Data Analysis on The Health of COVID19

Posted on Jun 16, 2021
The skills I demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.
Access to website


For this project, I created a website using R shiny to explore the data to find relationships between pre-existing health conditions and the number of hospitalization and mortality cases across each state in the US.

While there are numerous projects online that track the number of COVID19 cases on a global scale, I wanted to do something more than a superficial analysis on the number of COVID19 cases and examine the underlying health factors that potentially contribute towards the rise of COVID19 hospitalization and death rates. Using a combination of ggplot and plotly packages, I built a dynamic and interactive website where the user can navigate through COVID19 and health data visualizations.

Under the summary tab of the website, both the choropleth map and the line chart display the number of hospitalization or mortality cases in each state to which the user can choose between the two options under the drop down tab under "Select Item to Display". Here, I chose to focus on the hospitalization and mortality rates rather than the overall confirmed cases because my primary research question specifically sought to answer how pre-existing conditions might affect the rate of severe COVID cases.

Based on the available scientific literature, the general consensus within the science community is that people who develop severe symptoms usually have underlying health conditions that make them more susceptible to death by COVID19 whereas healthy people generally display little to no symptoms at all. Therefore, I believed that the hospitalization and mortality rates were more accurate representations of "severe" COVID19 cases than the overall confirmed COVID cases.

Data on COVID Hospitalization Rates

As we can see from the line plots for hospitalization rates below, the states exhibit a sinusoidal trend with the latest trend showing a dramatic rise in COVID hospitalization rates across most states as expected. On the brighter side, the death rates show a logarithmic trend where the death toll seems to be flattening out in the past few recent months. This is most likely due to better treatment protocols developed over time since the onset of the virus.

Data Analysis on The Health of COVID19

Choropleth map of COVID hospitalization and mortality rates in the US from 3/1/2020 - 11/25/2020.

Data Analysis on The Health of COVID19

Changes in COVID hospitalization rates from 3/1/2020 - 11/25/2020. 

Data Analysis on The Health of COVID19 Changes in COVID mortality rates from 3/1/2020 - 11/25/2020. 

In the pre-existing conditions section, I gathered various health data from the Kaiser Family Foundation ( and Department of Health websites of each state. The main conditions that I looked at were COPD (Chronic Obstructive Pulmonary Disorder), hypertension, obesity, diabetes, flu vaccination rates and cancer incidence rates. I specifically chose these health conditions since they are the most common comorbidities associated with severe COVID19 cases.

Data on Hospitalization and Mortality Rates By States

Here, I created a scatterplot of the hospitalization and mortality rates against the incidence rate of each conditions for each state with a regression line to visualize the correlation between the two variables. To visualize where each state lies relative to the regression line, the user can double click on the state of interest and click once to select other states for comparison. On the bottom of the page, the infobox displays the state with the lowest incidence rate as the "healthiest" state followed by the state with the highest rate as the "unhealthiest" state and the average US rate for the selected condition of interest.

This image has an empty alt attribute; its file name is screen-shot-2021-06-16-at-50518-pm-378363-Au4XlRNT-1024x572.png

The Correlation tab lists out the Pearson correlation coefficient of all the pairs between the response variables (hospitalization rate & mortality rate) and the explanatory variables (health conditions). One of the main reasons why the dataset yielded such low correlation values between the hospitalization/death rates and health condition incidence rates could be due to the fact that COVID related hospitalization and death rates are also comprised of people who are otherwise perfectly healthy.

Although it is true that people with severe symptoms tend to have preexisting health conditions, there remains a gap in knowledge about how the virus affects healthy people differently in terms of severity. Therefore, one would need to gather health data of individual COVID patients from hospitals in each state to better analyze the direct correlation between preexisiting health conditions and severe COVID cases.

Data Restrictions

However, due to obvious HIPAA regulations, I resorted to using the prevalence rates of each health condition as proxies for how likely each state would be susceptible to severe COVID cases. A key erroneous assumption here is that states with a higher prevalence of certain health conditions would have a higher hospitalization/death rate, which may not necessarily be the case.

For instance, states have employed different levels of gathering restrictions in response to COVID which may directly affect the likelihood of a person with preexisting condition to catch the disease. Geographic factors may also play a role where regions with lower population density may lower the risk of an unhealthy person being hospitalized in which case data analysis at a county level might be more insightful.

Death and Health Condition Rates

As for the low correlation results between death and health condition rates, the story becomes more complicated as there are other factors such as health infrastructure (availability of hospitals/beds, healthcare workers, access to healthcare), socieconomic status and treatment protocols.This image has an empty alt attribute; its file name is screen-shot-2021-06-16-at-50808-pm-796872-ITpPXDlu-1024x828.pngCorrelation heatmap between preexisting health conditions and COVID hospitalization/mortality rates. Blue represents positive correlation while red represents negative correlation. The circle size represents the correlation magnitude.

Due to the lack of available health data, I was unable to make meaningful conclusions regarding the relationship between underlying comorbidities and COVID related hospitalization/death rates. Therefore, I shifted my focus to understanding how COVID was affecting different demographic populations in the US.

Demographic Data

Interestingly, results showed that the white population had the highest number of mortalities BUT the black population was disproportionately affected in terms of mortality rates (cases per million) followed by the hispanic population. At the state level, I was able to see a consistent pattern where the black population had the highest mortality rates while the asian population had the lowest. While we cannot make conclusive statements with regards to why the black population is more susceptible to Covid-related mortality, it warrants further investigation into the relationship between COVID mortality and socioeconomic factors as mentioned above.

For future directions, it might be interesting to conduct a time series analysis in predicting regions/states of high risk based on socioeconomic data. Such machine learning models present opportunities to efficiently direct resources and aids to regions/populations that need them the most.

About Author

Thomas Kim

Goal-oriented data scientist with 4 years of quantitative background in biomedical research with a bachelors degree in Biology and a masters degree in Bioengineering. Demonstrated success in hypothesis testing, data analysis and visualization to communicate results to technical...
View all posts by Thomas Kim >

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI