Commentary: Mediation Analyses in the Real World

An assessment of the mediating role of hypertension in the effect of long-term air pollution exposure on dementia

Background

Growing evidence links air pollution exposure to the risk of dementia. We hypothesized that hypertension may partially mediate this effect.

Methods

We previously documented an association between air pollution and dementia in the Ginkgo Evaluation of Memory Study, a randomized, placebo-controlled trial of 3069 adults ≥75 years across four US sites who were evaluated for dementia every 6 months from 2000-2008. We utilized a two-stage regression approach for causal mediation analysis to decompose the total effect of air pollution on dementia into its natural direct and indirect effect through prevalent hypertension. Exposure to air pollution in the 10 or 20 years before enrollment was assigned using estimates from fine-scale spatial-temporal models for PM2.5, PM10, and NO2. We used Poisson regression models for hypertension and Cox proportional hazard models for time-to-incident all-cause dementia, adjusting for a priori confounders.

Results

Participants were free of mild cognitive impairment at baseline (n = 2564 included in analyses); 69% had prevalent hypertension at baseline. During follow-up, 12% developed all-cause dementia (Alzheimer's disease [AD] = 212; vascular dementia with or without AD [VaD/AD mixed] = 97). We did not find an adverse effect of any air pollutant on hypertension. Hypertension was associated with VaD/AD mixed (HR, 1.92 [95% CI = 1.14, 3.24]) but not AD. We did not observe mediation through hypertension for the effect of any pollutant on dementia outcomes.

Conclusions

The lack of mediated effect may be due to other mechanistic pathways and the minimal effect of air pollution on hypertension in this cohort of older adults.

Testing the Multiple Pathways of Residential Greenness to Pregnancy Outcomes Model in a Sample of Pregnant Women in the Metropolitan Area of Donostia-San Sebastián

Residential greenness may positively impact diverse human health indicators through the reduction of air pollution, the improvement of psychological health, and the promotion of physical activity. Previous studies indicate a weak but positive association with pregnancy outcomes. Our aim was to test the multiple pathways from residential greenness to pregnancy outcomes model, using residential NO₂ concentrations, psychological health, and moderate-to-vigorous physical activity (MVPA) during the first trimester of pregnancy, in a sample of 440 pregnant women residing in Donostia, Spain. Three metrics of residential greenness were calculated around each participant’s home address: normalized difference vegetation index (NDVI) within 300 m, and green space (>5000 m²) availability within 300 and 500 m. Residential NO₂ concentrations, psychological health, and MVPA were explored as mediators of the associations between these metrics and the following pregnancy outcomes: birth weight (BW), low birth weight (LBW), prematurity, small for gestational age (SGA), and large for gestational age (LGA). Educational attainment, parity, and body mass index (BMI) were treated as covariates. Counterfactual mediation analyses showed very low to null statistical support for an association between any of the greenspace metrics and pregnancy outcomes in the full sample. Green space availability (300 m) was associated with lower BW and showed a marginal protective effect against LGA.

Socio-economic trajectories and cardiovascular disease mortality in older people: the English Longitudinal Study of Ageing

Background

Socio-economic status from early life has been linked to cardiovascular disease risk, but the impact of life-course socio-economic trajectories, as well as the mechanisms underlying social inequalities in cardiovascular disease risk, is uncertain.

Objectives

We assessed the role of behavioural, psychosocial and physiological (including inflammatory) factors in the association between life-course socio-economic status and cardiovascular disease mortality in older adults.

Methods

Participants were 7846 individuals (44% women) from the English Longitudinal Study of Ageing, a representative study of individuals aged ≥ 50 years, established in 2002-03. Comprising four indicators of socio-economic status (father's social class, own education, occupational position and wealth), we computed an index of socio-economic trajectory and a lifetime cumulative socio-economic score. Behavioural (smoking, physical activity, alcohol consumption, body mass index) and psychosocial (social relations, loneliness) factors, physiological (blood pressure, total cholesterol, triglycerides) and inflammatory markers (C-reactive protein, fibrinogen), measured repeatedly over time, were potential explanatory variables. Cardiovascular disease mortality was ascertained by linkage of study members to a national mortality register. Mediation was calculated using the traditional 'change-in-estimate method' and alternative approaches such as counterfactual mediation modelling could not be applied in this context.

