Project Description
Improving patient experience in the ER
Waiting for medical help in an unexpected situation can be one of the most stressful and frustrating experiences that patients and their loved ones go through. It’s common for individuals to feel anxious and, in some cases, exhibit violent behaviors. To help ease their minds a bit, Zuyderland, a hospital in South Limburg, decided to introduce an innovative solution to improve the patient experience.
The idea was to develop monitors that visualize the expected waiting time in the emergency room. These kinds of monitors already exist in several Dutch hospitals, so what’s so special about this solution? Zuyderland decided to go a step further and provide patients with a waiting time monitor per department specialty based on a data model, now that’s a unique and innovative feature! During the development of this solution, our Data Science expert worked closely with Zuyderland to lay the groundwork and supported them with:
- Data preparation
Developing a suitable statistical model
Domain knowledge and expertise
A working prototype of the monitor
Theme
Intelligent Customer Interaction
Sector
Healthcare
Customer
Zuyderland
The concept
In the past, patients had no idea what was going on behind the scenes, but now they get a better idea of how busy it is. Based on the crowdedness of the emergency room, patients now have a better grasp of how much time they will spend waiting for treatment. Moreover, to safeguard their privacy, the calculated waiting time spans only from the moment they enter the emergency room until they begin receiving treatment.
The challenge
The first challenge was gathering and updating data from the SAP system in real-time to provide accurate information. We then built a model to predict ER wait times based on how many patients are currently there. Since some patient data might be missing or incomplete, the model sets a maximum wait time to give patients realistic expectations. Finally, in-depth knowledge of the medical domain is crucial to this project to ensure high quality data.
Techniques and technologies
This project required a close collaboration between the Data Science expert from Conclusion Intelligence and Zuyderland. During this project the following methods and technologies were used:
- Microsoft SQL
- Python
- Statistical Modeling
Machine learning
PowerBI
Result
The provided solution retrieves data every five minutes, depending on the type of patients and specialties. The created model incorporates a comprehensive analysis of patient flow patterns, taking into account the influx data from the past two years. This historical insight, combined with real-time updates, provides patients with a transparent and reliable estimation of their waiting times. Overall, this innovative solution has significantly improved the patient experience in the ER.
