Insights into the Cross-world Independence Assumption of Causal Mediation Analysis

Association of healthy diet score and adiposity with risk of colorectal cancer: findings from the UK Biobank prospective cohort study

PURPOSE

To explore the joint association of dietary patterns and adiposity with colorectal cancer (CRC), and whether adiposity mediates the relationship between dietary patterns and CRC risk, which could provide deeper insights into the underlying pathogenesis of CRC.

METHODS

The data of 307,023 participants recruited between 2006 and 2010 were extracted from the UK Biobank study. Healthy diet scores were calculated based on self-reported dietary data at baseline, and participants were categorized into three groups, namely, low, intermediate, and high diet score groups. Cox regression models with hazard ratios (HRs) and 95% confidence intervals (CIs) were used to estimate the effects of the healthy diet score on CRC incidence, adjusting for various covariates. Furthermore, the mediation roles of obesity and central obesity between the healthy diet score and CRC risk were assessed using a counterfactual causal analysis based on Cox regression model. Additionally, joint association between dietary patterns and adiposity on CRC risks was assessed on the additive and multiplicative scales.

RESULTS

Over a median 6.2-year follow-up, 3,276 participants developed CRC. After adjusting for sociodemographic and lifestyle factors, a lower risk of CRC incidence was found for participants with intermediate (HR = 0.83, 95% CI: 0.72 to 0.95) and high diet scores (HR = 0.73, 95% CI: 0.62 to 0.87) compared to those with low diet scores. When compared with the low diet score group, obesity accounted for 4.13% and 7.93% of the total CRC effect in the intermediate and high diet score groups, respectively, while central obesity contributed to 3.68% and 10.02% of the total CRC risk in the intermediate and high diet score groups, respectively. The mediating effect of adiposity on CRC risk was significant in men but not in women. Concurrent unhealthy diet and adiposity multiplied CRC risk.

CONCLUSION

Adiposity-mediated effects were limited in the link between dietary patterns and CRC incidence, implying that solely addressing adiposity may not sufficiently reduce CRC risk. Interventions, such as improving dietary quality in people with adiposity or promoting weight control in those with unhealthy eating habits, may provide an effective strategy to reduce CRC risk.

Proportion of treatment effect explained: An overview of interpretations

The selection of the primary endpoint in a clinical trial plays a critical role in determining the trial's success. Ideally, the primary endpoint is the clinically most relevant outcome, also termed the true endpoint. However, practical considerations, like extended follow-up, may complicate this choice, prompting the proposal to replace the true endpoint with so-called surrogate endpoints. Evaluating the validity of these surrogate endpoints is crucial, and a popular evaluation framework is based on the proportion of treatment effect explained (PTE). While methodological advancements in this area have focused primarily on estimation methods, interpretation remains a challenge hindering the practical use of the PTE. We review various ways to interpret the PTE. These interpretations-two causal and one non-causal-reveal connections between the PTE principal surrogacy, causal mediation analysis, and the prediction of trial-level treatment effects. A common limitation across these interpretations is the reliance on unverifiable assumptions. As such, we argue that the PTE is only meaningful when researchers are willing to make very strong assumptions. These challenges are also illustrated in an analysis of three hypothetical vaccine trials.

Practical causal mediation analysis: extending nonparametric estimators to accommodate multiple mediators and multiple intermediate confounders

