Final Poster & Reflection

My poster is finally complete and I am relieved to have completed all of the coursework for this semester!  In my previous course project, I had also wanted to compare a few different variables in a way where each layer could match up on one map, but unfamiliar with spatial analysis and weighted systems I instead created a poster with several different maps each only looking at the relationship between two variables.  I am glad that this course project gave me the opportunity to practice using my spatial analysis tools, and now feel more confident using ArcGIS and approaching different/more complicated methods necessary for the creation of effective and professional maps.  After looking at my analysis, I do not think that I have enough information to make a definitive conclusion about different variables that may impact people seeking asthma treatment, but I can say confidently that lower income areas that are further distances from hospitals are definitely at a greater risk than those living in more affluent and populated areas (even if there is more ozone that could irritate asthma symptoms!).  After completing two course projects over the course of a school year both looking at asthma within California, I would be interested to look in other areas of the United States to see if there are any similarities or surprises in the data.  I have attached the final version of my final undergraduate GIS map poster! (Sorry if it's blurry I had trouble uploading the PDF so I had to use a screenshot).

Update 5: Spatial Analysis & Weighted Overlay

Now that I have reclassified my rasters, I will assign each of them a weight based on how important I feel each category is for people who may suffer from asthma or may be at risk for not receiving treatment for their asthma.  I rated the distance to the hospital as the most important variable because if the condition is serious enough or life threatening, people are going to seek medical attention regardless of whether or not they have health insurance.  I weight I assigned to the hospitals was (0.35).  At a close second, I rated income as the second most important factor (0.3) because those who live below the poverty line may not be able to afford to seek medical help.  Rated third is ozone exceedance (0.25) because although ozone is an air pollutant that can aggravate asthma symptoms, I felt that income and hospitals have a greater impact on asthma at this time.  Lastly, is population density (0.1).  I chose to include population density because densely populated areas are more likely to have more hospitals, so although it is a piece of the puzzle, I did not feel that it was significant enough to give it a weight that more closely matched with the other three components.  After using the raster calculator to multiply the values,  I was able to produce a draft final output raster which I have attached below. 

Update 4: Reclassification of Rasters

Now that I have been able to convert all of my shapefiles to rasters, I will begin to reclassify each of the four variables so that I can assign each of them weights and eventually use a raster calculator to find areas within the state of California where asthma sufferers are at higher and lower risks of complications with their condition.  For ozone, areas that have lower numbers of ozone exceedance will be favored over those with more exceedances because higher levels of ozone may aggravate asthma symptoms due to poorer air quality.  Areas of higher income will be favored over areas of lower incomes because people with more money are more likely to be able to afford health insurances or the hospital bills incurred.  Counties with higher population densities are more likely to have more hospitals and because I did not see a correlation between higher population densities and ozone exceedances, I will not have to take the air quality into considereation. Before reclassifying hospitals, I will use the euclidean distance tool.  In studies, it has been shown that the closer the distance to the hospital, the less likely the patients condition will be fatal.  In the study that I gathered the information from, it was shown that hospitals within a distance of 6 miles have the lowest mortality rates with lower levels of survival at any distance larger than this.  Specifically, 12.5 miles was considered to be a far distance.  Therefore, any distance 6 miles and under was considered to be favorable.  The classifications were scaled from 1-3 with three being the most favorable and one being the least favorable.  Least favorable conditions are represented by lighter shades of blue while most favorable conditions are represented by darker shades of blue.  I have attached my initial reclassification results below.  The next step will be to assign each of the variables a weight and use the raster calculator to determine areas of high and low risk.
                                     

Ozone Reclassification

Population Density Reclassification

Income Reclassification 

         

                                    

Hospital Reclassification 

Update 3: Data & Initial Raster Results

I will now be looking at four different variables in order to determine the lowest and highest risk areas for asthma sufferers in the state of California: (1) Percentage of days ozone exceeded the national average, (2) Income, (3) Population Density, (4) Hospitals.  I will convert each of these shapefiles to rasters so that I can perform a spatial raster analysis using a weighted overlay.  After converting each shapefile to a raster, they will be reclassified for favorable criteria.  Because the ozone data that I had imported from excel are points rather than polygons, I performed a spatial join between ozone exceedances and counties and then converted the county polygon that now contains the ozone data to a raster.

 

Update #2 - Change of Plans & New Maps!

After seeing that there is not a visually significant relationship between ozone exceedance and population density, I began to think about the data that I was using and if it would  be more interesting to look at other variables.  When looking at the asthma data that I used in my previous course project, I realized that it was not looking at the number of cases of asthma within each county, but rather the number of people who were being hospitalized because of their asthma.  Therefore, the people who were being hospitalized may not even be from that particular county.  This made me think about what factors would influence people requiring hospitalization (other than the obvious severity of their asthma symptoms).  Thus, I decided that I would still look at ozone exceedance and population density, but I would also incorporate income and the number of hospitals within each county.  People who have lower incomes may be less likely to admit themselves to the hospital for their symptoms because they may not have health insurance or be able to afford the hospital bills.  Furthermore, in rural areas some hospitals may be too far away for people to be able to travel to.  The amount of hospitals in each county could also be related to population density, as more people would require more healthcare.  After importing data on medical facilities across the state of California, I narrowed down the facilities to general medical and surgical facilities to match the asthma hospitalization data I had also gathered online from the state of California.  I then used graduated colors to show the population density, and simple dots for the hospitals to show that areas of higher population densities are more likely to have a higher number of hospitals.  I also created a map showing the number of asthma hospitalization in relation to income to visually display how people in lower income areas may have less hospitalizations due to asthma.  I labeled major 4 major counties within California to serve as a location reference.

