2002: Urban Measurement of Aerosol an Asthma Project

Team members: Autumn Anderson, DeVoia Stewart, Jeff Waksman, Kristle Simms, Ankita Patel, Robert Gondolfo
Science Advisors: Barbara Carlson, Brian Cairns, Jacek Chowdhary

Introduction

UrbanMAAP is a project designed to examine the relationship between human health and the environment on a variety of space and time scales. Our current research is focused on understanding the connection between the dramatic global increase in the prevalence of asthma and changes in atmospheric composition. Numerous theories exist to explain the dramatic increase in asthma. According to these theories, asthma is increasing because of changes in indoor air quality; change in allergens and exposure to them related to lifestyle changes (e.g., increased exposure to mites due to increase in use of carpets); changes in hygiene (our environments are too clean and our immune systems are therefore underdeveloped); increased exposure to childhood vaccines; and changes in our environment. While air pollution is frequently rejected based on environmental data that indicate that the air is getting cleaner, numerous studies have found links between asthma and air pollution while numerous other studies have concluded that there is no link between asthma and air pollution. Obviously, they can't all be right.

Complicating the situation is the lack of data. Asthma prevalence has frequently been studies using asthma mortality or hospitalization data as a proxy for asthma prevalence. However, these data sources most likely contain a socioeconomic bias and therefore provided a distorted picture of asthma prevalence. Similarly, air quality is currently measured by the Environmental Protection Agency (EPA) aimed at certifying compliance to the Clean Air Acts and therefore not set up to measure exposure. In order to get around the lack of data, UrbanMAAP involves both data collection and data analysis. As in the past, we will continue to use an Asthma Survey to address the prevalence of asthma and handheld sunphotometer measurements to determine the aerosol load. However, this summer we will also investigate the feasibility of making ground-based measurements of some aerosol precursors as well as measuring some indicators of traffic pollution.

Background for 2002 Research Projects

Currently a comprehensive data set for asthma does not exist. Much of what is "known: about asthma has been determined from asthma mortality and hospitalization data that have been used as a proxy for asthma prevalence. There are several potential problems with these data sources namely the mortality and hospitalization data may be biased by socioeconomic differences in the availability of healthcare and therefore may over emphasize urban areas where the larger populations produce meaningful statistics. While these data sets may be biased, they nonetheless reveal a disturbing pattern of increases in asthma hospitalization and mortality despite advances in medical treatment.

The heightened awareness of the global asthma problem has led to a number of research/surveillance projects aimed at providing a more accurate picture of asthma and the factors that influence the incidence of asthma and are responsible for the dramatic increases in asthma over the past two decades. While there is no single accepted theory to explain the increase in asthma, the fact that it is a global problem suggests to us that it is environmental in nature. Several studies have rejected the idea that pollution is responsible. Two of the most frequently cited arguments against pollution are: (1) that the atmosphere has gotten cleaner over the past two decades and (2) studies comparing the prevalence of asthma in eastern and western European countries have found that asthma rates are lower in the more polluted eastern countries. However, when we look at the individual pollutants we find that while on a whole the atmosphere has gotten cleaner (e.g., lower levels of sulfur dioxide (SO2), carbon monoxide (CO) and lower levels of large particulates (PM10), levels of nitrogen oxides have increased. Coupled with this increase in the levels of nitrogen oxides has been a change in the chemical composition of the atmospheric aerosols, since there is less sulfur dioxide to produce sulfate aerosols, the nitrate fraction of the aerosols has increased. We believe that nitrogen oxides and nitrate aerosols are playing a role in the increase in asthma. But where do we go from here?

First while beyond the scope of any of the UrbanMAAP research projects we need a mechanism. Several studies have now been published that show that nitrogen oxides can trigger asthma. In addition, nitrogen oxides lower the threshold for allergic reactions. Thus, the concentration of allergen that is needed to trigger an allergic reaction is lower in the presence of increasing amounts of nitrogen oxides. There are additional physiological studies that have measured respiration and have found excess nitrogen oxides and lower pH values in the exhalations of asthmatics. These studies strongly suggest that nitrogen oxides may be playing a role. So what next?

This is where UrbanMAAP comes in. In this project we are looking for a relationship between asthma and pollution (namely nitrogen oxides and aerosols). If our hypothesis is correct then we would expect to see higher levels of asthma associated with higher levels of nitrogen oxides. While on the surface that seems to be a straightforward research task there are several confounding factors. First, there is the issue of what do we use as an asthma data set. Second, nitrogen oxide measurements are not available for all locations and certainly not available on the spatial scale that we need them. Third, nitrogen oxides are chemically reactive in the atmosphere.

