Estimating Causal Associations of Fine Particles With Daily Deaths in Boston

Long-term exposure to ambient ozone and adult-onset asthma: A prospective cohort study

BACKGROUND

The association between long-term exposure to ozone (O3) and adult-onset asthma (AOA) remains inconclusive, and analysis of causality is lacking.

OBJECTIVES

To examine the causal association between long-term O3 exposure and AOA.

METHODS

A prospective cohort study of 362,098 participants was conducted using the UK Biobank study. Incident cases of AOA were identified using health administrative data of the National Health Services. O3 exposure at participants' residential addresses was estimated by a spatio-temporal model. Instrumental variable (IV) modelling was used to analyze the causal association between O3 exposure and AOA, by incorporating wind speed and planetary boundary layer height as IVs into time-dependent Cox model. Negative control outcome (accidental injury) was also used to additionally evaluate unmeasured confounding.

RESULTS

During a mean follow-up of 11.38 years, a total of 10,973 incident AOA cases were identified. A U-shaped concentration-response relationship was observed between O3 exposure and AOA in the traditional Cox models with HR of 0.916 (95% CI: 0.888, 0.945) for O3 at low levels (<38.17 ppb), and 1.204 (95% CI: 1.168, 1.242) for O3 at high levels (≥38.17 ppb). However, in the IV analysis we only found a statistically significant association between high-level O3 exposure and AOA risk, but not for low-level O3 exposure. No significant associations between O3 exposure and accidental injury were observed.

CONCLUSION

Our findings suggest a potential causal relationship between long-term exposure to high-level ambient O3 and increased risks of AOA.

Causal association between long-term exposure to air pollution and incident Parkinson's disease

Epidemiological evidence for long-term air pollution exposure and Parkinson's disease (PD) is controversial, and analysis of causality is limited. We identified 293,888 participants who were free of PD at baseline in the UK Biobank (2006-2010). Time-varying air pollution [fine particulate (PM2.5) and ozone (O3)] exposures were estimated using spatio-temporal models. Incident cases of PD were identified using validated algorithms. Four methods were used to investigate the associations between air pollution and PD, including (1) standard time-varying Cox proportional-hazard model; (2) Cox models weighted by generalized propensity score (GPS) and inverse-probability weights (IPW); (3) instrumental variable (IV) analysis; and (4) negative control outcome analysis. During a median of 11.6 years of follow-up, 1822 incident PD cases were identified. Based on standard Cox regression, the hazard ratios (95% confidence interval) for a 1 µg/m3 or ppb increase in PM2.5 and O3 were 1.23 (1.17, 1.30) and 1.02 (0.98, 1.05), respectively. Consistent results were found in models weighted by GPS and IPW, and in IV analysis. There were no significant associations between air pollution and negative control outcomes. This study provides evidence to support a causal association between PM2.5 exposure and PD. Mitigation of air pollution could be a protective measure against PD.

Improving the design stage of air pollution studies based on wind patterns

A growing literature in economics and epidemiology has exploited changes in wind patterns as a source of exogenous variation to better measure the acute health effects of air pollution. Since the distribution of wind components is not randomly distributed over time and related to other weather parameters, multivariate regression models are used to adjust for these confounding factors. However, this type of analysis relies on its ability to correctly adjust for all confounding factors and extrapolate to units without empirical counterfactuals. As an alternative to current practices and to gauge the extent of these issues, we propose to implement a causal inference pipeline to embed this type of observational study within an hypothetical randomized experiment. We illustrate this approach using daily data from Paris, France, over the 2008-2018 period. Using the Neyman-Rubin potential outcomes framework, we first define the treatment of interest as the effect of North-East winds on particulate matter concentrations compared to the effects of other wind directions. We then implement a matching algorithm to approximate a pairwise randomized experiment. It adjusts nonparametrically for observed confounders while avoiding model extrapolation by discarding treated days without similar control days. We find that the effective sample size for which treated and control units are comparable is surprisingly small. It is however reassuring that results on the matched sample are consistent with a standard regression analysis of the initial data. We finally carry out a quantitative bias analysis to check whether our results could be altered by an unmeasured confounder: estimated effects seem robust to a relatively large hidden bias. Our causal inference pipeline is a principled approach to improve the design of air pollution studies based on wind patterns.

Mutual effects of fine particulate matter, nitrogen dioxide, and fireworks on cause-specific acute cardiovascular mortality: A case-crossover study in communities affected by aircraft noise

Ambient air pollution is the leading cause of environmental mortality and morbidity worldwide. However, the individual contributions to acute mortality of traffic-related air pollutants such as nitrogen dioxide (NO₂) and fine particulate matter (PM_2.5) are still debated. We conducted a time-stratified case-crossover study for a population located around Zurich airport in Switzerland, including 24,886 adult cardiovascular deaths from the Swiss National Cohort. We estimated the risk of cause-specific cardiovascular mortality associated with daily NO₂ and PM_2.5 concentrations at home using distributed lag models up to 7 days preceding death, adjusted for daily temperature, precipitation, acute night-time aircraft noise, firework celebrations, and holidays. Cardiovascular mortality was associated with NO₂, whereas the association with PM_2.5 disappeared upon adjustment for NO₂. The strongest association was observed between NO₂ and ischemic stroke mortality (odds ratio = 1.55 per 10 μg/m³, 95% confidence intervals = 1.20-2.00). Cause-specific mortality analyses showed differences in terms of delayed effect: odds ratios were highest at 1-3 days after exposure for most outcomes but at lags of 3-5 days for heart failure. Individual vulnerabilities to NO₂ associated cardiovascular mortality also varied by cause of death, possibly highlighting the role of different behaviours and risk factors in the most susceptible groups. The risk of cardiovascular mortality was also increased on firework days and after public holidays, independent from NO₂ and PM_2.5 concentrations. This study confirms the association between ambient NO₂, as a marker for primary emissions, and acute cardiovascular mortality in a specific setting around a major airport. Future research should clarify the role of additional air pollutants including ultra-fine particles on cardiovascular diseases to inform most efficient control measures.

Assessing additive effects of air pollutants on mortality rate in Massachusetts

BACKGROUND

We previously found additive effects of long- and short-term exposures to fine particulate matter (PM_2.5), ozone (O₃), and nitrogen dioxide (NO₂) on all-cause mortality rate using a generalized propensity score (GPS) adjustment approach. The study addressed an important question of how many early deaths were caused by each exposure. However, the study was computationally expensive, did not capture possible interactions and high-order nonlinearities, and omitted potential confounders.

METHODS

We proposed two new methods and reconducted the analysis using the same cohort of Medicare beneficiaries in Massachusetts during 2000-2012, which consisted of 1.5 million individuals with 3.8 billion person-days of follow-up. The first method, weighted least squares (WLS), leveraged large volume of data by aggregating person-days, which gave equivalent results to the linear probability model (LPM) method in the previous analysis but significantly reduced computational burden. The second method, m-out-of-n random forests (moonRF), implemented scaling random forests that captured all possible interactions and nonlinearities in the GPS model. To minimize confounding bias, we additionally controlled relative humidity and health care utilizations that were not included previously. Further, we performed low-level analysis by restricting to person-days with exposure levels below increasingly stringent thresholds.

RESULTS

We found consistent results between LPM/WLS and moonRF: all exposures were positively associated with mortality rate, even at low levels. For long-term PM_2.5 and O₃, the effect estimates became larger at lower levels. Long-term exposure to PM_2.5 posed the highest risk: 1 μg/m³ increase in long-term PM_2.5 was associated with 1053 (95% confidence interval [CI]: 984, 1122; based on LPM/WLS methods) or 1058 (95% CI: 988, 1127; based on moonRF method) early deaths each year among the Medicare population in Massachusetts.

CONCLUSIONS

This study provides more rigorous causal evidence between PM_2.5, O₃, and NO₂ exposures and mortality, even at low levels. The largest effect estimate for long-term PM_2.5 suggests that reducing PM_2.5 could gain the most substantial benefits. The consistency between LPM/WLS and moonRF suggests that there were not many interactions and high-order nonlinearities. In the big data context, the proposed methods will be useful for future scientific work in estimating causality on an additive scale.

Attributable Risk to Assess the Health Impact of Air Pollution: Advances, Controversies, State of the Art and Future Needs

Despite the increased attention given to the health impact assessment of air pollution and to the strategies to control it in both scientific literature and concrete interventions, the results of the implementations, especially those involving traffic, have not always been satisfactory and there is still disagreement about the most appropriate interventions and the methods to assess their effectiveness. This state-of-the-art article reviews the recent interpretation of the concepts that concern the impact assessment, and compares old and new measurements of attributable risk and attributable fraction. It also summarizes the ongoing discussion about the designs and methods for assessing the air pollution impact with particular attention to improvements due to spatio-temporal analysis and other new approaches, such as studying short term effects in cohorts, and the still discussed methods of predicting the values of attributable risk (AR). Finally, the study presents the more recent analytic perspectives and the methods for directly assessing the effects of not yet implemented interventions on air quality and health, in accordance with the suggestion in the strategic plan 2020-2025 from the Health Effect Institute.

Assessing short-term impact of PM10 on mortality using a semiparametric generalized propensity score approach

BACKGROUND

The shape of the exposure-response curve describing the effects of air pollution on population health has crucial regulatory implications, and it is important in assessing causal impacts of hypothetical policies of air pollution reduction.

METHODS

After having reformulated the problem of assessing the short-term impact of air pollution on health within the potential outcome approach to causal inference, we developed a method based on the generalized propensity score (GPS) to estimate the average dose-response function (aDRF) and quantify attributable deaths under different counterfactual scenarios of air pollution reduction. We applied the proposed approach to assess the impact of airborne particles with a diameter less than or equal to 10 μm (PM₁₀) on deaths from natural, cardiovascular and respiratory causes in the city of Milan, Italy (2003-2006).

RESULTS

As opposed to what is commonly assumed, the estimated aDRFs were not linear, being steeper for low-moderate values of exposure. In the case of natural mortality, the curve became flatter for higher levels; this behavior was less pronounced for cause-specific mortality. The effect was larger in days characterized by higher temperature. According to the curves, we estimated that a hypothetical intervention able to set the daily exposure levels exceeding 40 μg/m³ to exactly 40 would have avoided 1157 deaths (90%CI: 689, 1645) in the whole study period, 312 of which for respiratory causes and 771 for cardiovascular causes. These impacts were higher than those obtained previously from regression-based methods.

CONCLUSION

This novel method based on the GPS allowed estimating the average dose-response function and calculating attributable deaths, without requiring strong assumptions about the shape of the relationship. Its potential as a tool for investigating effect modification by temperature and its use in other environmental epidemiology contexts deserve further investigation.

The impact of PM2.5 on mortality in older adults: evidence from retirement of coal-fired power plants in the United States

BACKGROUND

Evidence of causal relationship between mortality of older adults and low- concentration PM_2.5 remains limited.

