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

Longitudinal mediation analysis with multilevel and latent growth models: a separable effects causal approach

BACKGROUND

Causal mediation analysis is widespread in applied medical research, especially in longitudinal settings. However, estimating natural mediational effects in such contexts is often difficult because of the presence of post-treatment confounding. Moreover, many models frequently used in applied research, like multilevel and latent growth models, present an additional difficulty, i.e. the presence of latent variables. In this paper, we propose a causal interpretation of these two classes of models based on a novel type of causal effects called separable, which overcome some of the issues of natural effects.

METHODS

We formally derive conditions for the identifiability of separable mediational effects and their analytical expressions based on the g-formula. We carry out a simulation study to investigate how moderate and severe model misspecification, as well as violation of the identfiability assumptions, affect estimates. We also present an application to real data.

RESULTS

The results show how model misspecification impacts the estimates of mediational effects, particularly in the case of severe misspecification, and that the bias worsens over time. The violation of assumptions affects separable effect estimates in a very different way for the mixed effect and the latent growth models.

CONCLUSION

Our approach allows us to give multilevel and latent growth models an appealing causal interpretation based on separable effects. The simulation study shows that model misspecification can heavily impact effect estimates, highlighting the importance of careful model choice.

Unraveling the Mechanisms Behind the Short-Term Effects of Dry Needling: New Insights From a Mediation Analysis With Repeatedly Measured Mediators and Outcomes

OBJECTIVE

To explore the causal pathways underlying the short-term effects of deep dry needling (DDN) in people with chronic neck pain.

DESIGN

Explanatory longitudinal mediation analysis with repeatedly measured mediators and outcomes.

SETTING

Primary care setting.

PARTICIPANTS

Patients (N=128) with chronic neck pain.

INTERVENTIONS

Participants were randomized into 2 groups; DDN of the neck muscles combined with stretching (n=64) and stretching alone (n=64).

MAIN OUTCOME MEASURES

Two outcomes (pain intensity and neck pain-related disability) and 3 candidate mediators (local pressure pain thresholds [PPTs], cervical range of motion [ROM], and neck muscle strength) were included. Pain intensity was also included as a competing mediator in the mediation analysis for disability. Mediators and outcomes were measured at 3 time points: after intervention and at 2- and 4-week follow-up. Age, sex, and the baseline values of the outcome and mediators were included as pretreatment mediator-outcome confounders.

RESULTS

Reductions in pain intensity strongly mediated the short-term effects of DDN on disability, from after intervention to 4-week follow-up. In addition, the attenuation of local hypersensitivity (via increasing PPTs) moderately mediated reductions in pain intensity at each time point. On the other hand, gains in cervical ROM contributed to reducing neck pain-related disability. Changes in muscle strength did not lead to better outcomes.

CONCLUSIONS

This novel study demonstrated that DDN effect on neck pain-related disability is strongly driven by the analgesic effects of this physical therapy modality. Increasing PPTs and cervical ROM seem to be also part of the mechanisms behind DDN's effect.

Effect of Semaglutide on Regression and Progression of Glycemia in People With Overweight or Obesity but Without Diabetes in the SELECT Trial

OBJECTIVE

To determine whether semaglutide slows progression of glycemia in people with cardiovascular disease and overweight or obesity but without diabetes.

RESEARCH DESIGN AND METHODS

In a multicenter, double-blind trial, participants aged ≥45 years, with BMI ≥27 kg/m2, and with preexisting cardiovascular disease but without diabetes (HbA1c <6.5%) were randomized to receive subcutaneous semaglutide (2.4 mg weekly) or placebo. Major glycemic outcomes were HbA1c and proportions achieving biochemical normoglycemia (HbA1c <5.7%) and progressing to biochemical diabetes (HbA1c ≥6.5%).

RESULTS

Of 17,604 participants, 8,803 were assigned to semaglutide and 8,801 to placebo. Mean ± SD intervention exposure was 152 ± 56 weeks and follow-up 176 ± 40 weeks. In both treatment arms mean nadir HbA1c for participants was at 20 weeks. Thereafter, HbA1c increased similarly in both arms, with a mean difference of -0.32 percentage points (95% CI -0.33 to -0.30; -3.49 mmol/mol [-3.66 to -3.32]) and with the difference favoring semaglutide throughout the study (P < 0.0001). Body weight plateaued at 65 weeks and was 8.9% lower with semaglutide. At week 156, a greater proportion treated with semaglutide were normoglycemic (69.5% vs. 35.8%; P < 0.0001) and a smaller proportion had biochemical diabetes by week 156 (1.5% vs. 6.9%; P < 0.0001). The number needed to treat was 18.5 to prevent a case of diabetes. Both regression and progression were dependent on glycemia at baseline, with the magnitude of weight reduction important in mediating 24.5% of progression and 27.1% of regression.