Mediation analysis is appealing for its ability to improve understanding of the mechanistic drivers of causal effects, but real-world data complexities challenge its successful implementation, including (i) the existence of post-exposure variables that also affect mediators and outcomes (thus, confounding the mediator-outcome relationship), that may also be (ii) multivariate, and (iii) the existence of multivariate mediators. All three challenges are present in the mediation analysis we consider here, where our goal is to estimate the indirect effects of receiving a Section 8 housing voucher as a young child on the risk of developing a psychiatric mood disorder in adolescence that operate through mediators related to neighborhood poverty, the school environment, and instability of the neighborhood and school environments, considered together and separately. Interventional direct and indirect effects (IDE/IIE) accommodate post-exposure variables that confound the mediator-outcome relationship, but currently, no readily implementable nonparametric estimator for IDE/IIE exists that allows for both multivariate mediators and multivariate post-exposure intermediate confounders. The absence of such an IDE/IIE estimator that can easily accommodate both multivariate mediators and post-exposure confounders represents a significant limitation for real-world analyses, because when considering each mediator subgroup separately, the remaining mediator subgroups (or a subset of them) become post-exposure intermediate confounders. We address this gap by extending a recently developed nonparametric estimator for the IDE/IIE to allow for easy incorporation of multivariate mediators and multivariate post-exposure confounders simultaneously. We apply the proposed estimation approach to our analysis, including walking through a strategy to account for other, possibly co-occurring intermediate variables when considering each mediator subgroup separately.

The role of psychosocial well-being and emotion-driven impulsiveness in food choices of European adolescents

BACKGROUND

It is unclear whether a hypothetical intervention targeting either psychosocial well-being or emotion-driven impulsiveness is more effective in reducing unhealthy food choices. Therefore, we aimed to compare the (separate) causal effects of psychosocial well-being and emotion-driven impulsiveness on European adolescents' sweet and fat propensity.

METHODS

We included 2,065 participants of the IDEFICS/I.Family cohort (mean age: 13.4) providing self-reported data on sweet propensity (score range: 0 to 68.4), fat propensity (range: 0 to 72.6), emotion-driven impulsiveness using the UPPS-P negative urgency subscale, and psychosocial well-being using the KINDLR Questionnaire. We estimated, separately, the average causal effects of psychosocial well-being and emotion-driven impulsiveness on sweet and fat propensity applying a semi-parametric doubly robust method (targeted maximum likelihood estimation). Further, we investigated a potential indirect effect of psychosocial well-being on sweet and fat propensity mediated via emotion-driven impulsiveness using a causal mediation analysis.

RESULTS

If all adolescents, hypothetically, had high levels of psychosocial well-being, compared to low levels, we estimated a decrease in average sweet propensity by 1.43 [95%-confidence interval: 0.25 to 2.61]. A smaller effect was estimated for fat propensity. Similarly, if all adolescents had high levels of emotion-driven impulsiveness, compared to low levels, average sweet propensity would be decreased by 2.07 [0.87 to 3.26] and average fat propensity by 1.85 [0.81 to 2.88]. The indirect effect of psychosocial well-being via emotion-driven impulsiveness was 0.61 [0.24 to 1.09] for average sweet propensity and 0.55 [0.13 to 0.86] for average fat propensity.

CONCLUSIONS

An intervention targeting emotion-driven impulsiveness, compared to psychosocial well-being, would be marginally more effective in reducing sweet and fat propensity in adolescents.

Mediation analysis with case-control sampling: Identification and estimation in the presence of a binary mediator

With reference to a stratified case-control (CC) procedure based on a binary variable of primary interest, we derive the expression of the distortion induced by the sampling design on the parameters of the logistic model of a secondary variable. This is particularly relevant when performing mediation analysis (possibly in a causal framework) with stratified case-control (SCC) data in settings where both the outcome and the mediator are binary. Despite being designed for parametric identification, our strategy is general and can be used also in a nonparametric context. With reference to parametric estimation, we derive the maximum likelihood (ML) estimator and the M-estimator of the joint outcome-mediator parameter vector. We then conduct a simulation study focusing on the main causal mediation quantities (i.e., natural effects) and comparing M- and ML estimation to existing methods, based on weighting. As an illustrative example, we reanalyze a German CC data set in order to investigate whether the effect of reduced immunocompetency on listeriosis onset is mediated by the intake of gastric acid suppressors.

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.

Examining the Causal Mediating Role of Cardiovascular Disease on the Effect of Subclinical Cardiovascular Disease on Cognitive Impairment via Separable Effects

BACKGROUND

An important epidemiological question is understanding how vascular risk factors contribute to cognitive impairment. Using data from the Cardiovascular Health Cognition Study, we investigated how subclinical cardiovascular disease (sCVD) relates to cognitive impairment risk and the extent to which the hypothesized risk is mediated by the incidence of clinically manifested cardiovascular disease (CVD), both overall and within apolipoprotein E-4 (APOE-4) subgroups.

