Mediation analysis for a survival outcome with time-varying exposures, mediators, and confounders

Estimation of separable direct and indirect effects in a continuous-time illness-death model

In this article we study the effect of a baseline exposure on a terminal time-to-event outcome either directly or mediated by the illness state of a continuous-time illness-death process with baseline covariates. We propose a definition of the corresponding direct and indirect effects using the concept of separable (interventionist) effects (Robins and Richardson in Causality and psychopathology: finding the determinants of disorders and their cures, Oxford University Press, 2011; Robins et al. in arXiv:2008.06019 , 2021; Stensrud et al. in J Am Stat Assoc 117:175-183, 2022). Our proposal generalizes Martinussen and Stensrud (Biometrics 79:127-139, 2023) who consider similar causal estimands for disentangling the causal treatment effects on the event of interest and competing events in the standard continuous-time competing risk model. Unlike natural direct and indirect effects (Robins and Greenland in Epidemiology 3:143-155, 1992; Pearl in Proceedings of the seventeenth conference on uncertainty in artificial intelligence, Morgan Kaufmann, 2001) which are usually defined through manipulations of the mediator independently of the exposure (so-called cross-world interventions), separable direct and indirect effects are defined through interventions on different components of the exposure that exert their effects through distinct causal mechanisms. This approach allows us to define meaningful mediation targets even though the mediating event is truncated by the terminal event. We present the conditions for identifiability, which include some arguably restrictive structural assumptions on the treatment mechanism, and discuss when such assumptions are valid. The identifying functionals are used to construct plug-in estimators for the separable direct and indirect effects. We also present multiply robust and asymptotically efficient estimators based on the efficient influence functions. We verify the theoretical properties of the estimators in a simulation study, and we demonstrate the use of the estimators using data from a Danish registry study.

Time-to-event surrogate endpoint validation using mediation analysis and meta-analytic data

With the ongoing development of treatments and the resulting increase in survival in oncology, clinical trials based on endpoints such as overall survival may require long follow-up periods to observe sufficient events and ensure adequate statistical power. This increase in follow-up time may compromise the feasibility of the study. The use of surrogate endpoints instead of final endpoints may be attractive for these studies. However, before a surrogate can be used in a clinical trial, it must be statistically validated. In this article, we propose an approach to validate surrogates when both the surrogate and final endpoints are censored event times. This approach is developed for meta-analytic data and uses a mediation analysis to decompose the total effect of the treatment on the final endpoint as a direct effect and an indirect effect through the surrogate. The meta-analytic nature of the data is accounted for in a joint model with random effects at the trial level. The proportion of the indirect effect over the total effect of the treatment on the final endpoint can be computed from the parameters of the model and used as a measure of surrogacy. We applied this method to investigate time-to-relapse as a surrogate endpoint for overall survival in resectable gastric cancer.

Exploring causal mechanisms and quantifying direct and indirect effects using a joint modeling approach for recurrent and terminal events

Recurrent events are commonly encountered in biomedical studies. In many situations, there exist terminal events, such as death, which are potentially related to the recurrent events. Joint models of recurrent and terminal events have been proposed to address the correlation between recurrent events and terminal events. However, there is a dearth of suitable methods to rigorously investigate the causal mechanisms between specific exposures, recurrent events, and terminal events. For example, it is of interest to know how much of the total effect of the primary exposure of interest on the terminal event is through the recurrent events, and whether preventing recurrent event occurrences could lead to better overall survival. In this work, we propose a formal causal mediation analysis method to compute the natural direct and indirect effects. A novel joint modeling approach is used to take the recurrent event process as the mediator and the survival endpoint as the outcome. This new joint modeling approach allows us to relax the commonly used "sequential ignorability" assumption. Simulation studies show that our new model has good finite sample performance in estimating both model parameters and mediation effects. We apply our method to an AIDS study to evaluate how much of the comparative effectiveness of the two treatments and the effect of CD4 counts on the overall survival are mediated by recurrent opportunistic infections.

Effect Measure Modification by Covariates in Mediation: Extending Regression-based Causal Mediation Analysis

Existing methods for regression-based mediation analysis assume that the exposure-mediator effect, exposure-outcome effect, and mediator-outcome effect are constant across levels of the baseline characteristics of patients. However, investigators often have insight into how these underlying effects may be modified by baseline characteristics and are interested in how the resulting mediation effects, such as the natural direct effect (NDE), the natural indirect effect. (NIE), and the proportion mediated, are modified by these baseline characteristics. Motivated by an empirical example of anti-interleukin-1 therapy's benefit on incident anemia reduction and its mediation by an early change in an inflammatory biomarker, we extended the closed-form regression-based causal mediation analysis with effect measure modification (EMM). Using a simulated numerical example, we demonstrated that naive analysis without considering EMM can give biased estimates of NDE and NIE and visually illustrated how baseline characteristics affect the presence and magnitude of EMM of NDE and NIE. We then applied the extended method to the empirical example informed by pathophysiologic insights into potential EMM by age, diabetes, and baseline inflammation. We found that the proportion modified through the early post-treatment inflammatory biomarker was greater for younger, nondiabetic patients with lower baseline level of inflammation, suggesting differential usefulness of the early post-treatment inflammatory biomarker in monitoring patients depending on baseline characteristics. To facilitate the adoption of EMM considerations in causal mediation analysis by the wider clinical and epidemiologic research communities, we developed a free- and open-source R package, regmedint.

A multistate approach for the study of interventions on an intermediate time-to-event in health disparities research

We propose a novel methodology to quantify the effect of stochastic interventions for a non-terminal intermediate time-to-event on a terminal time-to-event outcome. Investigating these effects is particularly important in health disparities research when we seek to quantify inequities in the timely delivery of treatment and its impact on patients' survival time. Current approaches fail to account for time-to-event intermediates and semi-competing risks arising in this setting. Under the potential outcome framework, we define causal contrasts relevant in health disparities research and provide identifiability conditions when stochastic interventions on an intermediate non-terminal time-to-event are of interest. Causal contrasts are estimated in continuous time within a multistate modeling framework and analytic formulae for the estimators of the causal contrasts are developed. We show via simulations that ignoring censoring in intermediate and/or terminal time-to-event processes or ignoring semi-competing risks may give misleading results. This work demonstrates that a rigorous definition of the causal effects and joint estimation of the terminal outcome and intermediate non-terminal time-to-event distributions are crucial for valid investigation of interventions and mechanisms in continuous time. We employ this novel methodology to investigate the role of delaying treatment uptake in explaining racial disparities in cancer survival in a cohort study of colon cancer patients.

