Bayesian measurement‐error‐driven hidden Markov regression model for calibrating the effect of covariates on multistate outcomes: Application to androgenetic alopecia

Multistate Markov regression models used for quantifying the effect size of state‐specific covariates pertaining to the dynamics of multistate outcomes have gained popularity. However, the measurements of multistate outcome are prone to the errors of classification, particularly when a population‐based survey/research is involved with proxy measurements of outcome due to cost consideration. Such a misclassification may affect the effect size of relevant covariates such as odds ratio used in the field of epidemiology. We proposed a Bayesian measurement‐error‐driven hidden Markov regression model for calibrating these biased estimates with and without a 2‐stage validation design. A simulation algorithm was developed to assess various scenarios of underestimation and overestimation given nondifferential misclassification (independent of covariates) and differential misclassification (dependent on covariates). We applied our proposed method to the community‐based survey of androgenetic alopecia and found that the effect size of the majority of covariate was inflated after calibration regardless of which type of misclassification. Our proposed Bayesian measurement‐error‐driven hidden Markov regression model is practicable and effective in calibrating the effects of covariates on multistate outcome, but the prior distribution on measurement errors accrued from 2‐stage validation design is strongly recommended.     

Parametric multistate survival models: Flexible modelling allowing transition‐specific distributions with application to estimating clinically useful measures of effect differences

Multistate models are increasingly being used to model complex disease profiles. By modelling transitions between disease states, accounting for competing events at each transition, we can gain a much richer understanding of patient trajectories and how risk factors impact over the entire disease pathway. In this article, we concentrate on parametric multistate models, both Markov and semi‐Markov, and develop a flexible framework where each transition can be specified by a variety of parametric models including exponential, Weibull, Gompertz, Royston‐Parmar proportional hazards models or log‐logistic, log‐normal, generalised gamma accelerated failure time models, possibly sharing parameters across transitions. We also extend the framework to allow time‐dependent effects. We then use an efficient and generalisable simulation method to calculate transition probabilities from any fitted multistate model, and show how it facilitates the simple calculation of clinically useful measures, such as expected length of stay in each state, and differences and ratios of proportion within each state as a function of time, for specific covariate patterns. We illustrate our methods using a dataset of patients with primary breast cancer. User‐friendly Stata software is provided.     

Analysis of censored discrete longitudinal data: Estimation of mean response

The study of longitudinal data is usually concerned with one or several response variables measured, possibly along with some covariates, at different points in time. In real‐life situations this is often complicated by missing observations due to what we usually refer to as ‘censoring’. In this paper we consider missingness of a monotone kind; subjects that dropout, i.e. are censored, fail to participate in the study at any of the subsequent observation times. Our scientific objective is to make inference about the mean response in a hypothetical population without any dropouts. There are several methods and approaches that address this problem, and we will present two existing methods (the linear‐increments method and the inverse‐probability‐weighting method), as well as propose a new method, based on a discrete Markov process. We examine the performance of the corresponding estimators and compare these with respect to bias and variability. To demonstrate the effectiveness of the approaches in estimating the mean of a response variable, we analyse simulated data of different multistate models with a Markovian structure. Analyses of substantive data from (1) a study of symptoms experienced after a traumatic brain injury, and (2) a study of cognitive function among the elderly, are used as illustrations of the methods presented. Copyright © 2008 John Wiley & Sons, Ltd.     

Blood pressure and the risk of chronic kidney disease progression using multistate marginal structural models in the CRIC Study

In patients with chronic kidney disease (CKD), clinical interest often centers on determining treatments and exposures that are causally related to renal progression. Analyses of longitudinal clinical data in this population are often complicated by clinical competing events, such as end‐stage renal disease (ESRD) and death, and time‐dependent confounding, where patient factors that are predictive of later exposures and outcomes are affected by past exposures. We developed multistate marginal structural models (MS‐MSMs) to assess the effect of time‐varying systolic blood pressure on disease progression in subjects with CKD. The multistate nature of the model allows us to jointly model disease progression characterized by changes in the estimated glomerular filtration rate (eGFR), the onset of ESRD, and death, and thereby avoid unnatural assumptions of death and ESRD as noninformative censoring events for subsequent changes in eGFR. We model the causal effect of systolic blood pressure on the probability of transitioning into 1 of 6 disease states given the current state. We use inverse probability weights with stabilization to account for potential time‐varying confounders, including past eGFR, total protein, serum creatinine, and hemoglobin. We apply the model to data from the Chronic Renal Insufficiency Cohort Study, a multisite observational study of patients with CKD.     

