Diabetes in America
Introduction
According to the Center for Disease Control and Prevention (CDC) report published in 2017, more than 100 million of US. adults are diagnosed with diabetes or prediabetes.The report finds that as of 2015, 30.3 million Americans, 9.4 percent of the U.S. population have diabetes. Another 84.1 million have prediabetes, a condition that if not treated often leads to type 2 diabetes within five years. The report confirms that the rate of new diabetes diagnoses remains steady. However, the disease continues to represent a growing health problem: Diabetes was the seventh leading cause of death in the U.S. in 2015. Some factors like Obesity and overweight, physical inactivity, high blood pressure, cholesterol and glucose was associated with people being diagnosed by diabetes.
Method
Data
The data for this study was obtained from the County Health Rankings & Roadmaps program. For the 2016 County Health Rankings, the CDC produced 2014 county estimates using single-year 2014 BRFSS data and a multilevel modeling approach based on respondent answers and their age, sex and race/ethnicity, combined with county-level poverty and county and state level contextual effects.
For the analysis below, the diabetes data included of 3141 observations for all the counties in the US with 14 variables. Below describes the 3 main variables examined for this study, scales for the variables are in percents:
- Diabetics: is the prevalence of diagnosed diabetes in a given county. Respondents were considered to have diagnosed diabetes if they responded “yes” to the question, “Has a doctor ever told you that you have diabetes?” Women who indicated that they only had diabetes during pregnancy were not considered to have diabetes.
- Limited Access to Healthy Foods is the percentage of the population who are low income and do not live close to a grocery store.
- Physical Inactivity is the percentage of adults aged 20 and over reporting no leisure-time physical activity.
The goal for this study is to examine:
- plot a spatial map to understand people diagnosed with diabetes in the US.
- linear regression to analyze the effect of limited access to unhealthy food and physical inactivity leads to diabetes.
Linear Regression Analysis
| Model 1 | ||
|---|---|---|
| (Intercept) | 4.16*** | |
| (0.17) | ||
| Healthy_Foods | 0.03*** | |
| (0.00) | ||
| Physcially_inactive | 0.25*** | |
| (0.01) | ||
| R2 | 0.36 | |
| Adj. R2 | 0.36 | |
| Num. obs. | 3107 | |
| RMSE | 1.82 | |
| p < 0.001, p < 0.01, p < 0.05 | ||
The diabetics in percent id normally distributed in our data. From the analysis above, one percent increase in limited access to healthy foods, a person has approximately 3% chance of being diagnosed diabetic. One percent increase in a person being less physically inactive, he or she has an approx of 25% chance of being diagnosed diabetic.
Spatial Mapping
Diabetes in Texas
From the interactive graph that gives the average percents of diabetics in the Texas state counties. There is a wide spread of people who are diabetics in the Texas. Brooks county has an average of 21.2% diabetics followed by Hudspeth county being second highest with average 20% diabetics.
Difference between Spatial and Non-Spatial Data
Spatial data defines location as it can be seen from the example above. Non-spatial data relates to a specific, precisely defined location. Spatial data mapping will help one understand the overall picture for example, from the U.S diabetes map, one can understand diabetes what States have a highest percents of diabetics.
Difference between cb = TRUE or FALSE in Tigris package
The US counties file in the Tigris package in R is used for spatial mapping as seen in the examples above. When retrieving the file if cb=True this downloads the file in a more generalized (1:500) counties file, its not a big file and easier to work with if your computer has less space. If cb=FALSE, a more detailed version of the file is downloaded.