CONCLUSIONS

In people with preexisting cardiovascular disease and overweight or obesity but without diabetes, long-term semaglutide increases regression to biochemical normoglycemia and reduces progression to biochemical diabetes but does not slow glycemic progression over time.

Direct and indirect treatment effects in the presence of semicompeting risks

Semicompeting risks refer to the phenomenon that the terminal event (such as death) can censor the nonterminal event (such as disease progression) but not vice versa. The treatment effect on the terminal event can be delivered either directly following the treatment or indirectly through the nonterminal event. We consider 2 strategies to decompose the total effect into a direct effect and an indirect effect under the framework of mediation analysis in completely randomized experiments by adjusting the prevalence and hazard of nonterminal events, respectively. They require slightly different assumptions on cross-world quantities to achieve identifiability. We establish asymptotic properties for the estimated counterfactual cumulative incidences and decomposed treatment effects. We illustrate the subtle difference between these 2 decompositions through simulation studies and two real-data applications in the Supplementary Materials.

Aetiology, ejection fraction and mortality in chronic heart failure: a mediation analysis

OBJECTIVE

Clinical decision making in chronic heart failure (CHF) is based primarily on left ventricular ejection fraction (LVEF), and only secondarily on aetiology of the underlying disease. Our aim was to investigate the mediating role of LVEF in the relationship between aetiology and mortality.

METHODS

Using data of 2056 Austrian patients with CHF (mean age 57.2 years; mean follow-up 8.8 years), effects of aetiology on LVEF and overall mortality were estimated using multivariable-adjusted linear and Cox regression models. In causal mediation analyses, we decomposed the total effect of aetiology on mortality into direct and indirect (mediated through LVEF) effects.

RESULTS

For the analysed aetiologies (dilated (DCM, n=1009) and hypertrophic (HCM, n=89) cardiomyopathy; ischaemic (IHD, n=529) and hypertensive (HHD, n=320) heart disease; cardiac amyloidosis (CA, n=109)), the effect of LVEF on mortality was similar (HR5%-points lower LVEF=1.07, 95% CI 1.04 to 1.10; pinteraction=0.718). HCM and CA were associated with significantly higher, and IHD and DCM with significantly lower LVEF compared with other aetiologies. Compared with respective other aetiologies, the corresponding total effect HRs for mortality were 0.77 (95% CI 0.67 to 0.89), 0.47 (95% CI 0.25 to 0.88), 1.40 (95% CI 1.21 to 1.62), 0.79 (95% CI 0.67 to 0.95) and 2.36 (95% CI 1.81 to 3.08) for DCM, HCM, IHD, HHD and CA, respectively. CA had the highest mortality despite a HRindirect effect of 0.74 (95% CI 0.65 to 0.83). For all other aetiologies, <20% of the total mortality effects were mediated through LVEF.

CONCLUSIONS

The direct effect of aetiology on mortality dominates the indirect effect through LVEF. Therefore, clarification of aetiology is as important as measurement of LVEF.

Model-based hypothesis tests for the causal mediation of semi-competing risks

Analyzing the causal mediation of semi-competing risks has become important in medical research. Semi-competing risks refers to a scenario wherein an intermediate event may be censored by a primary event but not vice versa. Causal mediation analyses decompose the effect of an exposure on the primary outcome into an indirect (mediation) effect: an effect mediated through a mediator, and a direct effect: an effect not through the mediator. Here we proposed a model-based testing procedure to examine the indirect effect of the exposure on the primary event through the intermediate event. Under the counterfactual outcome framework, we defined a causal mediation effect using counting process. To assess statistical evidence for the mediation effect, we proposed two tests: an intersection-union test (IUT) and a weighted log-rank test (WLR). The test statistic was developed from a semi-parametric estimator of the mediation effect using a Cox proportional hazards model for the primary event and a series of logistic regression models for the intermediate event. We built a connection between the IUT and WLR. Asymptotic properties of the two tests were derived, and the IUT was determined to be a size [Formula: see text] test and statistically more powerful than the WLR. In numerical simulations, both the model-based IUT and WLR can properly adjust for confounding covariates, and the Type I error rates of the proposed methods are well protected, with the IUT being more powerful than the WLR. Our methods demonstrate the strongly significant effects of hepatitis B or C on the risk of liver cancer mediated through liver cirrhosis incidence in a prospective cohort study. The proposed method is also applicable to surrogate endpoint analyses in clinical trials.