Estimation of markov chain transition probabilities and rates from fully and partially observed data: uncertainty propagation, evidence synthesis, and model calibration.

Markov transition models are frequently used to model disease progression. The authors show how the solution to Kolmogorov's forward equations can be exploited to map between transition rates and probabilities from probability data in multistate models. They provide a uniform, Bayesian treatment of estimation and propagation of uncertainty of transition rates and probabilities when 1) observations are available on all transitions and exact time at risk in each state (fully observed data) and 2) observations are on initial state and final state after a fixed interval of time but not on the sequence of transitions (partially observed data). The authors show how underlying transition rates can be recovered from partially observed data using Markov chain Monte Carlo methods in WinBUGS, and they suggest diagnostics to investigate inconsistencies between evidence from different starting states. An illustrative example for a 3-state model is given, which shows how the methods extend to more complex Markov models using the software WBDiff to compute solutions. Finally, the authors illustrate how to statistically combine data from multiple sources, including partially observed data at several follow-up times and also how to calibrate a Markov model to be consistent with data from one specific study.     

State‐space size considerations for disease‐progression models

Markov models of disease progression are widely used to model transitions in patients' health state over time. Usually, patients' health status may be classified according to a set of ordered health states. Modelers lump together similar health states into a finite and usually small, number of health states that form the basis of a Markov chain disease‐progression model. This increases the number of observations used to estimate each parameter in the transition probability matrix. However, lumping together observably distinct health states also obscures distinctions among them and may reduce the predictive power of the model. Moreover, as we demonstrate, precision in estimating the model parameters does not necessarily improve as the number of states in the model declines. This paper explores the tradeoff between lumping error introduced by grouping distinct health states and sampling error that arises when there are insufficient patient data to precisely estimate the transition probability matrix. Copyright © 2013 John Wiley & Sons, Ltd.     

A hidden Markov model approach to analyze longitudinal ternary outcomes when some observed states are possibly misclassified

Understanding the dynamic disease process is vital in early detection, diagnosis, and measuring progression. Continuous‐time Markov chain (CTMC) methods have been used to estimate state‐change intensities but challenges arise when stages are potentially misclassified. We present an analytical likelihood approach where the hidden state is modeled as a three‐state CTMC model allowing for some observed states to be possibly misclassified. Covariate effects of the hidden process and misclassification probabilities of the hidden state are estimated without information from a ‘gold standard’ as comparison. Parameter estimates are obtained using a modified expectation‐maximization (EM) algorithm, and identifiability of CTMC estimation is addressed. Simulation studies and an application studying Alzheimer's disease caregiver stress‐levels are presented. The method was highly sensitive to detecting true misclassification and did not falsely identify error in the absence of misclassification. In conclusion, we have developed a robust longitudinal method for analyzing categorical outcome data when classification of disease severity stage is uncertain and the purpose is to study the process' transition behavior without a gold standard. Copyright © 2016 John Wiley & Sons, Ltd.     

Bayesian analyses of longitudinal binary data using Markov regression models of unknown order

We present non-homogeneous Markov regression models of unknown order as a means to assess the duration of autoregressive dependence in longitudinal binary data. We describe a subject's transition probability evolving over time using logistic regression models for his or her past outcomes and covariates. When the initial values of the binary process are unknown, they are treated as latent variables. The unknown initial values, model parameters, and the order of transitions are then estimated using a Bayesian variable selection approach, via Gibbs sampling. As a comparison with our approach, we also implement the deviance information criterion (DIC) for the determination of the order of transitions. An example addresses the progression of substance use in a community sample of n = 242 American Indian children who were interviewed annually four times. An extension of the Markov model to account for subject-to-subject heterogeneity is also discussed.     

