Healthcare Accessibility in Cook County

Many individuals in the United States struggle to access necessary healthcare services. Healthy People 2030 defines Healthcare Accessibility as the capacity to obtain timely, high-quality, and affordable healthcare services. Aligning to enhance healthcare accessibility and promote healthier lives, our project aims to devise a systematic approach to improve access to healthcare within Cook County communities.

Healthcare Accessibility can be broken down into three key components:

• Healthcare Service Capacity: This refers to the availability of healthcare services, indicated by the number, costs, and quality ratings of hospitals, nursing homes, and home care agencies.
• Population Demand: This encompasses the demographic need for healthcare services, factoring in population size, growth, poverty levels, and vulnerability.
• Geographic Impedance: This assesses how easily residents can reach healthcare providers' locations, reflected by the gap between healthcare service capacity and population demand.

Our project employs a combination of weighting, machine learning, and geographical methods to create an index and predictive model for healthcare accessibility. We integrate data from various sources, primarily obtained through web scraping and searches, including information from the Centers for Medicare & Medicaid Services, US Census Bureau, and Healthgrades, which collectively provide details on over three million U.S. healthcare providers.

Our project offers three distinct applications beneficial to both policymakers and residents:

• Data Visualization: It enables the visualization of healthcare accessibility across various attributes, offering a clear view of the current landscape.
• Analysis and Prediction: This feature calculates a healthcare accessibility score for each community, identifying crucial factors influencing access to healthcare.
• Solutions Platform/Roadmap: An interactive dashboard allows policymakers to identify communities with significant healthcare gaps while enabling residents to locate timely, high-quality healthcare providers based on their location.

Team Member (Last Name in Alphabetic Order)

Hourui Guo, Yijia (Gaga) He, Yue (Luna) Jian, Qi Zhao

Package used

pandas
folium
plotly
matplotlib
pathlib
dash
json
numpy
base64
lxml
selenium
sklearn
webdriver-manager

Data Sources

Health Service Part

Name	Source	Collection Way	Responsible Team Members
Physicians' information on healthgrades	healthgrades website	Web Scraping	Gaga, Luna
Patient survey (HCAHPS) - In Patient Hospital	Data.CMS website	CSV file available	Hourui
Outpatient & Ambulatory Surgery CAHPS Survey	Data.CMS website	CSV file available	Hourui
Nursing Home Provider Information	Data.CMS website	CSV file available	Hourui
Health Expenses&Beds for Hospital - Hospital Provider Cost Report	Data.CMS website	CSV file available	Qi
Health Expenses&Beds for Nursing - Skilled Nursing Facility Cost Report	Data.CMS website	CSV file available	Qi
Health Expenses for Homecare- Home Health Agency Cost Report	Data.CMS website	CSV file available	Qi
CMS Manual System	CMS.gov	PDF file available	Hourui

Population Demand Part

Name	Source	Collection Way	Responsible Team Members
Demographic data	US Census Bureau	CSV file available	Qi
Income Data	US Census Bureau	CSV file available	Qi
Health Insurance Coverage	US Census Bureau	CSV file available	Qi
Employment	US Census Bureau	CSV file available	Qi
Social Characteristics	US Census Bureau	Shapefile available	Qi
Poverty Situation	US Census Bureau	Shapefile available	Qi

Geographic information

Name	Source	Collection Way	Responsible Team Members
ZCTA Code	US Census Bureau	Shapefile available	Qi
Zip Code with Longitude and Latitude	US Zip Codes from 2013 Government Data	GeoJson available	Qi

Dashboard Demo

Click the picture below to watch our dashboard demo video

Instruction to launch the application

1. Clone the repository

git clone [email protected]:gagahe-cx/Healthy-Cappybara.git

2. Navigate to the repository

cd ./ Healthy-Cappybara

3. set up and activate the virtual environment

poetry install
poetry shell

4. Launch the App

python3 -m HealthyCappybara

5. Engage with the App (Using Alphabetical Inputs)

• (a) The Dashboard,

• (b) Scraping Data,

• (c) Clean Data,

• (d) Quit App.

6. Option 2 has three sub-options. Users can input their specific criteria for conducting web scraping.

• Condition 1: How many medical categories do you want to crawl?
• Condition 2: How many cities do you want to crawl?
• Condition 3: Do you want to crawl now?
• Upon completion, the message " Congratulations! The data has been successfully crawled and saved to {file location}!" will be displayed.

7. Code Reference

• Scraping: https://www.zenrows.com/blog/selenium-python-web-scraping#headless-chrome
• Dashboard: https://github.com/plotly/dash-sample-apps