Spatiotemporally resolved air exchange rate as a modifier of acute air pollution-related morbidity in Atlanta

Detecting Elevated Air Pollution Levels by Monitoring Web Search Queries: Algorithm Development and Validation

BACKGROUND

Real-time air pollution monitoring is a valuable tool for public health and environmental surveillance. In recent years, there has been a dramatic increase in air pollution forecasting and monitoring research using artificial neural networks. Most prior work relied on modeling pollutant concentrations collected from ground-based monitors and meteorological data for long-term forecasting of outdoor ozone (O₃), oxides of nitrogen, and fine particulate matter (PM_2.5). Given that traditional, highly sophisticated air quality monitors are expensive and not universally available, these models cannot adequately serve those not living near pollutant monitoring sites. Furthermore, because prior models were built based on physical measurement data collected from sensors, they may not be suitable for predicting the public health effects of pollution exposure.

OBJECTIVE

This study aimed to develop and validate models to nowcast the observed pollution levels using web search data, which are publicly available in near real time from major search engines.

METHODS

We developed novel machine learning-based models using both traditional supervised classification methods and state-of-the-art deep learning methods to detect elevated air pollution levels at the US city level by using generally available meteorological data and aggregate web-based search volume data derived from Google Trends. We validated the performance of these methods by predicting 3 critical air pollutants (O₃, nitrogen dioxide, and PM_2.5) across 10 major US metropolitan statistical areas in 2017 and 2018. We also explore different variations of the long short-term memory model and propose a novel search term dictionary learner-long short-term memory model to learn sequential patterns across multiple search terms for prediction.

RESULTS

The top-performing model was a deep neural sequence model long short-term memory, using meteorological and web search data, and reached an accuracy of 0.82 (F₁-score 0.51) for O_3, 0.74 (F₁-score 0.41) for nitrogen dioxide, and 0.85 (F₁-score 0.27) for PM_2.5, when used for detecting elevated pollution levels. Compared with using only meteorological data, the proposed method achieved superior accuracy by incorporating web search data.

CONCLUSIONS

The results show that incorporating web search data with meteorological data improves the nowcasting performance for all 3 pollutants and suggest promising novel applications for tracking global physical phenomena using web search data.

Estimating the Impact of Air Pollution on Healthcare-Seeking Behaviour by Applying a Difference-in-Differences Method to Syndromic Surveillance Data

Syndromic surveillance data were used to estimate the direct impact of air pollution on healthcare-seeking behaviour, between 1 April 2012 and 31 December 2017. A difference-in-differences approach was used to control for spatial and temporal variations that were not due to air pollution and a meta-analysis was conducted to combine estimates from different pollution periods. Significant increases were found in general practitioner (GP) out-of-hours consultations, including a 98% increase (2–386, 95% confidence interval) in acute bronchitis and a 16% (3–30) increase in National Health Service (NHS) 111 calls for eye problems. However, the numbers involved are small; for instance, roughly one extra acute bronchitis consultation in a local authority on a day when air quality is poor. These results provide additional information for healthcare planners on the impacts of localised poor air quality. However, further work is required to identify the separate impact of different pollutants.

Estimating climate change-related impacts on outdoor air pollution infiltration

BACKGROUND

Rising temperatures due to climate change are expected to impact human adaptive response, including changes to home cooling and ventilation patterns. These changes may affect air pollution exposures via alteration in residential air exchange rates, affecting indoor infiltration of outdoor particles. We conducted a field study examining associations between particle infiltration and temperature to inform future studies of air pollution health effects.

METHODS

We measured indoor fine particulate matter (PM_2.5) in Atlanta in 60 homes (810 sampling-days). Indoor-outdoor sulfur ratios were used to estimate particle infiltration, using central site outdoor sulfur concentrations. Linear and mixed-effects models were used to examine particle infiltration ratio-temperature relationships, based on which we incorporated projected meteorological values (Representative Concentration Pathways intermediate scenario RCP 4.5) to estimate particle infiltration ratios in 20-year future (2046-2065) and past (1981-2000) scenarios.

RESULTS

The mean particle infiltration ratio in Atlanta was 0.70 ± 0.30, with a 0.21 lower ratio in summer compared to transition seasons (spring, fall). Particle infiltration ratios were 0.19 lower in houses using heating, ventilation, and air conditioning (HVAC) systems compared to those not using HVAC. We observed significant associations between particle infiltration ratios and both linear and quadratic models of ambient temperature for homes using natural ventilation and those using HVAC. Future temperature was projected to increase by 2.1 °C in Atlanta, which corresponds to an increase of 0.023 (3.9%) in particle infiltration ratios during cooler months and a decrease of 0.037 (6.2%) during warmer months.