A Causal Mediation Model for Longitudinal Mediators and Survival Outcomes with an Application to Animal Behavior

In animal behavior studies, a common goal is to investigate the causal pathways between an exposure and outcome, and a mediator that lies in between. Causal mediation analysis provides a principled approach for such studies. Although many applications involve longitudinal data, the existing causal mediation models are not directly applicable to settings where the mediators are measured on irregular time grids. In this paper, we propose a causal mediation model that accommodates longitudinal mediators on arbitrary time grids and survival outcomes simultaneously. We take a functional data analysis perspective and view longitudinal mediators as realizations of underlying smooth stochastic processes. We define causal estimands of direct and indirect effects accordingly and provide corresponding identification assumptions. We employ a functional principal component analysis approach to estimate the mediator process and propose a Cox hazard model for the survival outcome that flexibly adjusts the mediator process. We then derive a g-computation formula to express the causal estimands using the model coefficients. The proposed method is applied to a longitudinal data set from the Amboseli Baboon Research Project to investigate the causal relationships between early adversity, adult physiological stress responses, and survival among wild female baboons. We find that adversity experienced in early life has a significant direct effect on females' life expectancy and survival probability, but find little evidence that these effects were mediated by markers of the stress response in adulthood. We further developed a sensitivity analysis method to assess the impact of potential violation to the key assumption of sequential ignorability. Supplementary materials accompanying this paper appear on-line.

Effectiveness of a targeted brief intervention for recent suicide attempt survivors: a randomised controlled trial protocol

INTRODUCTION

Effective, brief, low-cost interventions for suicide attempt survivors are essential to saving lives and achieving the goals of the National Strategy for Suicide Prevention and Zero Suicide. This study aims to examine the effectiveness of the Attempted Suicide Short Intervention Program (ASSIP) in averting suicide reattempts in the United States healthcare system, its psychological mechanisms as predicted by the Interpersonal Theory of Suicide, and the potential implementation costs, barriers and facilitators for delivering it.

METHODS AND ANALYSIS

This study is a hybrid type 1 effectiveness-implementation randomised controlled trial (RCT). ASSIP is delivered at three outpatient mental healthcare clinics in New York State. Participant referral sites include three local hospitals with inpatient and comprehensive psychiatric emergency services, and outpatient mental health clinics. Participants include 400 adults who have had a recent suicide attempt. All are randomised to 'Zero Suicide-Usual Care plus ASSIP' or 'Zero Suicide-Usual Care'. Randomisation is stratified by sex and whether the index attempt is a first suicide attempt or not. Participants complete assessments at baseline, 6 weeks, and 3, 6, 12 and, 18 months. The primary outcome is the time from randomisation to the first suicide reattempt. Prior to the RCT, a 23-person open trial took place, in which 13 participants received 'Zero Suicide-Usual Care plus ASSIP' and 14 completed the first follow-up time point.

ETHICS AND DISSEMINATION

This study is overseen by the University of Rochester, with single Institutional Review Board (#3353) reliance agreements from Nathan Kline Institute (#1561697) and SUNY Upstate Medical University (#1647538). It has an established Data and Safety Monitoring Board. Results will be published in peer-reviewed academic journals, presented at scientific conferences, and communicated to referral organisations. Clinics considering ASSIP may use a stakeholder report generated by this study, including incremental cost-effectiveness data from the provider point of view.

TRIAL REGISTRATION NUMBER

NCT03894462.

Mediation analysis with multiple mediators under unmeasured mediator-outcome confounding

It is often of interest in the health and social sciences to investigate the joint mediation effects of multiple post-exposure mediating variables. Identification of such joint mediation effects generally require no unmeasured confounding of the outcome with respect to the whole set of mediators. As the number of mediators under consideration grows, this key assumption is likely to be violated as it is often infeasible to intervene on any of the mediators. In this article, we develop a simple two-step method of moments estimation procedure to assess mediation with multiple mediators simultaneously in the presence of potential unmeasured mediator-outcome confounding. Our identification result leverages heterogeneity of the population exposure effect on the mediators, which is plausible under a variety of empirical settings. The proposed estimators are illustrated through both simulations and an application to evaluate the mediating effects of post-traumatic stress disorder symptoms in the association between self-efficacy and fatigue among health care workers during the COVID-19 outbreak.

Dietary Intake and Body Mass Index Influence the Risk of Islet Autoimmunity in Genetically At-Risk Children: A Mediation Analysis Using the TEDDY Cohort

Background/Objective

Growth and obesity have been associated with increased risk of islet autoimmunity (IA) and progression to type 1 diabetes. We aimed to estimate the effect of energy-yielding macronutrient intake on the development of IA through BMI.

Research Design and Methods

Genetically at-risk children (n = 5,084) in Finland, Germany, Sweden, and the USA, who were autoantibody negative at 2 years of age, were followed to the age of 8 years, with anthropometric measurements and 3-day food records collected biannually. Of these, 495 (9.7%) children developed IA. Mediation analysis for time-varying covariates (BMI z-score) and exposure (energy intake) was conducted. Cox proportional hazard method was used in sensitivity analysis.

Results

We found an indirect effect of total energy intake (estimates: indirect effect 0.13 [0.05, 0.21]) and energy from protein (estimates: indirect effect 0.06 [0.02, 0.11]), fat (estimates: indirect effect 0.03 [0.01, 0.05]), and carbohydrates (estimates: indirect effect 0.02 [0.00, 0.04]) (kcal/day) on the development of IA. A direct effect was found for protein, expressed both as kcal/day (estimates: direct effect 1.09 [0.35, 1.56]) and energy percentage (estimates: direct effect 72.8 [3.0, 98.0]) and the development of GAD autoantibodies (GADA). In the sensitivity analysis, energy from protein (kcal/day) was associated with increased risk for GADA, hazard ratio 1.24 (95% CI: 1.09, 1.53), p = 0.042.

Conclusions

This study confirms that higher total energy intake is associated with higher BMI, which leads to higher risk of the development of IA. A diet with larger proportion of energy from protein has a direct effect on the development of GADA.

