A novel approach to joint prediction of preeclampsia and delivery timing using semicompeting risks.

BACKGROUND: Preeclampsia is a pregnancy complication that contributes substantially to perinatal morbidity and mortality worldwide. Existing approaches to modeling and prediction of preeclampsia typically focus either on predicting preeclampsia risk alone, or on the timing of delivery following a diagnosis of preeclampsia. As such, they are misaligned with typical healthcare interactions during which the 2 events are generally considered simultaneously. OBJECTIVE: This study aimed to describe the "semicompeting risks" framework as an innovative approach for jointly modeling the risk and timing of preeclampsia and the timing of delivery simultaneously. Through this approach, one can obtain, at any point during the pregnancy, clinically relevant summaries of an individual's predicted outcome trajectories in 4 risk categories: not developing preeclampsia and not having delivered, not developing preeclampsia but having delivered because of other causes, developing preeclampsia but not having delivered, and developing preeclampsia and having delivered. STUDY DESIGN: To illustrate the semicompeting risks methodology, we presented an example analysis of a pregnancy cohort from the electronic health record of an urban, academic medical center in Boston, Massachusetts (n=9161 pregnancies). We fit an illness-death model with proportional-hazards regression specifications describing 3 hazards for timings of preeclampsia, delivery in the absence of preeclampsia, and delivery following preeclampsia diagnosis. RESULTS: The results indicated nuanced relationships between a variety of risk factors and the timings of preeclampsia diagnosis and delivery, including maternal age, race/ethnicity, parity, body mass index, diabetes mellitus, chronic hypertension, cigarette use, and proteinuria at 20 weeks' gestation. Sample predictions for a diverse set of individuals highlighted differences in projected outcome trajectories with regard to preeclampsia risk and timing, and timing of delivery either before or after preeclampsia diagnosis. CONCLUSION: The semicompeting risks framework enables characterization of the joint risk and timing of preeclampsia and delivery, providing enhanced, meaningful information regarding clinical decision-making throughout the pregnancy.

An accelerated failure time regression model for illness-death data: A frailty approach.

This work presents a new model and estimation procedure for the illness-death survival data where the hazard functions follow accelerated failure time (AFT) models. A shared frailty variate induces positive dependence among failure times of a subject for handling the unobserved dependency between the nonterminal and the terminal failure times given the observed covariates. The motivation behind the proposed modeling approach is to leverage the well-known interpretability advantage of AFT models with respect to the observed covariates, while also benefiting from the simple and intuitive interpretation of the hazard functions. A semiparametric maximum likelihood estimation procedure is developed via a kernel smoothed-aided expectation-maximization algorithm, and variances are estimated by weighted bootstrap. We consider existing frailty-based illness-death models and place particular emphasis on highlighting the contribution of our current research. The breast cancer data of the Rotterdam tumor bank are analyzed using the proposed as well as existing illness-death models. The results are contrasted and evaluated based on a new graphical goodness-of-fit procedure. Simulation results and data analysis nicely demonstrate the practical utility of the shared frailty variate with the AFT regression model under the illness-death framework.

Nonparametric inference of general while-alive estimands for recurrent events.

Measuring the treatment effect on recurrent events like hospitalization in the presence of death has long challenged statisticians and clinicians alike. Traditional inference on the cumulative frequency unjustly penalizes survivorship as longer survivors also tend to experience more adverse events. Expanding a recently suggested idea of the "while-alive" event rate, we consider a general class of such estimands that adjust for the length of survival without losing causal interpretation. Given a user-specified loss function that allows for arbitrary weighting, we define as estimand the average loss experienced per unit time alive within a target period and use the ratio of this loss rate to measure the effect size. Scaling the loss rate by the width of the corresponding time window gives us an alternative, and sometimes more photogenic, way of showing the data. To make inferences, we construct a nonparametric estimator for the loss rate through the cumulative loss and the restricted mean survival time and derive its influence function in closed form for variance estimation and testing. As simulations and analysis of real data from a heart failure trial both show, the while-alive approach corrects for the false attenuation of treatment effect due to patients living longer under treatment, with increased statistical power as a result. The proposed methods are implemented in the R-package WA, which is publicly available from the Comprehensive R Archive Network (CRAN).

