# Modelling death rates due to COVID-19: A Bayesian approach

**Authors:** Cristian Bayes, Victor Sal y Rosas, Luis Valdivieso  
**arXiv:** [2004.02386](https://arxiv.org/abs/2004.02386)  
**Date:** 2020-04-06  
**Summary page:** https://www.summarizepaper.com/en/paper/2004.02386v2/

## Key points

- Authors conducted a study titled "Modelling death rates due to COVID-19: A Bayesian approach"
- Study aimed to estimate the number of deaths in Peru caused by COVID-19
- Researchers used prior information from daily deaths in China and Peruvian data
- Constructed a predictive Bayesian non-linear model for deaths in Peru
- Objective was to provide estimates to help authorities monitor and manage the pandemic
- Estimated total deaths in Peru around 612 persons, with a 95% confidence interval ranging from 604.3 to 833.7 persons
- Around 99% of expected deaths would be observed approximately 64 days after first reported death
- Inflection point in number of deaths estimated around day 26 after first reported death, with a confidence interval ranging from 25.1 to 26.8 days
- Estimates provide valuable insights for authorities in monitoring and managing the COVID-19 pandemic in Peru
- Findings can inform decision-making processes and help implement effective strategies to mitigate further spread and reduce mortality rates associated with COVID-19

## Summary

Authors Cristian Bayes, Victor Sal y Rosas, and Luis Valdivieso conducted a study titled "Modelling death rates due to COVID-19: A Bayesian approach" to estimate the number of deaths in Peru caused by COVID-19. The researchers used prior information from the daily number of deaths due to COVID-19 in China and data from Peruvian authorities to construct a predictive Bayesian non-linear model for the number of deaths in Peru. The objective was to provide estimates that could help authorities monitor the epidemic and implement strategies to manage the pandemic. The study focused on the exposure of COVID-19 and its outcome, which is the number of deaths. By assuming an intervention level similar to that implemented in China, the researchers estimated that the total number of deaths in Peru would be around 612 persons, with a 95% confidence interval ranging from 604.3 to 833.7 persons. Furthermore, the study found that approximately 64 days after the first reported death, around 99% of expected deaths would be observed. The inflection point in the number of deaths was estimated to occur around day 26 after the first reported death, with a confidence interval ranging from 25.1 to 26.8 days. Overall, these estimates provide valuable insights for authorities in monitoring and managing the COVID-19 pandemic in Peru. The findings can inform decision-making processes and help implement effective strategies to mitigate further spread and reduce mortality rates associated with COVID-19.