Causal inference with recurrent and competing events

Many research questions concern treatment effects on outcomes that can recur several times in the same individual. For example, medical researchers are interested in treatment effects on hospitalizations in heart failure patients and sports injuries in athletes. Competing events, such as death, complicate causal inference in studies of recurrent events because once a competing event occurs, an individual cannot have more recurrent events. Several statistical estimands have been studied in recurrent event settings, with and without competing events. However, the causal interpretations of these estimands, and the conditions that are required to identify these estimands from observed data, have yet to be formalized. Here we use a formal framework for causal inference to formulate several causal estimands in recurrent event settings, with and without competing events. When competing events exist, we clarify when commonly used classical statistical estimands can be interpreted as causal quantities from the causal mediation literature, such as (controlled) direct effects and total effects. Furthermore, we show that recent results on interventionist mediation estimands allow us to define new causal estimands with recurrent and competing events that may be of particular clinical relevance in many subject matter settings. We use causal directed acyclic graphs and single world intervention graphs to illustrate how to reason about identification conditions for the various causal estimands based on subject matter knowledge. Furthermore, using results on counting processes, we show that our causal estimands and their identification conditions, which are articulated in discrete time, converge to classical continuous time counterparts in the limit of fine discretizations of time. We propose estimators and establish their consistency for the various identifying functionals. Finally, we use the proposed estimators to compute the effect of blood pressure lowering treatment on the recurrence of acute kidney injury using data from the Systolic Blood Pressure Intervention Trial.

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.

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.

A Causal Approach to Functional Mediation Analysis with Application to a Smoking Cessation Intervention

The increase in the use of mobile and wearable devices now allows dense assessment of mediating processes over time. For example, a pharmacological intervention may have an effect on smoking cessation via reductions in momentary withdrawal symptoms. We define and identify the causal direct and indirect effects in terms of potential outcomes on the mean difference and odds ratio scales, and present a method for estimating and testing the indirect effect of a randomized treatment on a distal binary variable as mediated by the nonparametric trajectory of an intensively measured longitudinal variable (e.g., from ecological momentary assessment). Coverage of a bootstrap test for the indirect effect is demonstrated via simulation. An empirical example is presented based on estimating later smoking abstinence from patterns of craving during smoking cessation treatment. We provide an R package, funmediation, available on CRAN at https://cran.r-project.org/web/packages/funmediation/index.html, to conveniently apply this technique. We conclude by discussing possible extensions to multiple mediators and directions for future research.

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.

Early life adversity and adult social relationships have independent effects on survival in a wild primate

Adverse conditions in early life can have negative consequences for adult health and survival in humans and other animals. What variables mediate the relationship between early adversity and adult survival? Adult social environments represent one candidate: Early life adversity is linked to social adversity in adulthood, and social adversity in adulthood predicts survival outcomes. However, no study has prospectively linked early life adversity, adult social behavior, and adult survival to measure the extent to which adult social behavior mediates this relationship. We do so in a wild baboon population in Amboseli, Kenya. We find weak mediation and largely independent effects of early adversity and adult sociality on survival. Furthermore, strong social bonds and high social status in adulthood can buffer some negative effects of early adversity. These results support the idea that affiliative social behavior is subject to natural selection through its positive relationship with survival, and they highlight possible targets for intervention to improve human health and well-being.

Incidence of hypertension in young transgender people after a 5-year follow-up: association with gender-affirming hormonal therapy

In order to assess the risk of hypertension development, we performed a retrospective analysis of the clinical records of consecutive transgender patients who began gender-affirming hormonal therapy in our Outpatient Gender Identity Clinic with <30 years of age and had a follow-up >5 years. 149 transgender women treated with estradiol and 153 transgender men treated with testosterone were included; 129 of the transgender women received also androgen blockers (54 spironolactone, 49 cyproterone acetate and 26 LHRH agonists). The annual incidence of hypertension in young transgender men (1.18%) seemed comparable to that of the general population. In young transgender women, it seemed higher (2.14%); we found that the choice of androgen blocker had a remarkable effect, with a highly significant increase in patients treated with cyproterone acetate (4.90%) vs. the rest (0.80%); the adjusted hazard-ratio was 0.227 (p = 0.001). Correlation, logistic regression and mediation analyses were performed for the associations of the available clinical variables with the increase in systolic blood pressure and the onset of hypertension, but besides the use of cyproterone acetate, only the ponderal gain was found significant (Spearman's r: 0.361, p < 0.001); with a 36.7% mediation effect (31.2-42.3%). Cyproterone acetate has additional known risks, such as meningioma; although we cannot conclusively prove that it has a role in the development of hypertension, we conclude that the use of cyproterone acetate for this indication should be reconsidered.

Glucose-lowering drugs with cardiovascular benefits as modifiers of critical elements of the human life history

The life history theory assumes that all organisms are under selective pressure to harvest external resources and allocate them to maximise fitness: only organisms making the best use of energy obtain the greatest fitness benefits. The trade-off of energy spans four functions: maintenance, growth, reproduction, and defence against pathogens. The innovative antihyperglycaemic agents glucagon-like peptide 1 (GLP-1) receptor agonists and sodium-glucose cotransporter 2 (SGLT2) inhibitors decrease bodyweight and have the potential to counter low-grade inflammation. These key activities could rewire two components of the life history theory operative in adulthood-ie, maintenance and defence. In this Personal View, we postulate that the benefits of these medications on the cardiovascular system, beyond their glucose-lowering effects, could be mediated by the reduction of the maintenance cost driven by obesity and efforts spent on blunting low-grade inflammation.

Mediation of the total effect of cystic fibrosis-related diabetes on mortality: A UK Cystic Fibrosis Registry cohort study

AIM

To investigate whether the effect of cystic fibrosis-related diabetes (CFRD) on the composite outcome of mortality or transplant could act through lung function, pulmonary exacerbations and/or nutritional status.

METHODS

A retrospective cohort of adult cystic fibrosis (CF) patients who had not been diagnosed with CFRD were identified from the UK Cystic Fibrosis Registry (n = 2750). Rate of death or transplant was compared between patients who did and did not develop CFRD (with insulin use) during follow-up using Poisson regression, separately by sex. Causal mediation methods were used to investigate whether lung function, pulmonary exacerbations and nutritional status lie on the causal pathway between insulin-treated CFRD and mortality/transplant.

RESULTS

At all ages, the mortality/transplant rate was higher in both men and women diagnosed with CFRD. Pulmonary exacerbations were the strongest mediator of the effect of CFRD on mortality/transplant, with an estimated 15% [95% CI: 7%, 28%] of the effect at 2 years post-CFRD diagnosis attributed to exacerbations, growing to 24% [95% CI: 9%, 46%] at 4 years post-diagnosis. Neither lung function nor nutritional status were found to be significant mediators of this effect. Estimates were similar but with wider confidence intervals in a cohort that additionally included people with CFRD but not using insulin.

CONCLUSION

There is evidence that pulmonary exacerbations mediate the effect of CFRD on mortality but, as they are estimated to mediate less than one-quarter of the total effect, the mechanism through which CFRD influences survival may involve other factors.

Methods of analysis for survival outcomes with time-updated mediators, with application to longitudinal disease registry data

Mediation analysis is a useful tool to illuminate the mechanisms through which an exposure affects an outcome but statistical challenges exist with time-to-event outcomes and longitudinal observational data. Natural direct and indirect effects cannot be identified when there are exposure-induced confounders of the mediator-outcome relationship. Previous measurements of a repeatedly-measured mediator may themselves confound the relationship between the mediator and the outcome. To overcome these obstacles, two recent methods have been proposed, one based on path-specific effects and one based on an additive hazards model and the concept of exposure splitting. We investigate these techniques, focusing on their application to observational datasets. We apply both methods to an analysis of the UK Cystic Fibrosis Registry dataset to identify how much of the relationship between onset of cystic fibrosis-related diabetes and subsequent survival acts through pulmonary function. Statistical properties of the methods are investigated using simulation. Both methods produce unbiased estimates of indirect and direct effects in scenarios consistent with their stated assumptions but, if the data are measured infrequently, estimates may be biased. Findings are used to highlight considerations in the interpretation of the observational data analysis.