DISCUSSION

We estimated notable changes in particle infiltration ratio in Atlanta for different 20-year periods, with differential seasonal patterns. Moreover, when stratified by HVAC usage, increases in future ambient temperature due to climate change were projected to enhance seasonal differences in PM_2.5 infiltration in Atlanta. These analyses can help minimize exposure misclassification in epidemiologic studies of PM_2.5, and provide a better understanding of the potential influence of climate change on PM_2.5 health effects.

Evaluation of adverse effects of particulate matter on human life

Particulate matter (PM_2.5) has a severe impact on human health. The concentration of PM_2.5, related to air-quality changes, may be associated with perceptible effects on people's health. In this study, computer intelligence was used to assess the negative effects of PM_2.5. The input data, used for the evaluation, were grid definitions (shape-file), PM_2.5, air-quality data, incidence/prevalence rates, a population dataset, and the (Krewski) health-impact function. This paper presents a local (Pakistan) health-impact assessment of PM_2.5 in order to estimate the long-term effects on mortality. A rollback-to-a-standard scenario was based on the PM_2.5 concentration of 15 μg m^-3. Health benefits for a population of about 73 million people were calculated. The results showed that the estimated avoidable mortality, linked to ischemic heart disease and lung cancer, was 2,773 for every 100,000 people, which accounts for 2,024,290 preventable deaths of the total population. The total cost, related to the above mortality, was estimated to be US $ 1,000 million. Therefore, a policy for a PM_2.5-standard up to 15 μg m^-3 is suggested.

Exposure to household air pollution over 10 years is related to asthma and lung function decline

INTRODUCTION

We investigated if long-term household air pollution (HAP) is associated with asthma and lung function decline in middle-aged adults, and whether these associations were modified by glutathione S-transferase (GST) gene variants, ventilation and atopy.

MATERIALS AND METHODS

Prospective data on HAP (heating, cooking, mould and smoking) and asthma were collected in the Tasmanian Longitudinal Health Study (TAHS) at mean ages 43 and 53 years (n=3314). Subsamples had data on lung function (n=897) and GST gene polymorphisms (n=928). Latent class analysis was used to characterise longitudinal patterns of exposure. Regression models assessed associations and interactions.

RESULTS

We identified seven longitudinal HAP profiles. Of these, three were associated with persistent asthma, greater lung function decline and % reversibility by age 53 years compared with the "Least exposed" reference profile for those who used reverse-cycle air conditioning, electric cooking and no smoking. The "All gas" (OR 2.64, 95% CI 1.22-5.70), "Wood heating/smoking" (OR 2.71, 95% CI 1.21-6.05) and "Wood heating/gas cooking" (OR 2.60, 95% CI 1.11-6.11) profiles were associated with persistent asthma, as well as greater lung function decline and % reversibility. Participants with the GSTP1 Ile/Ile genotype were at a higher risk of asthma or greater lung function decline when exposed compared with other genotypes. Exhaust fan use and opening windows frequently may reduce the adverse effects of HAP produced by combustion heating and cooking on current asthma, presumably through increasing ventilation.

CONCLUSIONS

Exposures to wood heating, gas cooking and heating, and tobacco smoke over 10 years increased the risks of persistent asthma, lung function decline and % reversibility, with evidence of interaction by GST genes and ventilation.

A framework for estimating the US mortality burden of fine particulate matter exposure attributable to indoor and outdoor microenvironments

Exposure to fine particulate matter (PM_2.5) is associated with increased mortality. Although epidemiology studies typically use outdoor PM_2.5 concentrations as surrogates for exposure, the majority of PM_2.5 exposure in the US occurs in microenvironments other than outdoors. We develop a framework for estimating the total US mortality burden attributable to exposure to PM_2.5 of both indoor and outdoor origin in the primary non-smoking microenvironments in which people spend most of their time. The framework utilizes an exposure-response function combined with adjusted mortality effect estimates that account for underlying exposures to PM_2.5 of outdoor origin that likely occurred in the original epidemiology populations from which effect estimates are derived. We demonstrate the framework using several different scenarios to estimate the potential magnitude and bounds of the US mortality burden attributable to total PM_2.5 exposure across all non-smoking environments under a variety of assumptions. Our best estimates of the US mortality burden associated with total PM_2.5 exposure in the year 2012 range from ~230,000 to ~300,000 deaths. Indoor exposure to PM_2.5 of outdoor origin is typically the largest total exposure, accounting for ~40-60% of total mortality, followed by residential exposure to indoor PM_2.5 sources, which also drives the majority of variability in each scenario.

Fine-Scale Modeling of Individual Exposures to Ambient PM2.5, EC, NOx, CO for the Coronary Artery Disease and Environmental Exposure (CADEE) Study

Air pollution epidemiological studies often use outdoor concentrations from central-site monitors as exposure surrogates, which can induce measurement error. The goal of this study was to improve exposure assessments of ambient fine particulate matter (PM2.5), elemental carbon (EC), nitrogen oxides (NOx), and carbon monoxide (CO) for a repeated measurements study with 15 individuals with coronary artery disease in central North Carolina called the Coronary Artery Disease and Environmental Exposure (CADEE) Study. We developed a fine-scale exposure modeling approach to determine five tiers of individual-level exposure metrics for PM2.5, EC, NOx, CO using outdoor concentrations, on-road vehicle emissions, weather, home building characteristics, time-locations, and time-activities. We linked an urban-scale air quality model, residential air exchange rate model, building infiltration model, global positioning system (GPS)-based microenvironment model, and accelerometer-based inhaled ventilation model to determine residential outdoor concentrations (Cout_home, Tier 1), residential indoor concentrations (Cin_home, Tier 2), personal outdoor concentrations (Cout_personal, Tier 3), exposures (E, Tier 4), and inhaled doses (D, Tier 5). We applied the fine-scale exposure model to determine daily 24-h average PM2.5, EC, NOx, CO exposure metrics (Tiers 1-5) for 720 participant-days across the 25 months of CADEE. Daily modeled metrics showed considerable temporal and home-to-home variability of Cout_home and Cin_home (Tiers 1-2) and person-to-person variability of Cout_personal, E, and D (Tiers 3-5). Our study demonstrates the ability to apply an urban-scale air quality model with an individual-level exposure model to determine multiple tiers of exposure metrics for an epidemiological study, in support of improving health risk assessments.

The impact of air exchange rate on ambient air pollution exposure and inequalities across all residential parcels in Massachusetts

Individual housing characteristics can modify outdoor ambient air pollution infiltration through air exchange rate (AER). Time and labor-intensive methods needed to measure AER has hindered characterization of AER distributions across large geographic areas. Using publicly-available data and regression models associating AER with housing characteristics, we estimated AER for all Massachusetts residential parcels. We conducted an exposure disparities analysis, considering ambient PM_2.5 concentrations and residential AERs. Median AERs (h^-1) with closed windows for winter and summer were 0.74 (IQR: 0.47-1.09) and 0.36 (IQR: 0.23-0.57), respectively, with lower AERs for single family homes. Across residential parcels, variability of indoor PM_2.5 concentrations of ambient origin was twice that of ambient PM_2.5 concentrations. Housing parcels above the 90th percentile of both AER and ambient PM_2.5 (i.e., the leakiest homes in areas of highest ambient PM_2.5)-vs. below the 10 percentile-were located in neighborhoods with higher proportions of Hispanics (20.0% vs. 2.0%), households with an annual income of less than $20,000 (26.0% vs. 7.5%), and individuals with less than a high school degree (23.2% vs. 5.8%). Our approach can be applied in epidemiological studies to estimate exposure modifiers or to characterize exposure disparities that are not solely based on ambient concentrations.

Application of an indoor air pollution metamodel to a spatially-distributed housing stock

