Can Cigarette Taxes Reduce The Rate of Maternal Mortality?
I just completed my first four weeks of being a Data Science student and I wanted to use my cool new skills to revisit a college project. During my senior year at Pace University I wrote a thesis about the impact of state cigarette taxes on the rate of maternal mortality. All of the original analysis was done in Stata, but now that I know how to code in Python, I figured why not recreate my project in the new formal language I’m growing to love.
Female reproductive health is a topic that I’ve become very interested in, and it is the inspiration for this research. In order to dive deeper into the relationship between cigarette taxes and maternal mortality I’ve pulled data from National Vital Statistics which is published by the Center of Disease Control and Prevention, the American Community Survey, and The Tax Burden on Tobacco. This data will give us an idea of what maternal mortality is looking like in the states that are accounted for, demographics, education levels, and the variation of cigarette taxes from state to state. This data runs from 2003–2016, and it gives us 316 observations to analyze.
Below I will show snippets from my code so you can see how I decided to visualize and analyze my data.
As you can see from the graph above maternal mortality has been rising overtime. This finding indicates that maternal death is becoming a more prominent public health issue. The CDC says that the exact cause as to why this is happening is unclear. However, more and more pregnant women are faced with chronic health issues including heart disease and diabetes. According to the CDC, cardiovascular and cerebrovascular accidents were responsible for more than a third of pregnancy related deaths from 2011–2016.
This graph shows the crude death rate per state. I thought this would be an interesting visualization because it can show us the variability in maternal mortality among the states we are observing. Evidently, some states have higher rates of maternal mortality than others and perhaps further research can help us understand if certain geographic elements contribute to this problem.
Now for the best part, my model. I estimated one ordinary least squares regression model to examine the impact of state cigarette taxes on the rate of maternal mortality. The independent variables that I used for this study are the following: state cigarette tax, year, state. My dummy variables control for race including Hispanic, non-Hispanic Black, non-Hispanic White, and education including less than a high school degree, high school degree, and a college degree or more. In this study I used education to account for socioeconomic status because having an income may produce reverse causality within the model.
This study can help provide insight to policy implementation and future research endeavors. The two prominent findings are the statistically significant impact of state cigarette taxes on maternal mortality and the upward trend of maternal mortality over time. This study suggests that higher cigarette taxes may be a useful policy measure in reducing the rate of maternal mortality. However, future research should be conducted further examine this relationship to help us better understand how we can mitigate the rising rate of maternal death over time.
Sources: