The task is to analyse healthcare data from a hospital with the aim of answering critical questions that could help optimise operations, patient care, and resource management.By understanding patterns like the typical duration of patient stays or identifying potential disparities in treatment among different demographics, decisions about staffing, resource allocation, and hospital policy adjustments can be better informed.
The goal is to address key business questions around patient stays, resource utilization, and potential racial disparities in treatment. The primary focus is: 1.Understanding the average length of hospital stays. 2.Investigating correlations between lab procedures and the length of patient stays. 3.Identifying whether racial biases affected the number of lab procedures administered. 4.Analyzing which medical specialties conducted the most procedures. 5.Evaluating readmission rates and resource allocation in high-traffic areas.
The dataset used for this project was sourced from Kaggle(https://www.kaggle.com/datasets/brandao/diabetes?select=diabetic_data.csv).I manipulated the data using Python to preserve only the relevant attributes, and then exported these into CSV files. The final dataset is split into two tables: Demographics and Health.
We conduct an exploratory data analysis (EDA) using SQL to understand the dataset and extract meaningful insights.
1.Optimize Resource Allocation: Focus resources on efficiently managing the 1-4 day patient stays to improve hospital flow.
2.Reduce Readmissions: Create follow-up plans for high-risk patients to lower the 11% readmission rate.
3.Support High-Traffic Specialties: Boost staffing and resources in high-traffic departments like Internal Medicine.
4.Standardize Lab Procedures: Review lab procedures in departments like Internal Medicine to reduce unnecessary tests.
5.Monitor Outcomes by Specialty: Set KPIs for specialties to track patient outcomes and identify areas for improvement.