Complete effect decomposition for an arbitrary number of multiple ordered mediators with time-varying confounders: A method for generalized causal multi-mediation analysis

Causal mediation analysis is advantageous for mechanism investigation. In settings with multiple causally ordered mediators, path-specific effects have been introduced to specify the effects of certain combinations of mediators. However, most path-specific effects are unidentifiable. An interventional analog of path-specific effects is adapted to address the non-identifiability problem. Moreover, previous studies only focused on cases with two or three mediators due to the complexity of the mediation formula in a large number of mediators. In this study, we provide a generalized definition of traditional path-specific effects and interventional path-specific effects with a recursive formula, along with the required assumptions for nonparametric identification. Subsequently, a general approach is developed with an arbitrary number of multiple ordered mediators and with time-varying confounders. All methods and software proposed in this study contribute to comprehensively decomposing a causal effect confirmed by data science and help disentangling causal mechanisms in the presence of complicated causal structures among multiple mediators.

Causal Mediation Analysis with Multiple Time-varying Mediators

In longitudinal studies with time-varying exposures and mediators, the mediational g-formula is an important method for the assessment of direct and indirect effects. However, current methodologies based on the mediational g-formula can deal with only one mediator. This limitation makes these methodologies inapplicable to many scenarios. Hence, we develop a novel methodology by extending the mediational g-formula to cover cases with multiple time-varying mediators. We formulate two variants of our approach that are each suited to a distinct set of assumptions and effect definitions and present nonparametric identification results of each variant. We further show how complex causal mechanisms (whose complexity derives from the presence of multiple time-varying mediators) can be untangled. We implemented a parametric method, along with a user-friendly algorithm, in R software. We illustrate our method by investigating the complex causal mechanism underlying the progression of chronic obstructive pulmonary disease. We found that the effects of lung function impairment mediated by dyspnea symptoms accounted for 14.6% of the total effect and that mediated by physical activity accounted for 11.9%. Our analyses thus illustrate the power of this approach, providing evidence for the mediating role of dyspnea and physical activity on the causal pathway from lung function impairment to health status. See video abstract at, http://links.lww.com/EDE/B988 .

Quantifying direct and indirect effect for longitudinal mediator and survival outcome using joint modeling approach

Longitudinal biomarkers are widely used in biomedical and translational researches to monitor the progressions of diseases. Methods have been proposed to jointly model longitudinal data and survival data, but its causal mechanism is yet to be investigated rigorously. Understanding how much of the total treatment effect is through the biomarker is important in understanding the treatment mechanism and evaluating the biomarker. In this work, we propose a causal mediation analysis method to compute the direct and indirect effects, when a joint modeling approach is used to take the longitudinal biomarker as the mediator and the survival endpoint as the outcome. Such a joint modeling approach allows us to relax the commonly used "sequential ignorability" assumption. We demonstrate how to evaluate longitudinally measured biomarkers using our method with two case studies, an AIDS study and a liver cirrhosis study.

Longitudinal mediation analysis of time-to-event endpoints in the presence of competing risks

This proposal is motivated by an analysis of the English Longitudinal Study of Ageing (ELSA), which aims to investigate the role of loneliness in explaining the negative impact of hearing loss on dementia. The methodological challenges that complicate this mediation analysis include the use of a time-to-event endpoint subject to competing risks, as well as the presence of feedback relationships between the mediator and confounders that are both repeatedly measured over time. To account for these challenges, we introduce path-specific effect proportional (cause-specific) hazard models. These extend marginal structural proportional (cause-specific) hazard models to enable effect decomposition on either the cause-specific hazard ratio scale or the cumulative incidence function scale. We show that under certain ignorability assumptions, the path-specific direct and indirect effects indexing this model are identifiable from the observed data. We next propose an inverse probability weighting approach to estimate these effects. On the ELSA data, this approach reveals little evidence that the total effect of hearing loss on dementia is mediated through the feeling of loneliness, with a non-statistically significant indirect effect equal to 1.01 (hazard ratio (HR) scale; 95% confidence interval (CI) 0.99 to 1.05).

Transplantation Mediates Much of the Racial Disparity in Survival from Childhood-Onset Kidney Failure

BACKGROUND

The role of kidney transplantation in differential survival in Black and White patients with childhood-onset kidney failure is unexplored.

METHODS

We analyzed 30-year cohort data of children beginning RRT before 18 years of age between January 1980 and December 2017 (n=28,337) in the US Renal Data System. Cox regression identified transplant factors associated with survival by race. The survival mediational g-formula estimated the excess mortality among Black patients that could be eliminated if an intervention equalized their time with a transplant to that of White patients.

RESULTS

Black children comprised 24% of the cohort and their crude 30-year survival was 39% compared with 57% for White children (log rank P<0.001). Black children had 45% higher risk of death (adjusted hazard ratio [aHR], 1.45; 95% confidence interval [95% CI], 1.36 to 1.54), 31% lower incidence of first transplant (aHR, 0.69; 95% CI, 0.67 to 0.72), and 39% lower incidence of second transplant (aHR, 0.61; 95% CI, 0.57 to 0.65). Children and young adults are likely to require multiple transplants, yet even after their first transplant, Black patients had 11% fewer total transplants (adjusted incidence rate ratio [aIRR], 0.89; 95% CI, 0.86 to 0.92). In Black patients, grafts failed earlier after first and second transplants. Overall, Black patients spent 24% less of their RRT time with a transplant than did White patients (aIRR, 0.76; 95% CI, 0.74 to 0.78). Transplantation compared with dialysis strongly protected against death (aHR, 0.28; 95% CI, 0.16 to 0.48) by time-varying analysis. Mediation analyses estimated that equalizing transplant duration could prevent 35% (P<0.001) of excess deaths in Black patients.

CONCLUSIONS

Equalizing time with a functioning transplant for Black patients may equalize survival of childhood-onset ESKD with White patients.

