Causal inference in cumulative risk assessment: The roles of directed acyclic graphs

Assessment of associations between inhaled formaldehyde and lymphohematopoietic cancer through the integration of epidemiological and toxicological evidence with biological plausibility

Formaldehyde is recognized as carcinogenic for the portal of entry sites, though conclusions are mixed regarding lymphohematopoietic (LHP) cancers. This systematic review assesses the likelihood of a causal relationship between formaldehyde and LHP cancers by integrating components recommended by NASEM. Four experimental rodent bioassays and 16 observational studies in humans were included following the implementation of the a priori protocol. All studies were assessed for risk of bias (RoB), and meta-analyses were conducted on epidemiological studies, followed by a structured assessment of causation based on GRADE and Bradford Hill. RoB analysis identified systemic limitations precluding confidence in the epidemiological evidence due to inadequate characterization of formaldehyde exposure and a failure to adequately adjust for confounders or effect modifiers, thus suggesting that effect estimates are likely to be impacted by systemic bias. Mixed findings were reported in individual studies; meta-analyses did not identify significant associations between formaldehyde inhalation (when measured as ever/never exposure) and LHP outcomes, with meta-SMRs ranging from 0.50 to 1.51, depending on LHP subtype. No associations with LHP-related lesions were reported in reliable animal bioassays. No biologically plausible explanation linking the inhalation of FA and LHP was identified, supported primarily by the lack of systemic distribution and in vivo genotoxicity. In conclusion, the inconsistent associations reported in a subset of the evidence were not considered causal when integrated with the totality of the epidemiological evidence, toxicological data, and considerations of biological plausibility. The impact of systemic biases identified herein could be quantitatively assessed to better inform causality and use in risk assessment.

Residential greenness and chronic obstructive pulmonary disease in a large cohort in southern China: Potential causal links, risk trajectories, and mediation pathways

INTRODUCTION

Residential greenness may influence COPD mortality, but the causal links, risk trajectories, and mediation pathways between them remain poorly understood.

OBJECTIVES

We aim to comprehensively identify the potential causal links, characterize the dynamic progression of hospitalization or posthospital risk, and quantify mediation effects between greenness and COPD.

METHODS

This study was conducted using a community-based cohort enrolling individuals aged ≥ 18 years in southern China from January 1, 2009 to December 31, 2015. Greenness was characterized by normalized difference vegetation index (NDVI) around participants' residential addresses. We applied doubly robust Cox proportional hazards model, multi-state model, and multiple mediation method, to investigate the potential causal links, risk trajectories among baseline, COPD hospitalization, first readmission due to COPD or COPD-related complications, and all-cause death, as well as the multiple mediation pathways (particulate matter [PM], temperature, body mass index [BMI] and physical activity) connecting greenness exposure to COPD mortality.

RESULTS

Our final analysis included 581,785 participants (52.52% female; average age: 48.36 [Standard Deviation (SD): 17.56]). Each interquartile range (IQR: 0.06) increase in NDVI was associated with a reduced COPD mortality risk, yielding a hazard ratio (HR) of 0.88 (95 % CI: 0.81, 0.96). Furthermore, we observed per IQR (0.04) increase in NDVI was inversely associated with the risk of multiple transitions (baseline - COPD hospitalization, baseline - death, and readmission - death risks), especially a declined risk of all-cause death after readmission (HR = 0.66 [95 %CI: 0.44, 0.99]). Within the observed association between greenness and COPD mortality, three mediators were identified, namely PM, temperature, and BMI (HR for the total indirect effect: 0.773 [95 % CI: 0.703, 0.851]), with PM showing the highest mediating effect.

CONCLUSIONS

Our findings revealed greenness may be a beneficial factor for COPD morbidity, prognosis, and mortality. This protective effect is primarily attributed to the reduction in PM concentration.

Potential causal links and mediation pathway between urban greenness and lung cancer mortality: Result from a large cohort (2009 to 2020)

Urban greenness, as a vital component of the urban environment, plays a critical role in mitigating the adverse effects of rapid urbanization and supporting urban sustainability. However, the causal links between urban greenness and lung cancer mortality and its potential causal pathway remain poorly understood. Based on a prospective community-based cohort with 581,785 adult participants in southern China, we applied a doubly robust Cox proportional hazard model to estimate the causal associations between urban greenness exposure and lung cancer mortality. A general multiple mediation analysis method was utilized to further assess the potential mediating roles of various factors including particulate matter (PM1, PM2.5-1, and PM10-2.5), temperature, physical activity, and body mass index (BMI). We observed that each interquartile range (IQR: 0.06) increment in greenness exposure was inversely associated with lung cancer mortality, with a hazard ratio (HR) of 0.89 (95 % CI: 0.83, 0.96). The relationship between greenness and lung cancer mortality might be partially mediated by particulate matter, temperature, and physical activity, yielding a total indirect effect of 0.826 (95 % CI: 0.769, 0.887) for each IQR increase in greenness exposure. Notably, the protective effect of greenness against lung cancer mortality could be achieved primarily by reducing the particulate matter concentration.

