What I learned by analyzing Accidental Drug Related Deaths in Connecticut

Note from the Editor: This post is part of an inaugural series, written by students of the Spring 2019 Data Journalism I course in the Newmark Graduate School of Journalism at CUNY, sharing their work and thought process. Each week we have a Data Fest in which two of the class reporters present a data set, along with a brief critique and overview of what they did and discovered.

While looking for tons of data for my Data Fest presentation, my first set of data was the most popular baby names from 2011–2016 by gender and race. Who doesn’t love babies? I found I couldn’t do much with the data so I discarded it. I continued my search until I came across the “Accidental Drug Related Deaths 2012–2018” on data.gov.

The dataset records all the accidental deaths related to drug overdose in the state of Connecticut from January 2012 until December 31, 2018. It is the result of an investigation by the Office of the Chief Medical Examiner, “which includes the toxicity report, death certificate, as well as a scene investigation”.

There are several ideas that came to mind with all the information within the dataset. For example, a pivot table displaying the race and “cause of death” ranked by the most repeated “cause of death”. My initial idea was to create a chart displaying the deaths by drug type and gender, however, the way the data is divided made it difficult to create a chart of that sort on a deadline. With the help of my data professor, I was able to create a pivot table counting the number of deaths reported by sex and race.

Findings

The most reported deaths for both males and females is mostly common within the White race. However, in order to provide more context to the results, I would need to analyze census information such as population by race in Connecticut in the years 2012–2018 to make an accurate comparison of the deaths by race in proportion to the total state population and then get the number of deaths per 100,000 population by race.

After my presentation, which was this Friday morning, my professor shared resources to explore and go deeper:

1. The Kaiser Family Foundation has a great searchable database of birth and deaths per state. Here is the overall death rate by race/ethnicity per 100,000 people in Connecticut.
2. The source of the KFF website is the data from the Centers for Disease Control and Prevention (CDC), available on the CDC WONDER Online Database.The CDC WONDER is a set of “online databases” that “utilize a rich ad-hoc query system for the analysis of public health data. Reports and other query systems are also available”.
3. Paul Overberg, from The Wall Street Journal, hosted a session in the NICAR 19 conference, about using CDC WONDER to analyze birth and death data. Investigative Reporters and Editors (IRE) published Paul’s tipsheet. It’s very helpful for your work.

Other ideas with the Data

• Since the data provides latitude and longitude of the deaths, create a map for each year respectively, displaying the cities/counties with the most reported deaths. I could also create a map that shows the variation of accidental deaths over time to visually inspect patterns and outliers.
• Pivot table comparing race and description of injury (Ingestion, Injection, Inhalation, Drug use, etc.).
• Counta of the full dataset and description of injury by category: how many deaths per each category
• Focus on a specific year and create a chart based on the months with the most reported cases.
• Focus on a specific year and on the age range to see what ages were mostly reported.
• The dataset includes one column per substance detected in the body of the deceased. A “Y” value under the different substance columns indicates that particular substance was detected. This means that with a bit more work, I can get deeper insights about the drug/s detected (top, bottom, most combined drugs, etc.)

Further Questions

A brief analysis of this dataset can leads to questions such as how are drug prevention programs working or not working, how are laws on drug use and crime enforced in the state. It can also allow journalist to dive into the justice system of Connecticut. How accessible are drugs in each county? How does the police in Connecticut regulate drug policy? What programs are open to the public to practice safe drug use? Do the people reported have a history of drug use or a criminal background? Maybe the data on Connecticut is a microanalysis on what can extend to a deeper analysis on drug use throughout the U.S. based on each state.

Share tips and ideas: If you have any thoughts about this, you can contact me at noelis.ciriaco@journalism.cuny.edu. I am also part of the Spanish bilingual journalism program at our School and I am reporting on the increase of medical tourism of plastic surgeries in the Dominican Republic for El Deadline, our pop-up bilingual newsroom. If you have information, write me and I’ll get in touch with you.