Multi‐state Markov models for disease progression in the presence of informative examination times: An application to hepatitis C

In many chronic diseases it is important to understand the rate at which patients progress from infection through a series of defined disease states to a clinical outcome, e.g. cirrhosis in hepatitis C virus (HCV)‐infected individuals or AIDS in HIV‐infected individuals. Typically data are obtained from longitudinal studies, which often are observational in nature, and where disease state is observed only at selected examinations throughout follow‐up. Transition times between disease states are therefore interval censored. Multi‐state Markov models are commonly used to analyze such data, but rely on the assumption that the examination times are non‐informative, and hence the examination process is ignorable in a likelihood‐based analysis. In this paper we develop a Markov model that relaxes this assumption through the premise that the examination process is ignorable only after conditioning on a more regularly observed auxiliary variable. This situation arises in a study of HCV disease progression, where liver biopsies (the examinations) are sparse, irregular, and potentially informative with respect to the transition times. We use additional information on liver function tests (LFTs), commonly collected throughout follow‐up, to inform current disease state and to assume an ignorable examination process. The model developed has a similar structure to a hidden Markov model and accommodates both the series of LFT measurements and the partially latent series of disease states. We show through simulation how this model compares with the commonly used ignorable Markov model, and a Markov model that assumes the examination process is non‐ignorable. Copyright © 2010 John Wiley & Sons, Ltd.     

Nonparametric regression of state occupation,entry, exit,and waiting times with multistate right‐censored data

We construct nonparametric regression estimators of a number of temporal functions in a multistate system based on a continuous univariate baseline covariate. These estimators include state occupation probabilities, state entry, exit, and waiting (sojourn) time distribution functions of a general progressive (e.g., acyclic) multistate model. We subject the data to right censoring, and the censoring mechanism is explainable by observable covariates that could be time dependent. The resulting estimators are valid even if the multistate process is non‐Markov. We study the performance of the estimators in two simulation settings. We establish large sample consistency of these estimators. We illustrate our estimators using a data set on bone marrow transplant recipients. Copyright © 2012 John Wiley & Sons, Ltd.     

A Bayesian semiparametric Markov regression model for juvenile dermatomyositis

Juvenile dermatomyositis (JDM) is a rare autoimmune disease that may lead to serious complications, even to death. We develop a 2‐state Markov regression model in a Bayesian framework to characterise disease progression in JDM over time and gain a better understanding of the factors influencing disease risk. The transition probabilities between disease and remission state (and vice versa) are a function of time‐homogeneous and time‐varying covariates. These latter types of covariates are introduced in the model through a latent health state function, which describes patient‐specific health over time and accounts for variability among patients. We assume a nonparametric prior based on the Dirichlet process to model the health state function and the baseline transition intensities between disease and remission state and vice versa. The Dirichlet process induces a clustering of the patients in homogeneous risk groups. To highlight clinical variables that most affect the transition probabilities, we perform variable selection using spike and slab prior distributions. Posterior inference is performed through Markov chain Monte Carlo methods. Data were made available from the UK JDM Cohort and Biomarker Study and Repository, hosted at the UCL Institute of Child Health.     

Multi‐state models for colon cancer recurrence and death with a cured fraction

In cancer clinical trials, patients often experience a recurrence of disease prior to the outcome of interest, overall survival. Additionally, for many cancers, there is a cured fraction of the population who will never experience a recurrence. There is often interest in how different covariates affect the probability of being cured of disease and the time to recurrence, time to death, and time to death after recurrence. We propose a multi‐state Markov model with an incorporated cured fraction to jointly model recurrence and death in colon cancer. A Bayesian estimation strategy is used to obtain parameter estimates. The model can be used to assess how individual covariates affect the probability of being cured and each of the transition rates. Checks for the adequacy of the model fit and for the functional forms of covariates are explored. The methods are applied to data from 12 randomized trials in colon cancer, where we show common effects of specific covariates across the trials. Copyright © 2013 John Wiley & Sons, Ltd.     

