Something to Bandy About

Bandy Lee is a psychiatrist at Yale University and a violence expert.  In her brief Twitter-bio, she describes herself as "Violence expert and forensic psychiatrist of 20 years, uninvolved in politics until politics invaded her area of expertise."

Six years go, two years before there was a thought of Trump in the White House, she had done a research study of violence. She decided to bring it to more public attention on Sunday because she saw a comment from one of her social media followers about how a Democratic president would have responded to the coronavirus crisis differently than the Republican in office.  Unlike other people who may have had the same random thought, Dr. Lee had had the expertise to check it out.

In her response yesterday, she gave a cursory overview of the results in a series of tweets.  I selected a few, and will let her speak for herself, since she does it so much better.  And to repeat, this was not done as a way to condemn Trump, since it was completed two years before he took office.  What it does, though, is confirm the legendary adage by historian George Santayana, "Those who cannot remember the past are condemned to repeat it." 

​You can find the full study here.

For those here who don't want to read the full research study -- which I'm going to guess is pretty much everyone -- this is the opening Abstract introduction to the paper, with its summary of Objectives, Methods, Results and Conclusion.  I don't expect many (as in perhaps any) to read through all of even the Abstract, but I include it here so that you can at least see the foundation of what Dr. Lee was seriously dealing with.

ABSTRACT​​Objectives:Our goal was to identify if there might be associations between a major public health problem,i.e., violent deaths, and a potential macro-level determinant, i.e., political party in office.​Methods:Vital statistics, labor statistics, and GDP data were obtained for the years 1900–2010. Independent ttests were used to compare homicide, suicide, and total violent death rates during Republican and Democratic administrations and between states voting for Republican and Democratic presidential candidates. Correlation and hidden Markov modeling were used to assess the relationships among party of the president and changesi n unemployment rate, GDP, and violent death.Results:The party of the president was significantly associated with annual changes in suicide and homicide rates,unemployment rates, and GDP (pb0.001 to pb0.05, depending on the measure and time lag), with higher violent death and unemployment increases being associated with Republican presidencies and higher GDP with Democratic ones. Adjusting for unemployment and GDP reduced but did not eliminate party effect. Suicide and homicide rates were higher in states that voted for the Republican candidate for presidency than in states that voted for the Democratic candidate (pb0.0001 and pb0.07).​Conclusions:Violent deaths were associated with an increase under Republican presidents and a decrease under Democratic presidents, were higher in states that vote for the Republican than for the Democratic presidential candidates, and increased alongside increasing unemployment and falling national GDP. As with heart disease, obesity and cancer, identified associations with environmental factors can increase understanding of the public health problem and point to ways of reducing it. Future research beyond the boundaries of the United States could help elucidate the relationship between government, socioeconomic policy orientation, and violent death rates.