Irma deaths line graph
In September, October and November of 2017, 62 people died in the Virgin Islands, compared to an average of about 30 in that three-month period over the previous decade. Then the 2017 numbers dropped to more closely mirror the previous average. (Graphic: FREEMAN ROGERS)

Jared Lettsome, a friendly 30-year-old with Down syndrome, survived Hurricane Irma huddled in a closet with his family as the wind blew the roof off their East End home.

Sixteen days later, he died suddenly from a pulmonary embolism.

“He wasn’t sick or anything,” said his sister Jenel Lettsome, who explained that a blood clot formed in his leg before eventually traveling to his lungs and causing his death. “It came out of the blue.”

Ms. Lettsome is inclined to blame Irma, which left her previously active brother housebound with an uncle in Josiahs Bay.

“It could have been that he received a clot because my house got damaged and he was in the closet,” Ms. Lettsome said, adding that he was afraid, confused and stressed out during the storm’s aftermath. “It could have been a situation where he got a clot from sitting in the same position for too long, not being able to move around.”

Mr. Lettsome is not the only resident whose post-Irma death has been blamed on the hurricane.

Though the government’s official death toll is four — the number of people reportedly killed on the day Irma passed over the territory — the death rate spiked in the subsequent weeks.

In September, October and November of 2017, 62 people died in the Virgin Islands — more than double the average for that three-month period over the previous ten years — before the rate leveled back out in December.

Such numbers suggest that Irma’s total death toll could be closer to 34, according to Dr. Irwin Redlener, a Columbia University professor who has investigated mortality in Puerto Rico after Hurricane Maria.

“It’s a different perspective when you look long term at why people are dying in the months, weeks and years after the storm, and there’s reasons why there are late fatalities in addition to the early ones,” Dr. Redlener explained, adding that the VI numbers are “additional evidence on top of the studies done in Puerto Rico that we need to take a more sophisticated look at fatalities that occur in the aftermath of such disasters.”

Nearly two years after Irma, the VI government has not launched a comprehensive study of post-storm deaths. Until it does, the official toll will probably remain at four.

Jared Lettsome
Jared Lettsome celebrated his 30th birthday shortly before Hurricane Irma. Sixteen days after the storm, he died of a pulmonary embolism, which his sister believes could have been caused by his lack of mobility in the hurricane’s aftermath. (Photo: PROVIDED)


During the 10 years before Irma, an average of about 9.5 people died per month in the VI, according to government records. In the month of the storm, which struck on Sept. 6, 2017, a 10-year record of 24 people died. October 2017 was the third-deadliest month for the period, with 21 deaths, and another 17 died in November.

In December, the number dropped to 10, apparently marking a return to normal: In the subsequent year, the VI averaged about 9.6 deaths per month, only slightly higher than in the pre-storm decade.

Largely because of the post-Irma spike, 2017 saw a 10-year record of 152 deaths, nearly 40 higher than the average of slightly over 112 in the previous ten years. In 2018, the overall number dropped back down, to 115.

Dr. Redlener explained that Irma’s full toll can be estimated by adding the four “direct” fatalities that government blamed on Irma to the deaths that exceeded the average in the subsequent weeks.

This method brings the total up to about 34.

And the toll could be higher still: Because the government has no official tally of evacuees who died abroad, this estimate includes only deaths that occurred within the territory.

Such spikes in mortality are common after a major disaster, according to Dr. Ali Asgary, an associate professor of disaster and emergency management at York University in Toronto, Canada.

“Looking at your data, it didn’t really surprise me when I see it, based on research, based on my reading of previous disaster cases,” he said. “In fact, the same probably that you are facing here exists in almost all previous disasters, including in [the United States] and Canada.”

After reviewing the VI’s numbers, Dr. Asgary drew similar inferences as Dr. Redlener.

“We have a huge gap between what is called direct deaths of hurricanes and what you can refer to as indirect or overall,” Dr. Asgary said. “Why? It differs from community to community and situation to situation, but people are impacted by disasters and hurricanes in particular.”

In the VI, many Irma survivors now have a keen understanding of the risk factors, which disproportionately affect the elderly, the ill and other vulnerable groups, according to Dr. Redlener.

“If roads were down and public transportation was out of commission and that made it difficult for people to get to their regular care providers for anything — for heart disease, hypertension, lung disease, any of those kinds of issues — and they weren’t getting the regular care that they needed, then people could get complications and could die from those,” said the doctor, who is the founder and director of Columbia University’s National Center for Disaster Preparedness. “If communications are down and they have an acute situation two weeks after the storm and they have no way of contacting anybody or getting an ambulance … and don’t get the care they need acutely, that’s a cause of death.”

BVI Health Services Authority CEO Dr. Ronald Georges, who at the time of Irma was government’s medical officer of health, saw such problems first hand in the weeks after the storm.

“They would have negatively impacted logistics to maintain the health care system, and they would have also impacted the access of individuals who needed to access the health system,” he explained. “So those things were severely impacted, particularly in the districts on Tortola.

Another risk factor is stress, which can be particularly dangerous for people with pre-existing health conditions, the doctors explained.


Candida Woolhouse believes that her long-time partner Peter Morris might still be alive if it weren’t for Irma-related stress.