OBJECTIVES

This study investigates the effects of low-concentration PM_2.5 on the mortality of adults older than 65 using the closure of coal-fired power plants in the Eastern United States as a natural experiment.

METHODS

We investigated power plants in the Eastern United States (US) that had production changes through unit shutdown or plant retirement between 1999 and 2013. We included only non-clustered power plants without scrubbers and with capacities greater than 50 MW. We used instrumental variable (IV) and difference-in-differences (DID) approaches to estimate the causal impact of PM_2.5 concentrations on mortality among Medicare beneficiaries. We compared changes in monthly age-adjusted mortality before and after the retirement of coal-fired plants between the treated and control counties; we accounted for annual wind direction in our selection of treated and control counties. In the models, we initially included only county and monthly fixed effects, and then adjusted for covariates including: 1) only weather variables (temperature, dew point, pressure); and 2) weather variables and socio-economic variables (median household income and poverty rate).

RESULTS

The monthly age-adjusted mortality rate averaged across all plants was approximately 423 per 100,000 (SD = 69) and was higher for males than females. Mean PM_2.5 concentrations across all counties were 12 μg/m³ (SD = 3.78). Using the IV method, we found that reductions in PM_2.5 concentrations significantly decreased monthly mortality among older adults. IV results show that a 1-μg/m³ reduction in PM_2.5 concentrations leads to 7.17 fewer deaths per 100,000 per month, or a 1.7% lower monthly mortality rate among people older than 65 years. Using the DID approach, we found that power plant retirement significantly decreased: 1) monthly PM_2.5 levels by 2.1 μg/m³, and 2) monthly age-adjusted mortality by approximately 15 people per 100,000 (or 3.6%) in treated counties relative to control counties. The mortality effects were higher among males than females and its impact was the greatest among people older than 75 years.

CONCLUSION

These findings provide evidence of the effectiveness of local, plant-level control measures in reducing near-plant PM_2.5 and mortality among U.S. Medicare beneficiaries.

Causal association between ambient ozone concentration and mortality in Seoul, Korea

BACKGROUND

The linearity of concentration-response (C-R) curve between ambient ozone (O₃) concentration and mortality has been controversial. The aim of the present analysis was to examine the C-R curve between O₃ concentration and mortality with a causal framework approach.

METHODS

We extracted data of hourly meteorology, hourly O₃ concentration and daily non-accidental mortality in Seoul from 2001 to 2009. We divided the dataset into two, odd-number (training set) and even-number years (testing set). Using the training set, we constructed a prediction model from hourly O₃ concentration with support vector regression estimating the daily variations of mean O₃ concentration caused by sun irradiation, wind speed and direction, controlling temperature, barometric pressure and temporal trend. With this model we predicted variance of daily O₃ from the testing set, thus creating an instrumental variable. We analyzed the association between the instrumental variable and daily mortality. We also analyzed the association according to the quartiles of daily mean O₃ concentration to examine the linearity of the association.

RESULTS

The instrumental variable was significantly and negatively associated with daily mortality in the linear model. In the stratified analysis, the negative slope was observed in the lowest quartile and the negative slope of the association diminished as the quartile increased, and the slope became positive over the 3rd quartile (O₃ > 23.3 ppb). The interaction between quartiles and instrumental variable was significant (P = 0.0108).

CONCLUSION

We observed unequal effect of exposure to ambient O₃ concentration on mortality according to the different ranges of daily mean O₃ concentration with a causal framework approach.

The Potential Effects of Policy-driven Air Pollution Interventions on Childhood Lung Development

Rationale: Although elevated air pollution exposure impairs lung-function development in childhood, it remains a challenge to use this information to estimate the potential public health benefits of air pollution interventions in exposed populations.Objectives: Apply G-computation to estimate hypothetical effects of several realistic scenarios for future air pollution reductions on lung growth.Methods: Mixed-effects linear regression was used to estimate FEV1 and FVC from age 11 to 15 years in 2,120 adolescents across 3 cohorts (1993-2001, 1997-2004, and 2007-2011). Models included regional pollutants (nitrogen dioxide [NO2] or particulate matter with an aerodynamic diameter ≤2.5 μm [PM2.5]) and other important covariates. Using G-computation, a causal inference-based method, we then estimated changes in mean lung growth in our population for hypothetical interventions on either NO2 or PM2.5. Confidence intervals (CIs) were computed by bootstrapping (N = 1,000).Measurements and Main Results: Compared with the effects of exposure from observed NO2 concentrations during the study period, had communities remained at 1994 to 1997 NO2 levels, FEV1 and FVC growth were estimated to have been reduced by 2.7% (95% CI, -3.6 to -1.8) and 4.2% (95% CI, -5.2 to -3.4), respectively. If NO2 concentrations had been reduced by 30%, we estimated a 4.4% increase in FEV1 growth (95% CI, 2.8-5.9) and a 7.1% increase in FVC growth (95% CI, 5.7-8.6). Comparable results were observed for PM2.5 interventions.Conclusions: We estimated that substantial increases in lung function would occur as a result of interventions that reduce NO2 or PM2.5 concentrations. These findings provide a quantification of potential health benefits of air quality improvement.

Impact of Reductions in Emissions from Major Source Sectors on Fine Particulate Matter-Related Cardiovascular Mortality

BACKGROUND

Reductions in ambient concentrations of fine particulate matter (PM2.5) have contributed to reductions in cardiovascular (CV) mortality.

OBJECTIVES

We examined changes in CV mortality attributed to reductions in emissions from mobile, point, areal, and nonroad sources through changes in concentrations of PM2.5 and its major components [nitrates, sulfates, elemental carbon (EC), and organic carbon (OC)] in 2,132 U.S. counties between 1990 and 2010.

METHODS

Using Community Multiscale Air Quality model estimated PM2.5 total and component concentrations, we calculated population-weighted annual averages for each county. We estimated PM2.5 total- and component-related CV mortality, adjusted for county-level population characteristics and baseline PM2.5 concentrations. Using the index of Emission Mitigation Efficiency for primary emission-to-particle pathways, we expressed changes in particle-related mortality in terms of precursor emissions by each sector.

RESULTS

PM2.5 reductions represented 5.7% of the overall decline in CV mortality. Large point source emissions of sulfur dioxide accounted for 6.685 [95% confidence interval (CI): 5.703, 7.667] fewer sulfate-related CV deaths per 100,000 people. Mobile source emissions of primary EC and nitrous oxides accounted for 3.396 (95% CI: 2.772, 4.020) and 3.984 (95% CI: 2.472, 5.496) fewer CV deaths per 100,000 people respectively. Increased EC and OC emissions from areal sources increased carbon-related CV mortality by 0.788 (95% CI: -0.540, 2.116) and 0.245 (95% CI: -0.697, 1.187) CV deaths per 100,000 people.

DISCUSSION

In a nationwide epidemiological study of emission sector contribution to PM2.5-related mortality, we found that reductions in sulfur-dioxide emissions from large point sources and nitrates and EC emissions from mobile sources contributed the largest reduction in particle-related mortality rates respectively. https://doi.org/10.1289/EHP5692.

Synergic effect between high temperature and air pollution on mortality in Northeast Asia

High temperature and air pollutants have been reported as potential risk factors of mortality. Previous studies investigated interaction between the two variables; however, the excess death risk due to the synergic effect (i.e. interaction on the additive scale) between the two variables has not been investigated adequately on a multi-country scale. This study aimed to assess the excess death risk due to the synergism between high temperature and air pollution on mortality using a multicity time-series analysis. We collected time-series data on mortality, weather variables, and four air pollutants (PM₁₀, O₃, NO₂, and CO) for 16 metropolitan cities of three countries (Japan, Korea, and Taiwan) in Northeast Asia (1979-2015). Quasi-Poisson time-series regression and meta-analysis were used to estimate the additive interaction between high temperature and air pollution. The additive interaction was measured by relative excess risk due to interaction (RERI) index. We calculated RERI with relative risks (RR) of the 99th/10th, 90th/90th, and 99th/90th percentiles of temperature/air pollution metrics, where risk at the 90th/10th percentiles of temperature/air pollution metrics was the reference category. This study showed that there may exist positive and significant excess death risks due to the synergism between high temperature and air pollution in the total population for all pollutants (95% lower confidence intervals of all RERIs>0 or near 0). In final, we measured quantitatively the excess death risks due to synergic effect between high temperature and air pollution, and the synergism should be considered in public health interventions and a composite warning system.

Low Levels of Air Pollution and Health: Effect Estimates, Methodological Challenges, and Future Directions

PURPOSE OF REVIEW

Fine particle (PM2.5) levels have been decreasing in the USA over the past decades. Our goal was to assess the current literature to characterize the association between PM2.5 and adverse health at low exposure levels.

RECENT FINDINGS

We reviewed 26 papers that examined the association between short- and long-term exposure to PM2.5 and cardio-respiratory morbidity and mortality. There is evidence suggesting that these associations are stronger at lower levels. However, there are certain methodological and interpretational limitations specific to studies of low PM2.5 levels, and further methodological development is warranted. There is strong agreement across studies that air pollution effects on adverse health are still observable at low concentrations, even well below current US standards. These findings suggest that US standards need to be reevaluated, given that further improving air quality has the potential of benefiting public health.

Effects of policy-driven hypothetical air pollutant interventions on childhood asthma incidence in southern California

Childhood asthma is a major public health concern and has significant adverse impacts on the lives of the children and their families, and on society. There is an emerging link between air pollution, which is ubiquitous in our environment, particularly in urban centers, and incident childhood asthma. Here, using data from 3 successive cohorts recruited from the same 9 communities in southern California over a span of 20 y (1993 to 2014), we estimated asthma incidence using G-computation under hypothetical air pollution exposure scenarios targeting nitrogen dioxide (NO₂) and particulate matter <2.5 μm (PM_2.5) in separate interventions. We reported comparisons of asthma incidence under each hypothetical air pollution intervention with incidence under the observed natural course of exposure; results that may be more tangible for policymakers compared with risk ratios. Model results indicated that childhood asthma incidence rates would have been statistically significantly higher had the observed reduction in ambient NO₂ in southern California not occurred in the 1990s and early 2000s, and asthma incidence rates would have been significantly lower had NO₂ been lower than what it was observed to be. For example, compliance with a hypothetical standard of 20 ppb NO₂ was estimated to result in 20% lower childhood asthma incidence (95% CI, -27% to -11%) compared with the exposure that actually occurred. The findings for hypothetical PM_2.5 interventions, although statistically significant, were smaller in magnitude compared with results for the hypothetical NO₂ interventions. Our results suggest a large potential public health benefit of air pollutant reduction in reduced incidence of childhood asthma.

Causal Modeling in Environmental Health

The field of environmental health has been dominated by modeling associations, especially by regressing an observed outcome on a linear or nonlinear function of observed covariates. Readers interested in advances in policies for improving environmental health are, however, expecting to be informed about health effects resulting from, or more explicitly caused by, environmental exposures. The quantification of health impacts resulting from the removal of environmental exposures involves causal statements. Therefore, when possible, causal inference frameworks should be considered for analyzing the effects of environmental exposures on health outcomes.

Long-term NO2 exposures and cause-specific mortality in American older adults

BACKGROUND

The impact of long-term exposure to nitrogen dioxide (NO₂) on cause-specific mortality is poorly understood.

OBJECTIVE

To assess mortality risks associated with long-term NO₂ exposure and evaluate confounding of this association.

METHODS

We examined the association between 12-month moving average NO₂ exposure and cause-specific mortality in 14.1 million US Medicare beneficiaries between 2000 and 2008. Associations were examined using age, gender, and race-stratified and state-adjusted Poisson regression models. We assessed the potential for confounding by PM_2.5 and behavioral covariates and unmeasured confounding by decomposing NO₂ into its spatial and spatio-temporal components.

RESULTS

We found significant associations between 12-month NO₂ exposure and increased mortality from all-causes [risk ratio (RR): 1.052; 95% CI: 1.051, 1.054; per 10 ppb], cardiovascular (CVD) (1.133; 95% CI: 1.130, 1.137) and respiratory disease (1.050; 95% CI: 1.044, 1.056), all cancers (1.021; 95% CI: 1.017, 1.025), ischemic heart disease (IHD) (1.221; 95% CI: 1.217, 1.226), cerebrovascular (CBV) disease (1.092; 95% CI: 1.085, 1.100), and for the first time pneumonia (1.275; 95% CI: 1.263, 1.287). Associations generally remained positive and statistically significant after adjustment for PM_2.5 and behavioral factors.

CONCLUSIONS

Our findings provide additional evidence of the increased risk posed by long-term NO₂ exposures on increased mortality from all-causes, CVD, respiratory disease, IHD, CBV, and cancer and provide new evidence of their impact on mortality from pneumonia. Unmeasured confounding of these associations was present, however, demonstrating the need to understand sources of this confounding.

Current State of Research on the Risk of Morbidity and Mortality Associated with Air Pollution in Korea

PURPOSE

The effects of air pollution on health can vary regionally. Our goal was to comprehensively review previous epidemiological studies on air pollution and health conducted in Korea to identify future areas of potential study.

MATERIALS AND METHODS

We systematically searched all published epidemiologic studies examining the association between air pollution and occurrence of death, diseases, or symptoms in Korea. After classifying health outcomes into mortality, morbidity, and health impact, we summarized the relationship between individual air pollutants and health outcomes.

RESULTS

We analyzed a total of 27 studies that provided 104 estimates of the quantitative association between risk of mortality and exposure to air pollutants, including particulate matter with aerodynamic diameter less than 10 μm, particulate matter with aerodynamic diameter less than 2.5 μm, sulfur dioxide, nitrogen dioxide, ozone, and carbon monoxide in Korea between January 1999 and July 2018. Regarding the association with morbidity, there were 38 studies, with 98 estimates, conducted during the same period. Most studies examined the short-term effects of air pollution using a time series or case-crossover study design; only three cohort studies that examined long-term effects were found. There were four health impact studies that calculated the attributable number of deaths or disability-adjusted life years due to air pollution.

CONCLUSION

There have been many epidemiologic studies in Korea regarding air pollution and health. However, the present review shows that additional studies, especially cohort and experimental studies, are needed to provide more robust and accurate evidence that can be used to promote evidence-based policymaking.

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.

A National Multicity Analysis of the Causal Effect of Local Pollution, [Formula: see text], and [Formula: see text] on Mortality

BACKGROUND

Studies have long associated [Formula: see text] with daily mortality, but few applied causal-modeling methods, or at low exposures. Short-term exposure to [Formula: see text], a marker of local traffic, has also been associated with mortality but is less studied. We previously found a causal effect between local air pollution and mortality in Boston.

OBJECTIVES

We aimed to estimate the causal effects of local pollution, [Formula: see text], and [Formula: see text] on mortality in 135 U.S. cities.

METHODS

We used three methods which, under different assumptions, provide causal marginal estimates of effect: a marginal structural model, an instrumental variable analysis, and a negative exposure control. The instrumental approach used planetary boundary layer, wind speed, and air pressure as instruments for concentrations of local pollutants; the marginal structural model separated the effects of [Formula: see text] from the effects of [Formula: see text], and the negative exposure control provided protection against unmeasured confounders.

RESULTS

In 7.3 million deaths, the instrumental approach estimated that mortality increased 1.5% [95% confidence interval (CI): 1.1%, 2.0%] per [Formula: see text] increase in local pollution indexed as [Formula: see text]. The negative control exposure was not associated with mortality. Restricting our analysis to days with [Formula: see text] below [Formula: see text], we found a 1.70% (95% CI 1.11%, 2.29%) increase. With marginal structural models, we found positive significant increases in deaths with both [Formula: see text] and [Formula: see text]. On days with [Formula: see text] below [Formula: see text], we found a 0.83% (95% CI 0.39%, 1.27%) increase. Including negative exposure controls changed estimates minimally.

CONCLUSIONS

Causal-modeling techniques, each subject to different assumptions, demonstrated causal effects of locally generated pollutants on daily deaths with effects at concentrations below the current EPA daily [Formula: see text] standard. https://doi.org/10.1289/EHP2732.

Fine particulate matters: The impact of air quality standards on cardiovascular mortality

BACKGROUND

In 1997 the U.S. Environmental Protection Agency set the first annual National Ambient Air Quality Standard (NAAQS) for fine particulate matter (PM2.5). Although the weight of scientific evidence has determined that a causal relationship exists between PM2.5 exposures and cardiovascular effects, few studies have concluded whether NAAQS-related reductions in PM2.5 led to improvements in public health.

METHODS

We examined the change in cardiovascular (CV) mortality rate and the association between change in PM2.5 and change in CV-mortality rate before (2000-2004) and after implementation of the 1997 annual PM2.5 NAAQS (2005-2010) among U.S. counties. We further examined how the association varied with respect to two factors related to NAAQS compliance: attainment status and design values (DV). We used difference-in-differences and linear regression models, adjusted for sociodemographic confounders.

FINDINGS

Across 619 counties, there were 1.10 (95% CI: 0.37, 1.82) fewer CV-deaths per year per 100,000 people for each 1µg/m3 decrease in PM2.5. Nonattainment counties had a twofold larger reduction in mean annual PM2.5, 2.1µg/m3, compared to attainment counties, 0.97µg/m3. CV-mortality rate decreased by 0.59 (95% CI: -0.54, 1.71) in nonattainment and 1.96 (95% CI: 0.77, 3.15) deaths per 100,000 people for each 1µg/m3 decrease in PM2.5 in attainment counties. When stratifying counties by DV, results were similar: counties with DV greater than 15µg/m3 experienced the greatest decrease in mean annual PM2.5 (2.29µg/m3) but the smallest decrease in CV-mortality rate per unit decrease in PM2.5, 0.73 (95% CI: -0.57, 2.02).

INTERPRETATION

We report a significant association between the change in PM2.5 and the change in CV-mortality rate before and after the implementation of NAAQS and note that the health benefits per 1µg/m3 decrease in PM2.5 persist at levels below the current national standard.

FUNDING

US EPA intermural research.

The health impacts of weekday traffic: A health risk assessment of PM2.5 emissions during congested periods

Little work has accounted for congestion, using data that reflects driving patterns, traffic volume, and speed, to examine the association between traffic emissions and human health. In this study, we performed a health risk assessment of PM_2.5 emissions during congestion periods in the Greater Toronto and Hamilton Area (GTHA), Canada. Specifically, we used a micro-level approach that combines the Stochastic User Equilibrium Traffic Assignment Algorithm with a MOVES emission model to estimate emissions considering congestion conditions. Subsequently, we applied a concentration-response function to estimate PM_2.5-related mortality, and the associated health costs. Our results suggest that traffic congestion has a substantial impact on human health and the economy in the GTHA, especially at the most congested period (7:00am). Considering daily mortality, our results showed an impact of 206 (boundary test 95%: 116; 297) and 119 (boundary test 95%: 67; 171) deaths per year (all-cause and cardiovascular mortality, respectively). The economic impact from daily mortality is approximately $1.3 billion (boundary test 95%: 0.8; 1.9), and $778 million (boundary test 95%: 478; 981), for all-cause and cardiovascular mortality, respectively. Our study can guide reliable projections of transportation and air pollution levels, improving the capability of the medical community to prepare for future trends.

Causality test of ambient fine particles and human influenza in Taiwan: Age group-specific disparity and geographic heterogeneity

Influenza is a major global public health problem, with serious outcomes that can result in hospitalization or even death. We investigate the causal relationship between human influenza cases and air pollution, quantified by ambient fine particles <2.5μm in aerodynamic diameter (PM_2.5). A modified Granger causality test is proposed to ascertain age group-specific causal relationship between weekly influenza cases and weekly adjusted accumulative PM_2.5 from 2009 to 2015 in 11 cities and counties in Taiwan. We examine the causal relationship based on posterior probabilities of the log-linear integer-valued GARCH (generalized autoregressive conditional heteroscedastic) model with covariates, which enable us to handle characteristics of influenza data such as integer-value, lagged dependence, and over-dispersion. The resulting posterior probabilities show that the adult age group (25-64) and the elderly group in New Taipei in the north and cities in southwestern part of Taiwan are strongly affected by ambient fine particles. Moreover, the elderly group is clearly affected in all study sites. Globalization and economic growth have resulted in increased ambient air pollution (including PM_2.5) and subsequently substantial public health concerns in the West Pacific region. Minimizing exposure to air pollutants is particularly important for the elderly and susceptible individuals with respiratory diseases.

Best Practices for Gauging Evidence of Causality in Air Pollution Epidemiology

The contentious political climate surrounding air pollution regulations has brought some researchers and policy-makers to argue that evidence of causality is necessary before implementing more stringent regulations. Recently, investigators in an increasing number of air pollution studies have purported to have used "causal analysis," generating the impression that studies not explicitly labeled as such are merely "associational" and therefore less rigorous. Using 3 prominent air pollution studies as examples, we review good practices for how to critically evaluate the extent to which an air pollution study provides evidence of causality. We argue that evidence of causality should be gauged by a critical evaluation of design decisions such as 1) what actions or exposure levels are being compared, 2) whether an adequate comparison group was constructed, and 3) how closely these design decisions approximate an idealized randomized study. We argue that air pollution studies that are more scientifically rigorous in terms of the decisions made to approximate a randomized experiment are more likely to provide evidence of causality and should be prioritized among the body of evidence for regulatory review accordingly. Our considerations, although presented in the context of air pollution epidemiology, can be broadly applied to other fields of epidemiology.

