Hurricane Maria resulted in 1139 excess deaths in Puerto Rico between September and December 2017, when death counts returned to average, according to a new analysis of death records.
The storm was called “the biggest catastrophe in modern history” by Puerto Rico’s mayor, and the aftermath was devastating: extreme flooding, power outages for months, contaminated drinking water and food shortages.
Yet the US government says there were just 64 deaths between 20 September 2017 – when the hurricane made landfall in Puerto Rico – and the end of December 2017. That’s because they only counted people whose death certificate listed the hurricane as their cause of death.
New estimate
Now, Alexis Santos-Lozada at Pennsylvania State University and Jeffrey Howard at the University of Texas at San Antonio have a new estimate based on death records compiled by the government of Puerto Rico for each month of 2017.
They calculated the mean monthly deaths between 2010 and 2016 to establish a historical baseline, and then compared the 2017 statistics to see how many excess deaths were recorded. Based on the uncertainty of the statistics, they created a range of numbers of deaths for each month. Any death counts higher than the top of the range were considered excess deaths.
The researchers found that there were 459 excess deaths in September, 564 in October, and 116 in November. By December 2017, the death count in Puerto Rico had returned to a level that was at the high end of historical averages, but no longer significantly higher than normal.
Read more: How Atlantic upheaval will make future weather even more chaotic
Another study from July put the post-Maria death count at 4645. In that case, researchers surveyed more than 3000 randomly chosen households in Puerto Rico to gather information on deaths that may have happened outside of hospitals and were therefore potentially overlooked. They reported 56 deaths, 18 before the hurricane and 38 after, and then extrapolated to the whole population.
“Because these are relatively small numbers this resulted in a large amount of uncertainty,” says Howard. Based on their work, the death toll could have been anywhere from 793 to 8498 – the final number they reported is the mean of that range.
Howard and Santos-Lozada say their estimate is conservative and has less uncertainty, though it doesn’t include the cause of the deaths. “However, the individual death record data with causes of death has been released by the Puerto Rican government, and could be the subject of further studies. I hope that it is,” Howard says.
Journal reference: JAMA, DOI: 10.1001/jama.2018.10929
