Measurement error in environmental epidemiology and the shape of exposure-response curves

Model misspecification, measurement error, and apparent supralinearity in the concentration-response relationship between PM2.5 and mortality

A growing number of studies have produced results that suggest the shape of the concentration-response (C-R) relationship between PM2.5 exposure and mortality is "supralinear" such that incremental risk is higher at the lowest exposure levels than at the highest exposure levels. If the C-R function is in fact supralinear, then there may be significant health benefits associated with reductions in PM2.5 below the current US National Ambient Air Quality Standards (NAAQS), as each incremental tightening of the PM2.5 NAAQS would be expected to produce ever-greater reductions in mortality risk. In this paper we undertake a series of tests with simulated cohort data to examine whether there are alternative explanations for apparent supralinearity in PM2.5 C-R functions. Our results show that a linear C-R function for PM2.5 can falsely appear to be supralinear in a statistical estimation process for a variety of reasons, such as spatial variation in the composition of total PM2.5 mass, the presence of confounders that are correlated with PM2.5 exposure, and some types of measurement error in estimates of PM2.5 exposure. To the best of our knowledge, this is the first simulation-based study to examine alternative explanations for apparent supralinearity in C-R functions.

Long-term exposure to ambient air pollution and incident gout: A prospective cohort study in the UK Biobank

Gout is a chronic disorder characterized by the accumulation of uric acid in the body, leading to recurrent episodes of joint inflammation and pain. There remains a lack of studies investigating the association between long-term exposure to ambient air pollution and the incidence of gout. We conducted this prospective cohort study involving participants aged 38-70 from the UK Biobank who were enrolled in 2006-2010 and followed until 2023. Baseline residential concentrations of fine particulate matter (PM2.5), inhalable particulate matter (PM10), nitrogen dioxide (NO2) and nitrogen oxides (NOx) were predicted using land-use regression models. Cox proportional hazards models were employed to examine the relationship between air pollution and incident gout events. A total of 443,587 individuals were included in the analyses and a total of 6589 incident gout cases were identified over a follow-up of 6,130,439 person-years. There were significant associations between higher levels of air pollution and an increased incidence risk of gout. Higher risk of incident gout was associated with each interquartile range increase in concentrations of PM2.5 (hazard ratio:1.05, 95% confidence intervals: 1.02-1.09), PM10 (1.04, 1.00-1.07), NO2 (1.08, 1.05-1.12) and NOx (1.04, 1.02-1.07). The magnitude of associations was larger at higher concentrations. The association was more prominent among older adults, smokers, and individuals with lower and moderate physical activity. This prospective cohort study provides novel and compelling evidence of increased risk of incident gout associated with long-term air pollution exposures.

Effects of Ambient Air Pollutants on Hospital Admissions among Children Due to Asthma and Wheezing-Associated Lower Respiratory Infections in Mysore, India: A Time Series Study

Air pollutants are known to trigger asthma and wheezing-associated lower respiratory infections in children, but evidence regarding their effect on hospital admissions in India is limited. We conducted a time-series study over a period of five years to assess the role of ambient air pollutants in daily asthma-related hospital admissions in children in Mysore, India. Daily asthma and wheeze (associated with lower respiratory infections) admissions were modelled using a generalised additive model (GAM) to examine the non-linear effects and generalised linear models (GLM) for linear effects, if any. Models were adjusted by day of the week and lag days, with smooth terms for time, maximum temperature, and relative humidity, and they were stratified by sex and age group. Of the 362 children admitted, more than 50% were boys, and the mean age was 5.34 years (±4.66). The GAMs showed non-linear associations between NO2, PM2.5, and NH3. For example, a 10 µgm-3 (or 10%) increase in NO2 increased admissions by 2.42. These non-linear effects were more pronounced in boys. A linear effect was detected for PM10 with a relative risk (95% CI) of 1.028, 1.013, and 1.043 with admission. Further research is needed to explore whether these findings can be replicated in different cities in India. Air pollution needs to be controlled, and policies that focus on lower cut-off levels for vulnerable populations are necessary.

Heat-mortality relationship in North Carolina: Comparison using different exposure methods

BACKGROUND

Many studies have explored the heat-mortality relationship; however, comparability of results is hindered by the studies' use of different exposure methods.

OBJECTIVE

This study evaluated different methods for estimating exposure to temperature using individual-level data and examined the impacts on the heat-mortality relationship.

METHODS

We calculated different temperature exposures for each individual death by using a modeled, gridded temperature dataset and a monitoring station dataset in North Carolina for 2000-2016. We considered individual-level vs. county-level averages and measured vs. modeled temperature data. A case-crossover analysis was conducted to examine the heat-mortality risk under different exposure methods.

RESULTS

The minimum mortality temperature (MMT) (i.e., the temperature with the lowest mortality rate) for the monitoring station dataset was 23.87 °C and 22.67 °C (individual monitor and county average, respectively), whereas for the modeled temperature dataset the MMT was 19.46 °C and 19.61 °C (individual and county, respectively). We found higher heat-mortality risk while using temperature exposure estimated from monitoring stations compared to risk based on exposure using the modeled temperature dataset. Individual-aggregated monitoring station temperature exposure resulted in higher heat mortality risk (odds ratio (95% CI): 2.24 (95% CI: 2.21, 2.27)) for a relative temperature change comparing the 99th and 90th temperature percentiles, while modeled temperature exposure resulted in lower odds ratio of 1.27 (95% CI: 1.25, 1.29).

SIGNIFICANCE

Our findings indicate that using different temperature exposure methods can result in different temperature-mortality risk. The impact of using various exposure methods should be considered in planning health policies related to high temperatures, including under climate change. IMPACT STATEMENT: (1) We estimated the heat-mortality association using different methods to estimate exposure to temperature. (2) The mean temperature value among different exposure methods were similar although lower for the modeled data, however, use of the monitoring station temperature dataset resulted in higher heat-mortality risk than the modeled temperature dataset. (3) Differences in mortality risk from heat by urbanicity varies depending on the method used to estimate temperature exposure.

Systematic review of the association between long-term exposure to fine particulate matter and mortality

We used a transparent systematic review framework based on best practices for evaluating study quality and integrating evidence to conduct a review of the available epidemiology studies evaluating associations between long-term exposure to ambient concentrations of PM_2.5 and mortality (all-cause and non-accidental) conducted in North America. We found that while there is some consistency across studies for reporting positive associations, these associations are weak and several important methodological issues have led to uncertainties with regard to the evidence from these studies, including potential confounding by measured and unmeasured factors, exposue measurement error, and model misspecification. These uncertainties provide a plausible, alternative explanation to causality for the weakly positive findings across studies. Using a causality framework that incorporates best practices for making causal determinations, we concluded that the evidence for a causal relationship between long-term exposure to ambient PM_2.5 concentrations and mortality from these studies is inadequate.

A simulation-based assessment of the ability to detect thresholds in chronic risk concentration-response functions in the presence of exposure measurement error

An important question when setting appropriate air quality standards for fine particulate matter (PM_2.5) is whether there exists a “threshold” in the concentration-response (C-R) function, such that PM_2.5 levels below this threshold are not expected to produce adverse health effects. We hypothesize that measurement error may affect the recognition of a threshold in long-term cohort epidemiological studies. This study conducts what is, to the best of our knowledge, the first simulation of the effects of measurement error on the statistical models commonly employed in long-term cohort studies. We test the degree to which classical-type measurement error, such as differences between the true population-weighted exposure level to a pollutant and the observed measures of that pollutant, affects the ability to statistically detect a C-R threshold. The results demonstrate that measurement error can obscure the existence of a threshold in a cohort study’s C-R function for health risks from chronic exposures. With increased measurement error the ability to statistically detect a C-R threshold decreases, and both the estimated location of the C-R threshold and the estimated hazard ratio associated with PM_2.5 are attenuated. This result has clear implications for determining appropriate air quality standards for pollutants.

A review and analysis of personal and ambient PM2.5 measurements: Implications for epidemiology studies

BACKGROUND

In epidemiology studies, ambient measurements of PM_2.5 are often used as surrogates for personal exposures. However, it is unclear the degree to which ambient PM_2.5 reflects personal exposures.

OBJECTIVE

In order to examine potential sources of bias in epidemiology studies, we conducted a review and meta-analysis of studies to determine the extent to which short-term measurements of ambient PM_2.5 levels are related to short-term measurements of personal PM_2.5 levels.

METHODS

We conducted a literature search of studies reporting both personal and ambient measurements of PM_2.5 published in the last 10 years (2009-2019) and incorporated studies published prior to 2009 from reviews.

RESULTS

Seventy-one studies were identified. Based on 17 studies reporting slopes, a meta-analysis revealed an overall slope of 0.56 μg/m³ (95% CI: [0.39, 0.73]) personal PM_2.5 per μg/m³ increase in ambient PM_2.5. Slopes for summer months were higher (slope = 0.73, 95% CI: [0.64, 0.81]) than for winter (slope = 0.46, 95% CI: [0.36, 0.57]). Based on 44 studies reporting correlations, we calculated an overall personal-ambient PM_2.5 correlation of 0.63 (95% CI: [0.55, 0.71]). Correlations were stronger in studies conducted in Canada (r = 0.86, 95% CI: [0.67, 0.94]) compared to the USA (r = 0.60, 95% CI: [0.49, 0.70]) and China (r = 0.60, 95% CI: [0.46, 0.71]). Correlations also were stronger in urban areas (r = 0.53, 95% CI: [0.43, 0.62]) vs. suburban areas (r = 0.36, 95% CI: [0.21, 0.49]).

SIGNIFICANCE

Our results suggest a large degree of variability in the personal-ambient PM_2.5 association and the potential for exposure misclassification and measurement error in PM_2.5 epidemiology studies.

Projections of excess cardiovascular mortality related to temperature under different climate change scenarios and regionalized climate model simulations in Brazilian cities

BACKGROUND

There is an urgent need for more information about the climate change impact on health in order to strengthen the commitment to tackle climate change. However, few studies have quantified the health impact of climate change in Brazil and in the Latin America region. In this paper, we projected the impacts of temperature on cardiovascular (CVD) mortality according to two climate change scenarios and two regionalized climate model simulations in Brazilian cities.

METHODS

We estimated the temperature-CVD mortality relationship in 21 Brazilian cities, using distributed lag non-linear models in a two-stage time-series analysis. We combined the observed exposure-response functions with the daily temperature projected under two representative concentration pathways (RCP), RCP8.5 and RCP4.5, and two regionalized climate model simulations, Eta-HadGEM2-ES and Eta-MIROC5.

RESULTS

We observed a trend of reduction in mortality related to low temperatures and a trend of increase in mortality related to high temperatures, according to all the investigated models and scenarios. In most places, the increase in mortality related to high temperatures outweighed the reduction in mortality related to low temperatures, causing a net increase in the excess temperature-related mortality. These trends were steeper according to the higher emission scenario, RCP8.5, and to the Eta-HadGEM2-ES model. According to RCP8.5, our projections suggested that the temperature-related mortality fractions in 2090-99 compared to 2010-2019 would increase by 8.6% and 1.7%, under Eta-HadGEM2-ES and Eta-MIROC5, respectively. According to RCP4.5, these values would be 0.7% and -0.6%.

CONCLUSIONS

For the same climate model, we observed a greater increase trend in temperature-CVD mortality according to RCP8.5, highlighting a greater health impact associated with the higher emission scenario. Our results may be useful to support public policies and strategies for mitigation of and adaptation to climate change, particularly in the health sector.

Epigenetic Biomarkers of Prenatal Tobacco Smoke Exposure Are Associated with Gene Deletions in Childhood Acute Lymphoblastic Leukemia

BACKGROUND

Parental smoking is implicated in the etiology of acute lymphoblastic leukemia (ALL), the most common childhood cancer. We recently reported an association between an epigenetic biomarker of early-life tobacco smoke exposure at the AHRR gene and increased frequency of somatic gene deletions among ALL cases.

METHODS

Here, we further assess this association using two epigenetic biomarkers for maternal smoking during pregnancy-DNA methylation at AHRR CpG cg05575921 and a recently established polyepigenetic smoking score-in an expanded set of 482 B-cell ALL (B-ALL) cases in the California Childhood Leukemia Study with available Illumina 450K or MethylationEPIC array data. Multivariable Poisson regression models were used to test the associations between the epigenetic biomarkers and gene deletion numbers.

RESULTS

We found an association between DNA methylation at AHRR CpG cg05575921 and deletion number among 284 childhood B-ALL cases with MethylationEPIC array data, with a ratio of means (RM) of 1.31 [95% confidence interval (CI), 1.02-1.69] for each 0.1 β value reduction in DNA methylation, an effect size similar to our previous report in an independent set of 198 B-ALL cases with 450K array data [meta-analysis summary RM (sRM) = 1.32; 95% CI, 1.10-1.57]. The polyepigenetic smoking score was positively associated with gene deletion frequency among all 482 B-ALL cases (sRM = 1.31 for each 4-unit increase in score; 95% CI, 1.09-1.57).

CONCLUSIONS

We provide further evidence that prenatal tobacco-smoke exposure may influence the generation of somatic copy-number deletions in childhood B-ALL.

IMPACT

Analyses of deletion breakpoint sequences are required to further understand the mutagenic effects of tobacco smoke in childhood ALL.

Historical perspective on the role of cell proliferation in carcinogenesis for DNA-reactive and non-DNA-reactive carcinogens: Arsenic as an example

Our understanding of the etiology of cancer has developed significantly over the past fifty years, beginning with a single-hit linear no-threshold (LNT) conceptual model based on early studies conducted in Drosophila. Over the past several decades, multiple lines of evidence have accumulated to support a contemporary model of chemical carcinogenesis: a multi-hit model involving a prolonged stress environment that over time may drive the mutation of multiple cells into an injured state that ultimately could lead to uncontrolled proliferation via clonal expansion of mutation-carrying daughter cells. Arsenic carcinogenicity offers a useful case study for further exploration of advanced conceptual models for chemical carcinogenesis. A threshold for arsenic carcinogenicity is supported by its mode of action, characterized by repeating cycles of cytotoxicity and cellular regeneration. Furthermore, preliminary meta-analyses of epidemiology dose-response data for inorganic arsenic (iAs) and bladder cancer, correlated to dose-response data measured in vitro, support a threshold of effect in humans on the order of 50-100 μg/L in drinking water. In light of recent developments in our understanding of cancer etiology, we urge strong consideration of the existing mode-of-action evidence supporting a threshold of effect for arsenic carcinogenicity, as well as consideration of the potential methodological pitfalls in evaluating epidemiology dose-response data that could potentially bias in the direction of low-dose linearity.

Thermal Discomfort Levels, Building Design Concepts, and Some Heat Mitigation Strategies in Low-Income Communities of a South Asian City

Heat stress provokes thermal discomfort to people living in semiarid and arid climates. This study evaluates thermal discomfort levels, building design concepts, and some heat mitigation strategies in low-income neighborhoods of Faisalabad, Pakistan. The outdoor and indoor weather data are collected from April to August 2016 using a weather station installed ad hoc in urban settings, and the 52 houses of the five low-income participating communities living in congested and less environment-friendly areas of Faisalabad. The discomfort index values, related to the building design concepts, including (i) house orientation to sunlight and (ii) house ventilation, are calculated from outdoor and indoor dry-bulb and wet-bulb temperatures. Our results show that although June was the hottest month of summer 2016, based on the monthly mean temperature of the Faisalabad region, the month of May produced the highest discomfort levels, which were higher in houses exposed to sunlight and without ventilation. The study also identifies some popular heat mitigation strategies adopted by the five participating low-income communities during various heat-related health complaints. The strategies are gender-biased and have medical, cultural/customary backgrounds. For example, about 52% of the males and 28% of the females drank more water during dehydration, diarrhea, and eye infection. Over 11% and 19% of the males and females, respectively, moved to cooler places during fever. About 43% of the males and 51% of the females took water showers and rested to combat flu (runny nose), headache, and nosebleed. The people did not know how to cure muscular fatigue, skin allergy (from a type of Milia), and mild temperature. Planting trees in an area and developing open parks with greenery and thick canopy trees can be beneficial for neighborhoods resembling those evaluated in this study.

Association between past exposure to fine particulate matter (PM2.5) and peptic ulcer: A cross-sectional study in eastern China

Ambient fine particulate matter (particle diameter < 2.5 μm, or PM_2.5) is a major public health concern in China. Exposure to PM_2.5 has been associated with a wide range of adverse health outcomes. The current study aimed to estimate the association between exposure to PM_2.5 and the risk of peptic ulcer diseases (PUDs). We conducted a hospital-based cross-sectional study of seven major cities in Zhejiang Province, China (combined population > 57 million people), which included a total of 647,092 subjects who underwent gastroscopy examination (86,852 subjects were diagnosed with PUDs) recorded in 13 large hospitals from 2014 to 2018. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to evaluate the relationship between PM_2.5 and PUDs, including duodenal ulcers (DUs) and gastric ulcers (GUs). The overall estimated OR (95% CI) associated with every 10-μg/m³ increase in the 1-month average PM_2.5 before the detection of PUDs was 1.050 (95% CI: 1.038, 1.063). The association between PM_2.5 concentration and the prevalence of PUDs tended to be attenuated but remained significant when considering different exposure periods (OR = 1.030, 95% CI = 1.018-1.043 for the 3-month moving average; OR = 1.020, 95% CI = 1.005-1.037 for the 6-month moving average). Stronger associations were observed for DUs than GUs. The observed positive association of PM_2.5 exposure with PUDs remained significant in the two-pollutant models after adjusting for other air pollutants. Our findings could provide scientific evidence for a more general adverse role of air pollution on PUDs.

Intake of Alcohol and Tea and Risk of Nasopharyngeal Carcinoma: A Population-Based Case-Control Study in Southern China

BACKGROUND

The potential effect of alcohol or tea intake on the risk of nasopharyngeal carcinoma (NPC) remains controversial.

METHODS

In a population-based case-control study in southern China, we assessed alcohol or tea intake from 2,441 histopathologically confirmed NPC cases and 2,546 controls. We calculated mean daily ethanol (g/day) and tea intake (mL/day). Fully adjusted ORs with 95% confidence intervals (CI) were estimated using logistic regression; potential dose-response trends were evaluated using restricted cubic spline analysis.

RESULTS

Compared with nondrinkers, no significantly increased NPC risk in men was observed among current alcohol drinkers overall (OR, 1.08; 95% CI, 0.93-1.25), nor among current heavy drinkers (OR for ≥90 g/day ethanol vs. none, 1.32; 95% CI, 0.95-1.84) or former alcohol drinkers. Current tea drinking was associated with a decreased NPC risk (OR, 0.73; 95% CI, 0.64-0.84). Compared with never drinkers, those with the low first three quintiles of mean daily current intake of tea were at significantly lower NPC risk (OR, 0.53, 0.68, and 0.65, respectively), but not significant for the next two quintiles. Current daily tea intake had a significant nonlinear dose-response relation with NPC risk.

CONCLUSIONS

Our study suggests no significant association between alcohol and NPC risk. Tea drinking may moderately reduce NPC risk, but the lack of a monotonic dose-response association complicates causal inference.

IMPACT

Tea drinking might be a healthy habit for preventing NPC. More studies on biological mechanisms that may link tea with NPC risk are needed.

Temperature and cardiovascular mortality in Rio de Janeiro, Brazil: effect modification by individual-level and neighbourhood-level factors

BACKGROUND

Many factors related to susceptibility or vulnerability to temperature effects on mortality have been proposed in the literature. However, there is limited evidence of effect modification by some individual-level factors such as occupation, colour/race, education level and community-level factors. We investigated the effect modification of the temperature-cardiovascular mortality relationship by individual-level and neighbourhood-level factors in the city of Rio de Janeiro, Brazil.

METHODS

We used a case-crossover study to estimate the total effect of temperature on cardiovascular mortality in Rio de Janeiro between 2001 and 2018, and the effect modification by individual-level and neighbourhood-level factors. Individual-level factors included sex, age, colour/race, education, and place of death. Neighbourhood-level characteristics included social development index (SDI), income, electricity consumption and demographic change. We used conditional Poisson regression models combined with distributed lag non-linear models, adjusted for humidity and public holidays.

RESULTS

Our results suggest a higher vulnerability to high temperatures among the elderly, women, non-hospitalised deaths, and people with a lower education level. Vulnerability to low temperatures was higher among the elderly, men, non-white people, and for primary education level. As for neighbourhood-level factors, we identified greater vulnerability to low and high temperatures in places with lower SDI, lower income, lower consumption of electricity, and higher demographic growth.

CONCLUSION

The effects of temperature on cardiovascular disease mortality in Rio de Janeiro vary according to individual-level and neighbourhood-level factors. These findings are valuable to inform policymakers about the most vulnerable groups and places, in order to develop more effective and equitable public policies.

Implications of nonlinearity, confounding, and interactions for estimating exposure concentration-response functions in quantitative risk analysis

Recent advances in understanding of biological mechanisms and adverse outcome pathways for many exposure-related diseases show that certain common mechanisms involve thresholds and nonlinearities in biological exposure concentration-response (C-R) functions. These range from ultrasensitive molecular switches in signaling pathways, to assembly and activation of inflammasomes, to rupture of lysosomes and pyroptosis of cells. Realistic dose-response modeling and risk analysis must confront the reality of nonlinear C-R functions. This paper reviews several challenges for traditional statistical regression modeling of C-R functions with thresholds and nonlinearities, together with methods for overcoming them. Statistically significantly positive exposure-response regression coefficients can arise from many non-causal sources such as model specification errors, incompletely controlled confounding, exposure estimation errors, attribution of interactions to factors, associations among explanatory variables, or coincident historical trends. If so, the unadjusted regression coefficients do not necessarily predict how or whether reducing exposure would reduce risk. We discuss statistical options for controlling for such threats, and advocate causal Bayesian networks and dynamic simulation models as potentially valuable complements to nonparametric regression modeling for assessing causally interpretable nonlinear C-R functions and understanding how time patterns of exposures affect risk. We conclude that these approaches are promising for extending the great advances made in statistical C-R modeling methods in recent decades to clarify how to design regulations that are more causally effective in protecting human health.

Accounting for measurement error to assess the effect of air pollution on omic signals

Studies on the effects of air pollution and more generally environmental exposures on health require measurements of pollutants, which are affected by measurement error. This is a cause of bias in the estimation of parameters relevant to the study and can lead to inaccurate conclusions when evaluating associations among pollutants, disease risk and biomarkers. Although the presence of measurement error in such studies has been recognized as a potential problem, it is rarely considered in applications and practical solutions are still lacking. In this work, we formulate Bayesian measurement error models and apply them to study the link between air pollution and omic signals. The data we use stem from the "Oxford Street II Study", a randomized crossover trial in which 60 volunteers walked for two hours in a traffic-free area (Hyde Park) and in a busy shopping street (Oxford Street) of London. Metabolomic measurements were made in each individual as well as air pollution measurements, in order to investigate the association between short-term exposure to traffic related air pollution and perturbation of metabolic pathways. We implemented error-corrected models in a classical framework and used the flexibility of Bayesian hierarchical models to account for dependencies among omic signals, as well as among different pollutants. Models were implemented using traditional Markov Chain Monte Carlo (MCMC) simulative methods as well as integrated Laplace approximation. The inclusion of a classical measurement error term resulted in variable estimates of the association between omic signals and traffic related air pollution measurements, where the direction of the bias was not predictable a priori. The models were successful in including and accounting for different correlation structures, both among omic signals and among different pollutant exposures. In general, more associations were identified when the correlation among omics and among pollutants were modeled, and their number increased when a measurement error term was additionally included in the multivariate models (particularly for the associations between metabolomics and NO2).