Causal Effects of Motor Control on Gait Kinematics After Orthopedic Surgery in Cerebral Palsy: A Machine-Learning Approach

Background

Altered motor control is common in cerebral palsy (CP). Understanding how altered motor control affects movement and treatment outcomes is important but challenging due to complex interactions with other neuromuscular impairments. While regression can be used to examine associations between impairments and movement, causal modeling provides a mathematical framework to specify assumed causal relationships, identify covariates that may introduce bias, and test model plausibility. The goal of this research was to quantify the causal effects of altered motor control and other impairments on gait, before and after single-event multi-level orthopedic surgery (SEMLS).

Methods

We evaluated the impact of SEMLS on change in Gait Deviation Index (ΔGDI) between gait analyses. We constructed our causal model with a Directed Acyclic Graph that included the assumed causal relationships between SEMLS, ΔGDI, baseline GDI (GDIpre), baseline neurologic and orthopedic impairments (Imppre), age, and surgical history. We identified the adjustment set to evaluate the causal effect of SEMLS on ΔGDI and the impact of Imppre on ΔGDI and GDIpre. We used Bayesian Additive Regression Trees (BART) and accumulated local effects to assess relative effects.

Results

We prospectively recruited a cohort of children with bilateral CP undergoing SEMLS (N = 55, 35 males, age: 10.5 ± 3.1 years) and identified a control cohort with bilateral CP who did not undergo SEMLS (N = 55, 30 males, age: 10.0 ± 3.4 years). There was a small positive causal effect of SEMLS on ΔGDI (1.70 GDI points). Altered motor control (i.e., dynamic and static motor control) and strength had strong effects on GDIpre, but minimal effects on ΔGDI. Spasticity and orthopedic impairments had minimal effects on GDIpre or ΔGDI.

Conclusion

Altered motor control did have a strong effect on GDIpre, indicating that these impairments do have a causal effect on a child's gait pattern, but minimal effect on expected changes in GDI after SEMLS. Heterogeneity in outcomes suggests there are other factors contributing to changes in gait. Identifying these factors and employing causal methods to examine the complex relationships between impairments and movement will be required to advance our understanding and care of children with CP.

Inadequate Vitamin C Intake and Intestinal Inflammation Are Associated with Multiple Micronutrient Deficiency in Young Children: Results from a Multi-Country Birth Cohort Study

Children living in resource-limited settings often suffer from multiple micronutrient deficiencies (MMD). However, there lacks evidence on the correlates of MMD in young children. We investigated the role of diets, water, sanitation and hygiene practice, enteric infections, and impaired gut health on MMD in children at 24 months of age using data from the multi-country MAL-ED birth cohort study. Co-existence of more than one micronutrient deficiency (e.g., anemia, iron, zinc, or retinol deficiency) was considered as MMD. We characterized intestinal inflammation by fecal concentrations of myeloperoxidase (MPO) and neopterin (NEO) measured in the non-diarrheal stool samples. Bayesian network analysis was applied to investigate the factors associated with MMD. A total of 1093 children were included in this analysis. Overall, 47.6% of the children had MMD, with the highest prevalence in Pakistan (90.1%) and lowest in Brazil (6.3%). MMD was inversely associated with the female sex [OR: 0.72, 95% CI: 0.54, 0.92]. A greater risk of MMD was associated with lower vitamin C intake [OR: 0.70, 95% CI: 0.48, 0.94] and increased fecal concentrations of MPO [OR: 1.31, 95% CI: 1.08, 1.51]. The study results imply the importance of effective strategies to ameliorate gut health and improve nutrient intake during the early years of life.

Risk of cardiovascular disease is driven by different combinations of environmental, medical and behavioral factors: Building a conceptual model for cumulative risk assessment

The U.S. Environmental Protection Agency (EPA) and National Institute of Environmental Health (NIEHS) held a workshop in 2012 to discuss principles and applications of cumulative risk assessment (CRA). The workshop organizers chose cardiovascular disease (CVD) as an example health outcome for which cumulative risk considerations could illuminate environmental and health management strategies. To guide discussions, we developed a series of conceptual models illustrating factors influencing CVD. The CVD conceptual model represents complex processes across varying space and time scales, different causal pathways, and multiple chemical and non-chemical stressors and factors. It includes causal influences of environmental exposures and lifestyle choices, in the context of genetics and medical factors. The representation of cumulative risk as a network of interrelated nodes and arrows helps define and organize the problem and available information, determine the scope and scale, and creates a platform for analysis. It provides an interface for discussing how different entities (e.g., environmental versus health-driven organizations) can work together on different parts of the problem, and facilitates relative risk ranking and management triage. Color coding is used to distinguish categories of stressors and possible oversight responsibility. This work informs guidelines for CRA planning and assessment of factor combinations affecting real-world risk.

Association between blood cadmium levels and the risk of osteopenia and osteoporosis in Korean post-menopausal women

We aimed to investigate the association between cadmium levels and the risk of osteopenia and osteoporosis in Korean post-menopausal women. There was a significant positive association between cadmium levels and the risk of osteopenia and osteoporosis, but further studies for dose response are required.

PURPOSE

Cadmium exposure can exert detrimental effects on bone health, particularly in post-menopausal women. However, previous studies have failed to report an association in Korean post-menopausal women. We aimed to investigate the association between cadmium levels and the risk of osteopenia and osteoporosis in Korean post-menopausal women.

METHODS

In total, 5432 participants from the 4th and 5th Korean National Health and Nutrition Examination Survey (KNHANES) were randomly sampled for measurements of heavy metal concentrations in the blood, bone mass density (BMD), and nutrient intake. We analyzed data for 1031 post-menopausal women ≥50 years of age. Blood cadmium levels were categorized into quartiles, and a multinomial logistic regression model was used for analysis.

RESULTS

There was a significant positive association between cadmium levels and the risk of osteopenia and osteoporosis, but the odds ratio (OR) at the 4th level was lower than that at the 3rd level (OR and 95% confidence interval (CI) for osteopenia: 2nd quartile: 1.24, 0.88-1.74; 3rd quartile: 3.22, 2.24-4.64; 4th quartile: 1.27, 0.87-1.85; P for trend <0.001; OR and 95% CI for osteoporosis: 2nd quartile: 1.54, 1.05-2.25; 3rd quartile: 3.63, 2.31-5.69; 4th quartile: 1.70, 1.03-2.81; P for trend <0.001). This trend was consistent in the sensitivity analysis.

CONCLUSION

Our findings suggest that there is an association between blood cadmium levels and the risk of osteopenia and osteoporosis in Korean post-menopausal women. However, further prospective studies are required to determine whether there is a dose-response relationship and address potential selection bias, especially in patients with femoral neck osteoporosis.

Intestinal permeability and inflammation mediate the association between nutrient density of complementary foods and biochemical measures of micronutrient status in young children: results from the MAL-ED study

BACKGROUND

Environmental enteric dysfunction (EED) is thought to increase the risk of micronutrient deficiencies, but few studies adjust for dietary intakes and systemic inflammation.

OBJECTIVE

We tested whether EED is associated with micronutrient deficiency risk independent of diet and systemic inflammation, and whether it mediates the relation between intake and micronutrient status.

METHODS

Using data from 1283 children in the MAL-ED (Etiology, Risk Factors, and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health) birth cohort we evaluated the risk of anemia, low retinol, zinc, and ferritin, and high transferrin receptor (TfR) at 15 mo. We characterized gut inflammation and permeability by myeloperoxidase (MPO), neopterin (NEO), and α-1-antitrypsin (AAT) concentrations from asymptomatic fecal samples averaged from 9 to 15 mo, and averaged the lactulose:mannitol ratio z-score (LMZ) at 9 and 15 mo. Nutrient intakes from complementary foods were quantified monthly from 9 to 15 mo and densities were averaged for analyses. α-1-Acid glycoprotein at 15 mo characterized systemic inflammation. Relations between variables were modeled using a Bayesian network.