Unified semicompeting risks analysis of hepatitis natural history through mediation modeling.

Natural history of hepatitis B or C is comprised of multiple milestones such as liver cirrhosis and liver cancer. To fully characterize its natural course, semicompeting risks represent a common problem where liver cirrhosis and liver cancer are both of interest, but only the former may be censored by the latter. Copula, frailty and multistate models serve as well-established analytics for semicompeting risks. Here, we cast the semicompeting risks in a mediation framework, with liver cirrhosis as a mediator and liver cancer as an outcome. We define the indirect and direct effects as the effects of an exposure on the liver cancer incidence mediated and not mediated through liver cirrhosis, respectively. With the estimands derived as conditional probabilities, we derive respective expressions under the copula, frailty, and multistate models. Next, we propose estimators based on nonparametric maximum likelihood or U-statistics and establish their asymptotic results. Numerical studies demonstrate that the efficiency of copula models leads to potential bias due to model misspecification. Moreover, the robustness of frailty models is accompanied by a loss in efficiency, and multistate models balance the efficiency and robustness. We demonstrate the utility of the proposed methods by a hepatitis study, showing that hepatitis B and C lead to a higher incidence of liver cancer by increasing liver cirrhosis incidence. Thus, mediation modeling provides a unified framework that accommodates various semicompeting risks models.

Discussion on "Causal mediation of semicompeting risks" by Yen-Tsung Huang.

Huang proposes a method for assessing the impact of a point treatment on mortality either directly or mediated by occurrence of a nonterminal health event, based on data from a prospective cohort study in which the occurrence of the nonterminal health event may be preemptied by death but not vice versa. The author uses a causal mediation framework to formally define causal quantities known as natural (in)direct effects. The novelty consists of adapting these concepts to a continuous-time modeling framework based on counting processes. In an effort to posit "scientifically interpretable estimands," statistical and causal assumptions are introduced for identification. In this commentary, we argue that these assumptions are not only difficult to interpret and justify, but are also likely violated in the hepatitis B motivating example and other survival/time to event settings as well.

Invited Commentary: Opportunities That Come With Studying the Co-Occurrence of Multiple Outcomes.

In almost all clinical settings, patients are at risk for multiple potential events and, in consultation with health-care providers, must weigh the potential benefits and harms across these events when making decisions. As researchers seek to build an evidence base to inform these decisions, they must contend with a choice as to how they will handle the different events. One approach, arguably the standard approach in the literature, is to consider the events individually by conducting analyses and publishing results for each one at a time. Doing so, however, fails to acknowledge or exploit the inherent multivariate nature of the data, represents a lost opportunity, and results in an evidence base that is not aligned with how clinical decision-making is actually performed. The article by Prentice et al. (Am J Epidemiol. 2020;189(9):972-981) in this issue of the Journal moves beyond this standard by illustrating recently developed methods that directly take advantage of information on the co-occurrence of multiple events. Moreover, their article highlights the role of modern methods in deriving additional information and insight from studies of multiple clinical outcomes by making full use of multivariate data, with the goal being to complement, not replace, existing methods.

Semiparametric copula-based analysis for treatment effects in the presence of treatment switching.

In controlled trials, "treatment switching" occurs when patients in one treatment group switch to alternative treatments during the trial, and poses challenges to treatment effect evaluation owing to crossover of the treatments groups. In this work, we assume that treatment switching can occur after some disease progression event and view the progression and death events as two semicompeting risks. The proposed model consists of a copula model for the joint distribution of time-to-progression (TTP) and overall survival (OS) up to the earlier of the two events, as well as a conditional hazard model for OS subsequent to progression. The copula model facilitates assessing the marginal distributions of TTP and OS separately from the association between the two events, and, in particular, the treatment effect on OS in the absence of treatment switching. The proposed conditional hazard model for death subsequent to progression allows us to assess the treatment switching (crossover) effect on OS given occurrence of progression and covariates. Semiparametric proportional hazards models are employed in the marginal models for TTP and OS. A nonparametric maximum likelihood procedure is developed for model inference, which is verified through asymptotic theory and simulation studies. The proposed analysis is applied to a lung cancer dataset to illustrate its real utility.