Starting with the asthma data, before we began UrbanMAAP we analyzed the mortality data for the New York City metropolitan area (shown below). Consistent with the hospitalization data we found higher amounts of asthma in the Bronx and lower amounts of asthma in Queens. We also found that the death rate is increasing faster in the Bronx than the statewide average. In fact, the Bronx, Manhattan, and Brooklyn all have mortality rates considerably higher than the state average, so is asthma an urban problem? At the same time we found a relationship between the number of asthma hospitalizations and the nitrate fraction of inhalable particulates. This relationship predicts a higher prevalence of asthma in Manhattan than that based on hospitalization or mortality data.

Naturally, we thought that something had to be wrong with our relationship between asthma and nitrate fraction. But, is it possible that something is wrong with using mortality and/or hospitalization data as a proxy for asthma prevalence?

To address this question, we created the first UrbanMAAP survey. This survey was analyzed to determine the prevalence of asthma in the New York metropolitan area. Based on the survey data, we found that the prevalence of asthma is actually higher in Manhattan than it is in the Bronx. But when it comes to asthma burden (defined to be the number of asthma attacks divided by the number of people with asthma) we found the asthma burden to be higher in the Bronx than in Manhattan. This finding suggests that there are regional differences in asthma management and could be used by policy makers to develop better asthma management plans.

What then is responsible for the differences between the mortality and hospitalization data and the UrbanMAAP survey? For starters, we hypothesized that the hospitalization data may contain a socioeconomic bias. If, for example, more people in Manhattan visit their own doctor for asthma treatment and have their asthma effectively managed then they would not be going to the hospital for treatment. So, we added questions to the UrbanMAAP Asthma Survey to provide some insights into this question. To get at availability of medical care and type of medical care we asked the respondents to tell us where they are most likely to go for treatment when they have an asthma attack. While to get some information about how socioeconomic factors influence asthma statistics we asked the respondents to tell us their zip code and the cost of their school lunch. In the New York public schools subsidized lunches are available. This past summer, student participants analyzed the survey data and compared the type of lunch payment with zipcode level income information (available from the census data) and found a relationship between income and type of lunch payment. In addition, the students found a relation between income and type of medical care with higher income respondents more likely to have their asthma treated by a personal physician than lower income respondents. Finally, the survey data support the higher prevalence of asthma in Manhattan found in the analysis of our first survey and suggest that the hospitalization data underestimate asthma prevalence in Manhattan because a greater percentage of the people in Manhattan visit a personal physician for treatment than in the Bronx.

This is where you come in, while we have already been able to answer a number of interesting questions using the UrbanMAAP survey data, there are many more questions that remain to be answered.

Asthma Research Projects

This summer, we are going to focus on several research questions that aim to use the space-time variability of asthma and provide some insights into the factors responsible for the increase in asthma prevalence.

Investigation of Time Variability of Asthma

As part of this summer's project, we are going to examine the temporal trend in asthma using published data. We will begin by investigating the annual variation in asthma. Our primary interest is to determine whether or not the asthma rates have continued to rise. For this task, we will analyze mortality data for NY State, NYC, Washington DC, New Jersey, and California.

Basic questions:

• How do the trends at these sites compare?
• What do they suggest about asthma?
• How do these trends compare with air quality trends for these sites?

In addition, to analyzing interannual variations in asthma mortality, we are also going to use the survey data to investigate the seasonality of asthma.

Basic questions:

• In what season are asthma attacks most common?
• How does this compare with the seasonal variation in pollution concentrations?
• Does this result depend on when the survey was administered?
• Do the survey respondents give consistent answers to this question?

Questions that we should answer:

• Has the trend in asthma mortality changed? (Are fewer people dying of asthma?)
• Do any pollutants have an increasing trend?
• Do interannual variations tell us anything? (Are the interannual variations linked to other illness (e.g. colds/flu)

Investigation of Spatial Variability of Asthma

The primary question we seek to address is the determination of whether or not asthma is an urban problem. Based on the analysis of the mortality data, it is clear that Urban areas have an increased incidence of asthma mortality.

Basic questions:

• Based on the survey data, which location has the highest asthma prevalence and which has the lowest?
• What is different about these places? (Air quality, population, etc)
• How does this compare with the mortality data?
• Are there non-urban hot spots? (examine county level asthma hospitalization data)
• What do other surveys say?

Questions that we should answer:

• Is asthma an urban problem?
• Is asthma a traffic problem?

Investigation of Asthma Demographics

In these investigations we will examine racial and gender differences in the prevalence of asthma. As we have learned from our readings, the prevalence of asthma seems to depend both on gender and race. Focusing first on racial differences, we will use the survey data to investigate whether or not racial differences exist. Previous investigations have found that asthma prevalence varies with race but these studies have not considered that the dependence on race that they find might really be a dependence on space (i.e., where the people live). In this investigation, we will analyze the Asthma survey data to determine if there are racial and gender differences in the prevalence of asthma.

