Here, I provide the code for Duration of post-disaster mortality increments using placebo tests: Hurricane Maria in Puerto Rico.
Objectives: The purpose of this study is to determine the duration excess deaths attributable to Hurricanes Irma and Maria in Puerto Rico using a placebo test approach. Methods: Data for this study come from the Puerto Rico Vital Statistics System between 2000 and 2018. Linear regression models are used to: (1) test whether pre-hurricane death counts were consistent with the previous years, (2) measure the effect of Hurricanes Irma and Maria, and (3) determine whether deaths in 2018 were consistent with 2000-2016 patterns. Results: Monthly death counts for January - August 2017 were consistent to the previous years (2000-2016). A significant displacement is observed for September, October, and November 2017 in comparison to the reference period. No effect is observed for December 2017 or the first six months of 2018 in comparison to 2000-2016. Conclusions: Excess deaths attributable to Hurricane Maria were concentrated between September and November. The results suggest that no deaths occurred in excess of expected levels after November 2017. The use of placebo tests, to measure post-disaster displacements in mortality, can help determine the duration of the disruptions in events of interest caused by environmental disasters.
/SCRIPTS/- Script required to produce the Figures and Regression results included in the article./DATA/- Includes monthly death counts obtained from the Puerto Rico Vital Statistics System from 2000-2018.
For any issues with the functionality of these scripts, please create an issue.