Improvement of Midpoint Imputation for Estimation of Median Survival Time for Interval-Censored Time-to-Event Data

BACKGROUND

Progression-free survival (PFS) is used to evaluate treatment effects in cancer clinical trials. Disease progression (DP) in patients is typically determined by radiological testing at several scheduled tumor-assessment time points. This produces a discrepancy between the true progression time and the observed progression time. When the observed progression time is considered as the true progression time, a positively biased PFS is obtained for some patients, and the estimated survival function derived by the Kaplan-Meier method is also biased.

METHODS

While the midpoint imputation method is available and replaces interval-censored data with midpoint data, it unrealistically assumes that several DPs occur at the same time point when several DPs are observed within the same tumor-assessment interval. We enhanced the midpoint imputation method by replacing interval-censored data with equally spaced timepoint data based on the number of observed interval-censored data within the same tumor-assessment interval.

RESULTS

The root mean square error of the median of the enhanced method is almost always smaller than that of the midpoint imputation regardless of the tumor-assessment frequency. The coverage probability of the enhanced method is close to the nominal confidence level of 95% in most scenarios.

CONCLUSION

We believe that the enhanced method, which builds upon the midpoint imputation method, is more effective than the midpoint imputation method itself.

Evaluation of the natural history of disease by combining incident and prevalent cohorts: application to the Nun Study

The Nun study is a well-known longitudinal epidemiology study of aging and dementia that recruited elderly nuns who were not yet diagnosed with dementia (i.e., incident cohort) and who had dementia prior to entry (i.e., prevalent cohort). In such a natural history of disease study, multistate modeling of the combined data from both incident and prevalent cohorts is desirable to improve the efficiency of inference. While important, the multistate modeling approaches for the combined data have been scarcely used in practice because prevalent samples do not provide the exact date of disease onset and do not represent the target population due to left-truncation. In this paper, we demonstrate how to adequately combine both incident and prevalent cohorts to examine risk factors for every possible transition in studying the natural history of dementia. We adapt a four-state nonhomogeneous Markov model to characterize all transitions between different clinical stages, including plausible reversible transitions. The estimating procedure using the combined data leads to efficiency gains for every transition compared to those from the incident cohort data only.

Marginal proportional hazards models for multivariate interval-censored data

Multivariate interval-censored data arise when there are multiple types of events or clusters of study subjects, such that the event times are potentially correlated and when each event is only known to occur over a particular time interval. We formulate the effects of potentially time-varying covariates on the multivariate event times through marginal proportional hazards models while leaving the dependence structures of the related event times unspecified. We construct the nonparametric pseudolikelihood under the working assumption that all event times are independent, and we provide a simple and stable EM-type algorithm. The resulting nonparametric maximum pseudolikelihood estimators for the regression parameters are shown to be consistent and asymptotically normal, with a limiting covariance matrix that can be consistently estimated by a sandwich estimator under arbitrary dependence structures for the related event times. We evaluate the performance of the proposed methods through extensive simulation studies and present an application to data from the Atherosclerosis Risk in Communities Study.

Pre-pregnancy weight, the rate of gestational weight gain, and the risk of early gestational diabetes mellitus among women registered in a tertiary care hospital in India

BACKGROUND

The impact of pre-pregnancy weight and the rate of gestational weight gain (GWG) together on the risk of early GDM (< 24 weeks gestation; eGDM) has not been studied in the Indian context. We aimed to study the influence of (1) pre-pregnancy weight on the risk of eGDM diagnosed in two time intervals; and (2) in addition, the rate of GWG by 12 weeks on the risk of eGDM diagnosed in 19-24 weeks.

METHOD

Our study utilized real-world clinical data on pregnant women routinely collected at an antenatal care clinic at a private tertiary hospital, in Pune, India. Women registering before 12 weeks of gestation (v1), with a singleton pregnancy, and having a follow-up visit between 19-24 weeks (v2) were included (n = 600). The oral glucose tolerance test was conducted universally as per Indian guidelines (DIPSI) at v1 and v2 for diagnosing eGDM. The data on the onset time of eGDM were interval censored; hence, we modeled the risk of eGDM using binomial regression to assess the influence of pre-pregnancy weight on the risk of eGDM in the two intervals. The rate of GWG by 12 weeks was added to assess its impact on the risk of eGDM diagnosed in v2.

RESULT

Overall, 89 (14.8%) women (age 32 ± 4 years) were diagnosed with eGDM by 24 weeks, of which 59 (9.8%) were diagnosed before 12 weeks and 30 of 541 (5.5%) women were diagnosed between 19-24 weeks. Two-thirds (66%) of eGDM were diagnosed before 12 weeks of gestation. Women's pre-pregnancy weight was positively associated with the risk of GDM in both time intervals though the lower confidence limit was below zero in v1. The rate of GWG by 12 weeks was not observed to be associated with the risk of eGDM diagnosed between 19-24 weeks of gestation. These associations were independent of age, height, and parity.

CONCLUSION

Health workers may focus on pre-pregnancy weight, a modifiable risk factor for eGDM. A larger community-based study measuring weight and GDM status more frequently may be warranted to deepen the understanding of the role of GWG as a risk factor for GDM.

Combined estimating equation approaches for the additive hazards model with left-truncated and interval-censored data

Interval-censored failure time data arise commonly in various scientific studies where the failure time of interest is only known to lie in a certain time interval rather than observed exactly. In addition, left truncation on the failure event may occur and can greatly complicate the statistical analysis. In this paper, we investigate regression analysis of left-truncated and interval-censored data with the commonly used additive hazards model. Specifically, we propose a conditional estimating equation approach for the estimation, and further improve its estimation efficiency by combining the conditional estimating equation and the pairwise pseudo-score-based estimating equation that can eliminate the nuisance functions from the marginal likelihood of the truncation times. Asymptotic properties of the proposed estimators are discussed including the consistency and asymptotic normality. Extensive simulation studies are conducted to evaluate the empirical performance of the proposed methods, and suggest that the combined estimating equation approach is obviously more efficient than the conditional estimating equation approach. We then apply the proposed methods to a set of real data for illustration.

A pairwise pseudo-likelihood approach for regression analysis of left-truncated failure time data with various types of censoring

BACKGROUND

Failure time data frequently occur in many medical studies and often accompany with various types of censoring. In some applications, left truncation may occur and can induce biased sampling, which makes the practical data analysis become more complicated. The existing analysis methods for left-truncated data have some limitations in that they either focus only on a special type of censored data or fail to flexibly utilize the distribution information of the truncation times for inference. Therefore, it is essential to develop a reliable and efficient method for the analysis of left-truncated failure time data with various types of censoring.

METHOD

This paper concerns regression analysis of left-truncated failure time data with the proportional hazards model under various types of censoring mechanisms, including right censoring, interval censoring and a mixture of them. The proposed pairwise pseudo-likelihood estimation method is essentially built on a combination of the conditional likelihood and the pairwise likelihood that eliminates the nuisance truncation distribution function or avoids its estimation. To implement the presented method, a flexible EM algorithm is developed by utilizing the idea of self-consistent estimating equation. A main feature of the algorithm is that it involves closed-form estimators of the large-dimensional nuisance parameters and is thus computationally stable and reliable. In addition, an R package LTsurv is developed.

RESULTS

The numerical results obtained from extensive simulation studies suggest that the proposed pairwise pseudo-likelihood method performs reasonably well in practical situations and is obviously more efficient than the conditional likelihood approach as expected. The analysis results of the MHCPS data with the proposed pairwise pseudo-likelihood method indicate that males have significantly higher risk of losing active life than females. In contrast, the conditional likelihood method recognizes this effect as non-significant, which is because the conditional likelihood method often loses some estimation efficiency compared with the proposed method.

CONCLUSIONS

The proposed method provides a general and helpful tool to conduct the Cox's regression analysis of left-truncated failure time data under various types of censoring.

Sieve estimation of a class of partially linear transformation models with interval-censored competing risks data

In this paper, we consider a class of partially linear transformation models with interval-censored competing risks data. Under a semiparametric generalized odds rate specification for the cause-specific cumulative incidence function, we obtain optimal estimators of the large number of parametric and nonparametric model components via maximizing the likelihood function over a joint B-spline and Bernstein polynomial spanned sieve space. Our specification considers a relatively simpler finite-dimensional parameter space, approximating the infinite-dimensional parameter space as n → ∞, thereby allowing us to study the almost sure consistency, and rate of convergence for all parameters, and the asymptotic distributions and efficiency of the finite-dimensional components. We study the finite sample performance of our method through simulation studies under a variety of scenarios. Furthermore, we illustrate our methodology via application to a dataset on HIV-infected individuals from sub-Saharan Africa.

Maximum likelihood estimation for length-biased and interval-censored data with a nonsusceptible fraction

Left-truncated data are often encountered in epidemiological cohort studies, where individuals are recruited according to a certain cross-sectional sampling criterion. Length-biased data, a special case of left-truncated data, assume that the incidence of the initial event follows a homogeneous Poisson process. In this article, we consider an analysis of length-biased and interval-censored data with a nonsusceptible fraction. We first point out the importance of a well-defined target population, which depends on the prior knowledge for the support of the failure times of susceptible individuals. Given the target population, we proceed with a length-biased sampling and draw valid inferences from a length-biased sample. When there is no covariate, we show that it suffices to consider a discrete version of the survival function for the susceptible individuals with jump points at the left endpoints of the censoring intervals when maximizing the full likelihood function, and propose an EM algorithm to obtain the nonparametric maximum likelihood estimates of nonsusceptible rate and the survival function of the susceptible individuals. We also develop a novel graphical method for assessing the stationarity assumption. When covariates are present, we consider the Cox proportional hazards model for the survival time of the susceptible individuals and the logistic regression model for the probability of being susceptible. We construct the full likelihood function and obtain the nonparametric maximum likelihood estimates of the regression parameters by employing the EM algorithm. The large sample properties of the estimates are established. The performance of the method is assessed by simulations. The proposed model and method are applied to data from an early-onset diabetes mellitus study.

Estimating duration distribution aided by auxiliary longitudinal measures in presence of missing time origin

Understanding the distribution of an event duration time is essential in many studies. The exact time to the event is often unavailable, and thus so is the full event duration. By linking relevant longitudinal measures to the event duration, we propose to estimate the duration distribution via the first-hitting-time model (e.g. Lee and Whitmore in Stat Sci 21(4):501-513, 2006). The longitudinal measures are assumed to follow a Wiener process with random drift. We apply a variant of the MCEM algorithm to compute likelihood-based estimators of the parameters in the longitudinal process model. This allows us to adapt the well-known empirical distribution function to estimate the duration distribution in the presence of missing time origin. Estimators with smooth realizations can then be obtained by conventional smoothing techniques. We establish the consistency and weak convergence of the proposed distribution estimator and present its variance estimation. We use a collection of wildland fire records from Alberta, Canada to motivate and illustrate the proposed approach. The finite-sample performance of the proposed estimator is examined by simulation. Viewing the available data as interval-censored times, we show that the proposed estimator can be more efficient than the well-established Turnbull estimator, an alternative that is often applied in such situations.

Regression models for interval censored data using parametric pseudo-observations

Background

Time-to-event data that is subject to interval censoring is common in the practice of medical research and versatile statistical methods for estimating associations in such settings have been limited. For right censored data, non-parametric pseudo-observations have been proposed as a basis for regression modeling with the possibility to use different association measures. In this article, we propose a method for calculating pseudo-observations for interval censored data.

Methods

We develop an extension of a recently developed set of parametric pseudo-observations based on a spline-based flexible parametric estimator. The inherent competing risk issue with an interval censored event of interest necessitates the use of an illness-death model, and we formulate our method within this framework. To evaluate the empirical properties of the proposed method, we perform a simulation study and calculate pseudo-observations based on our method as well as alternative approaches. We also present an analysis of a real dataset on patients with implantable cardioverter-defibrillators who are monitored for the occurrence of a particular type of device failures by routine follow-up examinations. In this dataset, we have information on exact event times as well as the interval censored data, so we can compare analyses of pseudo-observations based on the interval censored data to those obtained using the non-parametric pseudo-observations for right censored data.

Results

Our simulations show that the proposed method for calculating pseudo-observations provides unbiased estimates of the cumulative incidence function as well as associations with exposure variables with appropriate coverage probabilities. The analysis of the real dataset also suggests that our method provides estimates which are in agreement with estimates obtained from the right censored data.

Conclusions

The proposed method for calculating pseudo-observations based on the flexible parametric approach provides a versatile solution to the specific challenges that arise with interval censored data. This solution allows regression modeling using a range of different association measures.

Supplementary Information

The online version contains supplementary material available at (10.1186/s12874-021-01227-8).

Semiparametric regression on cumulative incidence function with interval-censored competing risks data and missing event types

Competing risk data are frequently interval-censored, that is, the exact event time is not observed but only known to lie between two examination time points such as clinic visits. In addition to interval censoring, another common complication is that the event type is missing for some study participants. In this article, we propose an augmented inverse probability weighted sieve maximum likelihood estimator for the analysis of interval-censored competing risk data in the presence of missing event types. The estimator imposes weaker than usual missing at random assumptions by allowing for the inclusion of auxiliary variables that are potentially associated with the probability of missingness. The proposed estimator is shown to be doubly robust, in the sense that it is consistent even if either the model for the probability of missingness or the model for the probability of the event type is misspecified. Extensive Monte Carlo simulation studies show good performance of the proposed method even under a large amount of missing event types. The method is illustrated using data from an HIV cohort study in sub-Saharan Africa, where a significant portion of events types is missing. The proposed method can be readily implemented using the new function ciregic_aipw in the R package intccr.

A novel calibration framework for survival analysis when a binary covariate is measured at sparse time points

The goals in clinical and cohort studies often include evaluation of the association of a time-dependent binary treatment or exposure with a survival outcome. Recently, several impactful studies targeted the association between initiation of aspirin and survival following colorectal cancer (CRC) diagnosis. The value of this exposure is zero at baseline and may change its value to one at some time point. Estimating this association is complicated by having only intermittent measurements on aspirin-taking. Commonly used methods can lead to substantial bias. We present a class of calibration models for the distribution of the time of status change of the binary covariate. Estimates obtained from these models are then incorporated into the proportional hazard partial likelihood in a natural way. We develop non-parametric, semiparametric, and parametric calibration models, and derive asymptotic theory for the methods that we implement in the aspirin and CRC study. We further develop a risk-set calibration approach that is more useful in settings in which the association between the binary covariate and survival is strong.

Estimating menarcheal age distribution from partially recalled data

In a cross-sectional study, adolescent and young adult females were asked to recall the time of menarche, if experienced. Some respondents recalled the date exactly, some recalled only the month or the year of the event, and some were unable to recall anything. We consider estimation of the menarcheal age distribution from this interval-censored data. A complicated interplay between age-at-event and calendar time, together with the evident fact of memory fading with time, makes the censoring informative. We propose a model where the probabilities of various types of recall would depend on the time since menarche. For parametric estimation, we model these probabilities using multinomial regression function. Establishing consistency and asymptotic normality of the parametric maximum likelihood estimator requires a bit of tweaking of the standard asymptotic theory, as the data format varies from case to case. We also provide a non-parametric maximum likelihood estimator, propose a computationally simpler approximation, and establish the consistency of both these estimators under mild conditions. We study the small sample performance of the parametric and non-parametric estimators through Monte Carlo simulations. Moreover, we provide a graphical check of the assumption of the multinomial model for the recall probabilities, which appears to hold for the menarcheal data set. Our analysis shows that the use of the partially recalled part of the data indeed leads to smaller confidence intervals of the survival function.

Stochastic Functional Estimates in Longitudinal Models with Interval-Censored Anchoring Events

Timelines of longitudinal studies are often anchored by specific events. In the absence of fully observed the anchoring event times, the study timeline becomes undefined, and the traditional longitudinal analysis loses its temporal reference. In this paper, we considered an analytical situation where the anchoring events are interval-censored. We demonstrated that by expressing the regression parameter estimators as stochastic functionals of a plug-in estimate of the unknown anchoring event time distribution, the standard longitudinal models could be extended to accommodate the situation of less well-defined timelines. We showed that for a broad class of longitudinal models, the functional parameter estimates are consistent and asymptotically normally distributed with an convergence rate under mild regularity conditions. Applying the developed theory to linear mixed-effects models, we further proposed a hybrid computational procedure that combines the strengths of the Fisher's scoring method and the expectation-expectation (EM) algorithm, for model parameter estimation. We conducted a simulation study to validate the asymptotic properties and to assess the finite sample performance of the proposed method. A real data analysis was used to illustrate the proposed method. The method fills in a gap in the existing longitudinal analysis methodology for data with less well defined timelines.

Nonparametric estimators of survival function under the mixed case interval-censored model with left truncation

It is well known that the nonparametric maximum likelihood estimator (NPMLE) can severely underestimate the survival probabilities at early times for left-truncated and interval-censored (LT-IC) data. For arbitrarily truncated and censored data, Pan and Chappel (JAMA Stat Probab Lett 38:49-57, 1998a, Biometrics 54:1053-1060, 1998b) proposed a nonparametric estimator of the survival function, called the iterative Nelson estimator (INE). Their simulation study showed that the INE performed well in overcoming the under-estimation of the survival function from the NPMLE for LT-IC data. In this article, we revisit the problem of inconsistency of the NPMLE. We point out that the inconsistency is caused by the likelihood function of the left-censored observations, where the left-truncated variables are used as the left endpoints of censoring intervals. This can lead to severe underestimation of the survival function if the NPMLE is obtained using Turnbull's (JAMA 38:290-295, 1976) EM algorithm. To overcome this problem, we propose a modified maximum likelihood estimator (MMLE) based on a modified likelihood function, where the left endpoints of censoring intervals for left-censored observations are the maximum of left-truncated variables and the estimated left endpoint of the support of the left-censored times. Simulation studies show that the MMLE performs well for finite sample and outperforms both the INE and NPMLE.

Assessing the accuracy of predictive models with interval-censored data

We develop methods for assessing the predictive accuracy of a given event time model when the validation sample is comprised of case \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$K$\end{document} interval-censored data. An imputation-based, an inverse probability weighted (IPW), and an augmented inverse probability weighted (AIPW) estimator are developed and evaluated for the mean prediction error and the area under the receiver operating characteristic curve when the goal is to predict event status at a landmark time. The weights used for the IPW and AIPW estimators are obtained by fitting a multistate model which jointly considers the event process, the recurrent assessment process, and loss to follow-up. We empirically investigate the performance of the proposed methods and illustrate their application in the context of a motivating rheumatology study in which human leukocyte antigen markers are used to predict disease progression status in patients with psoriatic arthritis.

Estimation of life expectancies using continuous-time multi-state models

BACKGROUND AND OBJECTIVE

There is increasing interest in multi-state modelling of health-related stochastic processes. Given a fitted multi-state model with one death state, it is possible to estimate state-specific and marginal life expectancies. This paper introduces methods and new software for computing these expectancies.

METHODS

The definition of state-specific life expectancy given current age is an extension of mean survival in standard survival analysis. The computation involves the estimated parameters of a fitted multi-state model, and numerical integration. The new R package elect provides user-friendly functions to do the computation in the R software.

RESULTS

The estimation of life expectancies is explained and illustrated using the elect package. Functions are presented to explore the data, to estimate the life expectancies, and to present results.

CONCLUSIONS

State-specific life expectancies provide a communicable representation of health-related processes. The availability and explanation of the elect package will help researchers to compute life expectancies and to present their findings in an assessable way.

Comparison of a time-varying covariate model and a joint model of time-to-event outcomes in the presence of measurement error and interval censoring: application to kidney transplantation

Background

Tacrolimus (TAC) is an immunosuppressant drug given to kidney transplant recipients post-transplant to prevent antibody formation and kidney rejection. The optimal therapeutic dose for TAC is poorly defined and therapy requires frequent monitoring of drug trough levels. Analyzing the association between TAC levels over time and the development of potentially harmful de novo donor specific antibodies (dnDSA) is complex because TAC levels are subject to measurement error and dnDSA is assessed at discrete times, so it is an interval censored time-to-event outcome.

Methods

Using data from the University of Colorado Transplant Center, we investigated the association between TAC and dnDSA using a shared random effects (intercept and slope) model with longitudinal and interval censored survival sub-models (JM) and compared it with the more traditional interval censored survival model with a time-varying covariate (TVC). We carried out simulations to compare bias, level and power for the association parameter in the TVC and JM under varying conditions of measurement error and interval censoring. In addition, using Markov Chain Monte Carlo (MCMC) methods allowed us to calculate clinically relevant quantities along with credible intervals (CrI).

Results

The shared random effects model was a better fit and showed both the average TAC and the slope of TAC were associated with risk of dnDSA. The simulation studies demonstrated that, in the presence of heavy interval censoring and high measurement error, the TVC survival model underestimates the association between the survival and longitudinal measurement and has inflated type I error and considerably less power to detect associations.

Conclusions

To avoid underestimating associations, shared random effects models should be used in analyses of data with interval censoring and measurement error.

Electronic supplementary material

The online version of this article (10.1186/s12874-019-0773-1) contains supplementary material, which is available to authorized users.

Semiparametric competing risks regression under interval censoring using the R package intccr

BACKGROUND AND OBJECTIVE

Competing risk data are frequently interval-censored in real-world applications, that is, the exact event time is not precisely observed but is only known to lie between two time points such as clinic visits. This type of data requires special handling because the actual event times are unknown. To deal with this problem we have developed an easy-to-use open-source statistical software.

METHODS

An approach to perform semiparametric regression analysis of the cumulative incidence function with interval-censored competing risks data is the sieve maximum likelihood method based on B-splines. An important feature of this approach is that it does not impose restrictive parametric assumptions. Also, this methodology provides semiparametrically efficient estimates. Implementation of this methodology can be easily performed using our new R package intccr.

RESULTS

The R package intccr performs semiparametric regression analysis of the cumulative incidence function based on interval-censored competing risks data. It supports a large class of models including the proportional odds and the Fine-Gray proportional subdistribution hazards model as special cases. It also provides the estimated cumulative incidence functions for a particular combination of covariate values. The package also provides some data management functionality to handle data sets which are in a long format involving multiple lines of data per subject.

CONCLUSIONS

The R package intccr provides a convenient and flexible software for the analysis of the cumulative incidence function based on interval-censored competing risks data.

Bayes factors for choosing among six common survival models

A super model that includes proportional hazards, proportional odds, accelerated failure time, accelerated hazards, and extended hazards models, as well as the model proposed in Diao et al. (Biometrics 69(4):840-849, 2013) accounting for crossed survival as special cases is proposed for the purpose of testing and choosing among these popular semiparametric models. Efficient methods for fitting and computing fast, approximate Bayes factors are developed using a nonparametric baseline survival function based on a transformed Bernstein polynomial. All manner of censoring is accommodated including right, left, and interval censoring, as well as data that are observed exactly and mixtures of all of these; current status data are included as a special case. The method is tested on simulated data and two real data examples. The approach is easily carried out via a new function in the spBayesSurv R package.

Additive-multiplicative hazards regression models for interval-censored semi-competing risks data with missing intermediate events

BACKGROUND

In clinical trials and survival analysis, participants may be excluded from the study due to withdrawal, which is often referred to as lost-to-follow-up (LTF). It is natural to argue that a disease would be censored due to death; however, when an LTF is present it is not guaranteed that the disease has been censored. This makes it important to consider both cases; the disease is censored or not censored. We also note that the illness process can be censored by LTF. We will consider a multi-state model in which LTF is not regarded as censoring but as a non-fatal event.

METHODS

We propose a multi-state model for analyzing semi-competing risks data with interval-censored or missing intermediate events. More precisely, we employ the additive and multiplicative hazards model with log-normal frailty and construct the conditional likelihood to estimate the transition intensities among states in the multi-state model. Marginalization of the full likelihood is accomplished using adaptive importance sampling, and the optimal solution of the regression parameters is achieved through the iterative quasi-Newton algorithm.

RESULTS

Simulation is performed to investigate the finite-sample performance of the proposed estimation method in terms of the relative bias and coverage probability of the regression parameters. The proposed estimators turned out to be robust to misspecifications of the frailty distribution. PAQUID data have been analyzed and yielded somewhat prominent results.

CONCLUSIONS

We propose a multi-state model for semi-competing risks data for which there exists information on fatal events, but information on non-fatal events may not be available due to lost to follow-up. Simulation results show that the coverage probabilities of the regression parameters are close to a nominal level of 0.95 in most cases. Regarding the analysis of real data, the risk of transition from a healthy state to dementia is higher for women; however, the risk of death after being diagnosed with dementia is higher for men.

Estimating trends in the incidence rate with interval censored data and time-dependent covariates

We propose a multiple imputation method for estimating the incidence rate with interval censored data and time-dependent (and/or time-independent) covariates. The method has two stages. First, we use a semi-parametric G-transformation model to estimate the cumulative baseline hazard function and the effects of the time-dependent (and/or time-independent covariates) on the interval censored infection times. Second, we derive the participant's unique cumulative distribution function and impute infection times conditional on the covariate values. To assess performance, we simulated infection times from a Cox proportional hazards model and induced interval censoring by varying the testing rate, e.g., participants test 100%, 75%, 50% of the time, etc. We then compared the incidence rate estimates from our G-imputation approach with single random-point and mid-point imputation. By comparison, our G-imputation approach gave more accurate incidence rate estimates and appropriate standard errors for models with time-independent covariates only, time-dependent covariates only, and a mixture of time-dependent and time-independent covariates across various testing rates. We demonstrate, for the first time, a multiple imputation approach for incidence rate estimation with interval censored data and time-dependent (and/or time-independent) covariates.

Distribution-free estimation of local growth rates around interval censored anchoring events

Biological processes are usually defined on timelines that are anchored by specific events. For example, cancer growth is typically measured by the change in tumor size from the time of oncogenesis. In the absence of such anchoring events, longitudinal assessments of the outcome lose their temporal reference. In this paper, we considered the estimation of local change rates in the outcomes when the anchoring events are interval-censored. Viewing the subject-specific anchoring event times as random variables from an unspecified distribution, we proposed a distribution-free estimation method for the local growth rates around the unobserved anchoring events. We expressed the rate parameters as stochastic functionals of the anchoring time distribution and showed that under mild regularity conditions, consistent and asymptotically normal estimates of the rate parameters could be achieved, with a n convergence rate. We conducted a carefully designed simulation study to evaluate the finite sample performance of the method. To motivate and illustrate the use of the proposed method, we estimated the skeletal growth rates of male and female adolescents, before and after the unobserved pubertal growth spurt (PGS) times.

Closure time of ductus arteriosus after birth based on survival analysis

OBJECTIVES

The correct ductus arteriosus (DA) closure time is somewhere between the opening and closing time confirmed on echo, not on examination. We investigated DA closure time and factors affecting DA closure time using interval censoring analysis.

METHODS

This was an observational, retrospective study including 2611 healthy neonates. Echo was performed every 12-24 h after birth until DA closure. We investigated the DA closure time using interval censoring analysis. If the DA was closed on echo, we assumed that the DA was open at birth. We evaluated clinical factors affecting DA closure time.

RESULTS

Median DA closure time was 13.5 h (range, 7.7-18.7 h) after birth. DA closure time was associated with primipara status, maternal prostaglandin E2 (PGE2) administration, <2500 g birth weight, and diagnosis of congenital ductus arteriosus aneurysm (DAA). Using proportional hazards regression models, the interval-censored data (primipara, hazard ratio [HR] = 1.099, P = 0.04; PGE2, HR = 0.823, P = 0.03; <2500 g, HR = 1.413, P < 0.01; DAA, HR = 0.570, P < 0.01) were found to be significantly associated with DA closure time.

CONCLUSIONS

Estimation of DA closure time by interval censoring analysis is helpful to determine the optimal time to perform echo and to predict risk factors for patent DA.

Pseudo and conditional score approach to joint analysis of current count and current status data

We develop a joint analysis approach for recurrent and nonrecurrent event processes subject to case I interval censorship, which are also known in literature as current count and current status data, respectively. We use a shared frailty to link the recurrent and nonrecurrent event processes, while leaving the distribution of the frailty fully unspecified. Conditional on the frailty, the recurrent event is assumed to follow a nonhomogeneous Poisson process, and the mean function of the recurrent event and the survival function of the nonrecurrent event are assumed to follow some general form of semiparametric transformation models. Estimation of the models is based on the pseudo-likelihood and the conditional score techniques. The resulting estimators for the regression parameters and the unspecified baseline functions are shown to be consistent with rates of square and cubic roots of the sample size, respectively. Asymptotic normality with closed-form asymptotic variance is derived for the estimator of the regression parameters. We apply the proposed method to a fracture-osteoporosis survey data to identify risk factors jointly for fracture and osteoporosis in elders, while accounting for association between the two events within a subject.

Computationally Efficient Estimation for the Generalized Odds Rate Mixture Cure Model with Interval-Censored Data

For semiparametric survival models with interval censored data and a cure fraction, it is often difficult to derive nonparametric maximum likelihood estimation due to the challenge in maximizing the complex likelihood function. In this paper, we propose a computationally efficient EM algorithm, facilitated by a gamma-poisson data augmentation, for maximum likelihood estimation in a class of generalized odds rate mixture cure (GORMC) models with interval censored data. The gamma-poisson data augmentation greatly simplifies the EM estimation and enhances the convergence speed of the EM algorithm. The empirical properties of the proposed method are examined through extensive simulation studies and compared with numerical maximum likelihood estimates. An R package "GORCure" is developed to implement the proposed method and its use is illustrated by an application to the Aerobic Center Longitudinal Study dataset.

Regression analysis of current status data in the presence of a cured subgroup and dependent censoring

This paper discusses regression analysis of current status data, a type of failure time data where each study subject is observed only once, in the presence of dependent censoring. Furthermore, there may exist a cured subgroup, meaning that a proportion of study subjects are not susceptible to the failure event of interest. For the problem, we develop a sieve maximum likelihood estimation approach with the use of latent variables and Bernstein polynomials. For the determination of the proposed estimators, an EM algorithm is developed and the asymptotic properties of the estimators are established. Extensive simulation studies are conducted and indicate that the proposed method works well for practical situations. A motivating application from a tumorigenicity experiment is also provided.

Semiparametric regression analysis of interval-censored competing risks data

Interval-censored competing risks data arise when each study subject may experience an event or failure from one of several causes and the failure time is not observed directly but rather is known to lie in an interval between two examinations. We formulate the effects of possibly time-varying (external) covariates on the cumulative incidence or sub-distribution function of competing risks (i.e., the marginal probability of failure from a specific cause) through a broad class of semiparametric regression models that captures both proportional and non-proportional hazards structures for the sub-distribution. We allow each subject to have an arbitrary number of examinations and accommodate missing information on the cause of failure. We consider nonparametric maximum likelihood estimation and devise a fast and stable EM-type algorithm for its computation. We then establish the consistency, asymptotic normality, and semiparametric efficiency of the resulting estimators for the regression parameters by appealing to modern empirical process theory. In addition, we show through extensive simulation studies that the proposed methods perform well in realistic situations. Finally, we provide an application to a study on HIV-1 infection with different viral subtypes.

