Targeting Reduced Asthma Hospitalizations through Geospatial Statistics

Secondary Research of Preventable Asthma Hospitalizations

A Thesis Submitted to the Graduate Faculty of the National University, Schools of Businesss & Management and Engineering & Computing in partial fulfillment of the requirements for the degree of Masters of Science in Business Analytics

Prepared by Matthew C. Vanderbilt
MSBA Candidate & NU Scholar, National University
Director of Fiscal Affairs, Department of Medicine, UC San Diego School of Medicine
MatthewVanderbilt.com | GitHub | LinkedIn

See Project Wiki for complete details.

License

This code is licensed under GNU General Public License Version 3 - see the LICENSE.md file for details.

Project Output

Asthma Probability Model Dashboard on Tableau Public

SAS Code Execution

1. Load Data
2. Review Psychological Distress Scale
3. Recode & Normalize Variables
4. Restrict Data to Common Survey Variables
5. Create Final Table for Analysis
6. Create Subset Table for Decision Tree Analysis
7. Investigate Subject Characteristics
8. Investigate Subject Asthma Characteristics
9. Investigate Weighted Survey Population Frequencies
10. Perform Total Variable Correlation
11. Perform Multinomial Logistic Regression
12. Investigate Links between Healthy Food Accesss with Poverty and Descriptive BMI
13. Perform Binomial Logistic Regression
14. Perform Adjusted Binomial Logistic Regression
15. Perform Inverted Binomial Logistic Regression
16. Perform Adjusted Binomial Logistic Regression
17. Perform Final Binomial Logistic Regression
18. Perform Binomial Logistic Regression of Asthma Exacerbation
19. Perform Final Binomial Logistic Regression of Asthma Exacerbation