RESULTS

A greater risk of anemia was associated with LMZ [1.15 (95% CI: 1.01, 1.31)] and MPO [1.16 (1.01, 1.34)]. A greater risk of low ferritin was associated with AAT [1.19 (1.03, 1.37)] and NEO [1.22 (1.04, 1.44)]. A greater risk of low retinol was associated with LMZ [1.24 (1.08, 1.45)]. However, MPO was associated with a lower risk of high transferrin receptor [0.86 (0.74, 0.98)], NEO with a lower risk of low retinol [0.75 (0.62, 0.89)], and AAT with a lower risk of low plasma zinc [0.83 (0.70, 0.99)]. Greater nutrient intake densities (vitamins A and B6, calcium, protein, and zinc) were negatively associated with EED. Inverse associations between nutrient densities and micronutrient deficiency largely disappeared after adjustment for EED, suggesting that EED mediates these associations.

CONCLUSIONS

EED is independently associated with an increased risk of low ferritin, low retinol, and anemia. Greater nutrient density from complementary foods may reduce EED, and the control of micronutrient deficiencies may require control of EED.

A Quantitative Source-to-Outcome Case Study To Demonstrate the Integration of Human Health and Ecological End Points Using the Aggregate Exposure Pathway and Adverse Outcome Pathway Frameworks

Exposure to environmental contaminants can lead to adverse outcomes in both human and nonhuman receptors. The Aggregate Exposure Pathway (AEP) and Adverse Outcome Pathway (AOP) frameworks can mechanistically inform cumulative risk assessment for human health and ecological end points by linking together environmental transport and transformation, external exposure, toxicokinetics, and toxicodynamics. This work presents a case study of a hypothetical contaminated site to demonstrate a quantitative approach for implementing the AEP framework and linking this framework to AOPs. We construct an AEP transport and transformation model and then quantify external exposure pathways for humans, fishes, and small herbivorous mammals at the hypothetical site. A Monte Carlo approach was used to address parameter variability. Source apportionment was quantified for each species, and published pharmacokinetic models were used to estimate internal target site exposure from external exposures. Published dose-response data for a multispecies AOP network were used to interpret AEP results in the context of species-specific effects. This work demonstrates (1) the construction, analysis, and application of a quantitative AEP model, (2) the utility of AEPs for organizing mechanistic exposure data and highlighting data gaps, and (3) the advantages provided by a source-to-outcome construct for leveraging exposure data and to aid transparency regarding assumptions.

Bayesian network-based framework for exposure-response study design and interpretation

Conventional environmental-health risk-assessment methods are often limited in their ability to account for uncertainty in contaminant exposure, chemical toxicity and resulting human health risk. Exposure levels and toxicity are both subject to significant measurement errors, and many predicted risks are well below those distinguishable from background incident rates in target populations. To address these issues methods are needed to characterize uncertainties in observations and inferences, including the ability to interpret the influence of improved measurements and larger datasets. Here we develop a Bayesian network (BN) model to quantify the joint effects of measurement errors and different sample sizes on an illustrative exposure-response system. Categorical variables are included in the network to describe measurement accuracies, actual and measured exposures, actual and measured response, and the true strength of the exposure-response relationship. Network scenarios are developed by fixing combinations of the exposure-response strength of relationship (none, medium or strong) and the accuracy of exposure and response measurements (low, high, perfect). Multiple cases are simulated for each scenario, corresponding to a synthetic exposure response study sampled from the known scenario population. A learn-from-cases algorithm is then used to assimilate the synthetic observations into an uninformed prior network, yielding updated probabilities for the strength of relationship. Ten replicate studies are simulated for each scenario and sample size, and results are presented for individual trials and their mean prediction. The model as parameterized yields little-to-no convergence when low accuracy measurements are used, though progressively faster convergence when employing high accuracy or perfect measurements. The inferences from the model are particularly efficient when the true strength of relationship is none or strong with smaller sample sizes. The tool developed in this study can help in the screening and design of exposure-response studies to better anticipate where such outcomes can occur under different levels of measurement error. It may also serve to inform methods of analysis for other network models that consider multiple streams of evidence from multiple studies of cumulative exposure and effects.

Epidemiology: a foundation of environmental decision making

Many epidemiologic studies are designed so they can be drawn upon to provide scientific evidence for evaluating hazards of environmental exposures, conducting quantitative assessments of risk, and informing decisions designed to reduce or eliminate harmful exposures. However, experimental animal studies are often relied upon for environmental and public health policy making despite the expanding body of observational epidemiologic studies that could inform the relationship between actual, as opposed to controlled, exposures and health effects. This paper provides historical examples of how epidemiology has informed decisions at the U.S. Environmental Protection Agency, discusses some challenges with using epidemiology to inform decision making, and highlights advances in the field that may help address these challenges and further the use of epidemiologic studies moving forward.