The effect of ambient temperature on cardiovascular mortality in 27 Brazilian cities

BACKGROUND

There is limited evidence on the relationship between temperature and cardiovascular mortality in middle and low-income countries, particularly in Latin America. In this study, we investigated the total effect of temperature on cardiovascular mortality in 27 Brazilian cities, and the effect modification by geographic, socioeconomic, demographic and infrastructure characteristics within cities.

METHODS

In the city-specific analysis, we used time-series analyses to estimate the relationship between mean temperature and daily cardiovascular mortality using quasi-Poisson generalized linear models combined with distributed lag non-linear models. In the second stage, a meta-analysis was used to pool the effects of temperature on cardiovascular mortality for Brazil and its five regions (Central-West, North, Northeast, South, and Southeast). We used a meta-regression to examine the effect modification of city-specific geographic, socioeconomic, demographic and infrastructure-related variables.

RESULTS

The risks associated with temperature varied across the locations. Higher cardiovascular mortality was associated with low and high temperatures in most of the cities, Brazil and the Central-West, North, South, and Southeast regions. The overall relative risk (RR) for Brazil was 1.26 (95% confidence interval [CI]: 1.17-1.35) for the 1st percentile of temperature and 1.07 (95% CI: 1.01-1.13) for the 99th percentile of temperature versus the 79th percentile (27.7 °C), where RR was lowest. The temperature range was the variable that best explained the variation in effects among the cities, with greater effects in locations having a broader range.

CONCLUSIONS

The results indicate effects of low and high temperatures on the risk of cardiovascular mortality in most of Brazil's capital cities, besides a pooled effect for Brazil and the Central-West, North, South, and Southeast regions. These findings can help inform public policies addressing the health impact of temperature extremes, especially in the context of climate change.

Long-term effect of fine particulate matter on hospitalization with dementia

BACKGROUND

New evidence suggests that particulate matter less than 2.5 μm in diameter (PM_2.5) is associated with late-onset dementia (LOD). However, epidemiological studies for the entire population are lacking.

METHODS

We analyzed approximately 94 million follow-up records from fee-for-service Medicare records for 13 million Medicare beneficiaries residing in the southeastern United States (U.S.) from 2000 to 2013. We used spatially and temporally continuous PM_2.5 exposure data. To account for time-varying PM_2.5 levels, we applied an Andersen-Gill counting process proportional hazard model; we stratified our analyses by subtype of dementia and level of urbanization of residence.

RESULTS

During a median follow-up of 6 years, 1,409,599 hospitalizations with dementia occurred. The adjusted hazard ratio (HR) of hospitalization with dementia was 1.049 (95% confidence interval [CI], 1.048 to 1.051) per 1 μg/m³ increase in annual PM_2.5. The hazard ratio for vascular dementia was higher (HR, 1.086; 95% CI, 1.082 to 1.090). In large, the magnitude of the effect grew as the level of urbanization increased (HR, 1.036; 95% CI, 1.031 to 1.041 in rural areas versus HR, 1.052; 95% CI, 1.050 to 1.054 in metropolitan areas).

CONCLUSIONS

Long-term exposure to higher PM_2.5 was associated with increased hospitalizations with dementia.

Methods to account for uncertainties in exposure assessment in studies of environmental exposures

BACKGROUND

Accurate exposure estimation in environmental epidemiological studies is crucial for health risk assessment. Failure to account for uncertainties in exposure estimation could lead to biased results in exposure-response analyses. Assessment of the effects of uncertainties in exposure estimation on risk estimates received a lot of attention in radiation epidemiology and in several studies of diet and air pollution. The objective of this narrative review is to examine the commonly used statistical approaches to account for exposure estimation errors in risk analyses and to suggest how each could be applied in environmental epidemiological studies.

MAIN TEXT

We review two main error types in estimating exposures in epidemiological studies: shared and unshared errors and their subtypes. We describe the four main statistical approaches to adjust for exposure estimation uncertainties (regression calibration, simulation-extrapolation, Monte Carlo maximum likelihood and Bayesian model averaging) along with examples to give readers better understanding of their advantages and limitations. We also explain the advantages of using a 2-dimensional Monte-Carlo (2DMC) simulation method to quantify the effect of uncertainties in exposure estimates using full-likelihood methods. For exposures that are estimated independently between subjects and are more likely to introduce unshared errors, regression calibration and SIMEX methods are able to adequately account for exposure uncertainties in risk analyses. When an uncalibrated measuring device is used or estimation parameters with uncertain mean values are applied to a group of people, shared errors could potentially be large. In this case, Monte Carlo maximum likelihood and Bayesian model averaging methods based on estimates of exposure from the 2DMC simulations would work well. The majority of reviewed studies show relatively moderate changes (within 100%) in risk estimates after accounting for uncertainties in exposure estimates, except for the two studies which doubled/tripled naïve estimates.

CONCLUSIONS

In this paper, we demonstrate various statistical methods to account for uncertain exposure estimates in risk analyses. The differences in the results of various adjustment methods could be due to various error structures in datasets and whether or not a proper statistical method was applied. Epidemiological studies of environmental exposures should include exposure-response analyses accounting for uncertainties in exposure estimates.

Risk Analysis Implications of Dose-Response Thresholds for NLRP3 Inflammasome-Mediated Diseases: Respirable Crystalline Silica and Lung Cancer as an Example

Chronic inflammation mediates an extraordinarily wide range of diseases. Recent progress in understanding intracellular inflammasome assembly, priming, activation, cytokine signaling, and interactions with mitochondrial reactive oxygen species, lysosome disruption, cell death, and prion-like polymerization and spread of inflammasomes among cells, has potentially profound implications for dose-response modeling. This article discusses mechanisms of exposure concentration and duration thresholds for NOD-like receptor protein 3 (NLRP3)-mediated inflammatory responses and develops a simple biomathematical model of the onset of exposure-related tissue-level chronic inflammation and resulting disease risks, focusing on respirable crystalline silica (RCS) and lung cancer risk as an example. An inflammation-mediated 2-stage clonal expansion model of RCS-induced lung cancer is proposed that explains why relatively low estimated concentrations of RCS (eg, <1 mg/m³) do not increase lung cancer risk and why even high occupational concentrations increase risk only modestly (typically relative risk <2). The model of chronic inflammation implies a dose-response threshold for excess cancer risk, in contrast to traditional linear-no-threshold assumptions. If this implication is correct, then concentrations of crystalline silica (or amphibole asbestos fibers, or other environmental challenges that act via the NLRP3 inflammasome) below the threshold do not cause chronic inflammation and resulting elevated risks of inflammation-mediated diseases.

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.

Estimation of personal exposure to fine particles (PM2.5) of ambient origin for healthy adults in Hong Kong

Personal exposure and ambient fine particles (PM_2.5) measurements for 13 adult subjects (ages 19-57) were conducted in Hong Kong between April 2014 and June 2015. Six to 21 personal samples (mean = 19) per subject were obtained throughout the study period. Samples were analyzed for mass by gravimetric analysis, and 19 elements (from Na to Pb) were analyzed using X-Ray Fluorescence. Higher subject-specific correlations between personal and ambient sulfur (r_s = 0.92; p < 0.001) were found as compared to PM_2.5 mass (r_s = 0.79; p < 0.001) and other elements (0.06 < r_s < 0.86). Personal vs. ambient sulfur regression yielded an average exposure factor (F_pex) of 0.73 ± 0.02, supporting the use of sulfur as a surrogate to estimate personal exposure to PM_2.5 of ambient origin (E_a). E_a accounted for 41-82% and 57-73% of total personal PM_2.5 exposures (P) by season and by subject, respectively. The importance of both E_a and non-ambient exposures (E_na, 11.2 ± 5.6 μg/m³; 32.5 ± 10.9%) are noted. Mixed-effects models were applied to estimate the relationships between ambient PM_2.5 concentrations and their corresponding exposure variables (E_a, P). Higher correlations for E_a (0.90; p < 0.001) than for P (0.58; p < 0.01) were found. A calibration coefficient < 1 suggests an attenuation of 22% (ranging 16-28%) of the true effect estimates when using average ambient concentrations at central monitoring stations as surrogates for E_a. Stationary ambient data can be used to assess population exposure only if PM exposure is dominated by E_a.

Modernizing the Bradford Hill criteria for assessing causal relationships in observational data

Perhaps no other topic in risk analysis is more difficult, more controversial, or more important to risk management policy analysts and decision-makers than how to draw valid, correctly qualified causal conclusions from observational data. Statistical methods can readily quantify associations between observed variables using measures such as relative risk (RR) ratios, odds ratios (OR), slope coefficients for exposure or treatment variables in regression models, and quantities derived from these measures. Textbooks of epidemiology explain how to calculate population attributable fractions, attributable risks, burden-of-disease estimates, and probabilities of causation from relative risk (RR) ratios. Despite their suggestive names, these association-based measures have no necessary connection to causation if the associations on which they are based arise from bias, confounding, p-hacking, coincident historical trends, or other noncausal sources. But policy analysts and decision makers need something more: trustworthy predictions - and, later, evaluations - of the changes in outcomes caused by changes in policy variables. This concept of manipulative causation differs from the more familiar concepts of associational and attributive causation most widely used in epidemiology. Drawing on modern literature on causal discovery and inference principles and algorithms for drawing limited but useful causal conclusions from observational data, we propose seven criteria for assessing consistency of data with a manipulative causal exposure-response relationship - mutual information, directed dependence, internal and external consistency, coherent causal explanation of biological plausibility, causal mediation confirmation, and refutation of non-causal explanations - and discuss to what extent it is now possible to automate discovery of manipulative causal dependencies and quantification of causal effects from observational data. We compare our proposed principles for causal discovery and inference to the traditional Bradford Hill considerations from 1965. Understanding how old and new principles are related can clarify and enrich both.

Effects of exposure estimation errors on estimated exposure-response relations for PM2.5

Associations between fine particulate matter (PM2.5) exposure concentrations and a wide variety of undesirable outcomes, from autism and auto theft to elderly mortality, suicide, and violent crime, have been widely reported. Influential articles have argued that reducing National Ambient Air Quality Standards for PM2.5 is desirable to reduce these outcomes. Yet, other studies have found that reducing black smoke and other particulate matter by as much as 70% and dozens of micrograms per cubic meter has not detectably affected all-cause mortality rates even after decades, despite strong, statistically significant positive exposure concentration-response (C-R) associations between them. This paper examines whether this disconnect between association and causation might be explained in part by ignored estimation errors in estimated exposure concentrations. We use EPA air quality monitor data from the Los Angeles area of California to examine the shapes of estimated C-R functions for PM2.5 when the true C-R functions are assumed to be step functions with well-defined response thresholds. The estimated C-R functions mistakenly show risk as smoothly increasing with concentrations even well below the response thresholds, thus incorrectly predicting substantial risk reductions from reductions in concentrations that do not affect health risks. We conclude that ignored estimation errors obscure the shapes of true C-R functions, including possible thresholds, possibly leading to unrealistic predictions of the changes in risk caused by changing exposures. Instead of estimating improvements in public health per unit reduction (e.g., per 10 µg/m³ decrease) in average PM2.5 concentrations, it may be essential to consider how interventions change the distributions of exposure concentrations.

Towards establishing evidence-based guidelines on maximum indoor temperatures during hot weather in temperate continental climates

Rising environmental temperatures represent a major threat to human health. The activation of heat advisories using evidence-based thresholds for high-risk outdoor ambient temperatures have been shown to be an effective strategy to save lives during hot weather. However, although the relationship between weather and human health has been widely defined by outdoor temperature, corresponding increases in indoor temperature during heat events can also be harmful to health especially in vulnerable populations. In this review, we discuss our current understanding of the relationship between outdoor temperature and human health and examine how human health can also be adversely influenced by high indoor temperatures during heat events. Our assessment of the existing literature revealed a high degree of variability in what can be considered an acceptable indoor temperature because there are differences in how different groups of people may respond physiologically and behaviorally to the same living environment. Finally, we demonstrate that both non-physiological (e.g., geographical location, urban density, building design) and physiological (e.g., sex, age, fitness, state of health) factors must be considered when defining an indoor temperature threshold for preserving human health in a warming global climate.

Dermal exposure to toluene diisocyanate and respiratory cancer risk

Human exposure to toluene diisocyanate (TDI) occurs mainly through inhalation of vapors in occupational settings where TDI is produced or used, but dermal exposure to TDI is also possible during some operations. Because of a recent epidemiology study reporting a possible association with lung cancer risk in workers with potential dermal exposure to TDI, we evaluated the evidence from epidemiological, toxicological, and toxicokinetic studies to assess whether it is likely that dermal exposure to TDI can cause human respiratory cancers. We found that the reported associations with respiratory cancers in the epidemiology studies do not support TDI as a causal factor, as there are other explanations that are more likely than causation, such as confounding by smoking and low socioeconomic status. Experimental animal and genotoxicity studies indicate that the carcinogenic potential of TDI depends on its conversion to toluene diamine (TDA), and there is no evidence of systemic availability of TDA after dermal or inhalation exposure to TDI. Also, systemic uptake of TDI is very low after dermal exposure, and any absorbed TDI is more likely to react with biomolecules on or below the skin surface than to form TDA. Even if some TDA formation occurred after dermal exposure to TDI, TDA does not induce respiratory tract tumors in experimental animals after either dermal or oral exposure. We conclude that the available evidence indicates that dermal TDI exposure does not cause respiratory cancers in humans.

Impact of respiratory infections, outdoor pollen, and socioeconomic status on associations between air pollutants and pediatric asthma hospital admissions

Background

Epidemiology studies have shown that ambient concentrations of ozone and fine particulate matter (PM_2.5) are associated with increased emergency department (ED) visits and hospital admissions (HAs) for asthma.

Objective

Evaluate the impact of outdoor pollen, respiratory infections, and socioeconomic status (SES) on the associations between ambient ozone and PM_2.5 and asthma HAs in New York City.

Methods

Daily ozone, PM_2.5, meteorological factors, pollen, and hospitalization records during 1999 to 2009 were obtained for New York City residents. Daily counts of HAs for asthma and respiratory infections were calculated for all-age and specific age groups, and for high- and low-SES communities. Generalized additive models were used to examine ambient concentrations of ozone and PM_2.5 and asthma HAs, potential confounding effects of outdoor pollen and HAs for respiratory infections, and potential effect modification by neighborhood SES.

Results

Both ozone and PM_2.5 were statistically significantly associated with increased asthma HAs in children aged 6–18 years (per 10 ppb increase in ozone: RR = 1.0203, 95% CI: 1.0028–1.0382; per 10 μg/m³ increase in PM_2.5: RR = 1.0218, 95% CI: 1.0007–1.0434), but not with total asthma HAs, or asthma HAs in other age groups. These associations were stronger for children living in the high-SES areas. Adjustment for respiratory infection HAs at various lags did not result in changes greater than 10% in the risk estimates for either ozone or PM_2.5. In contrast, adjustment for outdoor pollen generally attenuated the estimated RRs for both ozone and PM_2.5.

Conclusions

Ambient ozone and PM_2.5 are associated with asthma HAs in school-age children, and these associations are not modified by SES. HAs for respiratory infections do not appear to be a confounder for observed ozone- and PM_2.5-asthma HAs associations, but pollen may be a weak confounder.

Relationship among environmental quality variables, housing variables, and residential needs: a secondary analysis of the relationship among indoor, outdoor, and personal air (RIOPA) concentrations database

Retrospective descriptive secondary analyses of data from relationships of indoor, outdoor, and personal air (RIOPA) study homes (in Houston, Texas; Los Angeles County, California; and, Elizabeth, New Jersey May 1999-February 2001) were conducted. Data included air exchange rates, associations between indoor and outdoor temperature and humidity, and calculated apparent temperature and humidex. Analyses examined if study homes provided optimum thermal comfort for residents during both heating and cooling seasons when compared to current American Society of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE) Standards 62/62.1 and 55. Results suggested outdoor temperature, humidex, and apparent temperature during the cooling season potentially served as indicators of indoor personal exposure to parameters of thermal comfort. Outdoor temperatures, humidex, and apparent temperature during the cooling season had statistically significant predictive abilities in predicting indoor temperature. During the heating season, only humidex in Texas and combined data across study states were statistically significant, but with weaker to moderate predicative ability. The high degree of correlation between outdoor and indoor environmental variables provided support for the validity of epidemiologic studies of weather relying on temporal comparisons. Results indicated most RIOPA study residents experienced thermal comfort; however, many values indicated how several residents may have experienced some discomfort depending on clothing and indoor activities. With climate change, increases in temperature are expected, with more days of extreme heat and humidity and, potentially harsher, longer winters. Homes being built or modernized should be created with the appropriate guidelines to provide comfort for residents daily and in extreme weather events.

Misclassified exposure in epigenetic mediation analyses. Does DNA methylation mediate effects of smoking on birthweight?

AIMS

Assessing whether epigenetic alterations mediate associations between environmental exposures and health outcomes is increasingly popular. We investigate the impact of exposure misclassification in such investigations.

MATERIALS & METHODS

We quantify bias and false-positive rates due to exposure misclassification in mediation analysis and assess the performance of the simulation extrapolation method (SIMEX). We evaluate whether DNA-methylation mediates smoking-birth weight relationship in the Norwegian Mother and Child Study birth cohort.

RESULTS

Ignoring exposure misclassification increases type I error in mediation analysis. The direct effect is underestimated and, when the mediator is a biomarker of the exposure, as is true for smoking, the indirect effect is overestimated.

CONCLUSION

Misclassification correction plus cautious interpretation are recommended for mediation analyses in the presence of exposure misclassification.

Misclassification errors from postal code-based geocoding to assign census geography in Nova Scotia, Canada

OBJECTIVES

Postal codes are often the only available geographic identifiers in many sources of health data in Canada. In order to conduct geographic analyses, postal codes are routinely geocoded to census geography to link to ecological data. Despite common use of this method, the extent of geographic misclassification errors is poorly understood. We estimated misclassification errors in the geocoding of postal codes to assign census geography in Nova Scotia, Canada.

METHODS

We examined differences between counts and match rates for postal-code geocoded and actual locations of buildings in Nova Scotia at two census administrative area levels: dissemination areas (DAs) and census subdivisions (CSDs). Actual locations were based on the data collected by the provincial government containing actual latitude/longitude of buildings. Variation in misclassification by rurality, using Statistics Canada's classification, was also assessed.

RESULTS

Outside two urban areas (Halifax Metro and Sydney) which had <10% differences in counts, many DAs had >30% differences. Match rates showed similar patterns, with the vast majority of non-urban DAs having <40% match rates. Even in major urban areas, 10% of DAs had large misclassification errors. Misclassification errors at the CSD level were still too great to estimate counts or rates without further area aggregation.

CONCLUSION

Routine use of postal code geocoding should be replaced with geocoding of location information using additional identifiers such as civic addresses or latitude and longitude. If data holders did this in-house before providing data to researchers, the accuracy and capacity of geographic analysis would be enhanced while protecting confidentiality.

How accurately and consistently do laboratories measure workplace concentrations of respirable crystalline silica?

Permissible exposure limits (PELs) for respirable crystalline silica (RCS) have recently been reduced from 0.10 to 0.05 mg/m³. This raises an important question: do current laboratory practices and standards for assessing RCS concentrations permit reliable discrimination between workplaces that are in compliance and workplaces that are not? To find out, this paper examines recent laboratory performance in quantifying RCS amounts on filters sent to them to assess their proficiency. A key finding is that accredited laboratories do not reliably (e.g., with 95% confidence) estimate RCS quantities to within a factor of 2. Thus, laboratory findings indicating that RCS levels are above or below a PEL provide little confidence that this is true. The current accreditation standard only requires laboratories to achieve estimates within three standard deviations of the correct (reference) value at least two thirds of the time, rather than a more usual standard such as within 25% of the correct value at least 95% of the time. Laboratory practices may improve as the new PEL is implemented, but they are presently essentially powerless to discriminate among RCS levels over most of the range of values that have been tested, leaving employers and regulators without a reliable means to ascertain when workplace RCS levels are above or below the PEL.

Serum anti-mullerian hormone and all-cause mortality in men

Several studies have examined an association of anti-mullerian hormone to various risk factors for mortality, however, to the best of our knowledge, no study has reported a direct relationship between anti-mullerian hormone and all-cause mortality. Therefore, we examined the relationship between baseline anti-mullerian hormone levels and subsequent all-cause mortality in men during median follow-up of 9.4 (range = 0-13) years. We used the continuous National Health and Nutrition Examination Surveys from 1999-2004 combined with National Death Index for vital status information through December 2011. Cox proportional hazard models were fit to estimate hazard ratios for all-cause mortality. Models were adjusted for age, ethnic background, body mass index, hypertension, diabetes, smoking status, c-reactive protein, total cholesterol, estimated glomerular filtration rate, serum estradiol, testosterone, sex hormone binding globulin, and androstenedione. Of the 989 men, 30 % were older than 65 years, 51 % were Caucasians, 33 % had hypertension, 27 % were active smokers, and 11 % had diabetes. Mean serum anti-mullerian hormone level of the population was 7.2 (6.3) ng/mL. During the 8943 person-years of follow-up, 164 (17 %) men died. In unadjusted analysis, each unit increase in serum anti-mullerian hormone level was associated with a 13 % lower risk of death (HR = 0.87; 95 %CI = 0.83-0.92). In multivariable models, the inverse association between serum anti-mullerian hormone levels and mortality remained significant (HR = 0.94; 95 %CI = 0.90-0.98) and was independent of confounding variables. Similarly, individuals in the highest quartile had significantly lower risk of death as compared to individuals in the lowest quartile (unadjusted HR = 0.13, 95 %CI = 0.07-0.25; adjusted HR = 0.36, 95 %CI = 0.16-0.81). We found an independent and inverse association between serum anti-mullerian hormone levels and all-cause mortality in men. The mechanism underlying this association is unknown. Further studies are needed to validate our findings in men and to examine this association in women.

Spatial measurement errors in the field of spatial epidemiology

Background

Spatial epidemiology has been aided by advances in geographic information systems, remote sensing, global positioning systems and the development of new statistical methodologies specifically designed for such data. Given the growing popularity of these studies, we sought to review and analyze the types of spatial measurement errors commonly encountered during spatial epidemiological analysis of spatial data.

Methods

Google Scholar, Medline, and Scopus databases were searched using a broad set of terms for papers indexed by a term indicating location (space or geography or location or position) and measurement error (measurement error or measurement inaccuracy or misclassification or uncertainty): we reviewed all papers appearing before December 20, 2014. These papers and their citations were reviewed to identify the relevance to our review.

Results

We were able to define and classify spatial measurement errors into four groups: (1) pure spatial location measurement errors, including both non-instrumental errors (multiple addresses, geocoding errors, outcome aggregations, and covariate aggregation) and instrumental errors; (2) location-based outcome measurement error (purely outcome measurement errors and missing outcome measurements); (3) location-based covariate measurement errors (address proxies); and (4) Covariate-Outcome spatial misaligned measurement errors. We propose how these four classes of errors can be unified within an integrated theoretical model and possible solutions were discussed.

Conclusion

Spatial measurement errors are ubiquitous threat to the validity of spatial epidemiological studies. We propose a systematic framework for understanding the various mechanisms which generate spatial measurement errors and present practical examples of such errors.

Spatiotemporal prediction of fine particulate matter using high-resolution satellite images in the Southeastern US 2003-2011

Numerous studies have demonstrated that fine particulate matter (PM2.5, particles smaller than 2.5 μm in aerodynamic diameter) is associated with adverse health outcomes. The use of ground monitoring stations of PM2.5 to assess personal exposure, however, induces measurement error. Land-use regression provides spatially resolved predictions but land-use terms do not vary temporally. Meanwhile, the advent of satellite-retrieved aerosol optical depth (AOD) products have made possible to predict the spatial and temporal patterns of PM2.5 exposures. In this paper, we used AOD data with other PM2.5 variables, such as meteorological variables, land-use regression, and spatial smoothing to predict daily concentrations of PM2.5 at a 1-km(2) resolution of the Southeastern United States including the seven states of Georgia, North Carolina, South Carolina, Alabama, Tennessee, Mississippi, and Florida for the years from 2003 to 2011. We divided the study area into three regions and applied separate mixed-effect models to calibrate AOD using ground PM2.5 measurements and other spatiotemporal predictors. Using 10-fold cross-validation, we obtained out of sample R(2) values of 0.77, 0.81, and 0.70 with the square root of the mean squared prediction errors of 2.89, 2.51, and 2.82 μg/m(3) for regions 1, 2, and 3, respectively. The slopes of the relationships between predicted PM2.5 and held out measurements were approximately 1 indicating no bias between the observed and modeled PM2.5 concentrations. Predictions can be used in epidemiological studies investigating the effects of both acute and chronic exposures to PM2.5. Our model results will also extend the existing studies on PM2.5 which have mostly focused on urban areas because of the paucity of monitors in rural areas.

Weight-of-evidence evaluation of long-term ozone exposure and cardiovascular effects

We conducted a weight-of-evidence (WoE) analysis to assess whether the current body of research supports a causal relationship between long-term ozone exposure (defined by EPA as at least 30 days in duration) at ambient levels and cardiovascular (CV) effects. We used a novel WoE framework based on the United States Environmental Protection Agency's National Ambient Air Quality Standards causal framework for this analysis. Specifically, we critically evaluated and integrated the relevant epidemiology and experimental animal data and classified a causal determination based on categories proposed by the Institute of Medicine's 2008 report, Improving the Presumptive Disability Decision-making Process for Veterans. We found that the risks of CV effects are largely null across human and experimental animal studies. The few positive associations reported in studies of CV morbidity and mortality are very small in magnitude, mainly reported in single-pollutant models, and likely attributable to bias, chance, or confounding. The few positive effects in experimental animal studies were observed mainly in ex vivo studies at high exposures, and even the in vivo findings are not likely relevant to humans. The available data also do not support a biologically plausible mechanism for the effects of ozone on the CV system. Overall, the current WoE provides no convincing case for a causal relationship between long-term exposure to ambient ozone and adverse effects on the CV system in humans, but the limitations of the available studies preclude definitive conclusions regarding a lack of causation; thus, we categorize the strength of evidence for a causal relationship between long-term exposure to ozone and CV effects as "below equipoise."

Weight-of-evidence evaluation of short-term ozone exposure and cardiovascular effects

There is a relatively large body of research on the potential cardiovascular (CV) effects associated with short-term ozone exposure (defined by EPA as less than 30 days in duration). We conducted a weight-of-evidence (WoE) analysis to assess whether it supports a causal relationship using a novel WoE framework adapted from the US EPA's National Ambient Air Quality Standards causality framework. Specifically, we synthesized and critically evaluated the relevant epidemiology, controlled human exposure, and experimental animal data and made a causal determination using the same categories proposed by the Institute of Medicine report Improving the Presumptive Disability Decision-making Process for Veterans ( IOM 2008). We found that the totality of the data indicates that the results for CV effects are largely null across human and experimental animal studies. The few statistically significant associations reported in epidemiology studies of CV morbidity and mortality are very small in magnitude and likely attributable to confounding, bias, or chance. In experimental animal studies, the reported statistically significant effects at high exposures are not observed at lower exposures and thus not likely relevant to current ambient ozone exposures in humans. The available data also do not support a biologically plausible mechanism for CV effects of ozone. Overall, the current WoE provides no convincing case for a causal relationship between short-term exposure to ambient ozone and adverse effects on the CV system in humans, but the limitations of the available studies preclude definitive conclusions regarding a lack of causation. Thus, we categorize the strength of evidence for a causal relationship between short-term exposure to ozone and CV effects as "below equipoise."

Are the elements of the proposed ozone National Ambient Air Quality Standards informed by the best available science?

The United States Environmental Protection Agency (US EPA) issues National Ambient Air Quality Standards (NAAQS) for six criteria pollutants, including ozone. Each standard has four elements: an indicator, level, averaging time, and form. Ozone levels (i.e., air concentrations) alone in scientific studies are not directly comparable to the "level" element of the NAAQS because the standard considers the level in the context of its relation to the remaining elements. Failure to appreciate this has led to misunderstandings regarding NAAQS that would be health-protective. This can be seen with controlled human ozone exposure studies, which often involved small numbers of people exercising quasi-continuously for a long duration at an intensity not common in the general population (and unlikely achievable by most sensitive individuals), under worst-case exposure profiles. In addition, epidemiology studies have used different averaging times and have had methodological limitations that may have biased results. Such considerations can make it difficult to compare ozone levels and results across studies and to appropriately apply them in a NAAQS evaluation. Relating patterns and circumstances of exposure, and exposure measurements, to all elements of the NAAQS can be challenging, but if US EPA fully undertook this, it would be evident that available evidence does not indicate that proposed lower ozone standards would be more health protective than the current one.

Mechanisms of action for arsenic in cardiovascular toxicity and implications for risk assessment

The possibility of an association between inorganic arsenic (iAs) exposure and cardiovascular outcomes has received increasing attention in the literature over the past decade. The United States Environmental Protection Agency (US EPA) is currently revising its Integrated Risk Assessment System (IRIS) review of iAs, and one of the non-cancer endpoints of interest is cardiovascular disease (CVD). Despite the increased interest in this area, substantial gaps remain in the available information, particularly regarding the mechanism of action (MOA) by which iAs could cause or exacerbate CVD. Few studies specifically address the plausibility of an association between iAs and CVD at the low exposure levels which are typical in the United States (i.e., below 100 μg As/L in drinking water). We have conducted a review and evaluation of the animal, mechanistic, and human data relevant to the potential MOAs of iAs and CVD. Specifically, we evaluated the most common proposed MOAs, which include disturbance of endothelial function and hepatic dysfunction. Our analysis of the available evidence indicates that there is not a well-established MOA for iAs in the development or progression of CVD. Few human studies of the potential MOAs have addressed plausibility at low doses and the applicability of extrapolation from animal studies to humans is questionable. However, the available evidence indicates that regardless of the specific MOA, the effects of iAs on physiological processes at the cellular level appear to operate via a threshold mechanism. This finding is consistent with the lack of association of CVD with iAs exposure in humans at levels below 100 μg/L, particularly when considering important exposure and risk modifiers such as nutrition and genetics. Based on this analysis, we conclude that there are no data supporting a linear dose-response relationship between iAs and CVD, indicating this relationship has a threshold.

