Hurricane Maria-Related Deaths May Be 70 Times Official Estimate

Hurricane Maria ravaged Puerto Rico almost a year ago, but scientists are still trying to unpack the many layers of damage it inflicted on the island.

Researchers are becoming more certain, though, of a sobering reality: Disaster deaths can last for far longer than we suspect.

A new study by a group of researchers from the Harvard T.H. Chan School of Public Health and other institutions suggests that the number of deaths caused by Maria in a three-month period (from impact on September 20 until that December) could figure around 4,600 people. Perhaps more importantly, this estimate (if accurate) would mean the mortality rate during this window of time was 62% higher than the mortality rate during the same period the year before.

Puerto Rican residents walk in flooded streets in Condado, San Juan, Puerto Rico

This unexpectedly large number of what scientists are calling “excess deaths” surpasses the number of people who died in Hurricane Katrina (1,833) and the September 11, 2001 terror attacks (2,996). Meanwhile, the government’s official casualty count from Hurricane Maria remains comparatively low (64).

The process of determining fatalities is a messy one, plagued with uncertainties. Much of the time, people struck by a catastrophe like Hurricane Maria die alone—maybe even forgotten. In other cases, they survive the initial disaster but die weeks or months later due to lack of medical supplies, food, or water.

In Puerto Rico, disaster-related deaths have to be confirmed by the Institute of Forensic Sciences, which means bodies need to be brought to San Juan or examiners themselves have to make the trek over to the deceased. Because of this, officials and researchers will often turn to door-to-door sampling of communities in order to get a sense of how many people were lost to a specific weather event.

For this particular analysis, researchers visited more than 3,000 residences across Puerto Rico and spoke with people who were still living there. They wanted to make their interviews as unbiased as possible, so they asked simple questions and let occupants report their difficulties from their own perspective. Collectively, these residents reported 38 deaths amongst their households. The team of scientists then converted that number into a mortality rate and then extrapolated the data to the overall population.

In addition to finding that the number of excess deaths in 2017 has a 95% chance of being between 800 and 8,000 (called a “confidence interval”), the team also discovered—and this is the crux of their paper—that the death rate stayed abnormally high through December.

“We think of these disasters as a one-time thing and then it’s over and you’re rebuilding,” said Satchit Balsai, a research fellow at the Harvard FXB Center for Health and Human Rights and one of the lead authors on the study. “But look at what happened to the medical system in Manhattan [after Hurricane Sandy]. We reeled from that upsurge in patient volume for at least one year. We lost hospitals, and patients were seeking care elsewhere. We really need to address this as a country and as a society.”

The study’s authors emphasize that the number being reported is just a calculated estimate, though. For a more detailed measure, researchers will need to do follow-up studies and probabilistic analyses (ones that convey the probability of each possible death count). The next big study that’s expected to hit the presses is one out of George Washington University this summer—it will assess the territory’s vital records as well as documentation from funeral homes, the medical system, and information from the public at large.

“The survey is a relatively small sample, and so the number itself is not a precise estimate. There are very large margins of error,” said Caroline Bukee, an associate professor of epidemiology at the Harvard T.H. Chan School of Public Health and one of the lead authors on the study. “But even our lowest estimate is 10 times higher than the official one.”

The data is open to the public, which means other researchers and organizations can dive into it with their own interpretations, expound on the findings, or use it to move in new policy directions. “We think it’s really important to put the numbers out there and then let people analyze it in multiple different ways,” Bukee said.

One of the most immediate concerns the study brings up is the lack of access to medical care in Puerto Rico after the hurricane. About 15% of people interview reported that they struggled to get the medicine they needed, and about 10% said they couldn’t use breathing equipment, which relies on electricity.

“The focus here is on the fact that the mortality rate was higher, and remained higher, for months after,” Satchit said. “We don’t know who in our neighborhoods is on dialysis, who’s on respiratory machines, who needs medical care. This is not the kind of granular information we have about our communities. That kind of community preparedness really needs to be built in.”