Donald Trump’s claim that there was a low death toll from Hurricane Maria is the most high-profile intervention in a long-running controversy.

The hurricane hit on September 20, 2017. On October 4, Trump went to the island and said there were six certified deaths related to the hurricane, then said ’16 people’ had died, citing that as evidence that it was not a catastrophe on the scale of Katrina.

Later that same day, after he left the government of Puerto Rico revised the figure of deaths to 34 – and for the first time specified how they had died. 

Governor Ricardo Rosello said 20 deaths resulted directly from the storm, including drownings and those killed in mudslides.

The total of 34 also includes sick and elderly who died in the aftermath of the hurricane, including some who died because oxygen could not be delivered amid power outages. There were also two suicides. That is the only breakdown ever given which showed how the island’s government was working out a death toll. None of the dead were named.

But the idea of including the sick and elderly and suicides is in line with the Centers for Disease Control guidelines on how to count deaths after a natural disaster.

They tell authorities to record deaths in two forms: directly related disaster death; and indirectly related disaster death.

The CDC say: ‘A directly related death is defined as a death directly attributable to the forces of the disaster or by the direct consequences of these forces, such as structural collapse, flying debris, or radiation exposure (

‘An indirectly related disaster death occurs when the unsafe or unhealthy conditions present during any phase of the disaster (i.e., pre-event or preparations, during the actual occurrence, or post-event during cleanup after a disaster) contribute to a death.’

That means that if someone is on life support and dies when it switches off because of a power cut, they are to be recorded as a disaster death. But someone who dies while on life support because of heart failure would not be.

Equally someone who dies in a car crash on a mud-covered road would be a disaster death. But someone who dies on the same road in a car crash after it is cleaned up would not be.

In November CNN reporters asked 112 funeral homes for an estimate of deaths related to the hurricane – and came up with 499. They only reached half the island’s funeral homes.

By December, the official death toll was stated as just 64, with no breakdown  of how that total was reached, but with names made public.

That total remained unchanged for months, with the governor saying it was calculated using the Centers for Disease Controls protocols for recording deaths. 

Check this out  MPs tweet pictures of themselves in Parliament after Supreme Court ruling on prorogation

But in February this year CNN and Puerto Rico’s Center for Investigative Journalism filed lawsuits demanding the island government release death certificates and related data – a demand the government fought for months, at taxpayer expense.

Among their claims was that the dead should have privacy.

In June the government lost and released data which showed that between September and December, compared to the previous four years, there were 1,427 deaths above the average – a huge increase in the death toll.

The government said it would wait for a study it had commissioned by George Washington University to decide on an official death toll revision.

But it did use the figure of 1,427 in a draft funding request to Congress in August.

At the end of August George Washington University published its report which put the number of deaths at 2,975 – not on the basis of counting bodies, or death certificates, but a statistical model of how many people would normally be expected to die versus how many people actually died. 

The model had to also take into account how many people fled the island, for which there were no official figures, to come to the ‘excess death’ total.

That means that its scientists did not definitively link 2,975 people’s deaths to the hurricane by examining their death certificates and looking for a cause of death like ‘drowning’, ‘impact of falling trees’, or ‘life support switched off during power failure’.

But the study did look at what happened with death certificates and was scathing. 

Doctors and forensic physicians, who sign death certificates, said they did not receive official guidance from the Puerto Rico Vital Statistics Register (PSVSR) on how to record deaths ‘during, or in conditions created by, a disaster’.

‘According to PSVSR personnel, a very small number of those completing death certificates did relate the deaths to the hurricane,’ the study warned.

‘Most other certificates lacked such information. This reduced the number of death certificates that indicated a relationship with the hurricane.’ 

Check this out  Credit Suisse boss who sometimes dresses as woman slammed for appearing in top 100 women list

Put simply, that means it was impossible for the academics to know how many people the hurricane really did kill – which is why the Georgetown academics turned to their statistical model. 

That model is accepted by FEMA and Trump’s own government agency accepted the estimate.

And there is another way of calculating the death toll: how many dead people’s families have asked FEMA for assistance with funeral and burial expenses.

An official revealed to CNN earlier this month that it has paid families of 247 people so far, has 2,350 cases ‘in various phases of review and processing’ and has rejected 420 applications. 

But just 11 rejections were because they were not related to the hurricane – 369 were because they had been unable to contact the person who filed the application.

And that FEMA figure does not include people who have not asked it for assistance. 

FEMA does not simply go by death certificates, but asks for further evidence to make sure that the deaths meet the CDC definition of being disaster related, either directly or indirectly.

That means that far from being a Democratic plot, his own government agency, run by a Republican appointee, has already concluded that the death toll is 247 – not the six to 18 he suggested – and could go up by another 2,350.