assessment times may be fixed upon study entry, individuals may become lost to follow‐up and miss visits subsequent to the time of loss to follow‐up. We consider a three‐state model to characterize a joint failure and loss to follow‐up process, and use it to investigate the impact of dependent loss to follow‐up on standard parametric, nonparametric, and semiparametric analysis. The effect of dependent loss to follow‐up is mitigated by fitting the joint model. The performance of standard methods is studied using the asymptotic theory of misspecified models, and the finite sample performance is examined for the standard and joint analyses through simulation studies. An application to data from a youth smoking prevention study is presented for illustration. [ABSTRACT FROM AUTHOR]

A framework is proposed for the joint modeling of life history and loss to follow-up (LTF) processes in cohort studies. This framework provides a basis for discussing independence conditions for LTF and censoring and examining the implications of dependent LTF. We consider failure time and more general life history processes. The joint models are based on multistate processes with expanded state spaces encompassing both the life history and LTF processes. Tracing studies are discussed as a means of investigating the presence of dependent censoring and providing valid estimates of transition intensities and state occupancy probabilities. Simulation studies and an illustration based on a cohort of individuals with systemic lupus erythematosus demonstrate the usefulness and properties of the proposed methods. [ABSTRACT FROM AUTHOR]

Failure time studies based on observational cohorts often have to deal with irregular intermittent observation of individuals, which produces interval-censored failure times. When the observation times depend on factors related to a person's failure time, the failure times may be dependently interval censored. Inverse-intensity-of-visit weighting methods have been developed for irregularly observed longitudinal or repeated measures data and recently extended to parametric failure time analysis. This article develops nonparametric estimation of failure time distributions using weighted generalized estimating equations and monotone smoothing techniques. Simulations are conducted for examination of the finite sample performance of proposed estimators. This research is motivated in part by the Toronto Psoriatic Arthritis Cohort Study, and the proposed methodology is applied to this study. [ABSTRACT FROM AUTHOR]

Event history studies based on disease clinic data often face several complications. Specifically, patients may visit the clinic irregularly, and the intermittent observation times could depend on disease-related variables; this can cause a failure time outcome to be dependently interval-censored. We propose a weighted estimating function approach so that dependently interval-censored failure times can be analysed consistently. A so-called inverse-intensity-of-visit weight is employed to adjust for the informative inspection times. Left truncation of failure times can also be easily handled. Additionally, in observational studies, treatment assignments are typically non-randomized and may depend on disease-related variables. An inverse-probability-of-treatment weight is applied to estimating functions to further adjust for measured confounders. Simulation studies are conducted to examine the finite sample performances of the proposed estimators. Finally, the Toronto Psoriatic Arthritis Cohort Study is used for illustration. Copyright © 2017 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

In this article, the author reflects on problems related to potential of big data in increasing scientific understanding. Topics discussed include use of big databases for auxiliary information, importance of using macro or aggregate data in econometric for enhancing conclusion from micro data on individuals and differences in potential efficiency gains across covariates while using case-control settings.

Response-dependent sampling is widely used in settings where certain variables are expensive to obtain. Estimation has been thoroughly investigated but recent applications have emphasized tests of association for expensive covariates and a response variable. We consider testing and provide easily implemented likelihood score tests for generalized linear models under a broad range of sampling plans. We show that when there are no additional covariates, the score statistics are identical for conditional and full likelihood approaches, and are of the same form as for ordinary random sampling. Applications in genetics are discussed briefly. [ABSTRACT FROM AUTHOR]

Intensity‐based multistate models provide a useful framework for characterizing disease processes, the introduction of interventions, loss to followup, and other complications arising in the conduct of randomized trials studying complex life history processes. Within this framework we discuss the issues involved in the specification of estimands and show the limiting values of common estimators of marginal process features based on cumulative incidence function regression models. When intercurrent events arise we stress the need to carefully define the target estimand and the importance of avoiding targets of inference that are not interpretable in the real world. This has implications for analyses, but also the design of clinical trials where protocols may help in the interpretation of estimands based on marginal features. [ABSTRACT FROM AUTHOR]

Life history studies collect information on events and other outcomes during people's lifetimes. For example, these may be related to childhood development, education, fertility, health, or employment. Such longitudinal studies have constraints on the selection of study members, the duration and frequency of follow-up, and the accuracy and completeness of information obtained. These constraints, along with factors associated with the definition and measurement of certain outcomes, affect our ability to understand, model, and analyze life history processes. My objective here is to discuss and illustrate some issues associated with the design and analysis of life history studies. Copyright © 2013 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Abstract: Many two-phase sampling designs have been applied in practice to obtain efficient estimates of regression parameters while minimizing the cost of data collection. This research investigates two-phase sampling designs for so-called expensive variable problems, and compares them with one-phase designs. Closed form expressions for the asymptotic relative efficiency of maximum likelihood estimators from the two designs are derived for parametric normal models, providing insight into the available information for regression coefficients under the two designs. We further discuss when we should apply the two-phase design and how to choose the sample sizes for two-phase samples. Our numerical study indicates that the results can be applied to more general settings. [Copyright &y& Elsevier]

Recurrent event data arise commonly in medical and public health studies. The analysis of such data has received extensive research attention and various methods have been developed in the literature. Depending on the focus of scientific interest, the methods may be broadly classified as intensity-based counting process methods, mean function-based estimating equation methods, and the analysis of times to events or times between events. These methods and models cover a wide variety of practical applications. However, there is a critical assumption underlying those methods-variables need to be correctly measured. Unfortunately, this assumption is frequently violated in practice. It is quite common that some covariates are subject to measurement error. It is well known that covariate measurement error can substantially distort inference results if it is not properly taken into account. In the literature, there has been extensive research concerning measurement error problems in various settings. However, with recurrent events, there is little discussion on this topic. It is the objective of this paper to address this important issue. In this paper, we develop inferential methods which account for measurement error in covariates for models with multiplicative intensity functions or rate functions. Both likelihood-based inference and robust inference based on estimating equations are discussed. The Canadian Journal of Statistics 40: 530-549; 2012 © 2012 Statistical Society of Canada [ABSTRACT FROM AUTHOR]

The authors comment on a study focusing on survey sampling and estimation. They cite previous studies proposing various ways for addressing missing data or for "strengthening" estimation of a target parameter using auxiliary data. They argue that while calibration can effectively incorporate auxiliary information into estimation of a target parameter, the approach appears less than transparent with respect to implicit assumptions, efficiency, or sensitivity to departures from assumptions.

In many chronic disease processes subjects are at risk of two or more types of events. We describe a bivariate mixed Poisson model in which a copula function is used to model the association between two gamma distributed random effects. The resulting model is a bivariate negative binomial process in which each type of event arises from a negative binomial process. Methods for parameter estimation are described for parametric and semiparametric models based on an EM algorithm. We also consider the issue of event-dependent censoring based on one type of event, which arises when one event is sufficiently serious that its occurence may influence the decision of whether to withdraw a patient from a study. The asymptotic biases of estimators of rate and mean functions from naive marginal analyses are discussed, as well as associated treatment effects. Because the joint model is fit based on a likelihood, consistent estimates are obtained. Simulation studies are carried out to evaluate the empirical performance of the proposed estimators with independent and event-dependent censoring and applications to a trial of breast cancer patients with skeletal metastases and a study of patients with chronic obstructive pulmonary disease illustrate the approach. Copyright © 2010 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

When statistical models are used to predict the values of unobserved random variables, loss functions are often used to quantify the accuracy of a prediction. The expected loss over some specified set of occasions is called the prediction error. This paper considers the estimation of prediction error when regression models are used to predict survival times and discusses the use of these estimates. Extending the previous work, we consider both point and confidence interval estimations of prediction error, and allow for variable selection and model misspecification. Different estimators are compared in a simulation study for an absolute relative error loss function, and results indicate that cross-validation procedures typically produce reliable point estimates and confidence intervals, whereas model-based estimates are sensitive to model misspecification. Links between performance measures for point predictors and for predictive distributions of survival times are also discussed. The methodology is illustrated in a medical setting involving survival after treatment for disease. Copyright © 2009 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