Applying Nonparametric Methods to Analyses of Short-Term Fine Particulate Matter Exposure and Hospital Admissions for Cardiovascular Diseases among Older Adults

Short-term exposure to fine particulate matter (PM_2.5) has been associated with increased risks of cardiovascular diseases (CVDs), but whether such associations are supportive of a causal relationship is unclear, and few studies have employed formal causal analysis methods to address this. We employed nonparametric methods to examine the associations between daily concentrations of PM_2.5 and hospital admissions (HAs) for CVD among adults aged 75 years and older in Texas, USA. We first quantified the associations in partial dependence plots generated using the random forest approach. We next used a Bayesian network learning algorithm to identify conditional dependencies between CVD HAs of older men and women and several predictor variables. We found that geographic location (county), time (e.g., month and year), and temperature satisfied necessary information conditions for being causes of CVD HAs among older men and women, but daily PM_2.5 concentrations did not. We also found that CVD HAs of disjoint subpopulations were strongly predictive of CVD HAs among older men and women, indicating the presence of unmeasured confounders. Our findings from nonparametric analyses do not support PM_2.5 as a direct cause of CVD HAs among older adults.

Temporal trends of PM10 and its impact on mortality in Lombardy, Italy

INTRODUCTION

Exposure to particulate matter with diameter ≤10 μm (PM₁₀) entails well documented adverse effects on human health. In the last decade, concentration of PM₁₀ in Lombardy (10 million inhabitants), Italy, has been gradually decreasing. We evaluated how the mortality burden due to PM₁₀ varied in that same period.

METHODS

We focused on 13 areas of the Region in 2003-2014: 11 cities with more than 50,000 inhabitants, 1 smaller alpine town and 1 agricultural province. For each area, we collected PM₁₀ annual average concentrations and natural mortality data, and we used the posterior area-specific effects from a previous Bayesian meta-analysis to estimate the short-term impact of PM₁₀ on mortality, in terms of deaths attributable (AD) to annual average exposures exceeding the WHO threshold of 20 μg/m³.

RESULTS

PM₁₀ annual average values showed a non-homogenous decreasing trend in the investigated time period in most of the areas. Overall, the population-weighted exposure levels decreased, except for a peak in 2011, but never met the WHO threshold. In 2003-2006, PM₁₀ levels were responsible, on average, for 343.0 annual AD from natural causes that decreased to 253.5 in 2007-2010 and to 208.3 in 2011-2014. Overall we estimated that PM₁₀ was responsible for about 1% of all natural deaths (min-max range: 0.86%-1.42%); the impact was heterogeneous among areas.

CONCLUSIONS

By collecting routinely available data for the most populated areas in Lombardy, we returned a picture of air pollution and health trends in the last decade. Notwithstanding the observed reduction in PM₁₀ between 2003 and 2014 and the resulting decline in the number of AD, the impact is still relevant. Hence, appropriate policies for emission reduction could have a further beneficial effect on population health. Studies based on routine data and local effect estimates are recommended to properly inform the policy-making process.

Associations between Environmental Quality and Mortality in the Contiguous United States, 2000-2005

BACKGROUND

Assessing cumulative effects of the multiple environmental factors influencing mortality remains a challenging task.

OBJECTIVES

This study aimed to examine the associations between cumulative environmental quality and all-cause and leading cause-specific (heart disease, cancer, and stroke) mortality rates.

METHODS

We used the overall Environmental Quality Index (EQI) and its five domain indices (air, water, land, built, and sociodemographic) to represent environmental exposure. Associations between the EQI and mortality rates (CDC WONDER) for counties in the contiguous United States (n = 3,109) were investigated using multiple linear regression models and random intercept and random slope hierarchical models. Urbanicity, climate, and a combination of the two were used to explore the spatial patterns in the associations.

RESULTS

We found 1 standard deviation increase in the overall EQI (worse environment) was associated with a mean 3.22% (95% CI: 2.80%, 3.64%) increase in all-cause mortality, a 0.54% (95% CI: -0.17%, 1.25%) increase in heart disease mortality, a 2.71% (95% CI: 2.21%, 3.22%) increase in cancer mortality, and a 2.25% (95% CI: 1.11%, 3.39%) increase in stroke mortality. Among the environmental domains, the associations ranged from -1.27% (95% CI: -1.70%, -0.84%) to 3.37% (95% CI: 2.90%, 3.84%) for all-cause mortality, -2.62% (95% CI: -3.52%, -1.73%) to 4.50% (95% CI: 3.73%, 5.27%) for heart disease mortality, -0.88% (95% CI: -2.12%, 0.36%) to 3.72% (95% CI: 2.38%, 5.06%) for stroke mortality, and -0.68% (95% CI: -1.19%, -0.18%) to 3.01% (95% CI: 2.46%, 3.56%) for cancer mortality. Air had the largest associations with all-cause, heart disease, and cancer mortality, whereas the sociodemographic index had the largest association with stroke mortality. Across the urbanicity gradient, no consistent trend was found. Across climate regions, the associations ranged from 2.29% (95% CI: 1.87%, 2.72%) to 5.30% (95% CI: 4.30%, 6.30%) for overall EQI, and larger associations were generally found in dry areas for both overall EQI and domain indices.

CONCLUSIONS

These results suggest that poor environmental quality, particularly poor air quality, was associated with increased mortality and that associations vary by urbanicity and climate region. Citation: Jian Y, Messer LC, Jagai JS, Rappazzo KM, Gray CL, Grabich SC, Lobdell DT. 2017. Associations between environmental quality and mortality in the contiguous United States, 2000-2005. Environ Health Perspect 125:355-362; http://dx.doi.org/10.1289/EHP119.

Estimating deaths attributable to airborne particles: sensitivity of the results to different exposure assessment approaches

BACKGROUND

Epidemiological evidences support the existence of an effect of airborne particulate on population health. However, few studies evaluated the robustness of the results to different exposure assessment approaches. In this paper, we estimated short term effects and impacts of high levels of particulate matter with aerodynamic diameter ≤10 μm (PM₁₀) and ≤2.5 μm (PM_2.5) in the Emilia-Romagna region (Northern Italy), one of the most polluted areas in Europe, in the period 2006-2010, and checked if the results changed when different exposure definitions were used.

