What Was the Actual Death Toll from Hurricane Maria in Puerto Rico?

Puerto Rico officials originally set the hurricane's death toll at 64, but a more recent assessment of the data yielded an estimate closer to 3,000.

  • Published 4 June 2018
  • Updated 28 August 2018

The governor of Puerto Rico has revised the territory’s official Hurricane Maria death toll to 2,975 following the release of a long-awaited study by the Milken Institute of Public Health at George Washington University on 28 August 2018.

The revised figure, which encompasses all Maria-related fatalities during the six months following the September 2017 storm, is more than 46 times higher than the previous official estimate of 64 and more than twice as high as the government’s later, unofficial estimate of 1,400. For comparison’s sake, Hurricane Katrina — a far costlier storm in terms of property damage — caused 1,833 deaths.

The original count had languished under a cloud of skepticism ever since it was announced by Puerto Rico’s Department of Public Safety in December 2017. The Category 4 hurricane was the worst natural disaster ever to strike the United States territory and the third costliest storm on record, having caused an estimated $90 billion in damage. Even before government officials locked the Hurricane Maria death toll in at 64, statistics-based estimates by a number of different sources indicated that the actual total could top 1,000.

The GWU study, conducted in collaboration with researchers from the University of Puerto Rico Graduate School of Public Health, was commissioned by Governor Ricardo Rosselló in February 2018 in response to widespread criticism of the earlier estimate. GWU’s results differ from those of a July 2018 study by Harvard University which concluded, based on survey data with a wide margin of uncertainty, that as many as 4,645 Puerto Ricans died as a result of the hurricane over a two-and-a-half month period. The newer findings relied on government mortality data and death certificates, which the researchers say afford a more accurate accounting of the death toll than survey data.

Both studies used their respective data to calculate overall mortality rates for the time periods covered, which were then compared to the expected mortality rate if the hurricane had not occurred. This yielded, in each instance, an “excess mortality rate” that could be attributed to the storm. GWU’s researchers arrived an excess mortality rate of 22 percent. That percentage was applied to the population as a whole to estimate a final death toll of 2,975.

The researchers also found that the poor and elderly were at greatest risk of failing to survive the hurricane:

All municipalities in Puerto Rico were hit hard by the hurricane and the aftermath, however, certain groups faced the biggest risk. In fact, this study showed that the risk of dying over this period was the highest (60 percent higher than expected) for people living in the poorest municipalities — and that the elevated risk persisted beyond February 2018.

Over this same period, older male Puerto Ricans had a risk of death that was 35 percent higher than expected and that elevated risk continued past the study observation period.

Their 69-page report, entitled “Ascertainment of the Estimated Excess Mortality from Hurricane Maria in Puerto Rico,” explained the marked disparity between the original death toll and the revised one thus:

The official government estimate of 64 deaths from the hurricane is low primarily because the conventions used for causal attribution only allowed for classification of deaths attributable directly to the storm, e.g., those caused by structural collapse, flying debris, floods and drownings (see below). During our broader study, we found that many physicians were not oriented in the appropriate certification protocol. This translated into an inadequate indicator for monitoring mortality in the hurricane’s aftermath. Verification of attribution takes time, while excess mortality estimation is a more immediate indicator.

Most importantly, the report proffered key recommendations for helping not only Puerto Rico but the mainland United States develop better methods for natural disaster preparedness and response. These include:

All jurisdictions, not only Puerto Rico and other parts of the U.S. but also globally, should develop methods to rapidly assess total excess mortality after natural disasters and to provide that information to the public. Monitoring should look not only at overall rates of death but also for spikes in death rates in certain areas and within subpopulations, such as the elderly.

Puerto Rico specifically needs to fully staff these public health functions within the Department of Health, including the Vital Statistics Registry and the Bureau of Forensic Sciences. In addition, the island must strengthen the coordination between the Vital Statistics Registry and the Bureau of Forensic Sciences with the goal of creating a timely and accurate surveillance system.

Puerto Rico needs to conduct after-action reviews and use those, along with the results of the study, to create a new crisis and emergency risk communication plan, one that is integrated with government agency and municipal plans, has community and stakeholder involvement, and is aligned with the possibility of catastrophic disasters.

“If enacted, the recommendations of this report could help save lives in Puerto Rico and beyond,” its co-authors said in a joint statement released to the press. It was published too late to have much of an impact on preparedness for the 2018 hurricane season, which was already nearing its peak (albeit a less severe one than predicted) in late August, but it is hoped that it will make a significant difference in the future.

Snopes.com
Since 1994
A Word to Our Loyal Readers

Support Snopes and make a difference for readers everywhere.

Editorial
  • David Mikkelson
  • Doreen Marchionni
  • David Emery
  • Bond Huberman
  • Jordan Liles
  • Alex Kasprak
  • Dan Evon
  • Dan MacGuill
  • Bethania Palma
  • Liz Donaldson
Operations
  • Vinny Green
  • Ryan Miller
  • Chris Reilly
  • Chad Ort
  • Elyssa Young

Most Snopes assignments begin when readers ask us, “Is this true?” Those tips launch our fact-checkers on sprints across a vast range of political, scientific, legal, historical, and visual information. We investigate as thoroughly and quickly as possible and relay what we learn. Then another question arrives, and the race starts again.

We do this work every day at no cost to you, but it is far from free to produce, and we cannot afford to slow down. To ensure Snopes endures — and grows to serve more readers — we need a different kind of tip: We need your financial support.

Support Snopes so we continue to pursue the facts — for you and anyone searching for answers.

Team Snopes