Sex-differential non-specific effects of adjuvanted and non-adjuvanted rabies vaccines versus placebo on all-cause mortality in dogs (NERVE-Dog study): a study protocol for a randomized controlled trial with a nested case-control study

BACKGROUND

It has been proposed that childhood vaccines in high-mortality populations may have substantial impacts on mortality rates that are not explained by the prevention of targeted diseases, nor conversely by typical expected adverse reactions to the vaccines, and that these non-specific effects (NSEs) are generally more pronounced in females. The existence of these effects, and any implications for the development of vaccines and the design of vaccination programs to enhance safety, remain controversial. One area of controversy is the reported association of non-live vaccines with increased female mortality. In a previous randomized controlled trial (RCT), we observed that non-live alum-adjuvanted animal rabies vaccine (ARV) was associated with increased female but not male mortality in young, free-roaming dogs. Conversely, non-live non-adjuvanted human rabies vaccine (NRV) has been associated with beneficial non-specific effects in children. Alum adjuvant has been shown to suppress Th1 responses to pathogens, leading us to hypothesize that alum-adjuvanted rabies vaccine in young dogs has a detrimental effect on female survival by modulating the immune response to infectious and/or parasitic diseases. In this paper, we present the protocol of a 3-arm RCT comparing the effect of alum-adjuvanted rabies vaccine, non-adjuvanted rabies vaccine and placebo on all-cause mortality in an owned, free-roaming dog population, with causal mediation analysis of the RCT and a nested case-control study to test this hypothesis.

METHODS

Randomised controlled trial with a nested case-control study.

DISCUSSION

We expect that, among the placebo group, males will have higher mortality caused by higher pathogen loads and more severe disease, as determined by haematological parameters and inflammatory biomarkers. Among females, we expect that there will be no difference in mortality between the NRV and placebo groups, but that the ARV group will have higher mortality, again mediated by higher pathogen loads and more severe disease. We anticipate that these changes are preceded by shifts in key serum cytokine concentrations towards an anti-inflammatory immune response in females. If confirmed, these results will provide a rational basis for mitigation of detrimental NSEs of non-live vaccines in high-mortality populations.

Sequential Evaluation of NT-proBNP in Heart Failure: Insights Into Clinical Outcomes and Efficacy of Vericiguat

BACKGROUND

The effect of vericiguat on sequential N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels and influence of this relationship on clinical outcomes is unknown.

OBJECTIVES

This study assessed the relationship between changes in NT-proBNP and the primary outcome (cardiovascular death or heart failure hospitalization); evaluated the effect of vericiguat on changes in NT-proBNP; and explored the association between the efficacy of vericiguat and changes in NT-proBNP.

METHODS

NT-proBNP was measured at randomization and at 16, 32, 48, and 96 weeks in 4,805 of 5,050 patients. The association between NT-proBNP change at week 16 and the primary outcome was assessed. The relationship between changes in NT-proBNP and the primary outcome according to treatment group was assessed by using joint modeling and mediation analysis.

RESULTS

A significant and sustained decline in NT-proBNP levels was seen in both treatment groups. After week 16, NT-proBNP levels decreased more with vericiguat vs placebo (any reduction: odds ratio [OR]: 1.45 [95% CI: 1.28-1.65]; P < 0.001; ≥50% reduction: OR: 1.27 [95% CI: 1.10-1.47]; P = 0.001) and were less likely to increase (≥20% increase: OR: 0.68 [95% CI: 0.59-0.78]; P < 0.001; ≥50% increase: OR: 0.70 [95% CI: 0.59-0.82]; P < 0.001). The treatment effect related to serial NT-proBNP on the primary composite outcome was HR: 0.96 (95% CI: 0.95-0.99) at week 16, which increased to HR: 0.90 (95% CI: 0.85-0.96) at week 48; the average extent of mediation of the composite outcome related to NT-proBNP was 45%.

CONCLUSIONS

In patients with worsening HFrEF, vericiguat significantly decreased NT-proBNP levels compared with placebo. This change appeared associated with a modest relative improvement in the primary outcome of cardiovascular death or heart failure hospitalization. (Vericiguat Global Study in Subjects With Heart Failure With Reduced Ejection Fraction [VICTORIA]; NCT02861534).

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).

Perfluoroalkyl Substances and Incident Natural Menopause in Midlife Women: The Mediating Role of Sex Hormones