A note on bias of measures of explained variation for survival data

Papers evaluating measures of explained variation, or similar indices, almost invariably use independence from censoring as the most important criterion. And they always end up suggesting that some measures meet this criterion, and some do not, most of the time leading to a conclusion that the first is better than the second. As a consequence, users are offered measures that cannot be used with time‐dependent covariates and effects, not to mention extensions to repeated events or multi‐state models. We explain in this paper that the aforementioned criterion is of no use in studying such measures, because it simply favors those that make an implicit assumption of a model being valid everywhere. Measures not making such an assumption are disqualified, even though they are better in every other respect. We show that if these, allegedly inferior, measures are allowed to make the same assumption, they are easily corrected to satisfy the ‘independent‐from‐censoring’ criterion. Even better, it is enough to make such an assumption only for the times greater than the last observed failure time τ, which, in contrast with the ‘preferred’ measures, makes it possible to use all the modeling flexibility up to τ and assume whatever one wants after τ. As a consequence, we claim that some of the measures being preferred as better in the existing reviews are in fact inferior. Copyright © 2015 John Wiley & Sons, Ltd.     

A semi-Markov model for multistate and interval-censored data with multiple terminal events. Application in renal transplantation

The semi-Markov assumption emphasizes the importance of time spent in a state. In order to compute this type of multistate model, most transition times are always considered to be exactly identified or right censored. However, in the longitudinal analysis of chronic diseases, investigators are often confronted with interval-censored data (transition times are known to have occurred in some interval). Thus, the two key issues are the modeling of the duration dependence and the interval censoring. In this article, we define a semi-Markov model, allowing for interval censoring, for parametric hazard functions with a union or logical sum- or intersection-shape and for determination of initial states according to covariates. Our modeling approach is specific to each transition, so as to obtain a more coherent model. Parallel to competing risks models, the multistate model takes into account several final events. We consider an example of kidney transplant recipient follow-up to illustrate the utility of the method.     

Multivariate generalized hidden Markov regression models with random covariates: Physical exercise in an elderly population

A time‐varying latent variable model is proposed to jointly analyze multivariate mixed‐support longitudinal data. The proposal can be viewed as an extension of hidden Markov regression models with fixed covariates (HMRMFCs), which is the state of the art for modelling longitudinal data, with a special focus on the underlying clustering structure. HMRMFCs are inadequate for applications in which a clustering structure can be identified in the distribution of the covariates, as the clustering is independent from the covariates distribution. Here, hidden Markov regression models with random covariates are introduced by explicitly specifying state‐specific distributions for the covariates, with the aim of improving the recovering of the clusters in the data with respect to a fixed covariates paradigm. The hidden Markov regression models with random covariates class is defined focusing on the exponential family, in a generalized linear model framework. Model identifiability conditions are sketched, an expectation‐maximization algorithm is outlined for parameter estimation, and various implementation and operational issues are discussed. Properties of the estimators of the regression coefficients, as well as of the hidden path parameters, are evaluated through simulation experiments and compared with those of HMRMFCs. The method is applied to physical activity data.     

The analysis of asthma control under a Markov assumption with use of covariates

In studies of disease states and their relation to evolution, data on the state are usually obtained at in frequent time points during follow-up. Moreover in many applications, there are measured covariates on each individual under study and interest centres on the relationship between these covariates and the disease evolution. We developed a continuous-time Markov model with use of time-dependent covariates and a Markov model with piecewise constant intensities to model asthma evolution. Methods to estimate the effect of covariates on transition intensities, to test the assumption of time homogeneity and to assess goodness-of-fit are proposed. We apply these methods to asthma control. We consider a three-state model and we discuss in detail the analysis of asthma control evolution.     