Exposure to polybrominated diphenyl ethers and phthalates in healthy men living in the greater Montreal area: A study of hormonal balance and semen quality

Studies investigating the associations between exposure of young men to polybrominated diphenyl ethers (PBDEs) or phthalates and hormone levels or semen quality have produced inconsistent results. Our goal was to investigate the association of exposure to PBDEs or phthalate metabolites with changes in markers of thyroid (TSH, free T3 and free T4) and reproductive function (sperm concentrations, motility, and quality; serum LH and testosterone) in 153 healthy young men from the greater Montreal area. Using covariate-adjusted models, we found that each 10-fold increase in BDE-47 was associated with lower TSH levels (-17.3%; 95% CI: -31.5, 0.0; p = 0.05). BDE-47 exposure was also associated with a decrease in sperm concentration (-19.7%; 95% CI: -36.8; 2.0; p = 0.07) and motility (-25.5%; 95% CI: -44.5, 0.1; p = 0.05). Trends towards decreases in these parameters were also observed in association with exposure to BDE-100 and the sum of BDE-47, -99, and -100 (∑₃BDEs). These associations were not accompanied by effects on sperm chromatin quality, as assessed with the HT-COMET assay. There were no substantial associations between urinary phthalate metabolite concentrations, either individually or grouped by molecular weight or parent compound, and sperm quality parameters; however, there was a positive association between elevated MECCP and free T4 (0.98; 95% CI: 0.02, 1.94; p = 0.05). Inverse associations between BDE-47 and ∑₃BDEs and free T3 and positive associations between MEHP and free T3 were stronger among individuals with BMI ≥ 25, suggesting that weight status may modify the effects of these endocrine disrupting chemicals.

Critical Care Resource Utilization and Outcomes of Children With Moderate Traumatic Brain Injury

OBJECTIVES

To characterize admission patterns, critical care resource utilization, and outcomes in moderate pediatric traumatic brain injury.

DESIGN

Retrospective cohort study.

SETTING

National Trauma Data Bank.

PATIENTS

Children under 18 years old with a diagnosis of moderate traumatic brain injury (admission Glasgow Coma Scale score of 9-13) in the National Trauma Data Bank between 2007 and 2014.

MEASUREMENT AND MAIN RESULTS

We examined clinical characteristics, critical care resource utilization, and discharge outcomes. Poor outcomes were defined as discharge to hospice, skilled nursing facility, long-term acute care, or death. We examined 20,010 patient records. Patients were 9 years old (interquartile range, 2-15 yr), male (64%) with isolated traumatic brain injury (81%), Glasgow Coma Scale score of 12, head Abbreviated Injury Scale score of 3, and Injury Severity Score of 10. Majority (34%) were admitted to nontrauma hospitals. Critical care utilization was 58.7% including 11.5% mechanical ventilation and 3.2% intracranial pressure monitoring. Compared to patients with Glasgow Coma Scale score of 13, admission Glasgow Coma Scale score of 9 was associated with greater critical care resource utilization, such as ICU admission (72% vs 50%), intracranial pressure monitoring (7% vs 1.8%), mechanical ventilation (21% vs 6%), and intracranial surgery (10% vs 5%). Most patients (70%) were discharged to home, but up to one third had poor outcomes. Older age group had a higher risk of poor outcomes (10-14 yr; adjusted relative risk, 1.32; 95% CI, 1.13-1.54; 15-17 yr; adjusted relative risk, 2.39; 95% CI, 2.12-2.70). Poor outcomes occurred with lower Glasgow Coma Scale (Glasgow Coma Scale score of 9 vs Glasgow Coma Scale score of 13: adjusted relative risk, 2.89; 95% CI, 2.47-3.38), higher Injury Severity Score (Injury Severity Score of ≥ 16 vs Injury Severity Score of < 9: adjusted relative risk, 8.10; 95% CI 6.27-10.45), and polytrauma (adjusted relative risk, 1.40; 95% CI, 1.22-1.61).

CONCLUSIONS

Critical care resources are used in more than half of all moderate pediatric traumatic brain injury, and many receive care at nontrauma hospitals. Up to one third of moderate pediatric traumatic brain injury have poor outcomes, risk factors for which include age greater than 10 years, lower admission Glasgow Coma Scale, higher Injury Severity Score, and polytrauma. There is urgent need to optimize triage, care, and outcomes in this vulnerable population.