Path-specific effects in the presence of a survival outcome and causally ordered multiple mediators with application to genomic data

Causal multimediation analysis (i.e. the causal mediation analysis with multiple mediators) is critical for understanding the effectiveness of interventions, especially in medical research. Deriving the path-specific effects of exposure on the outcome through a set of mediators can provide detail about the causal mechanism of interest However, existing models are usually restricted to partial decomposition, which can only be used to evaluate the cumulative effect of several paths. In genetics studies, partial decomposition fails to reflect the real causal effects mediated by genes, especially in complex gene regulatory networks. Moreover, because of the lack of a generalized identification procedure, the current multimediation analysis cannot be applied to the estimation of path-specific effects for any number of mediators. In this study, we derive the interventional analogs of path-specific effect for complete decomposition to address the difficulty of nonidentifiability. On the basis of two survival models of the outcome, we derive the generalized analytic forms for interventional analogs of path-specific effects by assuming the normal distributions of mediators. We apply the new methodology to investigate the causal mechanism of signature genes in lung cancer based on the cell cycle pathway, and the results clarify the gene pathway in cancer.

G-Computation to Causal Mediation Analysis With Sequential Multiple Mediators-Investigating the Vulnerable Time Window of HBV Activity for the Mechanism of HCV Induced Hepatocellular Carcinoma

Regression-based approaches are widely used in causal mediation analysis. The presence of multiple mediators, however, increases the complexity and difficulty of mediation analysis. In such cases, regression-based approaches cannot efficiently address estimation issues. Hence, a flexible approach to mediation analysis is needed. Therefore, we developed a method for using g-computation algorithm to conduct causal mediation analysis in the presence of multiple ordered mediators. Compared to regression-based approaches, the proposed simulation-based approach increases flexibility in the choice of models and increases the range of the outcome scale. The Taiwanese Cohort Study dataset was used to evaluate the efficacy of the proposed approach for investigating the mediating role of early and late HBV viral load in the effect of HCV infection on hepatocellular carcinoma (HCC) in HBV seropositive patients (n = 2,878; HCV carrier n = 123). Our results indicated that early HBV viral load had a negative mediating role in HCV-induced HCC. Additionally, early exposure to a low HBV viral load affected HCC through a lag effect on HCC incidence [OR = 0.873, 95% CI = (0.853, 0.893)], and the effect of early exposure to a low HBV viral load on HCC incidence was slightly larger than that of a persistently low viral load on HCC incidence [OR = 0.918, 95% CI = (0.896, 0.941)].

Neighborhood greenspace and risk of type 2 diabetes in a prospective cohort: the Multi-Ethncity Study of Atherosclerosis

BACKGROUND

Neighborhood greenspaces provide opportunities for increased physical activity and social interaction, and thus may reduce the risk of Type 2 diabetes. However, there is little robust research on greenspace and diabetes. In this study, we examine the longitudinal association between neighborhood greenspace and incident diabetes in the Multi-Ethnic Study of Atherosclerosis.

METHODS

A prospective cohort study (N = 6814; 2000-2018) was conducted to examine the association between greenspace, measured as annual and high vegetation season median greenness determined by satellite (Normalized Difference Vegetation Index) within 1000 m of participant homes, and incident diabetes assessed at clinician visits, defined as a fasting glucose level of at least 126 mg/dL, use of insulin or use of hypoglycemic medication, controlling for covariates in stages. Five thousand five hundred seventy-four participants free of prevalent diabetes at baseline were included in our analysis.

RESULTS

Over the study period, 886 (15.9%) participants developed diabetes. Adjusting for individual characteristics, individual and neighborhood-scale SES, additional neighborhood factors, and diabetes risk factors, we found a 21% decrease in the risk of developing diabetes per IQR increase in greenspace (HR: 0.79; 95% CI: 0.63, 0.99).

CONCLUSIONS

Overall, neighborhood greenspace provides a protective influence in the development of diabetes, suggesting that neighborhood-level urban planning that supports access to greenspace--along with healthy behaviors--may aid in diabetes prevention. Additional research is needed to better understand how an area's greenness influences diabetes risk, how to better characterize greenspace exposure and usage, and future studies should focus on robust adjustment for neighborhood-level confounders.

Educational note: causal decomposition of population health differences using Monte Carlo integration and the g-formula

One key objective of the population health sciences is to understand why one social group has different levels of health and well-being compared with another. Whereas several methods have been developed in economics, sociology, demography, and epidemiology to answer these types of questions, a recent method introduced by Jackson and VanderWeele (2018) provided an update to decompositions by anchoring them within causal inference theory. In this paper, we demonstrate how to implement the causal decomposition using Monte Carlo integration and the parametric g-formula. Causal decomposition can help to identify the sources of differences across populations and provide researchers with a way to move beyond estimating inequalities to explaining them and determining what can be done to reduce health disparities. Our implementation approach can easily and flexibly be applied for different types of outcome and explanatory variables without having to derive decomposition equations. We describe the concepts of the approach and the practical steps and considerations needed to implement it. We then walk through a worked example in which we investigate the contribution of smoking to sex differences in mortality in South Korea. For this example, we provide both pseudocode and R code using our package, cfdecomp. Ultimately, we outline how to implement a very general decomposition algorithm that is grounded in counterfactual theory but still easy to apply to a wide range of situations.

Causal inference for semi-competing risks data

The causal effects of Apolipoprotein E $\epsilon4$ allele (APOE) on late-onset Alzheimer's disease (AD) and death are complicated to define because AD may occur under one intervention but not under the other, and because AD occurrence may affect age of death. In this article, this dual outcome scenario is studied using the semi-competing risks framework for time-to-event data. Two event times are of interest: a nonterminal event time (age at AD diagnosis), and a terminal event time (age at death). AD diagnosis time is observed only if it precedes death, which may occur before or after AD. We propose new estimands for capturing the causal effect of APOE on AD and death. Our proposal is based on a stratification of the population with respect to the order of the two events. We present a novel assumption utilizing the time-to-event nature of the data, which is more flexible than the often-invoked monotonicity assumption. We derive results on partial identifiability, suggest a sensitivity analysis approach, and give conditions under which full identification is possible. Finally, we present and implement nonparametric and semiparametric estimation methods under right-censored semi-competing risks data for studying the complex effect of APOE on AD and death.