Estimating risk of emergency room visits for asthma from personal versus fixed site measurements of NO2

BACKGROUND

We examined the impact of data source and exposure measurement error for ambient NO2 on risk estimates derived from a case-crossover study of emergency room visits for asthma in Windsor, Canada between 2002 and 2009.

METHODS

Paired personal and fixed-site NO2 data were available from an independent population (47 children and 48 adults) in Windsor between 2005 and 2006. We used linear regression to estimate the relationship and measurement error variance induced between fixed site and personal measurements of NO2, and through a series of simulations, evaluated the potential for a Bayesian model to adjust for this change in scale and measurement error. Finally, we re-analyzed data from the previous case-crossover study adjusting for the estimated change in slope and measurement error.

RESULTS

Correlations between paired NO2 measurements were weak (R(2)≤0.08) and slopes were far from unity (0.0029≤β≤0.30). Adjusting the previous case-crossover analysis suggested a much stronger association between personal NO2 (per 1ppb) (Odds Ratio (OR)=1.276, 95% Credible Interval (CrI): 1.034, 1.569) and emergency room visits for asthma among children relative to the fixed-site estimate (OR=1.024, 95% CrI 1.004-1.045).

CONCLUSIONS

Our findings suggest that risk estimates based on fixed-site NO2 concentrations may differ substantially from estimates based on personal exposures if the change in scale and/or measurement error is large. In practice, one must always keep the scale being used in mind when interpreting risk estimates and not assume that coefficients for ambient concentrations reflect risks at the personal level.

A proposal for assessing study quality: Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (BEES-C) instrument

The quality of exposure assessment is a major determinant of the overall quality of any environmental epidemiology study. The use of biomonitoring as a tool for assessing exposure to ubiquitous chemicals with short physiologic half-lives began relatively recently. These chemicals present several challenges, including their presence in analytical laboratories and sampling equipment, difficulty in establishing temporal order in cross-sectional studies, short- and long-term variability in exposures and biomarker concentrations, and a paucity of information on the number of measurements required for proper exposure classification. To date, the scientific community has not developed a set of systematic guidelines for designing, implementing and interpreting studies of short-lived chemicals that use biomonitoring as the exposure metric or for evaluating the quality of this type of research for WOE assessments or for peer review of grants or publications. We describe key issues that affect epidemiology studies using biomonitoring data on short-lived chemicals and propose a systematic instrument--the Biomonitoring, Environmental Epidemiology, and Short-lived Chemicals (BEES-C) instrument--for evaluating the quality of research proposals and studies that incorporate biomonitoring data on short-lived chemicals. Quality criteria for three areas considered fundamental to the evaluation of epidemiology studies that include biological measurements of short-lived chemicals are described: 1) biomarker selection and measurement, 2) study design and execution, and 3) general epidemiological study design considerations. We recognize that the development of an evaluative tool such as BEES-C is neither simple nor non-controversial. We hope and anticipate that the instrument will initiate further discussion/debate on this topic.

The relationship between indoor and outdoor temperature, apparent temperature, relative humidity, and absolute humidity

Many studies report an association between outdoor ambient weather and health. Outdoor conditions may be a poor indicator of personal exposure because people spend most of their time indoors. Few studies have examined how indoor conditions relate to outdoor ambient weather. The average indoor temperature, apparent temperature, relative humidity (RH), and absolute humidity (AH) measured in 16 homes in Greater Boston, Massachusetts, from May 2011 to April 2012 was compared to measurements taken at Boston Logan airport. The relationship between indoor and outdoor temperatures is nonlinear. At warmer outdoor temperatures, there is a strong correlation between indoor and outdoor temperature (Pearson correlation coefficient, r = 0.91, slope, β = 0.41), but at cooler temperatures, the association is weak (r = 0.40, β = 0.04). Results were similar for outdoor apparent temperature. The relationships were linear for RH and AH. The correlation for RH was modest (r = 0.55, β = 0.39). Absolute humidity exhibited the strongest indoor-to-outdoor correlation (r = 0.96, β = 0.69). Indoor and outdoor temperatures correlate well only at warmer outdoor temperatures. Outdoor RH is a poor indicator of indoor RH, while indoor AH has a strong correlation with outdoor AH year-round.

Evaluation of the causal framework used for setting national ambient air quality standards

Abstract A scientifically sound assessment of the potential hazards associated with a substance requires a systematic, objective and transparent evaluation of the weight of evidence (WoE) for causality of health effects. We critically evaluated the current WoE framework for causal determination used in the United States Environmental Protection Agency's (EPA's) assessments of the scientific data on air pollutants for the National Ambient Air Quality Standards (NAAQS) review process, including its methods for literature searches; study selection, evaluation and integration; and causal judgments. The causal framework used in recent NAAQS evaluations has many valuable features, but it could be more explicit in some cases, and some features are missing that should be included in every WoE evaluation. Because of this, it has not always been applied consistently in evaluations of causality, leading to conclusions that are not always supported by the overall WoE, as we demonstrate using EPA's ozone Integrated Science Assessment as a case study. We propose additions to the NAAQS causal framework based on best practices gleaned from a previously conducted survey of available WoE frameworks. A revision of the NAAQS causal framework so that it more closely aligns with these best practices and the full and consistent application of the framework will improve future assessments of the potential health effects of criteria air pollutants by making the assessments more thorough, transparent, and scientifically sound.

Health effects research and regulation of diesel exhaust: an historical overview focused on lung cancer risk

The mutagenicity of organic solvent extracts from diesel exhaust particulate (DEP), first noted more than 55 years ago, initiated an avalanche of diesel exhaust (DE) health effects research that now totals more than 6000 published studies. Despite an extensive body of results, scientific debate continues regarding the nature of the lung cancer risk posed by inhalation of occupational and environmental DE, with much of the debate focused on DEP. Decades of scientific scrutiny and increasingly stringent regulation have resulted in major advances in diesel engine technologies. The changed particulate matter (PM) emissions in "New Technology Diesel Exhaust (NTDE)" from today's modern low-emission, advanced-technology on-road heavy-duty diesel engines now resemble the PM emissions in contemporary gasoline engine exhaust (GEE) and compressed natural gas engine exhaust more than those in the "traditional diesel exhaust" (TDE) characteristic of older diesel engines. Even with the continued publication of epidemiologic analyses of TDE-exposed populations, this database remains characterized by findings of small increased lung cancer risks and inconsistent evidence of exposure-response trends, both within occupational cohorts and across occupational groups considered to have markedly different exposures (e.g. truckers versus railroad shopworkers versus underground miners). The recently published National Institute for Occupational Safety and Health (NIOSH)-National Cancer Institute (NCI) epidemiologic studies of miners provide some of the strongest findings to date regarding a DE-lung cancer association, but some inconsistent exposure-response findings and possible effects of bias and exposure misclassification raise questions regarding their interpretation. Laboratory animal studies are negative for lung tumors in all species, except for rats under lifetime TDE-exposure conditions with durations and concentrations that lead to "lung overload." The species specificity of the rat lung response to overload, and its occurrence with other particle types, is now well-understood. It is thus generally accepted that the rat bioassay for inhaled particles under conditions of lung overload is not predictive of human lung cancer hazard. Overall, despite an abundance of epidemiologic and experimental data, there remain questions as to whether TDE exposure causes increased lung cancers in humans. An abundance of emissions characterization data, as well as preliminary toxicological data, support NTDE as being toxicologically distinct from TDE. Currently, neither epidemiologic data nor animal bioassay data yet exist that directly bear on NTDE carcinogenic potential. A chronic bioassay of NTDE currently in progress will provide data on whether NTDE poses a carcinogenic hazard, but based on the significant reductions in PM mass emissions and the major changes in PM composition, it has been hypothesized that NTDE has a low carcinogenic potential. When the International Agency for Research on Cancer (IARC) reevaluates DE (along with GEE and nitroarenes) in June 2012, it will be the first authoritative body to assess DE carcinogenic health hazards since the emergence of NTDE and the accumulation of data differentiating NTDE from TDE.