METHODS

We adopted a novel "separable effects" causal mediation framework that assumes that sCVD has separably intervenable atherosclerosis-related components. We then ran several mediation models, adjusting for key covariates.

RESULTS

We found that sCVD increased overall risk of cognitive impairment (risk ratio [RR] = 1.21, 95% confidence interval [CI]: 1.03, 1.44); however, there was little or no mediation by incident clinically manifested CVD (indirect effect RR = 1.02, 95% CI: 1.00, 1.03). We also found attenuated effects among APOE-4 carriers (total effect RR = 1.09, 95% CI: 0.81, 1.47; indirect effect RR = 0.99, 95% CI: 0.96, 1.01) and stronger findings among noncarriers (total effect RR = 1.29, 95% CI: 1.05, 1.60; indirect effect RR = 1.02, 95% CI: 1.00, 1.05). In secondary analyses restricting cognitive impairment to only incident dementia cases, we found similar effect patterns.

CONCLUSIONS

We found that the effect of sCVD on cognitive impairment does not seem to be mediated by CVD, both overall and within APOE-4 subgroups. Our results were critically assessed via sensitivity analyses, and they were found to be robust. Future work is needed to fully understand the relationship between sCVD, CVD, and cognitive impairment.

A Bayesian joint model for compositional mediation effect selection in microbiome data

Analyzing multivariate count data generated by high-throughput sequencing technology in microbiome research studies is challenging due to the high-dimensional and compositional structure of the data and overdispersion. In practice, researchers are often interested in investigating how the microbiome may mediate the relation between an assigned treatment and an observed phenotypic response. Existing approaches designed for compositional mediation analysis are unable to simultaneously determine the presence of direct effects, relative indirect effects, and overall indirect effects, while quantifying their uncertainty. We propose a formulation of a Bayesian joint model for compositional data that allows for the identification, estimation, and uncertainty quantification of various causal estimands in high-dimensional mediation analysis. We conduct simulation studies and compare our method's mediation effects selection performance with existing methods. Finally, we apply our method to a benchmark data set investigating the sub-therapeutic antibiotic treatment effect on body weight in early-life mice.

An exact regression-based approach for the estimation of natural direct and indirect effects with a binary outcome and a continuous mediator

In the causal mediation framework, a number of parametric regression-based approaches have been introduced in recent years for estimating natural direct and indirect effects for a binary outcome in an exact manner, without invoking simplifying assumptions based on the rareness or commonness of the outcome. However, most of these works have focused on a binary mediator. In this article, we aim at a continuous mediator and introduce an exact approach for the estimation of natural effects on the odds ratio, risk ratio, and risk difference scales. Our approach relies on logistic and linear models for the outcome and mediator, respectively, and uses numerical integration to calculate the nested counterfactual probabilities underlying the definition of natural effects. Formulas for the delta method standard errors for all effects estimators are provided. The performance of our proposed exact estimators was evaluated in simulation studies that featured scenarios with different levels of outcome rareness/commonness, including a marginally but not conditionally rare outcome scenario. Furthermore, we evaluated the merit of Firth's penalization to mitigate the bias in the logistic regression coefficients estimators for the smallest outcome prevalences and sample sizes investigated. Using a SAS macro provided, we implemented our approach to assess the effect of placental abruption on low birth weight mediated by gestational age. We found that our exact natural effects estimators worked properly in both simulated and real data applications.

