At least 4,645 people died in Puerto Rico after Hurricane Maria made landfall last fall, more than 70 times higher than the official government count, according to a new Harvard study. The study, published in the New England Journal of Medicine, says the government’s official death count is much too low.

“Our results indicate that the official death count of 64 is a substantial underestimate of the true burden of mortality after Hurricane Maria,” wrote the researchers.

Earlier this year, researchers surveyed 3,299 randomly selected households (of more than 1.1 million) across the US territory about displacement, infrastructure loss, and causes of death. Between September 20 and December 2018, one-third of deaths could be attributed to “delayed or interrupted healthcare”, including an inability to access medication, equipment needing electricity, closed facilities, and absent doctors. It’s unclear how many of those could have been prevented had access to adequate care been provided.

On average, households went 84 days without electricity, 68 without water, and 41 without telephone coverage during those months. The area also reported “substantial population displacement” – 2.8 percent of those sampled reporting having to leave their home because of the hurricane.

Gathering data on Puerto Rico's death counts has been a contentious and difficult task. In Puerto Rico, every disaster-related death must be confirmed by the Institute of Forensic Sciences by bringing the body to San Juan or having a medical examiner travel to verify the death. This causes a delay in death certificates as well as an increase in indirect deaths from worsening conditions that may not be captured.       

“These numbers will serve as an important independent comparison to official statistics from death-registry data, which are currently being re-evaluated, and underscore the inattention of the U.S. government to the frail infrastructure of Puerto Rico,” the research team wrote.

The Category 4 hurricane made landfall on September 20, 2017, following the destruction of Hurricane Irma two weeks earlier, further interrupting the water supply, electricity, telecommunications, and access to medical care. Thousands were displaced from their homes, seeking shelter across the territory and in the mainland US. Following the disaster, the Trump administration received criticism for its response to the growing humanitarian crisis. 

In December, Puerto Rico’s governor issued a review of the death toll that resulted in an estimate of more than 1,000 in the month following landfall. This study pulls on further data made available in November and December, as well as an expansion of how hurricane-related deaths are defined. According to the Centers for Disease Control and Prevention, deaths can be attributed to a tropical cyclone if they are “caused by forces related to the event, such as flying debris, or if they are caused by unsafe or unhealthy conditions” following the disaster.

Because the survey couldn’t account for people who lived alone and died following the storm, the total death count could be more than 5,700. However, researchers used a more conservative estimate to address recall bias and errors innate in self-reported surveys.

The researchers say accurate estimates of deaths, injuries, illness, and displacement following such a disaster are “critical to the immediate response” as well as the planning of future preparedness and risk reduction. They have made their data publicly available for additional analyses.