Data Analysis of Chronic Diseases in the US
The skills I demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.
Content source: National Center for Chronic Disease Prevention and Health Promotion
Background
According to the Centers for Disease Control and Prevention (CDC), chronic diseases are defined broadly as conditions that last 1 year or more and require ongoing medical attention or limit activities of daily living or both. They are the leading cause of mortality, disability, absenteeism at work and account for 90% of the national annual healthcare expenses in the US.
Six in ten adults American have one chronic condition such as depression,arthritis or diabetes and four in 10 have at least two or more.
Rationale
As reported by the Centers for Medicare & Medicaid Services (CMS), the national health spending growth is projected to average 5.7 percent, from 4.8 percent in 2019, and reach nearly $6.0 trillion by 2027. Moreover, personal health care prices are projected to grow by 2.7 percent per year on average over 2020-27. With more than half of the population in poor health and with skyrocketing healthcare costs, we should be aware of how much we spend to treat our health conditions and how much that cost varies across each state.
The shiny app provides a visual representation of the costs associated with chronic conditions within the US and identifies areas of interest (disease prevalence and highest costs) . The app can be found at: https://naekoue.shinyapps.io/Healthcare/
Dataset
The prevalence and Medicare utilization and spending are presented for 21 chronic conditions and for 210 dyads from 2007 to 2017. The data was provided standardized at the State and National level according to the CMS methodology.
Insights
Individual Chronic Conditions
After cleaning and analyzing the dataset, we observed that in 2017, the top 5 states with the highest alcohol abuse spending were Louisiana($27,569.33 ), Nevada($27,153.23 ),Texas($26,607.38 ), Florida($25,427.914) and the District of Columbia ($25,308.04).The national average for that condition was $22,450.97. Similarly, athma spending was the highest in Texas with $24,353.33 while the cost associated with this chronic disease was only $ 12,234.24 in Puerto Rico, $ 8,000 less than the national average.
Dyads
Dyads represent the combinations of two chronic conditions among Medicare beneficiaries with at least two of the conditions.
 |
 |
 |
 |
Conclusion
This shiny app was designed to provide a visual representation of the costs associated with chronic conditions within the US and identify areas of interest(disease prevalence and highest costs) . It serves as a starting point of a more in-depth analysis of the costs and long term implications of chronic conditions. Further work would include an analysis of healthcare costs associated with chronic conditions hospital readmissions, as well as an evaluation of hospital and long term care facilities geographical presence in zones with high risk Medicare beneficiaries.
Centers for Disease Control and Prevention.(2020, March 04).Retrieved from https://www.cdc.gov/chronicdisease/about/index.htm
https://www.cdc.gov/chronicdisease/about/costs/index.htm
https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Chronic-Conditions
About Author