In clinical trials featuring recurrent clinical events, the definition and estimation of treatment effects involves a number of interesting issues, especially when loss to follow-up may be event-related and when terminal events such as death preclude the occurrence of further events. This paper discusses a clinical trial of breast cancer patients with bone metastases where the recurrent events are skeletal complications, and where patients may die during the trial. We argue that treatment effects should be based on marginal rate and mean functions. When recurrent event data are subject to event-dependent censoring, however, ordinary marginal methods may yield inconsistent estimates. Incorporating correctly specified inverse probability of censoring weights into analyses can protect against dependent censoring and yield consistent estimates of marginal features. An alternative approach is to obtain estimates of rate and mean functions from models that involve some conditioning to render censoring conditionally independent. We consider three methods of estimating mean functions of recurrent event processes and examine the bias and efficiency of unweighted and inverse probability weighted versions of the methods with and without a terminating event. We compare the methods via simulation and use them to analyse the data from the breast cancer trial. [ABSTRACT FROM AUTHOR]

There has been extensive interest in discussing inference methods for survival data when some covariates are subject to measurement error. It is known that standard inferential procedures produce biased estimation if measurement error is not taken into account. With the Cox proportional hazards model a number of methods have been proposed to correct bias induced by measurement error, where the attention centers on utilizing the partial likelihood function. It is also of interest to understand the impact on estimation of the baseline hazard function in settings with mismeasured covariates. In this paper we employ a weakly parametric form for the baseline hazard function and propose simple unbiased estimating functions for estimation of parameters. The proposed method is easy to implement and it reveals the connection between the naive method ignoring measurement error and the corrected method with measurement error accounted for. Simulation studies are carried out to evaluate the performance of the estimators as well as the impact of ignoring measurement error in covariates. As an illustration we apply the proposed methods to analyze a data set arising from the Busselton Health Study [Knuiman, M.W., Cullent, K.J., Bulsara, M.K., Welborn, T.A., Hobbs, M.S.T., 1994. Mortality trends, 1965 to 1989, in Busselton, the site of repeated health surveys and interventions. Austral. J. Public Health 18, 129–135]. [Copyright &y& Elsevier]

nalyse de survie fondée sur le modèle des risques proportionnels et sur des données d'enqugte Les auteurs proposent des méthodes fondées sur une version stratifiée du modèle des risques proportionnels de Cox et adaptées aux situations où les données sont issues d'un plan d'échantillonnage complexe. Les méthodes qu'ils proposent s'appuient sur la théorie des équations d'estimation et celle des processus empiriques. Les auteurs discutent également des plans d'échantillonnage non infoimatifs en relation avec les méthodes pondérées et non pondérées. Ils illustrent leur méthodologie au moyen de l'Enquěte sur la dynamique du travail et du revenu de Statistique Canada. Ils abordent brièvement les questions relatives aux poids d'échantillonnage, aux méthodes d'adéquation et aux données manquantes ou entachées d'erreurs de mesure. Us mentionnent en outre quelques problèmes ouverts. [ABSTRACT FROM AUTHOR]

Une note concernant les durées de vie censurées par intervalle et la condition à somme constante de Oller, Gómez & Calle (2004) Oller, Gómez & Calle (2004) proposent une condition à somme constante pour certains processus générant des durées de vie censurées par intervalle. Ils montrent que dans les modèles répondant à cette condition, il est possible d'estimer de façon non paramétrique la loi du temps de survie au moyen d'une vraisemblance simplifiée bien connue. L'auteur montre que cette condition à somme constante équivaut à supposer l'existence d'un processus d'observation indépendant des durées de vie pour lequel la loi des données est la měme que sous le processus réel sous-jacent. [ABSTRACT FROM AUTHOR]