Evaluating the Mediating Role of Recall of Intervention Knowledge in the Relationship Between a Peer-Driven Intervention and HIV Risk Behaviors Among People Who Inject Drugs

Peer-driven interventions can be effective in reducing HIV injection risk behaviors among people who inject drugs (PWID). We employed a causal mediation framework to examine the mediating role of recall of intervention knowledge in the relationship between a peer-driven intervention and subsequent self-reported HIV injection-related risk behavior among PWID in the HIV Prevention Trials Network (HPTN) 037 study. For each intervention network, the index participant received training at baseline to become a peer educator, while non-index participants and all participants in the control networks received only HIV testing and counseling; recall of intervention knowledge was measured at the 6-month visit for each participant, and each participant was followed to ascertain HIV injection-related risk behaviors at the 12-month visit. We used inverse probability weighting to fit marginal structural models to estimate the total effect (TE) and controlled direct effect (CDE) of the intervention on the outcome. The proportion eliminated (PE) by intervening to remove mediation by the recall of intervention knowledge was computed. There were 385 participants (47% in intervention networks) included in the analysis. The TE and CDE risk ratios for the intervention were 0.47 [95% confidence interval (CI): 0.28, 0.78] and 0.73 (95% CI: 0.26, 2.06) and the PE was 49%. Compared to participants in the control networks, the peer-driven intervention reduced the risk of HIV injection-related risk behavior by 53%. The mediating role of recall of intervention knowledge accounted for less than 50% of the total effect of the intervention, suggesting that other potential causal pathways between the intervention and the outcome, such as motivation and skill, self-efficacy, social norms and behavior modeling, should be considered in future studies.

Applied causal inference methods for sequential mediators

BACKGROUND

Mediation analysis aims at estimating to what extent the effect of an exposure on an outcome is explained by a set of mediators on the causal pathway between the exposure and the outcome. The total effect of the exposure on the outcome can be decomposed into an indirect effect, i.e. the effect explained by the mediators jointly, and a direct effect, i.e. the effect unexplained by the mediators. However finer decompositions are possible in presence of independent or sequential mediators.

METHODS

We review four statistical methods to analyse multiple sequential mediators, the inverse odds ratio weighting approach, the inverse probability weighting approach, the imputation approach and the extended imputation approach. These approaches are compared and implemented using a case-study with the aim to investigate the mediating role of adverse reproductive outcomes and infant respiratory infections in the effect of maternal pregnancy mental health on infant wheezing in the Ninfea birth cohort.

RESULTS

Using the inverse odds ratio weighting approach, the direct effect of maternal depression or anxiety in pregnancy is equal to a 59% (95% CI: 27%,94%) increased prevalence of infant wheezing and the mediated effect through adverse reproductive outcomes is equal to a 3% (95% CI: -6%,12%) increased prevalence of infant wheezing. When including infant lower respiratory infections in the mediation pathway, the direct effect decreases to 57% (95% CI: 25%,92%) and the indirect effect increases to 5% (95% CI: -5%,15%). The estimates of the effects obtained using the weighting and the imputation approaches are similar. The extended imputation approach suggests that the small joint indirect effect through adverse reproductive outcomes and lower respiratory infections is due entirely to the contribution of infant lower respiratory infections, and not to an increased prevalence of adverse reproductive outcomes.

CONCLUSIONS

The four methods revealed similar results of small mediating role of adverse reproductive outcomes and early respiratory tract infections in the effect of maternal pregnancy mental health on infant wheezing. The choice of the method depends on what is the effect of main interest, the type of the variables involved in the analysis (binary, categorical, count or continuous) and the confidence in specifying the models for the exposure, the mediators and the outcome.

Understanding Etiologic Pathways Through Multiple Sequential Mediators: An Application in Perinatal Epidemiology

BACKGROUND

Causal mediation analysis facilitates decomposing the total effect into a direct effect and an indirect effect that operates through an intermediate variable. Recent developments in causal mediation analysis have clarified the process of evaluating how-and to what extent-different pathways via multiple causally ordered mediators link the exposure to the outcome.

METHODS

Through an application of natural effect models for multiple mediators, we show how placental abruption might affect perinatal mortality using small for gestational age (SGA) birth and preterm delivery as two sequential mediators. We describe methods to disentangle the total effect into the proportions mediated via each of the sequential mediators, when evaluating natural direct and natural indirect effects.