Mediation model with a categorical exposure and a censored mediator with application to a genetic study

Mediation analysis is a statistical method for evaluating the direct and indirect effects of an exposure on an outcome in the presence of a mediator. Mediation models have been widely used to determine direct and indirect contributions of genetic variants in clinical phenotypes. In genetic studies, the additive genetic model is the most commonly used model because it can detect effects from either recessive or dominant models (or any model in between). However, the existing approaches for mediation model cannot be directly applied when the genetic model is additive (e.g. the most commonly used model for SNPs) or categorical (e.g. polymorphic loci), and thus modification to measures of indirect and direct effects is warranted. In this study, we proposed overall measures of indirect, direct, and total effects for a mediation model with a categorical exposure and a censored mediator, which accounts for the frequency of different values of the categorical exposure. The proposed approach provides the overall contribution of the categorical exposure to the outcome variable. We assessed the empirical performance of the proposed overall measures via simulation studies and applied the measures to evaluate the mediating effect of a women’s age at menopause on the association between genetic variants and type 2 diabetes.

A survival mediation model with Bayesian model averaging

Determining the extent to which a patient is benefiting from cancer therapy is challenging. Criteria for quantifying the extent of "tumor response" observed within a few cycles of treatment have been established for various types of solids as well as hematologic malignancies. These measures comprise the primary endpoints of phase II trials. Regulatory approvals of new cancer therapies, however, are usually contingent upon the demonstration of superior overall survival with randomized evidence acquired with a phase III trial comparing the novel therapy to an appropriate standard of care treatment. With nearly two-thirds of phase III oncology trials failing to achieve statistically significant results, researchers continue to refine and propose new surrogate endpoints. This article presents a Bayesian framework for studying relationships among treatment, patient subgroups, tumor response, and survival. Combining classical components of a mediation analysis with Bayesian model averaging, the methodology is robust to model misspecification among various possible relationships among the observable entities. A posterior inference is demonstrated via an application to a randomized controlled phase III trial in metastatic colorectal cancer. Moreover, the article details posterior predictive distributions of survival and statistical metrics for quantifying the extent of direct and indirect, or tumor response mediated treatment effects.

Mediational g-formula for time-varying treatment and repeated-measured multiple mediators: Application to atorvastatin's effect on cardiovascular disease via cholesterol lowering and anti-inflammatory actions in elderly type 2 diabetics

Modern causal mediation theory has formalized several types of indirect and direct effects of treatment on outcomes regarding specific mediator variables. We reviewed and unified distinct approaches to estimate the "interventional" direct and indirect effects for multiple mediators and time-varying variables. This study was motivated by a clinical trial of elderly type-2 diabetic patients in which atorvastatin was widely prescribed to control patients' cholesterol levels to reduce diabetic complications, including cardiovascular disease. Among atorvastatin's preventive side-effects (pleiotropic effects), we focus on its anti-inflammatory action as measured by white blood cell counts. Hence, we estimate atorvastatin's interventional indirect effects through cholesterol lowering and through anti-inflammatory action, and interventional direct effect bypassing these two actions. In our analysis, total effect (six-year cardiovascular disease risk difference) estimated by standard plug-in g-formula of -3.65% (95% confidence interval: -10.29%, 4.38%) is decomposed into indirect effect via low-density lipoprotein cholesterol (-0.90% [-1.91%, -0.07%]), via white blood cell counts (-0.03% [-0.22%, 0.11%]), and direct effect (-2.84% [-9.71%, 5.41%]) by the proposed parametric mediational g-formula. The SAS program and its evaluation via simulated datasets are provided in the Supplemental materials.

Survival mediation analysis with the death-truncated mediator: The completeness of the survival mediation parameter

In medical research, the development of mediation analysis with a survival outcome has facilitated investigation into causal mechanisms. However, studies have not discussed the death-truncation problem for mediators, the problem being that conventional mediation parameters cannot be well defined in the presence of a truncated mediator. In the present study, we systematically defined the completeness of causal effects to uncover the gap, in conventional causal definitions, between the survival and nonsurvival settings. We propose a novel approach to redefining natural direct and indirect effects, which are generalized forms of conventional causal effects for survival outcomes. Furthermore, we developed three statistical methods for the binary outcome of survival status and formulated a Cox model for survival time. We performed simulations to demonstrate that the proposed methods are unbiased and robust. We also applied the proposed method to explore the effect of hepatitis C virus infection on mortality, as mediated through hepatitis B viral load.

Identification of causal intervention effects under contagion

Defining and identifying causal intervention effects for transmissible infectious disease outcomes is challenging because a treatment - such as a vaccine - given to one individual may affect the infection outcomes of others. Epidemiologists have proposed causal estimands to quantify effects of interventions under contagion using a two-person partnership model. These simple conceptual models have helped researchers develop causal estimands relevant to clinical evaluation of vaccine effects. However, many of these partnership models are formulated under structural assumptions that preclude realistic infectious disease transmission dynamics, limiting their conceptual usefulness in defining and identifying causal treatment effects in empirical intervention trials. In this paper, we propose causal intervention effects in two-person partnerships under arbitrary infectious disease transmission dynamics, and give nonparametric identification results showing how effects can be estimated in empirical trials using time-to-infection or binary outcome data. The key insight is that contagion is a causal phenomenon that induces conditional independencies on infection outcomes that can be exploited for the identification of clinically meaningful causal estimands. These new estimands are compared to existing quantities, and results are illustrated using a realistic simulation of an HIV vaccine trial.

Longitudinal Mediation Analysis Using Natural Effect Models

Mediation analysis is concerned with the decomposition of the total effect of an exposure on an outcome into the indirect effect, through a given mediator, and the remaining direct effect. This is ideally done using longitudinal measurements of the mediator, which capture the mediator process more finely. However, longitudinal measurements pose challenges for mediation analysis, because the mediators and outcomes measured at a given time point can act as confounders for the association between mediators and outcomes at a later time point; these confounders are themselves affected by the prior exposure and outcome. Such posttreatment confounding cannot be dealt with using standard methods (e.g., generalized estimating equations). Analysis is further complicated by the need for so-called cross-world counterfactuals to decompose the total effect. This work addresses these challenges. In particular, we introduce so-called natural effect models, which parameterize the direct and indirect effect of a baseline exposure with respect to a longitudinal mediator and outcome. These can be viewed as a generalization of marginal structural mean models to enable effect decomposition. We introduce inverse probability weighting techniques for fitting these models, adjusting for (measured) time-varying confounding of the mediator-outcome association. Application of this methodology uses data from the Millennium Cohort Study, a longitudinal study of children born in the United Kingdom between September 2000 and January 2002.

