Problem Statement

Heart disease and chronic lower respiratory illnesses are major causes of death for individuals under the age of 65 in the United States. SARS-CoV-2, the virus that causes COVID-19, is known to cause not only lung damage and can also lead to serious heart problems. During the first year of the COVID-19 pandemic there was an increase in the number of cardiovascular related deaths, and research has found an increase of heart-attack related deaths among people ages 25 to 44.

Medicaid is the United States’s public health insurance program that offers coverage to low-income Americans. In 2021, approximately one in six adults and four in 10 children were covered by Medicaid. Prior research has found that adults with Medicaid coverage have higher rates of cardiovascular and respiratory diseases compared to uninsured adults. This research project aimed to answer whether there was an association between state Medicaid per non-senior enrollee spending and heart/respiratory related mortality rates for non-seniors.

Methods

Description of Datasets

For this analysis, we used three different datasets : CDC WONDER Mortality (from the CDC WONDER, Wide-ranging Online Data for Epidemiologic Research), CDC WONDER Population, and KFF CMS Medicaid summary (Center for Medicare Services and KFF). All three datasets are for the United States and about US populations, disaggregated by age and state. The CDC mortality data are from 2021 and the KFF Medicaid data is from 2019. While all three datasets include information for individuals of all ages, in for this analysis we focused on those under the age of 65, since the majority (90%) of the Medicaid population is under the age of 65.

Analyzing the Data

CDC WONDER Population

This dataset contains the state populations for 50 states & District of Columbia for 2021. It has been cleaned to show the total population of people under the age of 65 per state and now contains only two columns, state and population.

CDC WONDER Mortality

This dataset contains causes of death in all 50 states & District of Columbia for 2021. It has been cleaned to show the number of heart and respiratory related causes of death for those under the age of 65 and now contains only two columns, state and heart_respiratory_deaths. The conditions included under the heart_respiratory_deaths are listed in the data dictionary appendix at the end of this brief.

KFF CMS Medicaid

This dataset contains non-senior enrollment in Medicaid, non-senior spending for Medicaid, and the spending per non-senior for all 50 states & District of Columbia for 2019. It has been cleaned to show the number of spending per state and now contains only four columns, state, non_senior_enrollment, non_senior_spending, and spending_per_non_senior.

Joined Dataframe

This data set merges together the KFF Medicaid data, the CDC Wonder Mortality data, and the CDC Wonder population data. This also includes two new variables: mortality_rate, that shows the mortality rate for heart and respiratory deaths for all 50 states per 100,000 non-seniors, and mortality_to_spending which shows the number of deaths per medicaid dollar spent.

Results

A Pearson’s correlation test showed no significant correlation between average state Medicaid spending per non-senior and mortality rates for heart and respiratory related deaths per 100,000 non-seniors, with a correlation of -0.145 and p-value of 0.315 (Figure 1).

The five states with the highest number of heart and respiratory deaths per Medicaid dollar spent were Alabama, West Virginia, South Carolina, Mississippi, and Georgia, in that order (Table1). In Alabama, there were about four times as many heart and respiratory deaths per medicaid dollar spent compared to the U.S. as a whole (Figure 2), with a mortality to spending ratio of .79 compared to the national ratio of 0.21. Alabama, West Virginia, and Mississippi also had the highest heart and respiratory mortality rates among non-seniors in the U.S.; West Virginia had 158.06 deaths per 100,000 people compared to the national rate of 75.91.

Table 1: Interactive Data Table: Top 10 States with the Most Deaths per Medicaid Dollar Spent

The table above shows the top 10 states with the highest number of deaths per medicaid dollar spent. The table shows population, respiratory/heart deaths, mortality rate, Medicaid enrollment, spending per Medicaid enrollee, and mortality to spending ratio.

Figure 1: Scatterplot of Medicaid spending and mortality rates

The scatter plot shows each state’s medicaid spending and mortality rate per non-senior, including the five highlighted states which have the highest number of deaths per Medicaid dollar spent among non-seniors. There appears to be no significant correlation between average state Medicaid spending per non-senior and mortality rates for heart and respiratory related deaths per 100,000 non-seniors (r = -0.145, p-value = .315).

Figure 2: Bar graph of 5 states with the highest mortality rate per million medicaid dollars spent

The bar graph supplements the information in the scatterplot by “zooming in” on the 5 states with the highest number of deaths per medicaid dollar spent.

Discussion

While the results did not show a correlation between average state Medicaid spending per non-senior and mortality rates for heart and respiratory related deaths per 100,000 non-seniors, our preliminary plots and analysis suggest there is a drastic difference between the five states with the highest number of deaths per medicaid dollar spent compared to the rest of the country (Figure 2). Given these plots and preliminary analyses, it appears both West Virginia and Alabama are examples of states that would benefit from increased spending in Medicaid programs, as witnessed by their rates compared to the rest of the country.

Future research should look more broadly into other diseases and perhaps overall mortality rates to see if there is any significance, as well as adjust for potential confounding variables Our analysis looked specifically at Medicaid spending rather than Medicare; for future studies, it would be interesting to see if there is a difference between Medicaid and Medicare spending and mortality rate.

Appendix: Data Dictionary

Variable Name Variable Type Description
state Character Includes all 50 states & DC
non_senior_spending Numeric Total state spending for children and adults under the age of 65.
non_senior_enrollment Numeric Total state enrollment for children and adults under the age of 65.
spending_per_non_senior Numeric non_senior_spending divided by non_senior_enrollment
heart_respiratory_deaths Numeric

Number of deaths occurring in each state due to heart or respiratory-related causes. Causes included under this category include:

  • Influenza and pneumonia
  • Chronic lower respiratory diseases
  • Lung diseases due to external agents
  • Other respiratory diseases principally affecting the interstitium
  • Other diseases of pleura
  • Other diseases of the respiratory system
  • Chronic rheumatic heart diseases
  • Hypertensive diseases
  • Ischaemic heart diseases
  • Pulmonary heart disease and diseases of pulmonary circulation
  • Other forms of heart disease
  • Diseases of arteries, arterioles and capillaries
  • Respiratory and cardiovascular disorders specific to the perinatal period
  • Other diseases of upper respiratory tract
  • Diseases of arteries, arterioles and capillaries
  • Suppurative and necrotic conditions of lower respiratory tract
  • Congenital malformations of the respiratory system
  • Symptoms and signs involving the circulatory and respiratory systems
Population Numeric Total state population of those under the age of 65.
mortality_rate Numeric Mortality rate heart and respiratory related deaths per 100,000 non-seniors
mortality_to_spending Numeric Number of heart or respiratory-related deaths per medicaid dollar spent