Estimation of causal mediation effects for a dichotomous outcome in multiple-mediator models using the mediation formula

Does health literacy mediate the relationship between socioeconomic status and health related outcomes in the Belgian adult population?

BACKGROUND

Health literacy (HL) has been put forward as a potential mediator through which socioeconomic status (SES) affects health. This study explores whether HL mediates the relation between SES and a selection of health or health-related outcomes.

METHODS

Data from the participants of the Belgian health interview survey 2018 aged 18 years or older were individually linked with data from the Belgian compulsory health insurance (n = 8080). HL was assessed with the HLS-EU-Q6. Mediation analyses were performed with health behaviour (physical activity, diet, alcohol and tobacco consumption), health status (perceived health status, mental health status), use of medicine (purchase of antibiotics), and use of preventive care (preventive dental care, influenza vaccination, breast cancer screening) as dependent outcome variables, educational attainment and income as independent variables of interest, age and sex as potential confounders and HL as mediating variable.

RESULTS

The study showed that unhealthy behaviours (except alcohol consumption), poorer health status, higher use of medicine and lower use of preventive care (except flu vaccination) were associated with low SES (i.e., low education and low income) and with insufficient HL. HL partially mediated the relationship between education and health behaviour, perceived health status and mental health status, accounting for 3.8-16.0% of the total effect. HL also constituted a pathway by which income influences health behaviour, perceived health status, mental health status and preventive dental care, with the mediation effects accounting for 2.1-10.8% of the total effect.

CONCLUSIONS

Although the influence of HL in the pathway is limited, our findings suggest that strategies for improving various health-related outcomes among low SES groups should include initiatives to enhance HL in these population groups. Further research is needed to confirm our results and to better explore the mediating effects of HL.

High adherence to the French dietary guidelines decreases type 2 diabetes risk in females through pathways of obesity markers: Evidence from the E3N-EPIC prospective cohort study

OBJECTIVE

Obesity and type 2 diabetes (T2D) have been associated with low adherence to the 2017 French food-based dietary guidelines, as assessed by the Programme National Nutrition Santé - guidelines score 2 (PNNS-GS2). Whether the association between T2D and PNNS-GS2 is direct or mediated by obesity has been little investigated.

RESEARCH METHODS

The study included 71,450 women from the E3N-EPIC cohort, mean age of 52.9 y (SD 6.7). The simplified PNNS-GS2 was derived via food history questionnaire. Multivariable Cox regression models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of T2D. Causal mediation analyses were used to decompose the total effect of sPNNS-GS2 on T2D into a direct effect and indirect effect mediated by body mass index (BMI) or the waist-hip ratio (WHR).

RESULTS

During a mean follow-up of 19 y, 3679 incident T2D cases were identified and validated. There was a linear association between adherence to sPNNS-GS2 and T2D (P-nonlinearity = 0.92). In the fully adjusted model, each 1-SD increase in the sPNNS-GS2 was associated with a lower T2D risk [HR (95% CI), 0.92 (0.89, 0.95)]. The overall associations were mainly explained by sPNNS-GS2-associated excess weight, with BMI and WHR mediating 52% and 58% of the associations, respectively.

CONCLUSIONS

Higher adherence to French food-based dietary guidelines was associated with a lower risk of T2D in women, and a significant portion of this effect could be attributed to excess weight measured by BMI or WHR. This finding helps better understand the mechanisms underlying the diet-T2D association.

A Bayesian nonparametric approach for multiple mediators with applications in mental health studies

Mediation analysis with contemporaneously observed multiple mediators is a significant area of causal inference. Recent approaches for multiple mediators are often based on parametric models and thus may suffer from model misspecification. Also, much of the existing literature either only allow estimation of the joint mediation effect or estimate the joint mediation effect just as the sum of individual mediator effects, ignoring the interaction among the mediators. In this article, we propose a novel Bayesian nonparametric method that overcomes the two aforementioned drawbacks. We model the joint distribution of the observed data (outcome, mediators, treatment, and confounders) flexibly using an enriched Dirichlet process mixture with three levels. We use standardization (g-computation) to compute all possible mediation effects, including pairwise and all other possible interaction among the mediators. We thoroughly explore our method via simulations and apply our method to a mental health data from Wisconsin Longitudinal Study, where we estimate how the effect of births from unintended pregnancies on later life mental depression (CES-D) among the mothers is mediated through lack of self-acceptance and autonomy, employment instability, lack of social participation, and increased family stress. Our method identified significant individual mediators, along with some significant pairwise effects.

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.

Risk of hip fracture in meat-eaters, pescatarians, and vegetarians: a prospective cohort study of 413,914 UK Biobank participants

BACKGROUND

Meat-free diets may be associated with a higher risk of hip fracture, but prospective evidence is limited. We aimed to investigate the risk of hip fracture in occasional meat-eaters, pescatarians, and vegetarians compared to regular meat-eaters in the UK Biobank, and to explore the role of potential mediators of any observed risk differences.

METHODS

Middle-aged UK adults were classified as regular meat-eaters (n = 258,765), occasional meat-eaters (n = 137,954), pescatarians (n = 9557), or vegetarians (n = 7638) based on dietary and lifestyle information at recruitment (2006-2010). Incident hip fractures were identified by record linkage to Hospital Episode Statistics up to September 2021. Multivariable Cox regression models were used to estimate associations between each diet group and hip fracture risk, with regular meat-eaters as the reference group, over a median follow-up time of 12.5 years.

RESULTS

Among 413,914 women, 3503 hip fractures were observed. After adjustment for confounders, vegetarians (HR (95% CI): 1.50 (1.18, 1.91)) but not occasional meat-eaters (0.99 (0.93, 1.07)) or pescatarians (1.08 (0.86, 1.35)) had a greater risk of hip fracture than regular meat-eaters. This is equivalent to an adjusted absolute risk difference of 3.2 (1.2, 5.8) more hip fractures per 1000 people over 10 years in vegetarians. There was limited evidence of effect modification by BMI on hip fracture risk across diet groups (p_interaction = 0.08), and no clear evidence of effect modification by age or sex (p_interaction = 0.9 and 0.3, respectively). Mediation analyses suggest that BMI explained 28% of the observed risk difference between vegetarians and regular meat-eaters (95% CI: 1.1%, 69.8%).

DISCUSSION

Vegetarian men and women had a higher risk of hip fracture than regular meat-eaters, and this was partly explained by their lower BMI. Ensuring adequate nutrient intake and weight management are therefore particularly important in vegetarians in the context of hip fracture prevention.

TRIAL REGISTRATION

NCT05554549, registered retrospectively.

Exploring the Mediating Role of Executive Function in the Relationship between Aerobic Fitness and Academic Achievement in Adolescents

(1) Background: The performance of physical activity at a sufficient volume can result in improvements in fitness. Aerobic fitness is a particular aspect of fitness that has consistently been shown to be related to both cognitive performance and academic achievement. Cognitive performance, particularly executive function, is itself predictive of academic achievement. It has been hypothesized that the benefits of aerobic fitness for academic achievement are due to improvements in executive function. This study explores the mediating role of cognitive performance on the relationship between physical fitness and academic achievement in high-school-aged adolescents. (2) Methods: High school students (N = 283, 127 male, 156 females, mean age = 16.05 years, SD = 0.41) performed a shuttle run test to assess aerobic fitness and completed the Stroop Color, Stroop Word, and Stroop Color–Word tests to measure information processing and inhibition. They also completed the National Union Academic Achievement Assessment (NUAAA) as part of their high school requirements. (3) Results: Mediation analyses showed that inhibition (performance in the Stroop Color–Word test) fully mediated the relationship between aerobic fitness and both mathematics and Korean performance. (4) Conclusions: This cross-sectional investigation suggests an important mediating role of cognitive performance related to executive function in understanding the relationship between aerobic fitness and the academic achievement of high-school-aged adolescents. This suggests that enhancements in performance related to executive function, which are attributed to increases in aerobic fitness, could explain the observed benefits for academic attainment.

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.

Chronic Exposure to Fine Particulate Matter Increases Mortality Through Pathways of Metabolic and Cardiovascular Disease: Insights From a Large Mediation Analysis

Background

Long‐term exposure to outdoor fine particulate matter (PM 2.5 ) is the leading environmental risk factor for premature mortality worldwide. Characterizing important pathways through which PM 2.5 increases individuals' mortality risk can clarify the PM 2.5 –mortality relationship and identify possible points of interventions. Recent evidence has linked PM 2.5 to the onset of diabetes and cardiovascular disease, but to what extent these associations contribute to the effect of PM 2.5 on mortality remains poorly understood.

Methods and Results

We conducted a population‐based cohort study to investigate how the effect of PM 2.5 on nonaccidental mortality is mediated by its impacts on incident diabetes, acute myocardial infarction, and stroke. Our study population comprised ≈200 000 individuals aged 20 to 90 years who participated in population‐based health surveys in Ontario, Canada, from 1996 to 2014. Follow‐up extended until December 2017. Using causal mediation analyses with Aalen additive hazards models, we decomposed the total effect of PM 2.5 on mortality into a direct effect and several path‐specific indirect effects mediated by diabetes, each cardiovascular event, or both combined. A series of sensitivity analyses were also conducted. After adjusting for various individual‐ and neighborhood‐level covariates, we estimated that for every 1000 adults, each 10 μg/m 3 increase in PM 2.5 was associated with ≈2 incident cases of diabetes, ≈1 major cardiovascular event (acute myocardial infarction and stroke combined), and ≈2 deaths annually. Among PM 2.5 ‐related deaths, 31.7% (95% CI, 17.2%–53.2%) were attributable to diabetes and major cardiovascular events in relation to PM 2.5 . Specifically, 4.5% were explained by PM 2.5 ‐induced diabetes, 22.8% by PM 2.5 ‐induced major cardiovascular events, and 4.5% through their interaction.

Conclusions

This study suggests that a significant portion of the estimated effect of long‐term exposure to PM 2.5 on deaths can be attributed to its effect on diabetes and cardiovascular diseases, highlighting the significance of PM 2.5 on deteriorating cardiovascular health. Our findings should raise awareness among professionals that improving metabolic and cardiovascular health may reduce mortality burden in areas with higher exposure to air pollution.

Bayesian network mediation analysis with application to the brain functional connectome

The brain functional connectome, the collection of interconnected neural circuits along functional networks, facilitates a cutting-edge understanding of brain functioning, and has a potential to play a mediating role within the effect pathway between an exposure and an outcome. While existing mediation analytic approaches are capable of providing insight into complex processes, they mainly focus on a univariate mediator or mediator vector, without considering network-variate mediators. To fill the methodological gap and accomplish this exciting and urgent application, in the article, we propose an integrative mediation analysis under a Bayesian paradigm with networks entailing the mediation effect. To parameterize the network measurements, we introduce individually specified stochastic block models with unknown block allocation, and naturally bridge effect elements through the latent network mediators induced by the connectivity weights across network modules. To enable the identification of truly active mediating components, we simultaneously impose a feature selection across network mediators. We show the superiority of our model in estimating different effect components and selecting active mediating network structures. As a practical illustration of this approach's application to network neuroscience, we characterize the relationship between a therapeutic intervention and opioid abstinence as mediated by brain functional sub-networks.

A randomized controlled trial of customized adherence enhancement (CAE-E): study protocol for a hybrid effectiveness-implementation project

BACKGROUND

Mood-stabilizing medications are a cornerstone of treatment for people with bipolar disorder, though approximately half of these individuals are poorly adherent with their medication, leading to negative and even severe health consequences. While a variety of approaches can lead to some improvement in medication adherence, there is no single approach that has superior adherence enhancement and limited data on how these approaches can be implemented in clinical settings. Existing data have shown an increasing need for virtual delivery of care and interactive telemedicine interventions may be effective in improving adherence to long-term medication.

METHODS

Customized adherence enhancement (CAE) is a brief, practical bipolar-specific approach that identifies and targets individual patient adherence barriers for intervention using a flexibly administered modular format that can be delivered via telehealth communications. CAE is comprised of up to four standard treatment modules including Psychoeducation, Communication with Providers, Medication Routines, and Modified Motivational Interviewing. Participants will attend assigned module sessions with an interventionist based on their reasons for non-adherence and will be assessed for adherence, functioning, bipolar symptoms, and health resource use across a 12-month period. Qualitative and quantitative data will also be collected to assess barriers and facilitators to CAE implementation and reach and adoption of CAE among clinicians in the community.

DISCUSSION

The proposed study addresses the need for practical adherence interventions that are effective, flexible, and designed to adapt to different settings and patients. By focusing on a high-risk, vulnerable group of people with bipolar disorder, and refining an evidence-based approach that will integrate into workflow of public-sector care and community mental health clinics, there is substantial potential for improving bipolar medication adherence and overall health outcomes on a broad level.

