Date of Graduation
5-2020
Document Type
Thesis
Degree Name
Bachelor of Science in Industrial Engineering
Degree Level
Undergraduate
Department
Industrial Engineering
Advisor/Mentor
Zhang, Shengfan
Committee Member/Reader
Sullivan, Kelly
Abstract
Diabetes is one of the leading causes of death in the United States and can cause severe impairments to those diagnosed. Prediabetes is a state when a patient has higher fasting plasma glucose levels than a non-diabetic person but is not quite high enough to be considered diabetes. Both diabetic and prediabetic patients are at higher risk for cardiovascular diseases (CVD), which is the leading cause of death in the United States. The primary form for prevention and treatment of CVD is through statin therapy. Statins are a class of medications used to treat and prevent CVD by limiting cholesterol production in the liver and stabilizing plaque in arteries. However, substantial research has found an association between statin use and the development of Type 2 diabetes. This is an important association to investigate because both statin use and diabetes are prevalent in the United States.
The association between statin use and the development of Type 2 diabetes poses a complicated risk for prediabetic patients. Because they are already at high risk for diabetes, taking a statin could further increase this risk. However, preventing CVD, which they are also at risk for, is critical as well. This research investigates the relationship between statin use and prediabetic subjects specifically.
An adult, prediabetic subpopulation was obtained from the National Health and Nutrition Examination Survey (NHANES), which is made publicly available through the Center for Disease Control and Prevention. Several random forest classifiers were built using this subpopulation to predict statin use among prediabetic patients. Analysis of the models found age, cholesterol levels, blood pressure levels, waist size, body mass index (BMI), and annual household income to be the best predictors of statin use in prediabetic subjects. Access to health insurance, gender, family history of heart attacks, and overall health rating were found to be the least impactful predictors of statin use among prediabetic subjects in all models. It appears the risk of CVD outweighs the risk of developing Type 2 diabetes, and doctors are continuing to prescribe statins for prediabetic patients despite the increased risk of developing diabetes.
Keywords
Prediabetes, Statin, Analytics, Machine Learning, Diabetes, NHANES
Citation
Gentile, A. (2020). Characterizing Statin use among Prediabetic Patients with Predictive Analytics. Industrial Engineering Undergraduate Honors Theses Retrieved from https://scholarworks.uark.edu/ineguht/69