RESULTS

Under the assumption that SGA births causally precedes preterm delivery, an analysis of 16.7 million singleton pregnancies is consistent with the hypothesis that abruption exerts powerful effects on perinatal mortality (adjusted risk ratio = 11.9; 95% confidence interval = 11.6, 12.1). The proportions of the estimated total effect mediated through SGA birth and preterm delivery were 2% and 58%, respectively. The proportion unmediated via either SGA or preterm delivery was 41%.

CONCLUSIONS

Through an application of causal mediation analysis with sequential mediators, we uncovered new insights into the pathways along which abruption impacts perinatal mortality.

Mediating role of psychological distress in the associations between neighborhood social environments and sleep health

STUDY OBJECTIVES

The characteristics of neighborhood social environments, such as safety and social cohesion, have been examined as determinants of poor sleep. The current study investigates associations between neighborhood social characteristics and sleep health, as well as the mediating role of psychological distress on these possible associations.

METHODS

Three waves of PHRESH Zzz (n = 2699), a longitudinal study conducted in two low-income, predominately Black neighborhoods, were utilized for this analysis. The characteristics of neighborhood social environments were measured using crime rates, a neighborhood social disorder index, and self-reported social cohesion. Sleep health was measured via 7 days of wrist-worn actigraphy as insufficient sleep, sleep duration, wake after sleep onset (WASO), and sleep efficiency. G-estimations based on structural nested mean models and mediation analyses were performed to estimate the effects of neighborhood social environments on sleep as well as direct/indirect effects through psychological distress.

RESULTS

Crime rate around residential addresses was associated with increased risk of insufficient sleep (risk ratio: 1.05 [1.02, 1.12]), increased WASO (β: 3.73 [0.26, 6.04]), and decreased sleep efficiency (β: -0.54 [-0.91, -0.09]). Perceived social cohesion was associated with decreased risk of insufficient sleep (OR: 0.93 [0.88, 0.97]). Psychological distress mediated part of the associations of crime and social cohesion with insufficient sleep.

CONCLUSIONS

Neighborhood social environments may contribute to poor sleep health in low-income, predominantly Black neighborhoods, and psychological distress can be a salient pathway linking these neighborhood characteristics and sleep health.

A principled approach to mediation analysis in perinatal epidemiology

For many research questions in perinatal epidemiology, gestational age is a mediator that features the causal pathway between exposure and outcome. A mediator is an intermediate variable between an exposure and outcome, which is influenced by the exposure on the causal pathway to the outcome. Therefore, conventional analyses that adjust, stratify, or match for gestational age or its proxy (eg, preterm vs term deliveries) are problematic. This practice, which is entrenched in perinatal research, induces an overadjustment bias. Depending on the causal question, it may be inappropriate to adjust (or condition) for a mediator, such as gestational age, by either design or statistical analysis, but its effect can be quantified through causal mediation analysis. In an exposition of such methods, we demonstrated the relationship between the exposure and outcome and provided a formal analytical framework to quantify the extent to which a causal effect is influenced by a mediator. We reviewed concepts of confounding and causal inference, introduced the concept of a mediator and illustrated the perils of adjusting for a mediator in an exposure-outcome paradigm for a given causal question, adopted causal methods that call for an evaluation of a mediator in a causal exposure effect on the outcome, and discussed unmeasured confounding assumptions in mediation analysis. Furthermore, we reviewed other developments in the causal mediation analysis literature, including decomposition of a total effect when the mediator interacts with the exposure (4-way decomposition), methods for multiple mediators, mediation methods for case-control studies, mediation methods for time-to-event outcomes, sample size and power analysis for mediation analysis, and available software to apply these methods. To illustrate these methods, we provided a clinical example to estimate the risk of perinatal mortality (outcome) concerning placental abruption (exposure) and to determine the extent to which preterm delivery (mediator; a proxy for gestational age) plays a role in this causal effect. We hoped that the adoption of mediation methods described in this review will move research in perinatal epidemiology away from biased adjustments of mediators toward a more nuanced quantification of effects that pose unique challenges and provide unique insights in our field.

Mediation analysis methods used in observational research: a scoping review and recommendations

Background

Mediation analysis methodology underwent many advancements throughout the years, with the most recent and important advancement being the development of causal mediation analysis based on the counterfactual framework. However, a previous review showed that for experimental studies the uptake of causal mediation analysis remains low. The aim of this paper is to review the methodological characteristics of mediation analyses performed in observational epidemiologic studies published between 2015 and 2019 and to provide recommendations for the application of mediation analysis in future studies.

Methods

We searched the MEDLINE and EMBASE databases for observational epidemiologic studies published between 2015 and 2019 in which mediation analysis was applied as one of the primary analysis methods. Information was extracted on the characteristics of the mediation model and the applied mediation analysis method.

Results

We included 174 studies, most of which applied traditional mediation analysis methods (n = 123, 70.7%). Causal mediation analysis was not often used to analyze more complicated mediation models, such as multiple mediator models. Most studies adjusted their analyses for measured confounders, but did not perform sensitivity analyses for unmeasured confounders and did not assess the presence of an exposure-mediator interaction.

Conclusions

To ensure a causal interpretation of the effect estimates in the mediation model, we recommend that researchers use causal mediation analysis and assess the plausibility of the causal assumptions. The uptake of causal mediation analysis can be enhanced through tutorial papers that demonstrate the application of causal mediation analysis, and through the development of software packages that facilitate the causal mediation analysis of relatively complicated mediation models.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-021-01426-3.

Integrated multiple mediation analysis: A robustness-specificity trade-off in causal structure

Recent methodological developments in causal mediation analysis have addressed several issues regarding multiple mediators. However, these developed methods differ in their definitions of causal parameters, assumptions for identification, and interpretations of causal effects, making it unclear which method ought to be selected when investigating a given causal effect. Thus, in this study, we construct an integrated framework, which unifies all existing methodologies, as a standard for mediation analysis with multiple mediators. To clarify the relationship between existing methods, we propose four strategies for effect decomposition: two-way, partially forward, partially backward, and complete decompositions. This study reveals how the direct and indirect effects of each strategy are explicitly and correctly interpreted as path-specific effects under different causal mediation structures. In the integrated framework, we further verify the utility of the interventional analogues of direct and indirect effects, especially when natural direct and indirect effects cannot be identified or when crossworld exchangeability is invalid. Consequently, this study yields a robustness-specificity trade-off in the choice of strategies. Inverse probability weighting is considered for estimation. The four strategies are further applied to a simulation study for performance evaluation and for analyzing the Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer dataset from Taiwan to investigate the causal effect of hepatitis C virus infection on mortality.

The impact of adjusting for pure predictors of exposure, mediator, and outcome on the variance of natural direct and indirect effect estimators

It is now well established that adjusting for pure predictors of the outcome, in addition to confounders, allows unbiased estimation of the total exposure effect on an outcome with generally reduced standard errors (SEs). However, no analogous results have been derived for mediation analysis. Considering the simplest linear regression setting and the ordinary least square estimator, we obtained theoretical results showing that adjusting for pure predictors of the outcome, in addition to confounders, allows unbiased estimation of the natural indirect effect (NIE) and the natural direct effect (NDE) on the difference scale with reduced SEs. Adjusting for pure predictors of the mediator increases the SE of the NDE's estimator, but may increase or decrease the variance of the NIE's estimator. Adjusting for pure predictors of the exposure increases the variance of estimators of the NIE and NDE. Simulation studies were used to confirm and extend these results to the case where the mediator or the outcome is binary. Additional simulations were conducted to explore scenarios featuring an exposure-mediator interaction as well as the relative risk and odds ratio scales for the case of binary mediator and outcome. Both a regression approach and an inverse probability weighting approach were considered in the simulation study. A real-data illustration employing data from the Canadian Study of Health and Aging is provided. This analysis is concerned with the mediating effect of vitamin D in the effect of physical activity on dementia and its results are overall consistent with the theoretical and empirical findings.