This article presents methodology for multivariate proportional hazards (PH)regression models. The methods employ flexible piecewise constant or spline specifications for baseline hazard functions in either marginal or conditional PH models, along with assumptions about the association among lifetimes. Because the models are parametric, ordinary maximum likelihood can be applied; it is able to deal easily with such data features as interval censoring or sequentially observed lifetimes, unlike existing semiparametric methods. A bivariate Clayton model (1978, Biometrika 65,141-151)is used to illustrate the approach taken. Because a parametric assumption about association is made, efficiency and robustness comparisons are made between estimation based on the bivariate Clayton model and "working independence" methods that specify only marginal distributions for each lifetime variable. [ABSTRACT FROM AUTHOR]

Les progrès de Finformatique et des technologies de [information ont profondément influencé beaucoup d'activités humaines. L'emploi d'outils statistiques se répand rapidement, dans des secteurs souvent nouveaux et non conventionnels. Le champ d'activité des statisticiens est en pleine expansion, mais ils n'en ont plus l'exclusivité. L'auteur décrit les défis qui en résultent et en tire des leçons concernant le travail du statisticien, sa formation et son sentiment d'appartenance à la profession. [ABSTRACT FROM AUTHOR]

Focuses on a nonparametric estimation of a lifetime distribution data in which censoring times for unfailed units are missing. Estimation of failure-time distributions from product warranty data; Distribution function of a lifetime variable; Nonparametric method of lifetime distribution by Suzuki and Kasashima; Effect of using an estimate of the censoring-time distribution.

In epidemiology, reliability and other areas one encounters situations where response times and covariates for individuals in a population are observed only if the response time does not exceed an associated censoring time. Methods for obtaining supplementary information about covariates and censoring times for nonresponding individuals are considered. Estimation procedures are developed for the resulting combination of truncated data and supplementary data. Theoretical and simulation results are given, and the methodology illustrated on some automobile failure data collected under a warranty plan. [ABSTRACT FROM AUTHOR]

Plusieurs situations pratiques impliquent une variable de réponse Y et des covariables [ABSTRACT FROM AUTHOR]

Plusieurs méthodes ont été proposées en vue de traiter le problème de la variation extra-poissonnienne dans une analyse de régression pour données de dénombrement. Cet article a pour objet l'étude de modèles de régression binomiale négative et se penche sur les propriétés d'efficacité et de robustesse des méthodes inférentielles découlant des modèles. Ces dernières sont comparées aux méthodes de quasi-vraisemblancce. [ABSTRACT FROM AUTHOR]

This article investigates data on recurrent events that arise from sources such as warranty claims, where the observation period for a unit is unknown until it experiences at least one event. This creates a type of truncation in the data. We consider nonparametric estimation of means and rates of the event occurrences with such "zero-truncated" data and examine the case where the population size and the distribution of observation times across units are at least approximately known. We study the behaviors of the proposed estimators by simulation. We examine some car warranty data by applying the methodology developed here and considering the underlying assumptions. [ABSTRACT FROM AUTHOR]

Disease progression is often monitored by intermittent follow‐up "visits" in longitudinal cohort studies, resulting in interval‐censored failure time outcomes. Furthermore, the timing and frequency of visits is often found related to a person's history of disease‐related variables in practice. This article develops a semiparametric estimation approach using weighted binomial regression and a kernel smoother to analyze interval‐censored failure time data. Visit times are allowed to be subject‐specific and outcome‐dependent. We consider a collection of widely used semiparametric regression models, including additive hazards and linear transformation models. For additive hazards models, the nonparametric component has a closed‐form estimator and the estimators of regression coefficients are shown to be asymptotically multivariate normal with sandwich‐type covariance matrices. Simulations are conducted to examine the finite sample performance of the proposed estimators. A data set from the Toronto Psoriatic Arthritis (PsA) Cohort Study is used to illustrate the proposed methodology. [ABSTRACT FROM AUTHOR]