On shared gamma-frailty conditional Markov model for semicompeting risks data.

Semicompeting risks data are a mixture of competing risks data and progressive state data. This type of data occurs when a nonterminal event is subject to truncation by a well-defined terminal event, but not vice versa. The shared gamma-frailty conditional Markov model (GFCMM) has been used to analyze semicompeting risks data because of its flexibility. There are two versions of this model: the restricted and the unrestricted model. Maximum likelihood estimation methodology has been proposed in the literature. However, we found through numerical experiments that the unrestricted model sometimes yields nonparametrically biased estimation. In this article, we provide a practical guideline for using the GFCMM in the analysis of semicompeting risk data that includes: (a) a score test to assess if the restricted model, which does not exhibit estimation problems, is reasonable under a proportional hazards assumption, and (b) a graphical illustration to justify whether the unrestricted model yields nonparametric estimation with substantial bias for cases where the test provides a statistical significant result against the restricted model. This guideline was applied to the Indianapolis-Ibadan Dementia Project data as an illustration to explore how dementia occurrence changes mortality risk.

Bayesian approach for flexible modeling of semicompeting risks data.

Semicompeting risks data arise when two types of events, non-terminal and terminal, are observed. When the terminal event occurs first, it censors the non-terminal event, but not vice versa. To account for possible dependent censoring of the non-terminal event by the terminal event and to improve prediction of the terminal event using the non-terminal event information, it is crucial to model their association properly. Motivated by a breast cancer clinical trial data analysis, we extend the well-known illness-death models to allow flexible random effects to capture heterogeneous association structures in the data. Our extension also represents a generalization of the popular shared frailty models that usually assume that the non-terminal event does not affect the hazards of the terminal event beyond a frailty term. We propose a unified Bayesian modeling approach that can utilize existing software packages for both model fitting and individual-specific event prediction. The approach is demonstrated via both simulation studies and a breast cancer data set analysis.

MEASURING PERFORMANCE FOR END-OF-LIFE CARE.

Although not without controversy, readmission is entrenched as a hospital quality metric with statistical analyses generally based on fitting a logistic-Normal generalized linear mixed model. Such analyses, however, ignore death as a competing risk, although doing so for clinical conditions with high mortality can have profound effects; a hospital's seemingly good performance for readmission may be an artifact of it having poor performance for mortality. in this paper we propose novel multivariate hospital-level performance measures for readmission and mortality that derive from framing the analysis as one of cluster-correlated semi-competing risks data. We also consider a number of profiling-related goals, including the identification of extreme performers and a bivariate classification of whether the hospital has higher-/lower-than-expected readmission and mortality rates via a Bayesian decision-theoretic approach that characterizes hospitals on the basis of minimizing the posterior expected loss for an appropriate loss function. in some settings, particularly if the number of hospitals is large, the computational burden may be prohibitive. To resolve this, we propose a series of analysis strategies that will be useful in practice. Throughout, the methods are illustrated with data from CMS on N = 17,685 patients diagnosed with pancreatic cancer between 2000-2012 at one of J = 264 hospitals in California.

Receiver operating characteristic curve estimation for time to event with semicompeting risks and interval censoring.

Semicompeting risks and interval censoring are frequent in medical studies, for instance when a disease may be diagnosed only at times of visit and disease onset is in competition with death. To evaluate the ability of markers to predict disease onset in this context, estimators of discrimination measures must account for these two issues. In recent years, methods for estimating the time-dependent receiver operating characteristic curve and the associated area under the ROC curve have been extended to account for right censored data and competing risks. In this paper, we show how an approximation allows to use the inverse probability of censoring weighting estimator for semicompeting events with interval censored data. Then, using an illness-death model, we propose two model-based estimators allowing to rigorously handle these issues. The first estimator is fully model based whereas the second one only uses the model to impute missing observations due to censoring. A simulation study shows that the bias for inverse probability of censoring weighting remains modest and may be less than the one of the two parametric estimators when the model is misspecified. We finally recommend the nonparametric inverse probability of censoring weighting estimator as main analysis and the imputation estimator based on the illness-death model as sensitivity analysis.