Predicting Diabetes Diagnoses

Authors

Sarah NetchertFollow

Faculty Advisor(s)

Michael Floren

Files

Abstract

This study explored the traits and health state of African Americans in central Virginia in order to determine what traits put people at a higher probability of being diagnosed with diabetes. We also want to know which traits will generate the highest probability a person will be diagnosed with diabetes. Traits that were included and used in this study were cholesterol, stabilized glucose, high density lipoprotein levels, age(years), gender, height(inches), weight(pounds), systolic blood pressure, diastolic blood pressure, waist size(inches), and hip size(inches). There were 403 individuals included in study since they were only ones screened for diabetes out of 1,046 African Americans being checked for obesity, diabetes, and other cardiovascular risk factors. Results indicated that a) the traits age and high density lipoprotein level have the greatest effect of the probability of being diagnosed and b) older women who are short with a larger waist, high cholesterol, low levels of high density lipoprotein, high glucose levels, and high systolic and diastolic blood pressure have a higher probability of being diagnosed with diabetes. The goal of the results it to bring attention to the traits that adults need to be aware of to keep themselves healthy. If someone finds themselves with many of these traits, they should seek life changes in order to lower cholesterol, glucose levels, systolic and diastolic blood pressure and raise their high density lipoprotein levels.

Publication Date

2020

Document Type

Poster

Department

Computer Science and Mathematics

Keywords

Diabetes, Statistics, Multiple Linear Regression

Disciplines

Applied Statistics | Biostatistics | Probability | Statistical Models

Recommended Citation

Netchert, Sarah, "Predicting Diabetes Diagnoses" (2020). Student Research Poster Presentations 2020. 48.
https://digitalcommons.misericordia.edu/research_posters2020/48