Update #1

In order to be able to visually see if there is possibly a correlation between population density and the percentage of days of Ozone exceedance I created a map showing the two data files.  For the population density, I chose a graduated quantity symbology which would show lower population densities in lighter colors and higher population densities in darker colors.  For the ozone exceedance, I chose to use a proportional symbol so that I could see which counties had higher occurrences and which had lower in relation to their respective population densities.  Before using the data files, I had expected that counties that had a higher population density would probably have higher occurrences of ozone exceedance days because high densely populated areas are normally cities, which I would expect to produce more pollutants than rural areas and thus have a larger number of asthma cases. With that being said, you cannot determine whether there is a significant relationship between two variables solely based on visual analysis, so I will have to use other analysis tools in order to determine if what I am seeing is significant or due to chance. 

Project Extension Final Proposal

Introduction:

Asthma is a chronic disease that affects 8 million people in the United States, or 8% of the population (Center for Disease Control and Prevention).  Involving the airway of the lungs, people who experience asthma have swelling and inflammation of the airways, making it difficult for air to navigate throughout the lungs and resulting in patient symptoms such as tightness of the chest, wheezing, coughing, shortness of breath, and difficulty breathing.  Asthma attack triggers can be separated into two categories; allergies such as pet dander and pollen, and non-allergic triggers that include environmental factors such as changes in the weather and air quality (AAAAI).  Poorer air quality may be associated with ozone and airborne particles, both of which are known to be triggers of asthma and are found in higher populated areas.  Ozone contributes to smog which can be irritating to the lungs, reduce lung function, and make it more difficult to breathe while airborne particles found in smog, dust, and smoke can enter the lungs and cause long-term and short-term problems such as reduced lung function, and more frequent asthma attacks (Air Pollution and Asthma).  Using the findings of my previous project, this project will aim to look at population density within each county, and whether that influences air quality.  It will then aim to look at the relationship between these three variables in order to determine if higher levels of each are found in clusters throughout California.

Objectives:

This project is set to use the data and conclusions drawn from my previous course project in order to determine if there is a relationship between population density of each county in California and air quality of each county.  I would expect for the more densely populated counties to have poorer air quality due to the assumption that more people would equate to more pollution being produced.  Furthermore, rather than just basing my air quality data off of my previous citations, I may choose to look at the number of factories within each county, as I presume that a higher number of factories would result in more air pollution and thus lower air quality. With that being said, as I have seen from my previous project, results do not always follow with what is expected.  I would also be interested in overlapping the air quality, population density, and asthma hospitalization data in order to see if there is a “hotspot” within California where these three variables correspond. 

Methodology:

I will use ArcMap and online GIS resources in order to display the following information:

• Create a new layer of data showing the population densities
• Determine if higher population densities have lower air quality
• Determine if there is overlap within California counties where population density, low air quality, and asthma hospitalizations correspond. 
• Analysis Used:
• Geoprocessing to intersect desired variables
• Hot-spot analysis to detect locations that are statistically significant in regards to their spatial clusters
• Correlation tools to detect relationships between the variables 

Data Sources:

Current data sources that will be used for this project are as follows:

• Asthma ED Visit Rates by County 2012 [Download ArcGIS online]. (2012) California Department Visit Rates by County in 2012, CA: CDPH-DATA [October, 2017]
• Emergency Department Visits Due to Asthma, Both Sexes, All Ages, All Races/Ethnicities, Spatially Modeled, Age-Adjusted Rates Per 10,000, 2010, Counties [Download]. (2010) California: Office of Statewide Health Planning and Development [October, 2017].
• CA Counties 2012 [Download ArcGIS online]. (2012) California US Census Bureau, FracTrackerAlliance [October, 2017]
• County Population [Download ArcGIS online]. U.S. Bureau of Reclamation (USBR), charles.schafer_CDPHDATA [October, 2017]
• EPA AirData Air Quality Monitors [Download]. CA, ESRI [October, 2017]
(https://epa.maps.arcgis.com/apps/webappviewer/index.html?id=5f239fd3e72f424f98ef3d5def547eb5&extent=-146.2334,13.1913,-46.3896,56.5319)

Work Plan:

Week of January 26th: Draft GIS Proposal
Week of February 2nd: Finalize draft proposal
Week of February 9th: Collect data (search for more in-depth air quality data)
Week of February 16th: Take a look at previous data/maps in order to determine county focus.  Should I include all counties of California or focus on the top counties that I had focused in on in my other project?
Week of February 23: Take new data and organize it into an excel sheet that can be imported into a GIS Map
Week of March 2nd: Import new layer data into GIS / select correct coordinate system/projections
Week of March 9th: Organize layers using correct symbology/appropriate color scheme 
Week of March 16th: Begin to visually analyze layers to see if there are any obvious overlapping/possible relationships
Week of March 23rd: Begin searching for spatial analysis tools that could accurately determine if there is a relationship between the data
Week of March 30th: Analyze data
Week of April 6th: Finalize maps/add final details (scale bar, legend, titles, north arrow etc.)
Week of April 13th: Work on final poster
Week of April 20th: Start finalizing project poster
Week of April 27th: Continue to work on finalizing the final poster
Week of April 30th: Turn in Final Project

Deliverables:
 A comprehensive map will be created illustrating if there is a connection between population density and air quality.  Another layer will be used in order to display if there is a relationship between all three variables (population density, asthma, air quality).  Analysis tools will also be used in order to support claims as well as show the data more clearly on the map. 


Sources:

"AAFA." Air Pollution and Asthma | AAFA.org. AAFA, n.d. Web.
"Asthma | AAAAI." The American Academy of Allergy, Asthma & Immunology. AAAAI Foundation, n.d. Web.
"Vital Signs." Centers for Disease Control and Prevention. Centers for Disease Control and Prevention, 03 May 2011. Web.