After the hurricane, Mr. Morris had to secure their devastated Havers home and his business, Cay Electronics at Nanny Cay, while Ms. Woolhouse and their 11-year-old daughter evacuated to another house along with her mother, who has Alzheimer’s disease.

“He stayed behind and saved the house really,” Ms. Woolhouse recalled. “We had to put tarpaulins up, and he did his Boy Scout thing. And then obviously the weeks after that he was trying to save his business.”

The effort, she believes, proved too much for the 73-year-old.

“It was just tremendous extra stress load,” she said, adding, “Peter was on his own — isolated — and I couldn’t even get back and see what was going on.”


Peter Morris
After struggling for weeks to secure his home and business, Peter Morris died of a stroke on Oct. 27, 2017. His long-time partner believes Irma-related stress was likely to blame. (Photo: PROVIDED)

On Oct. 27, 2017, Mr. Morris texted Ms. Woolhouse from his office to say that he wasn’t feeling well. Then he lay down on the floor with a pillow under his head and died of a stroke.

“I don’t think he would have ended up like that if it hadn’t been for the inordinate stress: trying to save the house, the home, the office,” Ms. Woolhouse said, adding that he didn’t have a history of serious health problems. “He was perfectly fine that early morning. He went out bouncing and happy, so it was very, very unexpected.”

In spite of Ms. Woolhouse’s suspicions, Irma is not listed as a cause on Mr. Morris’s death certificate. The same goes for others who died in the months after the hurricane.

In fact, the death register at the Civil Registry and Passport Office mentions Irma only twice, each time for one of the fatalities that government attributed to the storm: The causes for 55-year-old Xavier “Dag” Samuels and Derrick Ragnauth both cite “multiple trauma to the body during Hurricane Irma.” The register does not list the storm as a cause of the other two deaths that government blamed on Irma: 66-year-old Richard Benson, who drowned, and 54-year-old Charles Thomas, who died from pneumonia as a result of a respiratory tract infection, according to the register.

Sharleen DaBreo-Lettsome, the director of the Department of Disaster Management, said last week that these four deaths, which government confirmed in October 2017, still make up the official toll. However, in a post-storm review for Cabinet, DDM did note a spike in mortality in 2017 compared to the previous five years.

“It seems safe to say that at least some of those deaths were indirectly related to Hurricane Irma,” Ms. DaBreo-Lettsome wrote in an email.

But finding out more specific information, she said, would be difficult.

“Calculating deaths after a disaster is no easy task,” she explained, adding, “There is a lot of controversy over the best ways to determine deaths related to disasters, with some scholars preferring to visit households and interview a sample of the population, others choosing to analyse information from death certificates, and many comparing post-disaster death figures to available data from before a disaster.”


Currently, there is no internationally accepted standard for assessing post-disaster deaths or for calculating an indirect death toll, according to Dr. Redlener.

Doctors in the US sometimes assess individual deaths using guidelines devised by the US Centers for Disease Control, but this system is not consistently employed across the country.

Dr. Redlener, who described the CDC standards as “inadequate,” is currently helping to devise a new system as co-chairman of the recently established “9/20 Commission” in Puerto Rico.

“We’re obviously going to be looking at whatever’s out there, but we certainly think it’s appropriate to have new international standards for looking at these things,” he said.

The 9/20 Commission, named for the date Maria hit Puerto Rico, was launched after a study conducted last year by George Washington University researchers.

The team compared Puerto Rico’s post-Maria mortality to the pre-storm rate and estimated that the hurricane caused some 2,975 “excess” deaths over and above the 64 that the government originally reported.

“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, [such as] those caused by structural collapse, flying debris, floods and drownings,” stated the researchers, who studied the mortality rate in the six months after Maria. “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.”


Dr. Georges said he would like to see a similar study conducted here.

“It would be interesting to continue to look at the numbers, to see if some of the deaths were just brought forward: how many were brought forward versus how many were additional,” he told the Beacon last October, when he was the medical officer of health in the Ministry of Health and Social Development, adding that some residents “would have been in a situation where perhaps they might have had another two or three months of life: In that kind of situation [their deaths] could easily have been brought forward that two or three months.”

deaths by year

However, he said, such research will have to wait.

“My big issue is that the person that deals with pulling the data is sick and she hasn’t been in work for a while,” he said in October. “We’re supposed to be hiring a new surveillance person who has a strong [epidemic] background.”

The needed staff contingent still is not on board, but even with the requisite team the study likely would be particularly difficult in the VI: Since long before Irma, Dr. Georges admitted, the territory has struggled to adequately record causes of death.

“Traditionally, doctors don’t pay it much mind,” he said, adding, “We’ve had training in the past; we’ve improved the system, but it still has a long way to go.”

Dr. Georges stressed the difficulties of assessing a cause of death even in the best circumstances.

“It’s not just a matter of just writing it down,” he said. “You have to sit down and analyse and go and see what actually is underlying this cause. Because if somebody dies of a stroke, the underlying cause could be renal disease, but the renal disease was caused by diabetes and hypertension, so which do you put?”

Such questions are further complicated after a disaster.

“It’s not just crunching numbers, right?” he explained. “Because you also need to go into the story.”

Those stories, however, fade with time.

“We have to do it sooner rather than later because there’s a recall bias,” Dr. