University of Houston-Clear Lake    Search  Home  Login  
University of Houston-Clear Lake  
Welcome to EIH      
Environmental Institute of Houston > Publications > Annual Reports > 2003 Annual Report > Understanding Spatial Distribution of Asthma in the Harris County Area

Understanding Spatial Distribution of Asthma in the Harris County Area

DURING THE PAST SEVERAL YEARS, ASTHMA HAS BECOME a priority public health concern, especially among children in the United States. More Americans than ever before are suffering from asthma and it has become one of the most common and costly illness in this country. Besides AIDS and tuberculosis, asthma is the only chronic disease with an increasing death rate. Each day, 14 Americans die from asthma. Asthma death rates increased 58 percent from 1979 to 1992 and the death rate for children of 19 years and younger increased by 78 percent between 1980 and 1993. The cost of asthma in 1998 was estimated to be $11.3 billion. The purpose of this study is to determine whether income, race, and air pollution may influence the prevalence of asthma among the people living in the general area of Harris County, Texas.

Data Collection
In this research, we mainly used five sets of data and they are input either in the form of maps or tables: (a) asthma data for the year 1999 that provides asthma rate per thousand of population per zip code in the Harris County area;1 (b) income data that has median incomes per household for each zip code;2 (c) Race table that consists of populations of Black, White and other races for each zip code;2 (d) the zip code shape file of the Harris County area;5 and (e) air pollution data per zip code.2,3 Note that we have used Ozone and NO2 readings from six monitoring stations as there are insufficient data due to lack of monitoring stations in most of the areas of Harris County area.

Data Analysis and Results
ArcGIS and Microsoft Excel tools were used to analyze the data. ArcGIS is a Geographic Information System (GIS) that can be used to analyze and display data. Determination of correlation for various input data and plotting of charts were done using MS Excel. GIS is basically a powerful computer mapping and analysis technology that allows large quantities of information to be viewed and analyzed within a geographic context. Using GIS, the scope of study can be limitless in the sense that it can be local, regional, countrywide, or worldwide. The maps act as a visual representation of data. In our study, a zip code shapefile was used to represent various zip codes of Harris County.

Under one dataframe, various layers of data have been added to make comparative studies. We had layers such as asthma, income, race, ozone, and etc. Zip code column was considered as the primary key attribute while joining different data tables. The tables were imported from MS Excel in the form of database (.dbf) file and records with only matching attributes were considered in the final dataset. Once the layers were created by joining different tables, the Symbology property of each layer was changed to a Graduated color ramp. This color classification was done in order to compare and analyze the layers representing different input dataset. For instance, to see whether there is any correlation between average income and asthma in Harris County, we created a new layer for income by joining the zip code table and income table and choosing zip code as the common attribute from both tables. Then a comparison was made with a similar layer of asthma by changing the Symbology property to a Graduated color ramp. While making the comparison, it is important that the color ramp and its classification should be selected in such a way that any two layers under consideration should be able to compare and analyze. Figures 1 and 2 are examples of map that were created using this method.

Using MS Excel, the team found the correlation between the median household income and asthma as –0.2726. This shows an inverse correlation in the sense that the asthma prevalence rate is higher in lower income areas. We also found the correlation between race and asthma (see Table 1). It shows that the rate of asthmatic people decreases in a White dominant locality, whereas it increases in a predominantly Black locality. There does not seem to be a good correlation between asthma and other races. Lack of detailed data of all races in the county prevents us from getting more accurate findings as to what is the exact relation between asthma and other races such as Hispanic, etc.

Table 1. Correlation of Coefficient for Asthma and Race

White & Asthma 0.5597
Black & Asthma -0.5412
Others & Asthma -0.0108

The correlation between ozone and asthma are shown in the Table 2. It shows that there is no direct correlation between asthma and highest one hour ozone data (1st_1hr). Similarly, we do not find any good correlation for the average one-hour (Avg_1hr) and the Day_1hr Ozone reading, the latter being the number of days in the year when 1-hour values are expected to exceed the 1-hour standard and 1st_8hr stands for the highest eight hour ozone data. However, the asthma rate is relatively influenced by Avg_8hr (average four highest eight hour readings) and AllYrAveOzone (all year average ozone). Though ‘0.243’ and ‘0.2688’ are not very high correlation coefficients, it is shown that asthma starts correlating with Ozone when the amount of Ozone sampling time increases. Lack of Ozone readings due to lack of monitoring stations in most of the Harris County area prevent us from getting desired results.

Table 2. Correlation Coefficient for Asthma and Ozone

1st_1hr -0.3635
Avg_1hr -0.1784
Day_1hr 0.0896
1st_8hr 0.011
Avg_8hr 0.243
Day_8hr -0.1208
Ozone2000Annual 0.122
AllYrAveOzone 0.2688

Nevertheless, further and more detailed study can be done if sufficient ozone data is provided. The correlation coefficient between the 1hr_Max of NO2 and Asthma is -0.377 and that between annual mean of NO2 and Asthma is -0.421. Both of them are negative and therefore have inverse relation between the prevalence of Asthma and NO2. (See Fig. 3.) The study could not provide a cause for this phenomenon. Once again, lack of data due to lack of monitoring stations in most parts of the Harris County area prevented us from getting concluding results.

This study found that in Harris County, the asthma prevalence among people decreases as income increases. The rate of asthma is comparatively less in a White neighborhood than in a Black area. We could not find any conclusive answer for other races like Hispanic, Asian, etc. Further research can be carried out for all the races once detailed data is provided. Ozone density seems to have a strong correlation with asthma; however, NO2 density seems to have an inverse relation with asthma. Lack of Ozone and NO2 readings due to lack of monitoring stations in most of the areas of Harris County area prevented us from making concluding results.

1Texas Health Care Information Council’s Texas Hospital Inpatient Discharge Public Use Data File.
U.S. Census Bureau. <>
3U.S. Environmental Protection Agency (EPA). "AirData: Access to Air Pollution Data." <>
4Texas Natural Resource Conservation Commission (TNRCC). <>
5Texas Natural Resources Information System (TNRIS). <>

Mohammad A. Rob, Ph.D. is an Assistant Professor of
Mathematics at the University of Houston-Clear Lake.
He can be reached at Jin Zhaohui is a
graduate student at the University of Houston-Clear Lake.

 Print Print
2003 Annual Report      
Click to enlarge
Click to enlarge

Figure 1. Map of Asthma by Zip Code 
Click to enlarge
Click to enlarge

Figure 2. Map of Income Versus Zip Code 
Click to enlarge
Click to enlarge

Figure 3. NO2 Annual Mean and Asthma 
UHCL footer   

  Accessibility  |   Best Viewed  |   Clery Act  |   Compact with Texans  |   Emergency Information  |   Maps & Directions  
  State Website Linking and Privacy Policy  |   UH System  |   Public Information  |   Fraud Reporting  |   MySafeCampus