Estimates of population air pollution exposure typically rely on the outdoor component only, and rarely account for populations spending the majority of their time indoors. Housing is an important modifier of air pollution exposure due to outdoor pollution infiltrating indoors, and the removal of indoor-sourced pollution through active or passive ventilation. Here, we describe the application of an indoor air pollution modelling tool to a spatially distributed housing stock model for England and Wales, developed from Energy Performance Certificate (EPC) data and containing information for approximately 11.5 million dwellings. First, we estimate indoor/outdoor (I/O) ratios and total indoor concentrations of outdoor air pollution for PM_2.5 and NO₂ for all EPC dwellings in London. The potential to estimate concentration from both indoor and outdoor sources is then demonstrated by modelling indoor background CO levels for England and Wales pre- and post-energy efficient adaptation, including heating, cooking, and smoking as internal sources. In London, we predict a median I/O ratio of 0.60 (99% CIs; 0.53-0.73) for outdoor PM_2.5 and 0.41 (99%CIs; 0.34-0.59) for outdoor NO₂; Pearson correlation analysis indicates a greater spatial modification of PM_2.5 exposure by housing (ρ = 0.81) than NO₂ (ρ = 0.88). For the demonstrative CO model, concentrations ranged from 0.4-9.9 ppm (99%CIs)(median = 3.0 ppm) in kitchens and 0.3-25.6 ppm (median = 6.4 ppm) in living rooms. Clusters of elevated indoor concentration are found in urban areas due to higher outdoor concentrations and smaller dwellings with reduced ventilation potential, with an estimated 17.6% increase in the number of living rooms and 63% increase in the number of kitchens exceeding recommended exposure levels following retrofit without additional ventilation. The model has the potential to rapidly calculate indoor pollution exposure across large housing stocks and estimate changes to exposure under different pollution or housing policy scenarios.

Urban heat and air pollution: A framework for integrating population vulnerability and indoor exposure in health risk analyses

Urban growth and climate change will exacerbate extreme heat events and air pollution, posing considerable health challenges to urban populations. Although epidemiological studies have shown associations between health outcomes and exposures to ambient air pollution and extreme heat, the degree to which indoor exposures and social and behavioral factors may confound or modify these observed effects remains underexplored. To address this knowledge gap, we explore the linkages between vulnerability science and epidemiological conceptualizations of risk to propose a conceptual and analytical framework for characterizing current and future health risks to air pollution and extreme heat, indoors and outdoors. Our framework offers guidance for research on climatic variability, population vulnerability, the built environment, and health effects by illustrating how health data, spatially resolved ambient data, estimates of indoor conditions, and household-level vulnerability data can be integrated into an epidemiological model. We also describe an approach for characterizing population adaptive capacity and indoor exposure for use in population-based epidemiological models. Our framework and methods represent novel resources for the evaluation of health risks from extreme heat and air pollution, both indoors and outdoors.

Relationships between home ventilation rates and respiratory health in the Colorado Home Energy Efficiency and Respiratory Health (CHEER) study

BACKGROUND

As societies adopt green building practices to reduce energy expenditures and emissions that contribute to climate change, it is important to consider how such building design changes influence health. These practices typically focus on reducing air exchange rates between the building interior and the outdoor environment to minimize energy loss, the health effects of which are not well characterized. This study aims to evaluate the relationship between air exchange rates and respiratory health in a multi-ethnic population living in low-income, urban homes.

METHODS

The Colorado Home Energy Efficiency and Respiratory Health (CHEER) study is a cross-sectional study that enrolled 302 people in 216 non-smoking, low-income single-family homes, duplexes and town-homes from Colorado's Northern Front Range. A blower door test was conducted and the annual average air exchange rate (AAER) was estimated for each home. Respiratory health was assessed using a structured questionnaire based on standard instruments. We estimated the association between AAER and respiratory symptoms, adjusting for relevant confounders.

RESULTS

Air exchange rates in many homes were high compared to prior studies (median 0.54 air changes per hour, range 0.10, 2.17). Residents in homes with higher AAER were more likely to report chronic cough, asthma and asthma-like symptoms, including taking medication for wheeze, wheeze that limited activities and dry cough at night. Allergic symptoms were not associated with AAER in any models. The association between AAER and asthma-like symptoms was stronger for households located in areas with high potential exposure to traffic related pollutants, but this was not consistent across all health outcomes.

CONCLUSIONS

While prior studies have highlighted the potential hazards of low ventilation rates in residences, this study suggests high ventilation rates in single-family homes, duplexes and town-homes in urban areas may also have negative impacts on respiratory health, possibly due to the infiltration of outdoor pollutants.