The HostSeq initiative recruited 10,059 Canadians infected with SARS-CoV-2 between March 2020 and March 2023, obtained clinical information on their disease experience and whole genome sequenced (WGS) their DNA. We analyzed the WGS data for genetic contributors to severe COVID-19 (considering 3,499 hospitalized cases and 4,975 non-hospitalized after quality control). We investigated the evidence for replication of loci reported by the International Host Genetics Initiative (HGI); analyzed the X chromosome; conducted rare variant gene-based analysis and polygenic risk score testing. Population stratification was adjusted for using meta-analysis across ancestry groups. We replicated two loci identified by the HGI for COVID-19 severity: the LZTFL1/SLC6A20 locus on chromosome 3 and the FOXP4 locus on chromosome 6 (the latter with a variant significant at P < 5E-8). We found novel significant associations with MRAS and WDR89 in gene-based analyses, and constructed a polygenic risk score that explained 1.01% of the variance in severe COVID-19. This study provides independent evidence confirming the robustness of previously identified COVID-19 severity loci by the HGI and identifies novel genes for further investigation. Author summary: Host genetics determine how human genetics contribute to the response to infectious disease. HostSeq is a Canada-wide effort to contribute to our understanding of host genetics for COVID-19. The HostSeq study involves genetic and clinical data from individuals who tested positive for SARS-CoV-2 across Canada. This work examines locations in the human genome that have been reported to be involved in severe COVID-19 worldwide and determines if we also see the involvement of those genomic locations in Canadians. This work also explores several 'genetic quantities' such as a determination of how much of COVID-19 severity is due to genetics in Canada. [ABSTRACT FROM AUTHOR]

For regression with covariates missing not at random where the missingness depends on the missing covariate values, complete‐case (CC) analysis leads to consistent estimation when the missingness is independent of the response given all covariates, but it may not have the desired level of efficiency. We propose a general empirical likelihood framework to improve estimation efficiency over the CC analysis. We expand on methods in Bartlett et al. (2014, Biostatistics15, 719–730) and Xie and Zhang (2017, Int J Biostat13, 1–20) that improve efficiency by modeling the missingness probability conditional on the response and fully observed covariates by allowing the possibility of modeling other data distribution‐related quantities. We also give guidelines on what quantities to model and demonstrate that our proposal has the potential to yield smaller biases than existing methods when the missingness probability model is incorrect. Simulation studies are presented, as well as an application to data collected from the US National Health and Nutrition Examination Survey. [ABSTRACT FROM AUTHOR]

Multi-type recurrent event data arise when two or more different kinds of events may occur repeatedly over a period of observation. The scientific objectives in such settings are often to describe features of the marginal processes and to study the association between the different types of events. Interval-censored multi-type recurrent event data arise when the precise event times are unobserved, but intervals are available during which the events are known to have occurred. This type of data is common in studies of patients with advanced cancer, for example, where the events may represent the development of different types of metastatic lesions which are only detectable by conducting bone scans of the entire skeleton. In this setting it is of interest to characterize the incidence of the various types of bone lesions, to estimate the impact of treatment and other covariate effects on the development of new lesions, and to understand the relationship between the processes generating the bone lesions. We develop joint models for multi-type interval-censored recurrent events which accommodate dependencies between different types of events and enable one to examine the covariate effects via regression. However, since the marginal likelihood resulting from the multivariate random effect model is intractable, we describe a Gibbs sampling algorithm to facilitate model fitting and inference. We use generalized estimating equations for estimation and inference based on marginal models. The finite sample properties of the marginal approach are studied via simulation. The estimates of both the regression coefficients and the variance-covariance parameters are shown to have negligible bias and 95 per cent confidence intervals based on the asymptotic variance formula are shown to have excellent empirical coverage probabilities in all of the settings considered. The application of these methods to data from a trial of women with advanced breast cancer provides insight into the clinical course of bone metastases in this population. [ABSTRACT FROM AUTHOR]