Prior to this analysis, I did not hold strong opinions on the positive or negative effects of DST, but I fully acknowledged the role of adequate sleep in a healthy lifestyle. However, I believed scientific analysis and reporting could use statistical methods that are simpler, more explicit, less black-and-white, and more precise. Therefore, when I read about the results of Sandhu et al. (2014), which claimed a 20% decrease in the risk of heart attack on the Tuesday that follows an autumn DST change, I decided to investigate the issue through a Bayesian lens.
The results can be found on my blog, at https://treszkai.github.io/2019/11/11/dst-vs-ami, with accompanying software in the DSTvsAMI folder of this repository.