This project is on estimating the impact of opioid control policies. The analysis focuses on three states' policy chenges: Florida (February 2010), Texas (January 2007), and Washington (January 2012). For all three policy changes, pre-post analysis and difference-in-difference analysis are performed. For the Florida case, we will analyze the effect of its policy change on both opioid shipments and overdose death. For Texas and Washington, we will only analyze overdose deaths for the time being.
It is a subset of the US Vital Statistics Mortality Data. It includes data on nationwide drug/alcohol-induced death, and each observation includes the number of deaths for each county each year.
It is a dataset of all prescription opioid drug shipments in the United States from 2006 to 2012.
This data contained the population of each county in all US states on a yearly basis.
Florida’s drug policy was effective in decreasing the shipments of opioids as well as effective in shifting the overall growing trend of its mortality rate to a downward trend. Texas’s drug policy was potentially very successful once as the overall trend of the average mortality rate was controlled and started to decline after the policy went into effect. Washington’s drug policy was potentially not a very successful one since its average mortality rate continued to increase at the same rate after the policy took place.
The research design is overall beneficial in exploring the impact of drug regulation policies. Both the pre-post comparison and difference-in-difference analysis provided straightforward ways for us to see how policies changed the trend on opioids prescriptions and overdose deaths. The plots generated through the two methods are also simple enough for non-data-scientist readers to understand.