Cause-Specific Hazard Regression for Competing Risks Data Under Interval Censoring and Left Truncation

Inference for cause-specific hazards from competing risks data under interval censoring and possible left truncation has been understudied. Aiming at this target, a penalized likelihood approach for a Cox-type proportional cause-specific hazards model is developed, and the associated asymptotic theory is discussed. Monte Carlo simulations show that the approach performs very well for moderate sample sizes. An application to a longitudinal study of dementia illustrates the practical utility of the method. In the application, the age-specific hazards of AD, other dementia and death without dementia are estimated, and risk factors of all competing risks are studied.

Nonparametric estimation of time-to-event distribution based on recall data in observational studies

In a cross-sectional observational study, time-to-event distribution can be estimated from data on current status or from recalled data on the time of occurrence. In either case, one can treat the data as having been interval censored, and use the nonparametric maximum likelihood estimator proposed by Turnbull (J R Stat Soc Ser B 38:290-295, 1976). However, the chance of recall may depend on the time span between the occurrence of the event and the time of interview. In such a case, the underlying censoring would be informative, rendering the Turnbull estimator inappropriate. In this article, we provide a nonparametric maximum likelihood estimator of the distribution of interest, by using a model adapted to the special nature of the data at hand. We also provide a computationally simple approximation of this estimator, and establish the consistency of both the original and the approximate versions, under mild conditions. Monte Carlo simulations indicate that the proposed estimators have smaller bias than the Turnbull estimator based on incomplete recall data, smaller variance than the Turnbull estimator based on current status data, and smaller mean squared error than both of them. The method is applied to menarcheal data from a recent Anthropometric study of adolescent and young adult females in Kolkata, India.

Maximum likelihood estimation for semiparametric transformation models with interval-censored data

Interval censoring arises frequently in clinical, epidemiological, financial and sociological studies, where the event or failure of interest is known only to occur within an interval induced by periodic monitoring. We formulate the effects of potentially time-dependent covariates on the interval-censored failure time through a broad class of semiparametric transformation models that encompasses proportional hazards and proportional odds models. We consider nonparametric maximum likelihood estimation for this class of models with an arbitrary number of monitoring times for each subject. We devise an EM-type algorithm that converges stably, even in the presence of time-dependent covariates, and show that the estimators for the regression parameters are consistent, asymptotically normal, and asymptotically efficient with an easily estimated covariance matrix. Finally, we demonstrate the performance of our procedures through simulation studies and application to an HIV/AIDS study conducted in Thailand.

Joint latent class model for longitudinal data and interval-censored semi-competing events: Application to dementia

Joint models are used in ageing studies to investigate the association between longitudinal markers and a time-to-event, and have been extended to multiple markers and/or competing risks. The competing risk of death must be considered in the elderly because death and dementia have common risk factors. Moreover, in cohort studies, time-to-dementia is interval-censored since dementia is assessed intermittently. So subjects can develop dementia and die between two visits without being diagnosed. To study predementia cognitive decline, we propose a joint latent class model combining a (possibly multivariate) mixed model and an illness-death model handling both interval censoring (by accounting for a possible unobserved transition to dementia) and semi-competing risks. Parameters are estimated by maximum-likelihood handling interval censoring. The correlation between the marker and the times-to-events is captured by latent classes, homogeneous sub-groups with specific risks of death, dementia, and profiles of cognitive decline. We propose Markovian and semi-Markovian versions. Both approaches are compared to a joint latent-class model for competing risks through a simulation study, and applied in a prospective cohort study of cerebral and functional ageing to distinguish different profiles of cognitive decline associated with risks of dementia and death. The comparison highlights that among subjects with dementia, mortality depends more on age than on duration of dementia. This model distinguishes the so-called terminal predeath decline (among healthy subjects) from the predementia decline.

Sieve estimation in a Markov illness-death process under dual censoring

Semiparametric methods are well established for the analysis of a progressive Markov illness-death process observed up to a noninformative right censoring time. However, often the intermediate and terminal events are censored in different ways, leading to a dual censoring scheme. In such settings, unbiased estimation of the cumulative transition intensity functions cannot be achieved without some degree of smoothing. To overcome this problem, we develop a sieve maximum likelihood approach for inference on the hazard ratio. A simulation study shows that the sieve estimator offers improved finite-sample performance over common imputation-based alternatives and is robust to some forms of dependent censoring. The proposed method is illustrated using data from cancer trials.

Analysis of interval-censored recurrent event processes subject to resolution

Interval-censored recurrent event data arise when the event of interest is not readily observed but the cumulative event count can be recorded at periodic assessment times. In some settings, chronic disease processes may resolve, and individuals will cease to be at risk of events at the time of disease resolution. We develop an expectation-maximization algorithm for fitting a dynamic mover-stayer model to interval-censored recurrent event data under a Markov model with a piecewise-constant baseline rate function given a latent process. The model is motivated by settings in which the event times and the resolution time of the disease process are unobserved. The likelihood and algorithm are shown to yield estimators with small empirical bias in simulation studies. Data are analyzed on the cumulative number of damaged joints in patients with psoriatic arthritis where individuals experience disease remission.

