2013 Progress Report: Intra-Urban Variation of Air Pollution and Cardiovascular Health EffectsEPA Grant Number: R834898
Title: Intra-Urban Variation of Air Pollution and Cardiovascular Health Effects
Investigators: Ito, Kazuhiko , Chen, Yeh-hsin , Matte, Thomas , Ross, Zev
Current Investigators: Ito, Kazuhiko , Clougherty, Jane E. , Matte, Thomas , Ross, Zev
Institution: New York City Department of Health and Mental Hygiene , ZevRoss Spatial Analysis
Current Institution: New York University School of Medicine , Hunter College , New York University , University of Pittsburgh
EPA Project Officer: Ilacqua, Vito
Project Period: April 1, 2011 through March 31, 2013 (Extended to December 31, 2015)
Project Period Covered by this Report: April 1, 2013 through March 31,2014
Project Amount: $299,998
RFA: Exploring New Air Pollution Health Effects Links in Existing Datasets (2010) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Health Effects , Air
The objectives of this project are to: (1) determine the impacts of air pollution and weather effects on the cardiovascular health outcomes available at NYCDOHMH including cardiovascular emergency department syndrome data, hospitalizations, and mortality; (2) determine the effect of modification of the cardiovascular effects by intra-urban variation of combustion sources as measured by the NYC Community Air Survey (NYCCAS); and (3) determine the effect of modification of the cardiovascular effects of air pollution by socio-economic status. We have recently developed a cardiovascular ED syndrome indicator that is useful in determining near-real time impacts of weather and air pollution. NYC residents are exposed to multiple air pollutants coming from a variety of combustion sources, including transported secondary aerosols, local sources including traffic, building space-heating, and oil burning from ships in nearby ports. NYC residents also reflect a wide range of health and socio-economic status, and therefore likely present a range of susceptibility indicators associated with neighborhood characteristics. Thus, this study takes advantage of the unique databases that have been recently developed to determine the cardiovascular effects of air pollution in unique environmental and population settings of NYC to answer the relevant research questions.
Analysis of PM2.5 chemical constituents collected in NYCCAS:
We have completed land-use regression (LUR) analysis of 15 key elements from PM2.5 chemical constituents data collected 150 NYCCAS sites during the first year of NYCCAS measurements. The 15 key elements are (in elemental symbol): Na, Al, Si, S, K, Ca, Ti, V, Mn, Fe, Ni, Cu, Zn, Br, and Pb. Important emission indicators to predict spatial variation of these chemical constituents include: residual oil burning (Ni, Zn, Ca); near-road traffic (Cu, Fe, Ti); and ship emissions (V). A factor analysis including the NYCCAS gaseous pollutants confirmed the pattern of relationships between the elements and source types observed in the LUR results. We are preparing a manuscript to report these results and plan to submit this summer (2014).
Based on the LUR results mentioned above, we are now able to estimate concentrations of these PM2.5 chemical constituents at any location or an area in New York City, because we have computed values of emission indicators (i.e., predictor variables in the LUR models) at 100 m x 100 m grids within the city. For example, using the LUR model developed in this project, we estimated spatial exposures of Ni and Cu for a cohort of ~300,000 births that occurred in New York City during 2008-2010 (collaboration with Brown University to investigate the effects of air pollution on birth outcomes; see below). We also developed a routine to compute zip-code average values of any pollutant with a LUR model. We have computed zip-code level average NYCCAS core pollutants (PM2.5, black carbon, NO2, SO2, and O3) and Ni and expect to complete computing zip-code level averages for the rest of the constituents over this summer (2014).
Analysis of citywide air pollution and the health outcome data:
In last year’s progress report, we reported that we analyzed the relationship between the citywide daily average PM2.5 and CVD ED visits syndrome, hospitalizations, and mortality during the years 2002-2010 and found that the seasonal pattern, lag, and magnitude of associations were comparable to those we had published previously (years 2000-2006 in Ito et al. 2011; 2004-2006 in Mathes et al., 2011). Since then, we compiled and computed the citywide daily concentrations of NO2, SO2, CO, and O3 up to 2012, using available data from regulatory monitors. We observed that PM2.5 and these gaseous pollutants, with the exception of ozone, exhibited a monotonic decline in the mean levels the past 10 years. To determine if the declining pollution levels have influenced the short-term effects of air pollution, we analyzed these air pollutants’ associations with natural-cause mortality over a 13-year period divided into three periods: (1) 2000-2003; (2) 2004-2007; and (3) 2008-2012. The magnitude of reduction in air pollution across these periods ranged from ~20% (NO2) to over 50% (SO2 and CO). In time-series analysis, the significance of association with mortality was diminished in the most recent period for CO, NO2, SO2 and O3, but PM2.5 risk estimates and their significance have not declined. We are presenting the results at the 2014 ISEE conference (Ito et al.), and a manuscript is in preparation. A parallel analysis using CVD and respiratory hospitalizations will be conducted in the fall.