Perfluoroalkyl and polyfluoroalkyl substances (PFAS) have been associated with earlier natural menopause; however, the underlying mechanisms are not well understood, particularly the extent to which this relationship is mediated by sex hormones. We analyzed data (1999-2017) on 1,120 premenopausal women from the Study of Women's Health Across the Nation (SWAN). Causal mediation analysis was applied to quantify the degree to which follicle-stimulating hormone (FSH) and estradiol levels could mediate the associations between PFAS and incident natural menopause. Participants with higher PFAS concentrations had shorter times to natural menopause, with a relative survival of 0.82 (95% confidence interval (CI): 0.69, 0.96) for linear perfluorooctane sulfonate (n-PFOS), 0.84 (95% CI: 0.69, 1.00) for sum of branched-chain perfluorooctane sulfonate (Sm-PFOS), 0.79 (95% CI: 0.66, 0.93) for linear-chain perfluorooctanoate (n-PFOA), and 0.84 (95% CI: 0.71, 0.97) for perfluorononanoate (PFNA), comparing the highest tertile of PFAS concentrations with the lowest. The proportion of the effect mediated through FSH was 8.5% (95% CI: -11.7, 24.0) for n-PFOS, 13.2% (95% CI: 0.0, 24.5) for Sm-PFOS, 26.9% (95% CI: 15.6, 38.4) for n-PFOA, and 21.7% (6.8, 37.0) for PFNA. No significant mediation by estradiol was observed. The effect of PFAS on natural menopause may be partially explained by variations in FSH concentrations.

Modeling the underlying biological processes in Alzheimer's disease using a multivariate competing risk joint model

Many clinical trials repeatedly measure several longitudinal outcomes on patients. Patient follow‐up can discontinue due to an outcome‐dependent event, such as clinical diagnosis, death, or dropout. Joint modeling is a popular choice for the analysis of this type of data. Using example data from a prodromal Alzheimer's disease trial, we propose a new type of multivariate joint model in which longitudinal brain imaging outcomes and memory impairment ratings are allowed to be associated both with time to open‐label medication and dropout, and where the brain imaging outcomes may also directly affect the memory impairment ratings. Existing joint models for multivariate longitudinal outcomes account for the correlation between the longitudinal outcomes through the random effects, often by assuming a multivariate normal distribution. However, for these models, it is difficult to interpret how the longitudinal outcomes affect each other. We model the dependence between the longitudinal outcomes differently so that a first longitudinal outcome affects a second one. Specifically, for each longitudinal outcome, we use a linear mixed‐effects model to estimate its trajectory, where, for the second longitudinal outcome, we include the linear predictor of the first outcome as a time‐varying covariate. This facilitates an easy and direct interpretation of the association between the longitudinal outcomes and provides a framework for latent mediation analysis to understand the underlying biological processes. For the trial considered here, we found that part of the intervention effect is mediated through hippocampal brain atrophy. The proposed joint models are fitted using a Bayesian framework via MCMC simulation.

Causal mediation analysis with multiple causally non-ordered and ordered mediators based on summarized genetic data

Causal mediation analysis investigates the mechanism linking exposure and outcome. Dealing with the impact of unobserved confounders among exposure, mediator and outcome is an issue of great concern. Moreover, when multiple mediators exist, this causal pathway intertwines with other causal pathways, rendering it difficult to estimate the path-specific effects. In this study, we propose a method (PSE-MR) to identify and estimate path-specific effects of an exposure (e.g. education) on an outcome (e.g. osteoarthritis risk) through multiple causally ordered and non-ordered mediators (e.g. body mass index and pack-years of smoking) using summarized genetic data, when the sequential ignorability assumption is violated. Specifically, PSE-MR requires a specific rank condition in which the number of instrumental variables is larger than the number of mediators. Furthermore, we illustrate the utility of PSE-MR by providing guidance for practitioners and exploring the mediation effects of body mass index and pack-years of smoking in the causal pathways from education to osteoarthritis risk. Additionally, the results of simulation reveal that the causal estimates of path-specific effects are almost unbiased with good coverage and Type I error properties. Also, we summarize the least number of instrumental variables for the specific number of mediators to achieve 80% power.

Counterfactual mediation analysis in the multistate model framework for surrogate and clinical time-to-event outcomes in randomized controlled trials

In cancer randomized controlled trials, surrogate endpoints are frequently time-to-event endpoints, subject to the competing risk from the time-to-event clinical outcome. In this context, we introduce a counterfactual-based mediation analysis for a causal assessment of surrogacy. We use a multistate model for risk prediction to account for both direct transitions towards the clinical outcome and indirect transitions through the surrogate outcome. Within the counterfactual framework, we define natural direct and indirect effects with a causal interpretation. Based on these measures, we define the proportion of the treatment effect on the clinical outcome mediated by the surrogate outcome. We estimate the proportion for both the cumulative risk and restricted mean time lost. We illustrate our approach by using 18-year follow-up data from the SPCG-4 randomized trial of radical prostatectomy for prostate cancer. We assess time to metastasis as a surrogate outcome for prostate cancer-specific mortality.