The effect of hospital care on early survival after penetrating trauma

Background

The effectiveness of emergency medical interventions can be best evaluated using time-to-event statistical methods with time-varying covariates (TVC), but this approach is complicated by uncertainty about the actual times of death. We therefore sought to evaluate the effect of hospital intervention on mortality after penetrating trauma using a method that allowed for interval censoring of the precise times of death.

Methods

Data on persons with penetrating trauma due to interpersonal assault were combined from the 2008 to 2010 National Trauma Data Bank (NTDB) and the 2004 to 2010 National Violent Death Reporting System (NVDRS). Cox and Weibull proportional hazards models for survival time (t_SURV) were estimated, with TVC assumed to have constant effects for specified time intervals following hospital arrival. The Weibull model was repeated with t_SURV interval-censored to reflect uncertainty about the precise times of death, using an imputation method to accommodate interval censoring along with TVC.

Results

All models showed that mortality was increased by older age, female sex, firearm mechanism, and injuries involving the head/neck or trunk. Uncensored models showed a paradoxical increase in mortality associated with the first hour in a hospital. The interval-censored model showed that mortality was markedly reduced after admission to a hospital, with a hazard ratio (HR) of 0.68 (95% CI 0.63, 0.73) during the first 30 min declining to a HR of 0.01 after 120 min. Admission to a verified level I trauma center (compared to other hospitals in the NTDB) was associated with a further reduction in mortality, with a HR of 0.93 (95% CI 0.82, 0.97).

Conclusions

Time-to-event models with TVC and interval censoring can be used to estimate the effect of hospital care on early mortality after penetrating trauma or other acute medical conditions and could potentially be used for interhospital comparisons.

Electronic supplementary material

The online version of this article (doi:10.1186/s40621-014-0024-1) contains supplementary material, which is available to authorized users.

Estimation and residual analysis with R for a linear regression model with an interval-censored covariate

Interval-censored covariates are sometimes encountered in longitudinal studies and considered as possible predictors in a regression model. This paper, motivated by an AIDS study, proposes an implementation in R for the estimation of parameters and the assessment of the assumptions of a linear regression model with an interval-censored covariate. The properties of the parameters estimators and the behavior of three proposed residuals are addressed through two simulation studies. Also, guidelines are provided to check the goodness-of-fit of the fitted model in terms of the length of the censoring interval of the covariate. The methodology is illustrated with real data coming from the AIDS study. R functions and scripts are provided.

Composite likelihood for joint analysis of multiple multistate processes via copulas

A copula-based model is described which enables joint analysis of multiple progressive multistate processes. Unlike intensity-based or frailty-based approaches to joint modeling, the copula formulation proposed herein ensures that a wide range of marginal multistate processes can be specified and the joint model will retain these marginal features. The copula formulation also facilitates a variety of approaches to estimation and inference including composite likelihood and two-stage estimation procedures. We consider processes with Markov margins in detail, which are often suitable when chronic diseases are progressive in nature. We give special attention to the setting in which individuals are examined intermittently and transition times are consequently interval-censored. Simulation studies give empirical insight into the different methods of analysis and an application involving progression in joint damage in psoriatic arthritis provides further illustration.

Estimating the effect of emergency care on early survival after traffic crashes

INTRODUCTION

Traffic crash mortality is higher in rural areas, but it is unclear whether this is due to greater injury severity, time delays, or Emergency Medical Services (EMS) deficiencies.

METHODS

Data from 2002-2003 were combined from the Fatality Analysis Reporting System (FARS) and an "expanded version" of the National Automotive Sampling System (NASS) Crashworthiness Data System (CDS). Weighted Cox and Weibull models for survival time (tSURV) were estimated, with time-varying covariates (TVC) having constant effects for specified time intervals following EMS arrival time (tEMS) and hospital arrival time (tHOS). The Weibull model was repeated with tSURV interval-censored to reflect uncertainty about the exact time of death, using an imputation method to accommodate interval censoring along with TVC.

RESULTS

FARS contained records for 92,718 persons with fatal or incapacitating injuries, and NASS/CDS contained 5517 (weighted population of 642,716) with incapacitating injuries. All models associated mortality with increasing age, male sex, belt nonuse, higher speeds, and vehicle rollover. The interval-censored model associated EMS intervention with a beneficial effect until tEMS+30 min, but not thereafter; hospital intervention was associated with a strongly beneficial effect that increased with time. Rural location was associated with a higher baseline hazard; a 50% reduction in rural prehospital time would theoretically reduce 4-h mortality by about 7%.

CONCLUSION

Rural/urban disparity in crash mortality is mostly independent of time delays and EMS effects. However, survival models with TVC support clinical intuition of a "golden hour" in EMS care, and the importance of timely transport to a hospital.

Regression analysis of clustered interval-censored failure time data with informative cluster size

Correlated or clustered failure time data often occur in medical studies, among other fields (Cai and Prentice, 1995; Kalbfleisch and Prentice, 2002), and sometimes such data arise together with interval censoring (Wang et al., 2006). Furthermore, the failure time of interest may be related to the cluster size. For example, Williamson et al. (2008) discussed such an example arising from a lymphatic filariasis study. A simple and common approach to the analysis of these data is to simplify or convert interval-censored data to right-censored data due to the lack of proper inference procedures for direct analysis of these data. In this paper, two procedures are presented for regression analysis of clustered failure time data that allow both interval censoring and informative cluster size. Simulation studies are conducted to evaluate the presented approaches and they are applied to a motivating example.