Analyzing direct and indirect effects of treatment using dynamic path analysis applied to data from the Swiss HIV Cohort Study

When applying survival analysis, such as Cox regression, to data from major clinical trials or other studies, often only baseline covariates are used. This is typically the case even if updated covariates are available throughout the observation period, which leaves large amounts of information unused. The main reason for this is that such time‐dependent covariates often are internal to the disease process, as they are influenced by treatment, and therefore lead to confounded estimates of the treatment effect. There are, however, methods to exploit such covariate information in a useful way. We study the method of dynamic path analysis applied to data from the Swiss HIV Cohort Study. To adjust for time‐dependent confounding between treatment and the outcome ‘AIDS or death’, we carried out the analysis on a sequence of mimicked randomized trials constructed from the original cohort data. To analyze these trials together, regular dynamic path analysis is extended to a composite analysis of weighted dynamic path models. Results using a simple path model, with one indirect effect mediated through current HIV‐1 RNA level, show that most or all of the total effect go through HIV‐1 RNA for the first 4 years. A similar model, but with CD4 level as mediating variable, shows a weaker indirect effect, but the results are in the same direction. There are many reasons to be cautious when drawing conclusions from estimates of direct and indirect effects. Dynamic path analysis is however a useful tool to explore underlying processes, which are ignored in regular analyses. Copyright © 2011 John Wiley & Sons, Ltd.     

Semiparametric methods for multistate survival models in randomised trials

Transform methods have proved effective for networks describing a progression of events. In semi‐Markov networks, we calculated the transform of time to a terminating event from corresponding transforms of intermediate steps. Saddlepoint inversion then provided survival and hazard functions, which integrated, and fully utilised, the network data. However, the presence of censored data introduces significant difficulties for these methods. Many participants in controlled trials commonly remain event‐free at study completion, a consequence of the limited period of follow‐up specified in the trial design. Transforms are not estimable using nonparametric methods in states with survival truncated by end‐of‐study censoring. We propose the use of parametric models specifying residual survival to next event. As a simple approach to extrapolation with competing alternative states, we imposed a proportional incidence (constant relative hazard) assumption beyond the range of study data. No proportional hazards assumptions are necessary for inferences concerning time to endpoint; indeed, estimation of survival and hazard functions can proceed in a single study arm. We demonstrate feasibility and efficiency of transform inversion in a large randomised controlled trial of cholesterol‐lowering therapy, the Long‐Term Intervention with Pravastatin in Ischaemic Disease study. Transform inversion integrates information available in components of multistate models: estimates of transition probabilities and empirical survival distributions. As a by‐product, it provides some ability to forecast survival and hazard functions forward, beyond the time horizon of available follow‐up. Functionals of survival and hazard functions provide inference, which proves sharper than that of log‐rank and related methods for survival comparisons ignoring intermediate events. Copyright © 2013 John Wiley & Sons, Ltd.     

Simulating competing risks data in survival analysis

Competing risks analysis considers time‐to‐first‐event (‘survival time’) and the event type (‘cause’), possibly subject to right‐censoring. The cause‐, i.e. event‐specific hazards, completely determine the competing risk process, but simulation studies often fall back on the much criticized latent failure time model. Cause‐specific hazard‐driven simulation appears to be the exception; if done, usually only constant hazards are considered, which will be unrealistic in many medical situations. We explain simulating competing risks data based on possibly time‐dependent cause‐specific hazards. The simulation design is as easy as any other, relies on identifiable quantities only and adds to our understanding of the competing risks process. In addition, it immediately generalizes to more complex multistate models. We apply the proposed simulation design to computing the least false parameter of a misspecified proportional subdistribution hazard model, which is a research question of independent interest in competing risks. The simulation specifications have been motivated by data on infectious complications in stem‐cell transplanted patients, where results from cause‐specific hazards analyses were difficult to interpret in terms of cumulative event probabilities. The simulation illustrates that results from a misspecified proportional subdistribution hazard analysis can be interpreted as a time‐averaged effect on the cumulative event probability scale. Copyright © 2009 John Wiley & Sons, Ltd.