TRIAL REGISTRATION

The study was registered on ClinicalTrials.gov NCT04622150 on November 9, 2020.

Using mHealth to improve adherence and reduce blood pressure in individuals with hypertension and bipolar disorder (iTAB-CV): study protocol for a 2-stage randomized clinical trial

BACKGROUND

Cardiovascular disease in individuals with mental health conditions such as bipolar disorder is highly prevalent and often poorly managed. Individuals with bipolar disorder face significant medication adherence barriers, especially when they are prescribed multiple medications for other health conditions including hypertension. Poor adherence puts them at a disproportionate risk for poor health outcomes. As such, there is a need for effective interventions to improve hypertension medication adherence, particularly in patients that struggle with adherence due to mental health comorbidity.

METHODS

This 5-year project uses a 2-stage randomized controlled trial design to evaluate a brief, practical adherence intervention delivered via interactive text messaging (iTAB-CV) along with self-monitoring of medication taking, mood, and home blood pressure (N = 100) compared to self-monitoring alone (N = 100). Prior to randomization, all participants will view an educational video that emphasizes the importance of medication for the treatment of hypertension and bipolar disorder. Those randomized to the texting intervention will receive daily text messages with predetermined content to address 11 salient domains as well as targeted customized messages for 2 months. This group will then be re-randomized to receive either a high (gradual taper from daily to weekly texts) or low booster (weekly texts) phase for an additional 2 months. All participants will be monitored for 52 weeks. The primary outcomes are systolic blood pressure and adherence to antihypertensive medication as determined by a self-reported questionnaire and validated with an automated pill-monitoring device. Secondary outcomes include adherence to bipolar disorder medications, psychiatric symptoms, health status, self-efficacy for medication-taking behavior, illness beliefs, medication attitudes, and habit strength.

DISCUSSION

This study specifically targets blood pressure and mental health symptom control in people with bipolar and includes implementation elements in the study design intended to inform future scale-up. Promising pilot data and a theoretical model, which views sustained medication-taking behavior in the context of habit formation, suggests that this remotely delivered intervention may help advance care for this high-risk population and is amenable to both scale up and easy adaptation for other groups with poor medication adherence.

TRIAL REGISTRATION

The study was registered at ClinicalTrials.gov ( NCT04675593 ) on December 19, 2020.

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.

Tobacco control policies and smoking cessation treatment utilization: A moderated mediation analysis

BACKGROUND

Tobacco policies, including clean indoor air laws and cigarette taxes, increase smoking cessation in part by stimulating the use of cessation treatments. We explored whether the associations between tobacco policies and treatment use varies across sociodemographic groups.

METHODS

We used data from 62,165 U.S. adult participants in the 2003 and 2010/11 Tobacco Use Supplement to the Current Population Survey (TUS-CPS) who reported smoking cigarettes during the past-year. We built on prior structural equation models used to quantify the degree to which smoking cessation treatment use (prescription medications, nicotine replacement therapy, counseling/support groups, quitlines, and internet resources) mediated the association between clean indoor air laws, cigarette excise taxes, and recent smoking cessation. In the current study, we added selected moderators to each model to investigate whether associations between tobacco polices and smoking cessation treatment use varied by sex, race/ethnicity, education, income, and health insurance status.

RESULTS

Associations between clean indoor air laws and the use of prescription medication and nicotine replacement therapies varied significantly between racial/ethnic, age, and education groups in 2003. However, none of these moderation effects remained significant in 2010/11. Higher cigarette excise taxes in 2010/2011 were associated with higher odds of using counseling among older adults and higher odds of using prescription medications among younger adults. No other moderator reached statistical significance. Smoking cessation treatments did not mediate the effect of taxes on smoking cessation in 2003 and were not included in these analyses.

CONCLUSIONS

Sociodemographic differences in associations between clean indoor air laws and smoking cessation treatment use have decreased from 2003 to 2010/11. In most cases, policies appear to stimulate smoking cessation treatment use similarly across varied sociodemographic groups.

Assessing Potential Outcomes Mediation in HIV Interventions

Knowledge of causal processes through mediation analysis can help improve the effectiveness and reduce costs of public health programs, like HIV prevention and treatment interventions. Advancements in mediation using the potential outcomes framework provide a method for estimating the causal effect of interventions on outcomes via a mediating variable. The purpose of this paper is to provide practical information about mediation and the potential outcomes framework that can enhance data analysis and causal inference for intervention studies. Causal mediation effects are defined and then estimated using data from an HIV intervention randomized trial among people who inject drugs (PWID) in Ukraine. Results from a potential outcomes mediation analysis show that the intervention had a total causal effect on incident HIV infection such that participants in the experimental group were 36% less likely to become infected during the 12-month study than those in the control arm, but that neither self-efficacy nor network communication mediated this effect. Because neither putative mediator was significant, measurement and confounding issues should be investigated to rule out these mediators. Other putative mediators, such as injection frequency, route of administration, or HIV knowledge can be considered. Future research is underway to examine additional, multiple mediators explaining efficacy of the current intervention and sensitivity to confounding effects.

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.

Latent class mediator for multiple indicators of mediation

This paper demonstrates the utility of latent classes in evaluating the effect of an intervention on an outcome through multiple indicators of mediation. These indicators are observed intermediate variables that identify an underlying latent class mediator, with each class representing a different mediating pathway. The use of a latent class mediator allows us to avoid modeling the complex interactions between the multiple indicators and ensures the decomposition of the total mediating effects into additive effects from individual mediating pathways, a desirable feature for evaluating multiple indicators of mediation. This method is suitable when the goal is to estimate the total mediating effects that can be decomposed into the additive effects of distinct mediating pathways. Each indicator may be involved in multiple mediation pathways and at the same time multiple indicators may contribute to a single mediating pathway. The relative importance of each pathway may vary across subjects. We applied this method to the analysis of the first 6 months of data from a 2-year clustered randomized trial for adults in their first episode of schizophrenia. Four indicators of mediation are considered: individual resiliency training; family psychoeducation; supported education and employment; and a structural assessment for medication. The improvement in symptoms was found to be mediated by the latent class mediator derived from these four service indicators. Simulation studies were conducted to assess the performance of the proposed model and showed that the simultaneous estimation through the maximum likelihood yielded little bias when the entropy of the indicators was high.

Gene-based mediation analysis in epigenetic studies

Mediation analysis has been a useful tool for investigating the effect of mediators that lie in the path from the independent variable to the outcome. With the increasing dimensionality of mediators such as in (epi)genomics studies, high-dimensional mediation model is needed. In this work, we focus on epigenetic studies with the goal to identify important DNA methylations that act as mediators between an exposure disease outcome. Specifically, we focus on gene-based high-dimensional mediation analysis implemented with kernel principal component analysis to capture potential nonlinear mediation effect. We first review the current high-dimensional mediation models and then propose two gene-based analytical approaches: gene-based high-dimensional mediation analysis based on linearity assumption between mediators and outcome (gHMA-L) and gene-based high-dimensional mediation analysis based on nonlinearity assumption (gHMA-NL). Since the underlying true mediation relationship is unknown in practice, we further propose an omnibus test of gene-based high-dimensional mediation analysis (gHMA-O) by combing gHMA-L and gHMA-NL. Extensive simulation studies show that gHMA-L performs better under the model linear assumption and gHMA-NL does better under the model nonlinear assumption, while gHMA-O is a more powerful and robust method by combining the two. We apply the proposed methods to two datasets to investigate genes whose methylation levels act as important mediators in the relationship: (1) between alcohol consumption and epithelial ovarian cancer risk using data from the Mayo Clinic Ovarian Cancer Case-Control Study and (2) between childhood maltreatment and comorbid post-traumatic stress disorder and depression in adulthood using data from the Gray Trauma Project.

Multimodal neuroimaging data integration and pathway analysis

With advancements in technology, the collection of multiple types of measurements on a common set of subjects is becoming routine in science. Some notable examples include multimodal neuroimaging studies for the simultaneous investigation of brain structure and function and multi-omics studies for combining genetic and genomic information. Integrative analysis of multimodal data allows scientists to interrogate new mechanistic questions. However, the data collection and generation of integrative hypotheses is outpacing available methodology for joint analysis of multimodal measurements. In this article, we study high-dimensional multimodal data integration in the context of mediation analysis. We aim to understand the roles that different data modalities play as possible mediators in the pathway between an exposure variable and an outcome. We propose a mediation model framework with two data types serving as separate sets of mediators and develop a penalized optimization approach for parameter estimation. We study both the theoretical properties of the estimator through an asymptotic analysis and its finite-sample performance through simulations. We illustrate our method with a multimodal brain pathway analysis having both structural and functional connectivity as mediators in the association between sex and language processing.

Parenting stress and family resilience affect the association of adverse childhood experiences with children's mental health and attention-deficit/hyperactivity disorder

INTRODUCTION

Exposure to adverse childhood experience (ACE) has harmful consequences for children's health and well-being. However, it is less clear how different social processes may amplify or mitigate the effects of ACE on children's mental health. We examined how parenting stress mediates and family resilience moderates the associations of ACE with children's mental health and attention-deficit/hyperactivity disorder (ADHD) outcomes.

METHODS

This secondary data analysis included 44,684 children aged 6-17 years from the 2016-17 National Survey of Children's Health (NSCH). Logistic regression with survey weights was used to account for the complex survey design and obtain odds ratios (OR) and 95% CI adjusted for sociodemographics.

RESULTS

Overall, 7.3% of children had any mental health condition and 10.4% had ADHD. A higher ACE score (per 1-unit increase) was associated with a higher prevalence of any mental health condition (OR = 1.33, 95% CI: 1.27-1.40) and ADHD (OR = 1.21; 95% CI: 1.15-1.27) after adjustment for sociodemographics. Parenting stress mediated 57% of the total effect of ACE on any mental health condition and 60% of the total effect of ACE on ADHD diagnosis. The effect of ACE on mental health and ADHD outcomes was stronger among children with low levels of family resilience and connection index (FRCI) than among those with higher levels of FRCI.

CONCLUSIONS

Parenting stress may be a potential mechanism through which ACE impacts a child's mental health and behavioral outcomes. Family resilience can lessen the impact of ACE on children's mental health and behavioral disorders.

Cigarette excise taxes, clean indoor air laws, and use of smoking cessation treatments: A mediation analysis

The role of smoking cessation treatments in the link between clean indoor air laws and cigarette taxes with smoking cessation is not known. This study examined whether the use of smoking cessation treatments mediates the association between clean indoor air laws and cigarette excise taxes, on the one hand, and recent smoking cessation, on the other hand. Using data on 62,165 adult participants in the 2003 and 2010-2011 Current Population Survey-Tobacco Use Supplement who reported smoking cigarettes in the past year, we employed structural equation models to quantify the degree to which smoking cessation treatments (prescription medications, nicotine replacement therapy, counseling/support groups, quitlines, and internet-based resources) mediate the association between clean indoor air laws, cigarette excise taxes and recent smoking cessation. Recent smoking cessation was associated with clean indoor air laws in 2003 and with both clean indoor air laws and excise taxes in 2010-2011. Smoking cessation treatments explained between 29% to 39% of the effect of clean indoor air laws and taxes on recent smoking cessation. While clean indoor air laws remained significantly associated with the recent smoking cessation over the first decade of the 2000s, excise taxes gained a more prominent role in later years of that decade. The influence of these policies was partly mediated through the use of smoking cessation treatments, underscoring the importance of policies that make these treatments more widely available.

Association of Socioeconomic Status and Comorbidities with Racial Disparities during Kidney Transplant Evaluation

BACKGROUND AND OBJECTIVES

Black patients referred for kidney transplantation have surpassed many obstacles but likely face continued racial disparities before transplant. The mechanisms that underlie these disparities are unclear. We determined the contributions of socioeconomic status (SES) and comorbidities as mediators to disparities in listing and transplant.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

We studied a cohort (n=1452 black; n=1561 white) of patients with kidney failure who were referred for and started the transplant process (2009-2018). We estimated the direct and indirect effects of SES (self-reported income, education, and employment) and medical comorbidities (self-reported and chart-abstracted) as mediators of racial disparities in listing using Cox proportional hazards analysis with inverse odds ratio weighting. Among the 983 black and 1085 white candidates actively listed, we estimated the direct and indirect effects of SES and comorbidities as mediators of racial disparities on receipt of transplant using Poisson regression with inverse odds ratio weighting.