Explaining the link between adiposity and colorectal cancer risk in men and postmenopausal women in the UK Biobank: A sequential causal mediation analysis

Mechanisms underlying adiposity-colorectal cancer (CRC) association are incompletely understood. Using UK Biobank data, we investigated the role of C-reactive protein (CRP), hemoglobin-A1c (HbA1c) and (jointly) sex hormone-binding globulin (SHBG) and testosterone, in explaining this association. Total effect of obesity versus normal-weight (based on waist circumference, body mass index, waist-hip ratio) on CRC risk was decomposed into natural direct (NDE) and indirect (NIE) effects using sequential mediation analysis. After a median follow-up of 7.1 years, 2070 incident CRC cases (men = 1,280; postmenopausal women = 790) were recorded. For men, the adjusted risk ratio (RR) for waist circumference (≥102 vs. ≤94 cm) was 1.37 (95% confidence interval [CI], 1.19-1.58). The RRsNIE were 1.08 (95% CI: 1.01-1.16) through all biomarkers, 1.06 (95% CI: 1.01-1.11) through pathways influenced by CRP, 0.99 (95% CI: 0.97-1.01) through HbA1c beyond (the potential influence of) CRP and 1.03 (95% CI: 0.99-1.08) through SHBG and testosterone combined beyond CRP and HbA1c. The RRNDE was 1.26 (95% CI: 1.09-1.47). For women, the RR for waist circumference (≥88 vs. ≤80 cm) was 1.27 (95% CI: 1.07-1.50). The RRsNIE were 1.08 (95% CI: 0.94-1.22) through all biomarkers, 1.08 (95% CI: 0.99-1.17) through CRP, 1.00 (95% CI: 0.98-1.02) through HbA1c beyond CRP and 1.00 (95% CI: 0.92-1.09) through SHBG and testosterone combined beyond CRP and HbA1c. The RRNDE was 1.18 (95% CI: 0.96-1.45). For men and women, pathways influenced by CRP explained a small proportion of the adiposity-CRC association. Testosterone and SHBG also explained a small proportion of this association in men. These results suggest that pathways marked by these obesity-related factors may not explain a large proportion of the adiposity-CRC association.

Adiposity and estrogen receptor-positive, postmenopausal breast cancer risk: Quantification of the mediating effects of fasting insulin and free estradiol

Adiposity increases estrogen receptor (ER)-positive postmenopausal breast cancer risk. While mechanisms underlying this relationship are uncertain, dysregulated sex-steroid hormone production and insulin signaling are likely pathways. Our aim was to quantify mediating effects of fasting insulin and free estradiol in the adiposity and ER-positive postmenopausal breast cancer association. We used data from a case-cohort study of sex hormones and insulin signaling nested within the Melbourne Collaborative Cohort Study. Eligible women, at baseline, were not diagnosed with cancer, were postmenopausal, did not use hormone therapy and had no history of diabetes or diabetes medication use. Women with ER-negative disease or breast cancer diagnosis within the first follow-up year were excluded. We analyzed the study as a cumulative sampling case-control study with 149 cases and 1,029 controls. Missing values for insulin and free estradiol were multiply imputed with chained equations. Interventional direct (IDE) and indirect (IIE) effects were estimated using regression-based multiple-mediator approach. For women with body mass index (BMI) >30 kg/m2 compared to women with BMI 18.5-25 kg/m2 , the risk ratio (RR) of breast cancer was 1.75 (95% confidence interval [CI] 1.05-2.91). The estimated IDE (RR) not through the mediators was 1.03 (95% CI 0.43-2.48). Percentage mediated effect through free estradiol was 72% (IIE-RR 1.56; 95% CI 1.11-2.19). There was no evidence for an indirect effect through insulin (IIE-RR 1.12; 95% CI 0.68-1.84; 28% mediated). Our results suggest that circulating free estradiol plays an important mediating role in the adiposity-breast cancer relationship but does not explain all of the association.

Racial/Ethnic Disparities in Atrial Fibrillation Treatment and Outcomes among Dialysis Patients in the United States

BACKGROUND

Because stroke prevention is a major goal in the management of ESKD hemodialysis patients with atrial fibrillation, investigating racial/ethnic disparities in stroke among such patients is important to those who could benefit from strategies to maximize preventive measures.

METHODS

We used the United States Renal Data System to identify ESKD patients who initiated hemodialysis from 2006 to 2013 and then identified those with a subsequent atrial fibrillation diagnosis and Medicare Part A/B/D. Patients were followed for 1 year for all-cause stroke, mortality, prescription medications, and cardiovascular disease procedures. The survival mediational g-formula quantified the percentage of excess strokes attributable to lower use of atrial fibrillation treatments by race/ethnicity.

RESULTS

The study included 56,587 ESKD hemodialysis patients with atrial fibrillation. Black, white, Hispanic, and Asian patients accounted for 19%, 69%, 8%, and 3% of the population, respectively. Compared with white patients, black, Hispanic, or Asian patients were more likely to experience stroke (13%, 15%, and 16%, respectively) but less likely to fill a warfarin prescription (10%, 17%, and 28%, respectively). Warfarin prescription was associated with decreased stroke rates. Analyses suggested that equalizing the warfarin distribution to that in the white population would prevent 7%, 10%, and 12% of excess strokes among black, Hispanic, and Asian patients, respectively. We found no racial/ethnic disparities in all-cause mortality or use of cardiovascular disease procedures.

CONCLUSIONS

Racial/ethnic disparities in all-cause stroke among hemodialysis patients with atrial fibrillation are partially mediated by lower use of anticoagulants among black, Hispanic, and Asian patients. The reasons for these disparities are unknown, but strategies to maximize stroke prevention in minority hemodialysis populations should be further investigated.

Sedentary Behavior and Chronic Disease: Mechanisms and Future Directions

Background: Recent updates to physical activity guidelines highlight the importance of reducing sedentary time. However, at present, only general recommendations are possible (ie, “Sit less, move more”). There remains a need to investigate the strength, temporality, specificity, and dose–response nature of sedentary behavior associations with chronic disease, along with potential underlying mechanisms. Methods: Stemming from a recent research workshop organized by the Sedentary Behavior Council themed “Sedentary behaviour mechanisms—biological and behavioural pathways linking sitting to adverse health outcomes,” this paper (1) discusses existing challenges and scientific discussions within this advancing area of science, (2) highlights and discusses emerging areas of interest, and (3) points to potential future directions. Results: A brief knowledge update is provided, reflecting upon current and evolving thinking/discussions, and the rapid accumulation of new evidence linking sedentary behavior to chronic disease. Research “action points” are made at the end of each section—spanning from measurement systems and analytic methods, genetic epidemiology, causal mediation, and experimental studies to biological and behavioral determinants and mechanisms. Conclusion: A better understanding of whether and how sedentary behavior is causally related to chronic disease will allow for more meaningful conclusions in the future and assist in refining clinical and public health policies/recommendations.

Effect decomposition through multiple causally nonordered mediators in the presence of exposure-induced mediator-outcome confounding

Avin et al (2005) showed that, in the presence of exposure-induced mediator-outcome confounding, decomposing the total causal effect (TCE) using standard conditional exchangeability assumptions is not possible even under a nonparametric structural equation model with all confounders observed. Subsequent research has investigated the assumptions required for such a decomposition to be identifiable and estimable from observed data. One approach was proposed by VanderWeele et al (2014). They decomposed the TCE under three different scenarios: (1) treating the mediator and the exposure-induced confounder as joint mediators; (2) generating path-specific effects albeit without distinguishing between multiple distinct paths through the exposure-induced confounder; and (3) using so-called randomised interventional analogues where sampling values from the distribution of the mediator within the levels of the exposure effectively marginalises over the exposure-induced confounder. In this paper, we extend their approach to the case where there are multiple mediators that do not influence each other directly but which are all influenced by an exposure-induced mediator-outcome confounder. We provide a motivating example and results from a simulation study based on from our work in dental epidemiology featuring the 1982 Pelotas Birth Cohort in Brazil.

Mediation analysis of time-to-event endpoints accounting for repeatedly measured mediators subject to time-varying confounding

In this article, we will present statistical methods to assess to what extent the effect of a randomised treatment (versus control) on a time-to-event endpoint might be explained by the effect of treatment on a mediator of interest, a variable that is measured longitudinally at planned visits throughout the trial. In particular, we will show how to identify and infer the path-specific effect of treatment on the event time via the repeatedly measured mediator levels. The considered proposal addresses complications due to patients dying before the mediator is assessed, due to the mediator being repeatedly measured, and due to posttreatment confounding of the effect of the mediator by other mediators. We illustrate the method by an application to data from the LEADER cardiovascular outcomes trial.

Defining causal mediation with a longitudinal mediator and a survival outcome

In the context of causal mediation analysis, prevailing notions of direct and indirect effects are based on nested counterfactuals. These can be problematic regarding interpretation and identifiability especially when the mediator is a time-dependent process and the outcome is survival or, more generally, a time-to-event outcome. We propose and discuss an alternative definition of mediated effects that does not suffer from these problems, and is more transparent than the current alternatives. Our proposal is based on the extended graphical approach of Robins and Richardson (in: Shrout (ed) Causality and psychopathology: finding the determinants of disorders and their cures, Oxford University Press, Oxford, 2011), where treatment is decomposed into different components, or aspects, along different causal paths corresponding to real world mechanisms. This is an interesting alternative motivation for any causal mediation setting, but especially for survival outcomes. We give assumptions allowing identifiability of such alternative mediated effects leading to the familiar mediation g-formula (Robins in Math Model 7:1393, 1986); this implies that a number of available methods of estimation can be applied.

Role of Nonsteroidal Antiinflammatory Drugs in the Association Between Osteoarthritis and Cardiovascular Diseases: A Longitudinal Study

OBJECTIVE

To elucidate the role of nonsteroidal antiinflammatory drugs (NSAIDs) in the increased risk of cardiovascular disease (CVD) among osteoarthritis (OA) patients.

METHODS

This longitudinal study was based on linked health administrative data from British Columbia, Canada. From a population-based cohort of 720,055 British Columbians, we selected 7,743 OA patients and 23,229 age- and sex-matched non-OA controls. We used multivariable Cox proportional hazards models to estimate the risk of developing incident CVD (primary outcome) as well as ischemic heart disease, congestive heart failure, and stroke (secondary outcomes). To estimate the mediating effect of NSAIDs, defined as current use of an NSAID according to linked PharmaNet data, in the OA-CVD relationship, we implemented a marginal structural model.

RESULTS

OA patients had a higher risk of developing CVD than controls without OA. After adjusting for socioeconomic status, body mass index, hypertension, diabetes, hyperlipidemia, chronic obstructive pulmonary disease, and Romano comorbidity score, the adjusted hazard ratio (HR) was 1.23 (95% confidence interval [95% CI] 1.17-1.28). The adjusted HRs for congestive heart failure, ischemic heart disease, and stroke were 1.42 (95% CI 1.33-1.51), 1.17 (95% CI 1.10-1.26), and 1.14 (95% CI 1.07-1.22), respectively. Approximately 41% of the total effect of OA on increased CVD risk was mediated through NSAIDs. For the secondary outcomes, the proportion mediated through NSAIDs was 23%, 56%, and 64% for congestive heart failure, ischemic heart disease, and stroke, respectively.

CONCLUSION

The findings of this first study to evaluate the mediating role of NSAIDs in the relationship between OA and CVD suggest that NSAID use contributes substantially to the OA-CVD association.

Mediation analysis for new recognition criteria, working hours and overwork-related disease: a nationwide ecological study using 11-year follow-up data in Taiwan

OBJECTIVES

Taiwan revised its criteria for overwork-related cerebrovascular and cardiovascular disease (CCVD) in 2010. A new definition of overwork increased the number of recognised cases. Meanwhile, actual average working hours decreased. We estimated the effects of the revised criteria on the number of overwork-related CCVD cases and the mediation effect through reduced working hours.

METHODS

From the Labor Insurance of Taiwan, we collected data on the total number of overwork-related CCVD cases from 2006 to 2016 and average monthly working hours for 13 industry groups. We conducted causal mediation analysis to investigate the mechanism of the effect of new criteria on CCVD mediated by working hours.

RESULTS

From 2006 to 2016, 594 overwork-related cases of CCVD were recognised across 13 industry groups. After introducing the new criteria, overwork-related CCVD increased by 8.40 cases (per one million person-years) (95% CI 4.53 to 15.05), which resulted from a decrease of 1.54 (95% CI 0.22 to 3.82) cases due to reduced working hours (mediation effect) and an increase of 9.93 (95% CI 5.24 to 18.17) cases related to the effect of the criteria change and other covariates excluding working hours (alternative effect).

CONCLUSIONS

Working hours are an important mediator of the effect of policy on the rate of overwork-related CCVD. Introducing new criteria for recognising overwork-related disease might raise awareness and prompt reductions in working hours, which also help to reduce CCVD. Our findings suggest that understanding mediation effects is important to evaluating national health policies.

Impact of comorbidities and functional impairment on 5-year loss of health utility in patients with lower-limb osteoarthritis in the KHOALA cohort

PURPOSE

To examine the respective and combined impact of "hypothetical" functional impairment (FI) and burden of comorbidities accrual on a 5-year risk of health utility (HU) loss in osteoarthritis (OA).

METHODS

Participants of the Knee and Hip Osteoarthritis Long-term Assessment (KHOALA) study with a 5-year follow-up were included. FI, number of comorbidities and HU were measured annually by the WOMAC, Functional Comorbidity Index and Short-Form 6D, respectively. We estimated the population risk of HU loss (PRD: population risk difference, PRR: population risk ratio) under hypothetical FI and comorbidities using the parametric G-formula. Then, mediation analysis investigated the causal mechanism of comorbidities on HU through FI by estimating total, direct and indirect effects.

RESULTS

We examined data from 767 patients (68.8% women; 61.6 years). The estimated 5-year risk of HU loss was 47.5% [41.9; 52.2] under natural course and 24.9% [15.5; 34.2] when imposing "Patient acceptable function and No comorbidity" corresponding to a PRD = - 22.6 [- 26.5; - 21.2] and a PRR = 0.5 [0.4; 0.6]. The estimated total risk of HU loss comparing "Two comorbidities" versus "No comorbidity" was significant without mediation effect of FI: Total = 10.1% [6.8; 12.9]; direct = 8.0% [2.7; 13.1]; indirect = 2.1% [- 2.0; 5.2].

CONCLUSIONS

FI and comorbidities are important and independent determinants of HU loss in patient with OA. Half of cases (50%) of HU loss during 5 years could be avoided by preventing comorbidities (30%) and limiting FI under patient acceptable function (20%). Caregivers should additionally pay close attention to the prevention and the treatment of comorbidities in routine management of OA.

Marginal Time-Dependent Causal Effects in Mediation Analysis With Survival Data

The main aim of mediation analysis is to study the direct and indirect effects of an exposure on an outcome. To date, the literature on mediation analysis with multiple mediators has mainly focused on continuous and dichotomous outcomes. However, the development of methods for multiple mediation analysis of survival outcomes is still limited. Here we extend to survival outcomes a method for multiple mediation analysis based on the computation of appropriate weights. The approach considered has the advantages of not requiring specific models for mediators, allowing nonindependent mediators of any nature, and not relying on the assumption of rare outcomes. Simulation studies show good performance of the proposed estimator in terms of bias and coverage probability. The method is further applied to an example from a published study on prostate cancer mortality aimed at understanding the extent to which the effect of DNA methyltransferase 3b (DNMT3b) genotype on mortality was explained by DNA methylation and tumor aggressiveness. This approach can be used to quantify the marginal time-dependent direct and indirect effects carried by multiple indirect pathways, and software code is provided to facilitate its application.

Development of a mobile application (App) to delineate "digital chronotype" and the effects of delayed chronotype by bedtime smartphone use

The widespread use and deep reach of smartphones motivate the use of mobile applications to continuously monitor the relationship between circadian system, individual sleep patterns, and environmental effects. We selected 61 adults with 14-day data from the "Know Addiction" database. We developed an algorithm to identify the "sleep time" based on the smartphone behaviors. The total daily smartphone use duration and smartphone use duration prior to sleep onset were identified respectively. We applied mediation analysis to investigate the effects of total daily smartphone use on sleep through pre-sleep use (PS). The results showed participants' averaged pre-sleep episodes within 1 h prior to sleep are 2.58. The duration of three pre-sleep uses (PS_1∼3) maybe a more representative index for smartphone use before sleep. Both total daily duration and the duration of the last three uses prior to sleep of smartphone use significantly delayed sleep onset, midpoint of sleep and reduced total sleep time. One hour of increased smartphone use daily, delays the circadian rhythm by 3.5 min, and reduced 5.5 min of total sleep time (TST). One hour of increased pre-sleep smartphone use delayed circadian rhythm by 1.7 min, and reduced 39 s of TST. The mediation effects of PS_1∼3 significantly impacted on these three sleep indicators. PS_1∼3 accounted for 14.3% of total daily duration, but the proportion mediated of delayed circadian rhythm was 44.0%. We presented "digital chronotype" with an automatic system that can collect high temporal resolution data from naturalistic settings with high ecological validity. Smartphone screen time, mainly mediated by pre-sleep use, delayed the circadian rhythm and reduced the total sleep time.

A review of time scale fundamentals in the g-formula and insidious selection bias

PURPOSE OF REVIEW

We review recent examples of data analysis with the g-formula, a powerful tool for analyzing longitudinal data and survival analysis. Specifically, we focus on the common choices of time scale and review inferential issues that may arise.

RECENT FINDINGS

Researchers are increasingly engaged with questions that require time scales subject to left-truncation and right-censoring. The assumptions necessary for allowing right-censoring are well defined in the literature, whereas similar assumptions for left-truncation are not well defined. Policy and biologic considerations sometimes dictate that observational data must be analyzed on time scales that are subject to left-truncation, such as age.

SUMMARY

Further consideration of left-truncation is needed, especially when biologic or policy considerations dictate that age is the relevant time scale of interest. Methodologic development is needed to reduce potential for bias when left-truncation may occur.

Exploring causality mechanism in the joint analysis of longitudinal and survival data

In many biomedical studies, disease progress is monitored by a biomarker over time, eg, repeated measures of CD4 in AIDS and hemoglobin in end-stage renal disease patients. The endpoint of interest, eg, death or diagnosis of a specific disease, is correlated with the longitudinal biomarker. In this paper, we examine and compare different models of longitudinal and survival data to investigate causal mechanisms, specifically, those related to the role of random effects. We illustrate the methods by data from two clinical trials: an AIDS study and a liver cirrhosis study.