Modeling individual exposures to ambient PM2.5 in the diabetes and the environment panel study (DEPS)

Air pollution epidemiology studies of ambient fine particulate matter (PM_2.5) often use outdoor concentrations as exposure surrogates, which can induce exposure error. The goal of this study was to improve ambient PM_2.5 exposure assessments for a repeated measurements study with 22 diabetic individuals in central North Carolina called the Diabetes and Environment Panel Study (DEPS) by applying the Exposure Model for Individuals (EMI), which predicts five tiers of individual-level exposure metrics for ambient PM_2.5 using outdoor concentrations, questionnaires, weather, and time-location information. Using EMI, we linked a mechanistic air exchange rate (AER) model to a mass-balance PM_2.5 infiltration model to predict residential AER (Tier 1), infiltration factors (F_{inf_home}, Tier 2), indoor concentrations (C_in, Tier 3), personal exposure factors (F_pex, Tier 4), and personal exposures (E, Tier 5) for ambient PM_2.5. We applied EMI to predict daily PM_2.5 exposure metrics (Tiers 1-5) for 174 participant-days across the 13 months of DEPS. Individual model predictions were compared to a subset of daily measurements of F_pex and E (Tiers 4-5) from the DEPS participants. Model-predicted F_pex and E corresponded well to daily measurements with a median difference of 14% and 23%; respectively. Daily model predictions for all 174 days showed considerable temporal and house-to-house variability of AER, F_{inf_home}, and C_in (Tiers 1-3), and person-to-person variability of F_pex and E (Tiers 4-5). Our study demonstrates the capability of predicting individual-level ambient PM_2.5 exposure metrics for an epidemiological study, in support of improving risk estimation.

Probabilistic estimation of residential air exchange rates for population-based human exposure modeling

Residential air exchange rates (AERs) are a key determinant in the infiltration of ambient air pollution indoors. Population-based human exposure models using probabilistic approaches to estimate personal exposure to air pollutants have relied on input distributions from AER measurements. An algorithm for probabilistically estimating AER was developed based on the Lawrence Berkley National Laboratory Infiltration model utilizing housing characteristics and meteorological data with adjustment for window opening behavior. The algorithm was evaluated by comparing modeled and measured AERs in four US cities (Los Angeles, CA; Detroit, MI; Elizabeth, NJ; and Houston, TX) inputting study-specific data. The impact on the modeled AER of using publically available housing data representative of the region for each city was also assessed. Finally, modeled AER based on region-specific inputs was compared with those estimated using literature-based distributions. While modeled AERs were similar in magnitude to the measured AER they were consistently lower for all cities except Houston. AERs estimated using region-specific inputs were lower than those using study-specific inputs due to differences in window opening probabilities. The algorithm produced more spatially and temporally variable AERs compared with literature-based distributions reflecting within- and between-city differences, helping reduce error in estimates of air pollutant exposure.

Influence of exposure differences on city-to-city heterogeneity in PM2.5-mortality associations in US cities

BACKGROUND

Multi-city population-based epidemiological studies have observed heterogeneity between city-specific fine particulate matter (PM_2.5)-mortality effect estimates. These studies typically use ambient monitoring data as a surrogate for exposure leading to potential exposure misclassification. The level of exposure misclassification can differ by city affecting the observed health effect estimate.

METHODS

The objective of this analysis is to evaluate whether previously developed residential infiltration-based city clusters can explain city-to-city heterogeneity in PM_2.5 mortality risk estimates. In a prior paper 94 cities were clustered based on residential infiltration factors (e.g. home age/size, prevalence of air conditioning (AC)), resulting in 5 clusters. For this analysis, the association between PM_2.5 and all-cause mortality was first determined in 77 cities across the United States for 2001-2005. Next, a second stage analysis was conducted evaluating the influence of cluster assignment on heterogeneity in the risk estimates.

RESULTS

Associations between a 2-day (lag 0-1 days) moving average of PM_2.5 concentrations and non-accidental mortality were determined for each city. Estimated effects ranged from -3.2 to 5.1% with a pooled estimate of 0.33% (95% CI: 0.13, 0.53) increase in mortality per 10 μg/m³ increase in PM_2.5. The second stage analysis determined that cluster assignment was marginally significant in explaining the city-to-city heterogeneity. The health effects estimates in cities with older, smaller homes with less AC (Cluster 1) and cities with newer, smaller homes with a large prevalence of AC (Cluster 3) were significantly lower than the cluster consisting of cities with older, larger homes with a small percentage of AC.

CONCLUSIONS

This is the first study that attempted to examine whether multiple exposure factors could explain the heterogeneity in PM_2.5-mortality associations. The results of this study were found to explain a small portion (6%) of this heterogeneity.

Modifications of exposure to ambient particulate matter: Tackling bias in using ambient concentration as surrogate with particle infiltration factor and ambient exposure factor

Numerous epidemiological studies explored health risks attributed to outdoor particle pollution. However, a number of these studies routinely utilized ambient concentration as a surrogate for personal exposure to ambient particles. This simplification ignored the difference between indoor and outdoor concentrations of outdoor originated particles and may bias the estimate of particle-health associations. Intending to avoid the bias, particle infiltration factor (F_inf), which describes the penetration of outdoor particles in indoor environment, and ambient exposure factor (α), which represents the fraction of outdoor particles people are truly exposed to, are utilized as modification factors to modify outdoor particle concentration. In this study, the probabilistic distributions of annually-averaged and seasonally-averaged F_inf and α were assessed for residences and residents in Beijing. F_inf of a single residence and α of an individual was estimated based on the mechanisms governing particle outdoor-to-indoor migration and human time-activity pattern. With this as the core deterministic model, probabilistic distributions of F_inf and α were estimated via Monte Carlo Simulation. Annually-averaged F_inf of PM_2.5 and PM₁₀ for residences in Beijing tended to be log-normally distributed as lnN(-0.74,0.14) and lnN(-0.94,0.15) with geometric mean value as 0.47 and 0.39, respectively. Annually-averaged α of PM_2.5 and PM₁₀ for Beijing residents also tended to be log-normally distributed as lnN(-0.59,0.12) and lnN(-0.73,0.13) with geometric mean value as 0.55 and 0.48, respectively. As for seasonally-averaged results, F_inf and α of PM_2.5 and PM₁₀ were largest in summer and smallest in winter. The obvious difference between these modification factors and unity suggested that modifications of ambient particle concentration need to be considered in epidemiological studies to avoid misclassifications of personal exposure to ambient particles. Moreover, considering the inter-individual difference of F_inf and α may lead to a brand new perspective of particle-health associations in further epidemiological study.

Air quality co-benefits of subnational carbon policies

To mitigate climate change, governments ranging from city to multi-national have adopted greenhouse gas (GHG) emissions reduction targets. While the location of GHG reductions does not affect their climate benefits, it can impact human health benefits associated with co-emitted pollutants. Here, an advanced modeling framework is used to explore how subnational level GHG targets influence air pollutant co-benefits from ground level ozone and fine particulate matter. Two carbon policy scenarios are analyzed, each reducing the same total amount of GHG emissions in the Northeast US: an economy-wide Cap and Trade (CAT) program reducing emissions from all sectors of the economy, and a Clean Energy Standard (CES) reducing emissions from the electricity sector only. Results suggest that a regional CES policy will cost about 10 times more than a CAT policy. Despite having the same regional targets in the Northeast, carbon leakage to non-capped regions varies between policies. Consequently, a regional CAT policy will result in national carbon reductions that are over six times greater than the carbon reduced by the CES in 2030. Monetized regional human health benefits of the CAT and CES policies are 844% and 185% of the costs of each policy, respectively. Benefits for both policies are thus estimated to exceed their costs in the Northeast US. The estimated value of human health co-benefits associated with air pollution reductions for the CES scenario is two times that of the CAT scenario.

IMPLICATIONS

In this research, an advanced modeling framework is used to determine the potential impacts of regional carbon policies on air pollution co-benefits associated with ground level ozone and fine particulate matter. Study results show that spatially heterogeneous GHG policies have the potential to create areas of air pollution dis-benefit. It is also shown that monetized human health benefits within the area covered by policy may be larger than the model estimated cost of the policy. These findings are of particular interest both as U.S. states work to develop plans to meet state-level carbon emissions reduction targets set by the EPA through the Clean Power Plan, and in the absence of comprehensive national carbon policy.

Summer indoor heat exposure and respiratory and cardiovascular distress calls in New York City, NY, U.S

Most extreme heat studies relate outdoor weather conditions to human morbidity and mortality. In developed nations, individuals spend ~90% of their time indoors. This pilot study investigated the indoor environments of people receiving emergency medical care in New York City, NY, U.S., from July to August 2013. The first objective was to determine the relative influence of outdoor conditions as well as patient characteristics and neighborhood sociodemographics on indoor temperature and specific humidity (N = 764). The second objective was to determine whether cardiovascular or respiratory cases experience hotter and more humid indoor conditions as compared to controls. Paramedics carried portable sensors into buildings where patients received care to passively monitor indoor temperature and humidity. The case-control study compared 338 respiratory cases, 291 cardiovascular cases, and 471 controls. Intuitively, warmer and sunnier outdoor conditions increased indoor temperatures. Older patients who received emergency care tended to occupy warmer buildings. Indoor-specific humidity levels quickly adjusted to outdoor conditions. Indoor heat and humidity exposure above a 26 °C threshold increased (OR: 1.63, 95% CI: 0.98-2.68, P = 0.056), but not significantly, the proportion of respiratory cases. Indoor heat exposures were similar between cardiovascular cases and controls.