RESULTS

Within the first year, 876 (60%) black and 1028 (66%) white patients were waitlisted. The relative risk of listing for black compared with white patients was 0.76 (95% confidence interval [95% CI], 0.69 to 0.83); after adjustment for SES and comorbidity, the relative risk was 0.90 (95% CI, 0.83 to 0.97). The proportion of the racial disparity in listing was explained by SES by 36% (95% CI, 26% to 57%), comorbidity by 44% (95% CI, 35% to 61%), and SES with comorbidity by 58% (95% CI, 44% to 85%). There were 409 (42%) black and 496 (45%) white listed candidates transplanted, with a median duration of follow-up of 3.9 (interquartile range, 1.2-7.1) and 2.8 (interquartile range, 0.8-6.3) years, respectively. The incidence rate ratio for black versus white candidates was 0.87 (95% CI, 0.79 to 0.96); SES and comorbidity did not explain the racial disparity.

CONCLUSIONS

SES and comorbidity partially mediated racial disparities in listing but not for transplant.

Bayesian Causal Mediation Analysis with Multiple Ordered Mediators

Causal mediation analysis provides investigators insight into how a treatment or exposure can affect an outcome of interest through one or more mediators on causal pathway. When multiple mediators on the pathway are causally ordered, identification of mediation effects on certain causal pathways requires a sensitivity parameter to be specified. A mixed model-based approach was proposed in the Bayesian framework to connect potential outcomes at different treatment levels, and identify mediation effects independent of a sensitivity parameter, for the natural direct and indirect effects on all causal pathways. The proposed method is illustrated in a linear setting for mediators and outcome, with mediator-treatment interactions. Sensitivity analysis was performed for the prior choices in the Bayesian models. The proposed Bayesian method was applied to an adolescent dental health study, to see how social economic status can affect dental caries through a sequence of causally ordered mediators in dental visit and oral hygiene index.

Causality and Speech, Language and Hearing Sciences: epidemiological approach

PURPOSE

To review the Speech-Language Pathology literature studies from the epidemiology and causality perspective.

RESEARCH STRATEGIES

A national and international literature survey was carried out with searches from PubMed, SciELO and gray literature bases, conducted according to the instructions of the Cochrane Collaboration and published until January 9th, 2019. The review guiding question asks if Speech-Language Pathology uses methods in their evidence to infer causality.

SELECTION CRITERIA

All studies that presented a causal epidemiological approach in speech therapy were included, as well as excluded those that did not present an appropriate methodological approach for cause and effect analysis.

DATA ANALYSIS

Two authors of this study independently reviewed all citations. A priori determined form was used to extract the following data: author, year of publication, country of origin, theoretical conception, application or not of the study and central discussion addressed in the article.

RESULTS

From the search performed 3842 articles were found. However, none of them investigated their outcomes from the causality point of view, not allowing cause and effect inference.

CONCLUSION

There is a shortage of studies that evidence causality in Speech-Language Pathology, which may alter the effectiveness and reliable handling of diagnosis and speech-language therapy, since it is still based on association and not on cause and effect based on studies designed to that.

Continuous-time causal mediation analysis

While causal mediation analysis has seen considerable recent development for a single measured mediator (M) and final outcome (Y), less attention has been given to repeatedly measured M and Y. Previous methods have typically involved discrete-time models that limit inference to the particular measurement times used and do not recognize the continuous nature of the mediation process over time. To overcome such limitations, we present a new continuous-time approach to causal mediation analysis that uses a differential equations model in a potential outcomes framework to describe the causal relationships among model variables over time. A connection between the differential equation models and standard repeated measures models is made to provide convenient model formulation and fitting. A continuous-time extension of the sequential ignorability assumption allows for identifiable natural direct and indirect effects as functions of time, with estimation based on a two-step approach to model fitting in conjunction with a continuous-time mediation formula. Novel features include a measure of an overall mediation effect based on the "area between the curves," and an approach for predicting the effects of new interventions. Simulation studies show good properties of estimators and the new methodology is applied to data from a cohort study to investigate sugary drink consumption as a mediator of the effect of socioeconomic status on dental caries in children.

Design-based single-mediator approach for complex survey data

We discuss a two-step approach to test for a mediated effect using data gathered via complex sampling. The approach incorporates design-based multiple linear regressions and a generalized Sobel’s method to test for significance of a mediated effect. We illustrate the applications to a study of nicotine dependence, race/ethnicity and cigarette purchase price among daily smokers in the U.S. The study goal was to assess significance of cigarette purchase price as a mediator in the association between race/ethnicity (non-Hispanic Black/African American, non-Hispanic White) and nicotine dependence measured in terms of the average number of cigarettes smoked per day. The single-mediator model incorporated 18 covariates as control factors. The results indicated a significant mediated effect of cigarette purchase price on the association. However, the relative effect size of 5% indicated low practical significance of the cigarette purchase price as a mediator in the association between race/ethnicity and nicotine dependence. The approach can be modified to studies where data are gathered via other types of complex sampling.

Sparse Principal Component based High-Dimensional Mediation Analysis

Causal mediation analysis aims to quantify the intermediate effect of a mediator on the causal pathway from treatment to outcome. When dealing with multiple mediators, which are potentially causally dependent, the possible decomposition of pathway effects grows exponentially with the number of mediators. An existing approach incorporated the principal component analysis (PCA) to address this challenge based on the fact that the transformed mediators are conditionally independent given the orthogonality of the principal components (PCs). However, the transformed mediator PCs, which are linear combinations of original mediators, can be difficult to interpret. A sparse high-dimensional mediation analysis approach is proposed which adopts the sparse PCA method to the mediation setting. The proposed approach is applied to a task-based functional magnetic resonance imaging study, illustrating its ability to detect biologically meaningful results related to an identified mediator.

BAYESIAN METHODS FOR MULTIPLE MEDIATORS: RELATING PRINCIPAL STRATIFICATION AND CAUSAL MEDIATION IN THE ANALYSIS OF POWER PLANT EMISSION CONTROLS

Emission control technologies installed on power plants are a key feature of many air pollution regulations in the US. While such regulations are predicated on the presumed relationships between emissions, ambient air pollution, and human health, many of these relationships have never been empirically verified. The goal of this paper is to develop new statistical methods to quantify these relationships. We frame this problem as one of mediation analysis to evaluate the extent to which the effect of a particular control technology on ambient pollution is mediated through causal effects on power plant emissions. Since power plants emit various compounds that contribute to ambient pollution, we develop new methods for multiple intermediate variables that are measured contemporaneously, may interact with one another, and may exhibit joint mediating effects. Specifically, we propose new methods leveraging two related frameworks for causal inference in the presence of mediating variables: principal stratification and causal mediation analysis. We define principal effects based on multiple mediators, and also introduce a new decomposition of the total effect of an intervention on ambient pollution into the natural direct effect and natural indirect effects for all combinations of mediators. Both approaches are anchored to the same observed-data models, which we specify with Bayesian nonparametric techniques. We provide assumptions for estimating principal causal effects, then augment these with an additional assumption required for causal mediation analysis. The two analyses, interpreted in tandem, provide the first empirical investigation of the presumed causal pathways that motivate important air quality regulatory policies.

