Authors:

Sia Rajput

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

The opioid epidemic has become a national health challenge with a staggering 80,411 reported overdose deaths in 2021. Despite the excessive number of opioid deaths in Florida, the opioid reversal agent Narcan has been readily available as a lifesaving agent to restore respiratory depression. However, opioid overdose, specifically overdose on the highly potent opioid fentanyl, continues to be a burden due to illicit fentanyl abuse with many critical factors still unknown. This interdisciplinary research project seeks to uncover the intricate relationships between provoking and palliative factors contributing to reported fentanyl overdoses across the state of Florida using machine learning models, random forests, and gradient boosting models to identify areas at high risk of fentanyl overdoses. Our cohort will consist of two groups: people who die from fentanyl overdose vs. people who survive. We'll use linear and logistic regression coefficients to get a baseline of the importance of the predictors in determining fentanyl overdoses. After determining what specific factors are associated with high rates of fentanyl overdose deaths, we'll create a heatmap indicating what regions are at high risk as well as the features that could be improved to reduce the risk in particular regions. Some key factors that we'll be analyzing are police presence, access to harm reduction sites, demographics, income levels, traffic and road accessibility, and the quality of healthcare services, specifically emergency medical services (EMS), emergency departments (ED), and intensive care unit (ICU) locations. We expect to find factors responsible for fentanyl overdose deaths across Florida and identify high risk areas to permit accessibility and alleviate factors associated with fentanyl overdoses to mitigate the opioid crisis. While our work is aimed at Florida, our work can serve as a potential generalizable model at the national level provided that reliable data is available.

Source:

University of Florida / Sia Rajput, Hunor Vajda, Alaguvalliappan Thiagarajan, / 2023

Topics:

No topics listed

Co-authors:

Sia Rajput