We are currently analyzing specific causes of CVD hospitalizations (hypertension, myocardial infarction, ischemic heart disease, dysrhythmia, heart failure, and stroke) for their associations with air pollution to ensure that the evaluation of the impacts of intra-urban variation of air pollutants and socio-economic factors can focus on the most relevant CVD sub-categories. We are finding that some sub-categories (e.g., heart failure) of CVD hospitalizations are more strongly associated with air pollutants than other categories.
Investigating the effects of weather on CVD and other outcomes:
Estimating the air pollution effects requires an adequate adjustment for weather variables, particularly temperature, as its associations with the health outcomes are often stronger than those for air pollution, making it an important confounder. The impacts of weather on CVD and other health outcomes are also of interest to us for policy reasons at NYCDOHMH. Therefore, for all the health outcome models developed for air pollution effects, we are also evaluating the impact of temperature (heat-wave and cold, as well as mild range temperature). In the past year, we conducted two sets of analyses, in parallel to the analysis of air pollution effects, as described below.
We estimated annual excess natural cause deaths associated with heat waves in New York City during the months of May through September from 1997-2013, accounting for seasonal and secular trends and multi-day lagged effects of high temperature on risk. Heat waves were defined retrospectively, applying National Weather Service criteria for issuing heat advisories in New York City to computed maximum daily heat index values from LaGuardia Airport. Percent excess deaths on heat-wave days and the three days following each heat wave were estimated in a Poisson time-series model using an unconstrained distributed lag for days 0-3, and adjusting for within-season trends and day of week, and accommodating over-dispersion. The mean number of annual heat wave days was 7.8 (range 0-19) and a random-effects estimate of annual excess deaths attributable to heat waves were 94.0 (95% CI 57, 131), several-fold greater than hyperthermia deaths. An abstract of this analysis was accepted for a poster presentation at 2014 ISEE.
Using the Revised Digital Forecast (RDF) dataset obtained from the National Climatic Data Center (NCDC) for LaGuardia Airport during the warm season (May through September) for years 2004-2011, we are conducting analyses to: (1) investigate the correlations of forecast weather data with observed weather measurements and the accuracy of the heat advisory and excessive heat watch or warning identified by forecast weather products for predicting extreme heat events occur in the observations; and (2) assess the consistency of the associations between extreme heat events and natural-cause mortality based on observed and forecast weather products. We do not report specific results here (because that requires a within-agency review), but we have produced sufficient results for a manuscript.
Collaboration with other researchers using the data and/or methods from this project:
Through our collaborations with other within-agency researchers as well as outside investigators from Columbia University (climate change) and Brown University (a birth outcome study, for which we have a separate approved IRB protocol), we have synergistic research efforts using the data or methods developed in this EPA project. These include: (1) an analysis of daily spring tree pollen (measured by Fordham and Columbia investigators) and asthma emergency department visits and over-the-counter allergy medication sales (Ito et al., a manuscript has been written and reviewed by co-authors; to be submitted shortly); (2) an analysis of birth-weight and PM2.5 chemical constituents (based on the LUR model developed from this EPA project), and emission indicators in New York City (Ito et al., a manuscript in preparation; an abstract accepted for an oral presentation at 2014 ISEE); and (3) an analysis of the short-term effects of weather and air pollution on spontaneous term and preterm birth and birth with premature rupture of the fetal membranes (the birth outcome study; McAlexander et al., a manuscript in preparation; an abstract accepted for a poster presentation at 2014 ISEE).
The following activities are planned for the coming months and year: (1) completing and submitting manuscripts with the results from analyses conducted to date; (2) spatially stratified time-series (and case-crossover) analysis of CVD outcomes using zip code level PM2.5 chemical constituents and socio-economic variables; (3) analysis of the linked mortality-hospitalizations data to investigate the difference in PM2.5 impacts on CVD hospitalizations and mortality; and (4) parallel analyses of asthma morbidity outcomes using the methods applied to the CVD outcomes, as time allows.