Early life socioeconomic position and mortality from cardiovascular diseases: an application of causal mediation analysis in the Stockholm Public Health Cohort

OBJECTIVE

We aimed to quantify the mediating impact of adult social and behavioural mechanisms in the association between childhood socioeconomic position (SEP) and cardiovascular disease (CVD) mortality by employing a weighting approach to mediation analysis.

DESIGN

Prospective cohort study.

SETTING

Stockholm County, Sweden.

PARTICIPANTS

19 720 individuals who participated in the Stockholm Public Health Cohort survey in 2002 and were older than 40 years.

PRIMARY AND SECONDARY OUTCOME MEASURES

The primary outcome was CVD mortality. Non-CVD mortality was additionally analysed for comparison.

METHODS

Study subjects were followed in routine registers from 2002 to 2011 for mortality. Data on father's SEP and adult social and behavioural factors came from questionnaire survey. The inverse odds weighting method was used to estimate the total effect, the natural direct effect and the natural indirect effect (NIE) in Poisson regression models. All results were adjusted for gender, age, country of birth and marital status. Multiple imputation was used to handle missing data.

RESULTS

The total effect of manual versus non-manual father's SEP on CVD mortality was estimated as an incidence rate ratio (IRR) of 1.24 (95% CI 1.09 to 1.41). When the social and behavioural factors were accounted for, the IRR for the NIE was 1.09 (95% CI 1.04 to 1.14), suggesting a mediation of 44% of the total effect. As for non-CVD mortality, father's manual SEP was associated with 1.15 fold excess risk (IRR: 1.15; 95% CI 1.04 to 1.27) of which the effect represented by the whole set of mediators was 1.06 (95% CI 1.01 to 1.10).

CONCLUSION

Adult social and behavioural factors had a considerable mediating effect on the early life social origin of mortality from CVDs and other causes. Future research employing causal mediation analysis may nevertheless have to consider additional factors for a fuller understanding of the mechanisms.

Generalized causal mediation and path analysis: Extensions and practical considerations

Causal mediation analysis seeks to decompose the effect of a treatment or exposure among multiple possible paths and provide casually interpretable path-specific effect estimates. Recent advances have extended causal mediation analysis to situations with a sequence of mediators or multiple contemporaneous mediators. However, available methods still have limitations, and computational and other challenges remain. The present paper provides an extended causal mediation and path analysis methodology. The new method, implemented in the new R package, gmediation (described in a companion paper), accommodates both a sequence (two stages) of mediators and multiple mediators at each stage, and allows for multiple types of outcomes following generalized linear models. The methodology can also handle unsaturated models and clustered data. Addressing other practical issues, we provide new guidelines for the choice of a decomposition, and for the choice of a reference group multiplier for the reduction of Monte Carlo error in mediation formula computations. The new method is applied to data from a cohort study to illuminate the contribution of alternative biological and behavioral paths in the effect of socioeconomic status on dental caries in adolescence.

High-dimensional multivariate mediation with application to neuroimaging data

Mediation analysis is an important tool in the behavioral sciences for investigating the role of intermediate variables that lie in the path between a treatment and an outcome variable. The influence of the intermediate variable on the outcome is often explored using a linear structural equation model (LSEM), with model coefficients interpreted as possible effects. While there has been significant research on the topic, little work has been done when the intermediate variable (mediator) is a high-dimensional vector. In this work, we introduce a novel method for identifying potential mediators in this setting called the directions of mediation (DMs). DMs linearly combine potential mediators into a smaller number of orthogonal components, with components ranked based on the proportion of the LSEM likelihood each accounts for. This method is well suited for cases when many potential mediators are measured. Examples of high-dimensional potential mediators are brain images composed of hundreds of thousands of voxels, genetic variation measured at millions of single nucleotide polymorphisms (SNPs), or vectors of thousands of variables in large-scale epidemiological studies. We demonstrate the method using a functional magnetic resonance imaging study of thermal pain where we are interested in determining which brain locations mediate the relationship between the application of a thermal stimulus and self-reported pain.

Anti-chlamydia IgG and IgA are insufficient to prevent endometrial chlamydia infection in women, and increased anti-chlamydia IgG is associated with enhanced risk for incident infection

PROBLEM

Chlamydia infections in women can ascend to the upper genital tract, and repeated infections are common, placing women at risk for sequelae. The protective role of anti-chlamydia antibodies to surface exposed antigens in ascending and incident infection is unclear.

METHOD OF STUDY

A whole-bacterial ELISA was used to quantify chlamydia-specific IgG and IgA in serum and cervical secretions of 151 high-risk women followed longitudinally. Correlations were determined between antibody and cervical burden, and causal mediation analysis investigated the effect of antibody on ascension. We examined the relationship of antibody to incident infection using the marginal Cox model.

RESULTS

Serum and cervical anti-chlamydia IgG and cervical IgA levels correlated inversely with cervical burden. While lower burden was associated with reduced ascension, causal mediation analysis revealed that the indirect effects of antibody mediated through reductions in bacterial burden were insufficient to prevent ascension. Analysis of women uninfected at enrollment revealed that serum and cervical anti-chlamydia IgG were associated with increased risk of incident infection; hazard ratio increased 3.6-fold (95% CI, 1.3-10.3), and 22.6-fold (95% CI, 3.1-165.2) with each unit of serum and cervical IgG, respectively.

CONCLUSION

Although anti-chlamydia IgG and IgA correlated with reduced cervical chlamydia burden, they failed to prevent ascension and increased levels of anti-chlamydia IgG were associated with increased risk for incident infection.

Early Origins of Later Life Psychological Well-Being? A Novel Application of Causal Mediation Analysis to Life Course Research

Objectives

This study employs a novel approach to mediation analysis to clarify the influence of interrelated indicators of life course socioeconomic status (SES) on later life psychological well-being in India. Contrary to traditional approaches (i.e., use of product and difference-in-coefficients), we recognize the role of confounders in the estimation of total, direct, and indirect effects of parental education on respondents' psychological well-being.

