Learning about the specifics of how, when, and why people end up in the emergency department can help to tackle many common problems faced in healthcare seen today. This includes the upgrade of medical quality, reduction of patient complaints, and less waste of medical resources. It can also unveil hidden medical knowledge through looking at the correlation and association of apparently independent variables. This new knowledge can be used to combat many common complaints and issues in the emergency department such as increased wait time, ambulance diversion, reduced staff morale, and adverse patient outcomes. Though this correlation does not seem to be linear, there are commonalities between different demographics and the injuries they receive. Finding these correlations will help to make emergency department care more beneficial, as well as can be used to help prevent injuries that can be avoided.
This project uses a dataset from Kaggle called How People Get Hurt which details demographics about who ends up in the Emergency Department, as well as why they are there. This dataset is compiled from the National Electronic Injury Surveillance System (NEISS) and includes data from January 1, 2016, to December 31, 2020.
- Fork repo and clone into local dir
- Download dataset and convert from .txt to .csv file using excel or similar program, saving to datasets folder
- Install all requirements using
pip install -r requirements.txt - Run Jupyter Notebook cells for different algorithms
My findings are presented in my term paper titled Investigating Emergency Department Statistics to Prevent Hospitalizations.