High-Dimensional Mediation Analysis Based on Additive Hazards Model for Survival Data

Mediation analysis has been extensively used to identify potential pathways between exposure and outcome. However, the analytical methods of high-dimensional mediation analysis for survival data are still yet to be promoted, especially for non-Cox model approaches. We propose a procedure including "two-step" variable selection and indirect effect estimation for the additive hazards model with high-dimensional mediators. We first apply sure independence screening and smoothly clipped absolute deviation regularization to select mediators. Then we use the Sobel test and the BH method for indirect effect hypothesis testing. Simulation results demonstrate its good performance with a higher true-positive rate and accuracy, as well as a lower false-positive rate. We apply the proposed procedure to analyze DNA methylation markers mediating smoking and survival time of lung cancer patients in a TCGA (The Cancer Genome Atlas) cohort study. The real data application identifies four mediate CpGs, three of which are newly found.

Potential kidney protection with liraglutide and semaglutide: Exploratory mediation analysis

AIMS

To investigate whether effects on chronic kidney disease risk factors could explain the apparent reduction in kidney outcomes (composite of macroalbuminuria, doubling of serum creatinine, renal replacement therapy, or renal death), primarily driven by changes in albuminuria, after treatment with the glucagon-like peptide-1 receptor agonists (GLP-1RAs) liraglutide and semaglutide in patients with type 2 diabetes in the LEADER and SUSTAIN 6 trials.

MATERIALS AND METHODS

We evaluated the mediation effect of glycated haemoglobin (HbA1c), systolic blood pressure (BP), and body weight on the kidney effects of GLP-1RAs. Diastolic BP, haemoglobin, heart rate, low-density lipoprotein and total cholesterol, and white blood cell count were also investigated. The mediation effect was estimated by the novel Vansteelandt statistical method. Subgroups with estimated glomerular filtration rate (eGFR) <60 and ≥60 mL/min/1.73 m² were examined in LEADER.

RESULTS

We observed that HbA1c mediated 25% (95% confidence interval [CI] -7.1; 67.3) and 26% (95% CI noncalculable), and systolic BP 9% (95% CI 2.8; 22.7) and 22% (95% CI noncalculable) of kidney effects of GLP-1RAs in LEADER and SUSTAIN 6, respectively. Small or no mediation was observed for the other parameters; for example, body weight mediated 9% (95% CI -7.9; 35.5) in the former and did not mediate effects in the latter study. Mediation by HbA1c was greater in patients with eGFR ≥60 mL/min/1.73 m² (57%) versus those with eGFR <60 mL/min/1.73 m² (no mediation).

CONCLUSIONS

Our results suggest that HbA1c and systolic BP may moderately mediate kidney benefits of liraglutide and semaglutide, with all other variables having a small to no effect. Potential kidney benefits may be driven by other mediators or potentially by direct mechanisms.

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.

Is the effect of Mediterranean diet on hip fracture mediated through type 2 diabetes mellitus and body mass index?

BACKGROUND

We examined whether the inverse association between adherence to a Mediterranean diet and hip fracture risk is mediated by incident type 2 diabetes mellitus (T2DM) and body mass index (BMI).

METHODS

We included 50 755 men and women from the Cohort of Swedish Men and the Swedish Mammography Cohort who answered lifestyle and medical questionnaires in 1997 and 2008 (used for calculation of the Mediterranean diet score 9mMED; low, medium, high) and BMI in 1997, and incident T2DM in 1997-2008). The cumulative incidence of hip fracture from the National Patient Register (2009-14) was considered as outcome.

RESULTS

We present conditional odds ratios (OR) 9[95% confidence interval, CI) of hip fracture for medium and high adherence to mMED, compared with low adherence. The total effect ORs were 0.82 (0.71, 0.95) and 0.75 (0.62, 0.91), respectively. The controlled direct effect of mMED on hip fracture (not mediated by T2DM, considering BMI as an exposure-induced confounder), calculated using inverse probability weighting of marginal structural models, rendered ORs of 0.82 (0.72, 0.95) and 0.73 (0.60, 0.88), respectively. The natural direct effect ORs (not mediated by BMI or T2DM, calculated using flexible mediation analysis) were 0.82 (0.71, 0.95) and 0.74(0.61, 0.89), respectively. The path-specific indirect and partial indirect natural effects ORs (through BMI or T2DM) were close to 1.