Method

Drawing from the first wave (2007-2010) of the Study on Global Ageing and Adult Health (SAGE) and adopting a counterfactual approach, we estimate both natural direct and indirect effects of parental education through individual educational attainment (secondarily, through household assets as an additional mediator) on respondents' life-satisfaction and quality of life (QOL).

Results

Findings document a statistically not significant positive total effect of parental education on life satisfaction and QOL. While lower for women, significant indirect effects suggest that the positive influence of parental education operates primarily through the individual's education. Notably, we found negative direct effect of parental education on psychological well-being outcomes.

Discussion

Contrary to prior literature, we found no positive direct influence of parental education on later life psychological well-being, but established its influence through socioeconomic positioning over the life course.

A multiple mediator analysis approach to quantify the effects of the ADH1B and ALDH2 genes on hepatocellular carcinoma risk

Previous work suggested a genetic component affecting the risk of hepatocellular carcinoma (HCC) and mediation analyses have elucidated potential indirect pathways of these genetic effects. Specifically, the effects of alcohol dehydrogenase (ADH1B) and aldehyde dehydrogenase (ALDH2) genes on HCC risk vary based on alcohol consumption habits. However, alcohol consumption may not be the only mediator in the identified pathway: factors related to alcohol consumption may contribute to the same indirect pathway. Thus, we developed a multimediator model to quantify the genetic effects on HCC risk through sequential dichotomous mediators under the counterfactual framework. Our method provided a closed form formula for the mediation effects through different indirect paths, which requires no assumption for the rarity of outcome. In simulation studies of a finite sample, we presented the utility of the method with the variance of the effects estimated using the delta method and bootstrapping. We applied our method to data from participants in Taiwan (580 cases and 3,207 controls) and quantified the mediation effects of single nucleotide polymorphisms (SNPs) in the ADH1B and ALDH2 genes on HCC through alcohol consumption (yes/no) and high alanine transaminase (ALT) levels (greater than or equal to 45 U/L or below 45 U/L). Assuming a dominant risk model, we identified that the SNPs' effects through alcohol consumption is more significant than through ALT levels on HCC risk. This new method provides insight to the magnitude of various casual mechanisms as a closed form solution and can be readily applied in other genomic studies.

A Tutorial in Bayesian Potential Outcomes Mediation Analysis

Statistical mediation analysis is used to investigate intermediate variables in the relation between independent and dependent variables. Causal interpretation of mediation analyses is challenging because randomization of subjects to levels of the independent variable does not rule out the possibility of unmeasured confounders of the mediator to outcome relation. Furthermore, commonly used frequentist methods for mediation analysis compute the probability of the data given the null hypothesis, which is not the probability of a hypothesis given the data as in Bayesian analysis. Under certain assumptions, applying the potential outcomes framework to mediation analysis allows for the computation of causal effects, and statistical mediation in the Bayesian framework gives indirect effects probabilistic interpretations. This tutorial combines causal inference and Bayesian methods for mediation analysis so the indirect and direct effects have both causal and probabilistic interpretations. Steps in Bayesian causal mediation analysis are shown in the application to an empirical example.

Do baby teeth really matter? Changing parental perception and increasing dental care utilization for young children

INTRODUCTION

Parent/caregivers' inability to recognize the importance of baby teeth has been associated with inadequate self-management of children's oral health (i.e. lower likelihood of preventive dental visits) which may result in dental caries and the need for more expensive caries-related restorative treatment under general anesthesia. Health behavior theories aid researchers in understanding the impact and effectiveness of interventions on changing health behaviors and health outcomes. One example is the Common-Sense Model of Self-Regulation (CSM) which focuses on understanding an individual's illness perception (i.e. illness and treatment representations), and subsequently has been used to develop behavioral interventions to change inaccurate perceptions and describe the processes involved in behavior change.

METHODS

We present two examples of randomized clinical trials that are currently testing oral health behavioral interventions to change parental illness perception and increase dental utilization for young children disproportionately impacted by dental caries in elementary schools and pediatric primary care settings. Additionally, we compared empiric data regarding parent/caregiver perception of the chronic nature of dental caries (captured by the illness perception questionnaire revised for dental: IPQ-RD constructs: identity, consequences, control, timeline, illness coherence, emotional representations) between parent/caregivers who did and did not believe baby teeth were important.

RESULTS

Caregivers who believed that baby teeth don't matter had significantly (P<0.05) less accurate perception in the majority of the IPQ-RD constructs (except timeline construct) compared to caregivers who believed baby teeth do matter.

CONCLUSION

These findings support our CSM-based behavioral interventions to modify caregiver caries perception, and improve dental utilization for young children.

A Multiyear Assessment of Public Response to a Statewide Drug Take-Back and Disposal Campaign, 2010 to 2012

This study is the first to analyze public response to a drug take-back program, the American Medicine Chest Challenge, in a single state over a period of 3 years (2010-2012). The study utilized a three-wave repeated cross-sectional design and an annual phone survey conducted with a representative sample of adults ( N = 906 in 2010, N = 907 in 2011, and N = 906 in 2012), which assessed exposure to the campaign, drug disposal behaviors, possible mediators of campaign effects (risk appraisal, personal agency, normative influence, and interpersonal talk), and potential confounders. Logistic regression and causal mediation analysis were employed to estimate confounder-adjusted direct and mediated effects of the campaign. Results showed that the campaign reached a sizable portion (50% to 60%) of state adults and that campaign exposure was associated with increased likelihood of having conversations with others about this topic. About 55% of all adults in the state reported taking at least one of the actions recommended by the campaign, and campaign exposure was associated with increased likelihood of disposing of prescription drugs at a drug collection day event (adjusted odds ratio = 4) and of talking to a child about the risks associated with prescription drug abuse (adjusted odds ratio = 2). The causal mediation analysis demonstrated that the campaign influenced audiences by reinforcing their efficacy to safely dispose of prescription drugs, but also potentially by stimulating conversations among community members about this topic. Drug take-back campaigns can be an effective mechanism to decrease the availability of prescription drugs in communities.

Assessing natural direct and indirect effects for a continuous exposure and a dichotomous outcome

Recent advances in the literature on mediation have extended from traditional linear structural equation modeling approach to causal mediation analysis using potential outcomes framework. Pearl proposed a mediation formula to calculate expected potential outcomes used in the natural direct and indirect effects definition under the key sequential ignorability assumptions. Current methods mainly focused on binary exposure variables, and in this article, this approach is further extended to settings in which continuous exposures may be of interest. Focusing on a dichotomous outcome, we give precise definitions of the natural direct and indirect effects on both the risk difference and odds ratio scales utilizing the empirical joint distribution of the exposure and baseline covariates from the whole sample analysis population. A mediation-formula based approach is proposed to estimate the corresponding causal quantities. Simulation study is conducted to assess the statistical properties of the proposed method and we illustrate our approach by applying it to the Jackson Heart Study to estimate the mediation effects of diabetes on the relation between obesity and chronic kidney disease. Sensitivity analysis is performed to assess the impact of violation of no unmeasured mediator-outcome confounder assumption.

Life Course Pathways of Adversities Linking Adolescent Socioeconomic Circumstances and Functional Somatic Symptoms in Mid-Adulthood: A Path Analysis Study

While research examining the health impact of early socioeconomic conditions suggests that effects may exist independently of or jointly with adult socioeconomic position, studies exploring other potential pathways are few. Following a chain of risk life course model, this prospective study seeks to examine whether pathways of occupational class as well as material and social adversities across the life course link socioeconomic disadvantage in adolescent to functional somatic symptoms in mid-adulthood. Applying path analysis, a multiple mediator model was assessed using prospective data collected during 26 years through the Northern Swedish Cohort. The sample contained 987 individuals residing in the municipality of Luleå, Sweden, who participated in questionnaire surveys at age 16, 21, 30 and 42. Socioeconomic conditions (high/low) in adolescence (age 16) were operationalized using the occupation of the parents, while occupational class in adulthood (manual/non-manual) was measured using the participant’s own occupation at age 21 and 30. The adversity measurements were constructed as separate age specific parcels at age 21 and 30. Social adversity included items pertaining to stressful life events that could potentially harm salient relationships, while material adversity was operationalized using items concerning unfavorable financial and material circumstances. Functional somatic symptoms at age 42 was a summary measure of self-reported physical symptoms, palpitation and sleeping difficulties that had occurred during the last 12 months. An association between socioeconomic conditions at age 16 and functional somatic symptoms at age 42 (r = 0.068) which was partially explained by people’s own occupational class at age 21 and then material as well as social adversity at age 30 was revealed. Rather than proposing a direct and independent health effect of the socioeconomic conditions of the family, the present study suggests that growing up in an unfavorable socioeconomic environment might be a source for a chain of adverse material and social living situations, which in turn affects adult health.

Causal mediation analysis with a latent mediator

Health researchers are often interested in assessing the direct effect of a treatment or exposure on an outcome variable, as well as its indirect (or mediation) effect through an intermediate variable (or mediator). For an outcome following a nonlinear model, the mediation formula may be used to estimate causally interpretable mediation effects. This method, like others, assumes that the mediator is observed. However, as is common in structural equations modeling, we may wish to consider a latent (unobserved) mediator. We follow a potential outcomes framework and assume a generalized structural equations model (GSEM). We provide maximum-likelihood estimation of GSEM parameters using an approximate Monte Carlo EM algorithm, coupled with a mediation formula approach to estimate natural direct and indirect effects. The method relies on an untestable sequential ignorability assumption; we assess robustness to this assumption by adapting a recently proposed method for sensitivity analysis. Simulation studies show good properties of the proposed estimators in plausible scenarios. Our method is applied to a study of the effect of mother education on occurrence of adolescent dental caries, in which we examine possible mediation through latent oral health behavior.

Sensitivity analyses for parametric causal mediation effect estimation

Causal mediation analysis uses a potential outcomes framework to estimate the direct effect of an exposure on an outcome and its indirect effect through an intermediate variable (or mediator). Causal interpretations of these effects typically rely on sequential ignorability. Because this assumption is not empirically testable, it is important to conduct sensitivity analyses. Sensitivity analyses so far offered for this situation have either focused on the case where the outcome follows a linear model or involve nonparametric or semiparametric models. We propose alternative approaches that are suitable for responses following generalized linear models. The first approach uses a Gaussian copula model involving latent versions of the mediator and the final outcome. The second approach uses a so-called hybrid causal-observational model that extends the association model for the final outcome, providing a novel sensitivity parameter. These models, while still assuming a randomized exposure, allow for unobserved (as well as observed) mediator-outcome confounders that are not affected by exposure. The methods are applied to data from a study of the effect of mother education on dental caries in adolescence.

Understanding treatment effect mechanisms of the CAMBRA randomized trial in reducing caries increment

The Caries Management By Risk Assessment (CAMBRA) randomized controlled trial showed that an intervention featuring combined antibacterial and fluoride therapy significantly reduced bacterial load and suggested reduced caries increment in adults with 1 to 7 baseline cavitated teeth. While trial results speak to the overall effectiveness of an intervention, insight can be gained from understanding the mechanism by which an intervention acts on putative intermediate variables (mediators) to affect outcomes. This study conducted mediation analyses on 109 participants who completed the trial to understand whether the intervention reduced caries increment through its action on potential mediators (oral bacterial load, fluoride levels, and overall caries risk based on the composite of bacterial challenge and salivary fluoride) between the intervention and dental outcomes. The primary outcome was the increment from baseline in decayed, missing, and filled permanent surfaces (ΔDMFS) 24 mo after completing restorations for baseline cavitated lesions. Analyses adjusted for baseline overall risk, bacterial challenge, and fluoride values under a potential outcome framework using generalized linear models. Overall, the CAMBRA intervention was suggestive in reducing the 24-mo DMFS increment (reduction in ΔDMFS: -0.96; 95% confidence interval [CI]: -2.01 to 0.08; P = 0.07); the intervention significantly reduced the 12-mo overall risk (reduction in overall risk: -19%; 95% CI, -7 to -41%;], P = 0.005). Individual mediators, salivary log10 mutans streptococci, log10 lactobacilli, and fluoride level, did not represent statistically significant pathways alone through which the intervention effect was transmitted. However, 36% of the intervention effect on 24-mo DMFS increment was through a mediation effect on 12-mo overall risk (P = 0.03). These findings suggest a greater intervention effect carried through the combined action on multiple aspects of the caries process rather than through any single factor. In addition, a substantial portion of the total effect of the CAMBRA intervention may have operated through unanticipated or unmeasured pathways not included among the potential mediators studied.