METHODS

Short-term impact of particles on population mortality was assessed, both considering the 9 provincial capitals of the Emilia-Romagna and the region as a whole. We estimated the effects of PM₁₀ and PM_2.5 on natural mortality by combining city-specific results in a Bayesian random-effects meta-analysis, and we used these estimates to calculate impacts in terms of attributable deaths. For PM₁₀, we considered different definitions of exposure, based on the use of the air pollutant levels measured by different monitoring stations (background or traffic monitors) or predicted by a dispersion model.

RESULTS

Annual average concentrations of PM₁₀ and PM_2.5 exceeding the WHO limits of 20 and 10 μg/m³ were respectively responsible for 5.9 and 3.0 deaths per 100 000 inhabitants per year in the provincial capitals, during the period 2006-2010. The total impact in the region in 2010 amounted to 4.4 and 2.8 deaths per 100 000 for PM₁₀ and PM_2.5, respectively. The impact estimates for PM₁₀ did not substantially change when the exposure levels were derived from background or traffic monitoring stations, or arose from the dispersion model, in particular when the counterfactual value of 20 μg/m³ was considered. The effect estimates appeared more sensitive to the exposure definition.

CONCLUSIONS

A reduction in particle concentrations could have produced significant health benefits in the region. This general conclusion did not change when different exposure definitions were used, provided that the same exposure assessment approach was used for both effect and impact estimations. Caution is therefore recommended when using effect estimates from the literature to assess health impacts of air pollution in actual contexts.

Assessing the short term impact of air pollution on mortality: a matching approach

BACKGROUND

The opportunity to assess short term impact of air pollution relies on the causal interpretation of the exposure-response association. However, up to now few studies explicitly faced this issue within a causal inference framework. In this paper, we reformulated the problem of assessing the short term impact of air pollution on health using the potential outcome approach to causal inference. We considered the impact of high daily levels of particulate matter ≤10 μm in diameter (PM₁₀) on mortality within two days from the exposure in the metropolitan area of Milan (Italy), during the period 2003-2006. Our research focus was the causal impact of a hypothetical intervention setting daily air pollution levels under a pre-fixed threshold.

METHODS

We applied a matching procedure based on propensity score to estimate the total number of attributable deaths (AD) during the study period. After defining the number of attributable deaths in terms of difference between potential outcomes, we used the estimated propensity score to match each high exposure day, namely each day with a level of exposure higher than 40 μg/m³, with a day with similar background characteristics but a level of exposure lower than 40 μg/m³. Then, we estimated the impact by comparing mortality between matched days.

RESULTS

During the study period daily exposures larger than 40 μg/m³ were responsible for 1079 deaths (90% CI: 116; 2042). The impact was more evident among the elderly than in the younger age classes. Exposures ≥ 40 μg/m³ were responsible, among the elderly, for 1102 deaths (90% CI: 388, 1816), of which 797 from cardiovascular causes and 243 from respiratory causes. Clear evidence of an impact on respiratory mortality was found also in the age class 65-74, with 87 AD (90% CI: 11, 163).

CONCLUSIONS

The propensity score matching turned out to be an appealing method to assess historical impacts in this field, which guarantees that the estimated total number of AD can be derived directly as sum of either age-specific or cause-specific AD, unlike the standard model-based procedure. For this reason, it is a promising approach to perform surveillance focusing on very specific causes of death or diseases, or on susceptible subpopulations. Finally, the propensity score matching is free from issues concerning the exposure-confounders-mortality modeling and does not involve extrapolation. On the one hand this enhances the internal validity of our results; on the other, it makes the approach scarcely appropriate for estimating future impacts.

STROBE-Long-Term Exposure to Ambient Fine Particulate Air Pollution and Hospitalization Due to Peptic Ulcers

Little is known about the effect of air pollution on the gastrointestinal (GI) system. We investigated the association between long-term exposures to outdoor fine particles (PM2.5) and hospitalization for peptic ulcer diseases (PUDs) in a large cohort of Hong Kong Chinese elderly.A total of 66,820 subjects aged ≥65 years who were enrolled in all 18 Government Elderly Health Service centers of Hong Kong participated in the study voluntarily between 1998 and 2001. They were prospectively followed up for more than 10 years. Annual mean exposures to PM2.5 at residence of individuals were estimated by satellite data through linkage with address details including floor level. All hospital admission records of the subjects up to December 31, 2010 were retrieved from the central database of Hospital Authority. We used Cox regression to estimate the hazard ratio (HR) for PUD hospitalization associated with PM2.5 exposure after adjustment for individual and ecological covariates.A total of 60,273 subjects had completed baseline information including medical, socio-demographic, lifestyle, and anthropometric data at recruitment. During the follow-up period, 1991 (3.3%) subjects had been hospitalized for PUD. The adjusted HR for PUD hospitalization per 10 μg/m of PM2.5 was 1.18 (95% confidence interval: 1.02-1.36, P = 0.02). Further analysis showed that the associations with PM2.5 were significant for gastric ulcers (HR 1.29; 1.09-1.53, P = 0.003) but not for duodenal ulcers (HR 0.98; 0.78 to 1.22, P = 0.81).Long-term exposures to PM2.5 were associated with PUD hospitalization in elder population. The mechanism underlying the PM2.5 in the development of gastric ulcers warrants further research.