Results

During the 8.4-year follow-up, 1301 study members died (438 from cardiovascular disease). A stable low-social-class trajectory was associated with around double the risk of cardiovascular disease mortality (hazard ratio; 95% confidence interval: 1.94, 1.37; 2.75) compared with a stable high social class across the life course. Individuals in the lowest relative to the highest life-course cumulative socio-economic status group were also more than twice as likely to die of cardiovascular disease (2.57, 1.81; 3.65). Behavioural factors and inflammatory markers contributed most to explaining this gradient, whereas the role of psychosocial and other physiological risk factors was modest.

Conclusions

In a population-based cohort of older individuals living in England, we provide evidence that disadvantage across the life course is linked to cardiovascular mortality. That behavioural factors and inflammatory markers partially explain this gradient may provide insights into the potential for intervention.

Mediation analysis with a time-to-event outcome: a review of use and reporting in healthcare research

BACKGROUND

Mediation analysis tests whether the relationship between two variables is explained by a third intermediate variable. We sought to describe the usage and reporting of mediation analysis with time-to-event outcomes in published healthcare research.

METHODS

A systematic search of Medline, Embase, and Web of Science was executed in December 2016 to identify applications of mediation analysis to healthcare research involving a clinically relevant time-to-event outcome. We summarized usage over time and reporting of important methodological characteristics.

RESULTS

We included 149 primary studies, published from 1997 to 2016. Most studies were published after 2011 (n = 110, 74%), and the annual number of studies nearly doubled in the last year (from n = 21 to n = 40). A traditional approach (causal steps or change in coefficient) was most commonly taken (n = 87, 58%), and the majority of studies (n = 114, 77%) used a Cox Proportional Hazards regression for the outcome. Few studies (n = 52, 35%) mentioned any of the assumptions or limitations fundamental to a causal interpretation of mediation analysis.

CONCLUSION

There is increasing use of mediation analysis with time-to-event outcomes. Current usage is limited by reliance on traditional methods and the Cox Proportional Hazards model, as well as low rates of reporting of underlying assumptions. There is a need for formal criteria to aid authors, reviewers, and readers reporting or appraising such studies.

Mechanism evaluation of a lifestyle intervention for patients with musculoskeletal pain who are overweight or obese: protocol for a causal mediation analysis

INTRODUCTION

Low back pain (LBP) and knee osteoarthritis (OA) are highly prevalent and disabling conditions that cause societal and economic impact worldwide. Two randomised controlled trials (RCTs) will evaluate the effectiveness of a multicomponent lifestyle intervention for patients with LBP and knee OA who are overweight or obese. The key targets of this intervention are to improve physical activity, modify diet and correct pain beliefs. These factors may explain how a lifestyle intervention exerts its effects on key patient-relevant outcomes: pain, disability and quality of life. The aim of this protocol is to describe a planned analysis of a mechanism evaluation for a lifestyle intervention for overweight or obese patients with LBP and knee OA.

METHODS AND ANALYSIS

Causal mediation analyses of 2 two-armed RCTs. Both trials are part of a cohort-multiple RCT, embedded in routine health service delivery. In each respective trial, 160 patients with LBP and 120 patients with knee OA waiting for orthopaedic consultation will be randomised to a lifestyle intervention, or to remain part of the original cohort. The intervention consists of education and advice about the benefits of weight loss and physical activity, and the Australian New South Wales Get Healthy Service. All outcome measures including patient characteristics, primary and alternative mediators, outcomes, and potential confounders will be measured at baseline (T0). The primary mediator, weight, will be measured at 6 months post randomisation; alternative mediators including diet, physical activity and pain beliefs will be measured at 6 weeks post randomisation. All outcomes (pain, disability and quality of life) will be measured at 6 months post randomisation. Data will be analysed using causal mediation analysis with sensitivity analyses for sequential ignorability. All mediation models were specified a priori before completing data collection and without prior knowledge about the effectiveness of the intervention.

ETHICS AND DISSEMINATION

The study is approved by the Hunter New England Health Human Research Ethics Committee (13/12/11/5.18) and the University of Newcastle Human Research Ethics Committee (H-2015-0043). The results will be disseminated in peer-reviewed journals and at scientific conferences.

TRIAL REGISTRATION NUMBER

ACTRN12615000490572 and ACTRN12615000478516; Pre-results.