Air Pollution Exposure Model for Individuals (EMI) in Health Studies: Evaluation for Ambient PM2.5 in Central North Carolina

Air pollution health studies of fine particulate matter (diameter ≤2.5 μm, PM2.5) often use outdoor concentrations as exposure surrogates. Failure to account for variability of indoor infiltration of ambient PM2.5 and time indoors can induce exposure errors. We developed and evaluated an exposure model for individuals (EMI), which predicts five tiers of individual-level exposure metrics for ambient PM2.5 using outdoor concentrations, questionnaires, weather, and time-location information. We linked a mechanistic air exchange rate (AER) model to a mass-balance PM2.5 infiltration model to predict residential AER (Tier 1), infiltration factors (Tier 2), indoor concentrations (Tier 3), personal exposure factors (Tier 4), and personal exposures (Tier 5) for ambient PM2.5. Using cross-validation, individual predictions were compared to 591 daily measurements from 31 homes (Tiers 1-3) and participants (Tiers 4-5) in central North Carolina. Median absolute differences were 39% (0.17 h(-1)) for Tier 1, 18% (0.10) for Tier 2, 20% (2.0 μg/m(3)) for Tier 3, 18% (0.10) for Tier 4, and 20% (1.8 μg/m(3)) for Tier 5. The capability of EMI could help reduce the uncertainty of ambient PM2.5 exposure metrics used in health studies.

Comparison of Highly Resolved Model-Based Exposure Metrics for Traffic-Related Air Pollutants to Support Environmental Health Studies

Human exposure to air pollution in many studies is represented by ambient concentrations from space-time kriging of observed values. Space-time kriging techniques based on a limited number of ambient monitors may fail to capture the concentration from local sources. Further, because people spend more time indoors, using ambient concentration to represent exposure may cause error. To quantify the associated exposure error, we computed a series of six different hourly-based exposure metrics at 16,095 Census blocks of three Counties in North Carolina for CO, NO_x, PM_2.5, and elemental carbon (EC) during 2012. These metrics include ambient background concentration from space-time ordinary kriging (STOK), ambient on-road concentration from the Research LINE source dispersion model (R-LINE), a hybrid concentration combining STOK and R-LINE, and their associated indoor concentrations from an indoor infiltration mass balance model. Using a hybrid-based indoor concentration as the standard, the comparison showed that outdoor STOK metrics yielded large error at both population (67% to 93%) and individual level (average bias between −10% to 95%). For pollutants with significant contribution from on-road emission (EC and NO_x), the on-road based indoor metric performs the best at the population level (error less than 52%). At the individual level, however, the STOK-based indoor concentration performs the best (average bias below 30%). For PM_2.5, due to the relatively low contribution from on-road emission (7%), STOK-based indoor metric performs the best at both population (error below 40%) and individual level (error below 25%). The results of the study will help future epidemiology studies to select appropriate exposure metric and reduce potential bias in exposure characterization.

Modeling spatial and temporal variability of residential air exchange rates for the Near-Road Exposures and Effects of Urban Air Pollutants Study (NEXUS)