CONCLUSIONS

Mediterranean diet has a direct effect on hip fracture risk via pathways other than through T2DM and BMI. We cannot exclude mediating effects of T2DM or BMI, or that their effects cancel each other out.

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.

Effects of glucagon-like peptide-1 receptor agonists liraglutide and semaglutide on cardiovascular and renal outcomes across body mass index categories in type 2 diabetes: Results of the LEADER and SUSTAIN 6 trials

Associations between body mass index (BMI) and the cardiovascular (CV) and kidney efficacy of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) in patients with type 2 diabetes (T2D) are uncertain; therefore, data analysed separately from the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial and the Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes (SUSTAIN 6) were examined. These international, randomized, placebo-controlled trials investigated liraglutide and semaglutide (both subcutaneous) in patients with T2D and at high risk of CV events. In post hoc analyses, patients were categorized by baseline BMI (<25, ≥25-<30, ≥30-<35 and ≥35 kg/m2 ), and CV and kidney outcomes with GLP-1 RA versus placebo were analysed. All baseline BMI data from LEADER (n = 9331) and SUSTAIN 6 (n = 3290) were included (91% and 92% of patients with overweight or obesity, respectively). In SUSTAIN 6, nominally significant heterogeneity of semaglutide efficacy by baseline BMI was observed for CV death/myocardial infarction/stroke (major adverse CV events, primary outcome of both; Pinteraction = .02); otherwise, there was no statistical heterogeneity for either GLP-1 RA versus placebo across BMI categories for key CV and kidney outcomes. The lack of statistical heterogeneity from these cardiorenal outcomes implies that liraglutide and semaglutide may be beneficial for many patients and is probable not to depend on their baseline BMI, but further study is needed.

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.

Estimation of Controlled Direct Effects in Longitudinal Mediation Analyses with Latent Variables in Randomized Studies

In a randomized study with longitudinal data on a mediator and outcome, estimating the direct effect of treatment on the outcome at a particular time requires adjusting for confounding of the association between the outcome and all preceding instances of the mediator. When the confounders are themselves affected by treatment, standard regression adjustment is prone to severe bias. In contrast, G-estimation requires less stringent assumptions than path analysis using SEM to unbiasedly estimate the direct effect even in linear settings. In this article, we propose a G-estimation method to estimate the controlled direct effect of treatment on the outcome, by adapting existing G-estimation methods for time-varying treatments without mediators. The proposed method can accommodate continuous and noncontinuous mediators, and requires no models for the confounders. Unbiased estimation only requires correctly specifying a mean model for either the mediator or the outcome. The method is further extended to settings where the mediator or outcome, or both, are latent, and generalizes existing methods for single measurement occasions of the mediator and outcome to longitudinal data on the mediator and outcome. The methods are utilized to assess the effects of an intervention on physical activity that is possibly mediated by motivation to exercise in a randomized study.

Cardiovascular Risk Reduction With Liraglutide: An Exploratory Mediation Analysis of the LEADER Trial

OBJECTIVE

The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial (ClinicalTrials.gov reg. no. NCT01179048) demonstrated a reduced risk of cardiovascular (CV) events for patients with type 2 diabetes who received the glucagon-like peptide 1 receptor agonist liraglutide versus placebo. The mechanisms behind this CV benefit remain unclear. We aimed to identify potential mediators for the CV benefit observed with liraglutide in the LEADER trial.

RESEARCH DESIGN AND METHODS

We performed exploratory analyses to identify potential mediators of the effect of liraglutide on major adverse CV events (MACE; composite of CV death, nonfatal myocardial infarction, or nonfatal stroke) from the following candidates: glycated hemoglobin (HbA1c), body weight, urinary albumin-to-creatinine ratio (UACR), confirmed hypoglycemia, sulfonylurea use, insulin use, systolic blood pressure, and LDL cholesterol. These candidates were selected as CV risk factors on which liraglutide had an effect in LEADER such that a reduction in CV risk might result. We used two methods based on a Cox proportional hazards model and the new Vansteelandt method designed to use all available information from the mediator and to control for confounding factors.

RESULTS

Analyses using the Cox methods and Vansteelandt method indicated potential mediation by HbA1c (up to 41% and 83% mediation, respectively) and UACR (up to 29% and 33% mediation, respectively) on the effect of liraglutide on MACE. Mediation effects were small for other candidates.

CONCLUSIONS

These analyses identify HbA1c and, to a lesser extent, UACR as potential mediators of the CV effects of liraglutide. Whether either is a marker of an unmeasured factor or a true mediator remains a key question that invites further investigation.

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.