Basic questions:

• Using the NYC data, are there racial differences in asthma prevalence?
• How does that picture change we you analyze all of the survey data?
• What does this say about asthma and race? (i.e., does asthma depend on race or where the people live?).

Questions that we should answer:

• Does asthma prevalence depend on race?
• Does asthma prevalence depend on gender?

Investigation of Socio-economic Factors

In these investigations we will seek to identify the socio-economic factors that influence our perception of asthma and investigate whether or not socio-economic factors are responsible for the differences between the asthma survey data and hospitalization and mortality data.

Basic questions:

• Based on mortality data, which borough has the highest prevalence of asthma?
• Based on survey data:
  • Which borough has the highest prevalence of asthma?
  • Are there borough-level differences in access to medical care?
  • Which borough has the highest percentage of respondents that go to the ER for treatment?
  • Which borough has the highest asthma burden (total number of attacks/number of people with asthma)?
  • How does asthma prevalence change with income?

Question that we should answer:

• oes poverty cause asthma?

Investigation of Other Factors that Influence Asthma Prevalence

This investigation builds on the spatial differences revealed above. If there are spatial differences in asthma prevalence and asthma burden, what are the factors that contribute to these differences?

Basic questions:

• Are there spatial differences in asthma prevalence and asthma burden?
• Do these differences reflect the spatial pattern of the availability of medical care?
• Do these differences reflect differences in asthma management/control due to parents' level of education?

Question that we should answer:

• What factors contribute to asthma management?

Investigation of Asthma and Allergy

This investigation begins to examine the relationship between asthma and allergy. For this investigation, feel free to use additional sources of information.

Basic questions:

• What is the prevalence of asthma and allergy?
• Do asthma sufferers have a higher prevalence of allergy than the general public?
• What are most of the respondents allergic to?
• Does their allergen have a seasonal cycle?
• Is the allergen pattern consistent with the seasonal pattern of asthma?

Questions we should answer:

• What is our survey estimates prevalence of allergic asthma?
• Are the allergens consistent with the seasonal cycle of asthma? Asthma burden?

Other Asthma Investigations

Feel free to suggest other asthma investigations.

Sunphotometer/Environmental Measurement Projects

This section of the UrbanMAAP project focuses on the analysis of trends in several environmental parameters. As discussed briefly above, one of the difficulties encountered in epidemiological studies is determining a persons exposure to a pollutant since most pollution monitoring sites have been established for compliance with clean-air regulations. To partially address this issue, we have been investigating the utility of hand-held sunphotometer measurements. This summer we are going to focus on characterizing some details of the instrument's design and response as well as investigate other types of measurements that we could make.

Investigation of the Spectral Response of the Sunphotometer

In order for the sunphotometer measurements to be useful, the spectral response of the sunphotometer has to be temporally stable. In order to examine this, we will be making measurements of the spectral response of the LED in the CCNY laboratory.

Basic research questions:

• What is the spectral response of the LED?
• Does the spectral response change with temperature and if so by how much?

Question that we should answer:

• What is the spectral response of the LED and how does it change with changing temperature?

Investigation of Sunphotometer Measurement Strategy

Since the sunphotometer measurements are analyzed using Langley Regression, plots of air mass versus the log of the voltages, how does the optical depth determination depend on the sampling of different air masses? In this investigation, we will use MFRSR data obtained under a variety of atmospheric conditions.

Basic research questions:

• Does the accuracy of the optical depth retrieval depend on temporal sampling? Is there a minimum number of observations that are required to obtain an accurate estimate of aerosol optical depth?
• Does the number of observations depend on optical depth?
• Does the number depend on optical depth stability?

Question that we should answer:

• What is the optimal observation strategy?

Investigation of Sunphotometer Design

We have in the past tried several different versions of the hand-held sunphotometer. As part of this investigation, we should compare the different instrument designs and recommend a design for future IOPs.

Basic research questions:

• Which instrument design provides the best (most accurate and most stable) measurements of aerosol optical depth? Include in this comparison the GLOBE 2 LED version.

Question that we should answer:

• Which sunphotometer should be used for future IOPs?

Investigation of Other Measurements

Since we believe that the asthma increases are due to changing levels of nitrogen oxides, are there other measurements that we could be making that would allow us to more directly examine these relationships?

Basic research questions:

• What other measurements could we be making?
• What instruments are available to make those measurements?

Question that we should answer:

• What other measurements should we be making?