Air pollution health studies often use outdoor concentrations as exposure surrogates. Failure to account for variability of residential infiltration of outdoor pollutants can induce exposure errors and lead to bias and incorrect confidence intervals in health effect estimates. The residential air exchange rate (AER), which is the rate of exchange of indoor air with outdoor air, is an important determinant for house-to-house (spatial) and temporal variations of air pollution infiltration. Our goal was to evaluate and apply mechanistic models to predict AERs for 213 homes in the Near-Road Exposures and Effects of Urban Air Pollutants Study (NEXUS), a cohort study of traffic-related air pollution exposures and respiratory effects in asthmatic children living near major roads in Detroit, Michigan. We used a previously developed model (LBL), which predicts AER from meteorology and questionnaire data on building characteristics related to air leakage, and an extended version of this model (LBLX) that includes natural ventilation from open windows. As a critical and novel aspect of our AER modeling approach, we performed a cross validation, which included both parameter estimation (i.e., model calibration) and model evaluation, based on daily AER measurements from a subset of 24 study homes on five consecutive days during two seasons. The measured AER varied between 0.09 and 3.48 h(-1) with a median of 0.64 h(-1). For the individual model-predicted and measured AER, the median absolute difference was 29% (0.19 h‑1) for both the LBL and LBLX models. The LBL and LBLX models predicted 59% and 61% of the variance in the AER, respectively. Daily AER predictions for all 213 homes during the three year study (2010-2012) showed considerable house-to-house variations from building leakage differences, and temporal variations from outdoor temperature and wind speed fluctuations. Using this novel approach, NEXUS will be one of the first epidemiology studies to apply calibrated and home-specific AER models, and to include the spatial and temporal variations of AER for over 200 individual homes across multiple years into an exposure assessment in support of improving risk estimates.

An empirical assessment of exposure measurement error and effect attenuation in bipollutant epidemiologic models

BACKGROUND

Using multipollutant models to understand combined health effects of exposure to multiple pollutants is becoming more common. However, complex relationships between pollutants and differing degrees of exposure error across pollutants can make health effect estimates from multipollutant models difficult to interpret.

OBJECTIVES

We aimed to quantify relationships between multiple pollutants and their associated exposure errors across metrics of exposure and to use empirical values to evaluate potential attenuation of coefficients in epidemiologic models.

METHODS

We used three daily exposure metrics (central-site measurements, air quality model estimates, and population exposure model estimates) for 193 ZIP codes in the Atlanta, Georgia, metropolitan area from 1999 through 2002 for PM2.5 and its components (EC and SO4), as well as O3, CO, and NOx, to construct three types of exposure error: δspatial (comparing air quality model estimates to central-site measurements), δpopulation (comparing population exposure model estimates to air quality model estimates), and δtotal (comparing population exposure model estimates to central-site measurements). We compared exposure metrics and exposure errors within and across pollutants and derived attenuation factors (ratio of observed to true coefficient for pollutant of interest) for single- and bipollutant model coefficients.

RESULTS

Pollutant concentrations and their exposure errors were moderately to highly correlated (typically, > 0.5), especially for CO, NOx, and EC (i.e., "local" pollutants); correlations differed across exposure metrics and types of exposure error. Spatial variability was evident, with variance of exposure error for local pollutants ranging from 0.25 to 0.83 for δspatial and δtotal. The attenuation of model coefficients in single- and bipollutant epidemiologic models relative to the true value differed across types of exposure error, pollutants, and space.

CONCLUSIONS

Under a classical exposure-error framework, attenuation may be substantial for local pollutants as a result of δspatial and δtotal with true coefficients reduced by a factor typically < 0.6 (results varied for δpopulation and regional pollutants).

Who is more affected by ozone pollution? A systematic review and meta-analysis

Ozone is associated with adverse health; however, less is known about vulnerable/sensitive populations, which we refer to as sensitive populations. We systematically reviewed epidemiologic evidence (1988-2013) regarding sensitivity to mortality or hospital admission from short-term ozone exposure. We performed meta-analysis for overall associations by age and sex; assessed publication bias; and qualitatively assessed sensitivity to socioeconomic indicators, race/ethnicity, and air conditioning. The search identified 2,091 unique papers, with 167 meeting inclusion criteria (73 on mortality and 96 on hospitalizations and emergency department visits, including 2 examining both mortality and hospitalizations). The strongest evidence for ozone sensitivity was for age. Per 10-parts per billion increase in daily 8-hour ozone concentration, mortality risk for younger persons, at 0.60% (95% confidence interval (CI): 0.40, 0.80), was statistically lower than that for older persons, at 1.27% (95% CI: 0.76, 1.78). Findings adjusted for publication bias were similar. Limited/suggestive evidence was found for higher associations among women; mortality risks were 0.39% (95% CI: -0.22, 1.00) higher than those for men. We identified strong evidence for higher associations with unemployment or lower occupational status and weak evidence of sensitivity for racial/ethnic minorities and persons with low education, in poverty, or without central air conditioning. Findings show that some populations, especially the elderly, are particularly sensitive to short-term ozone exposure.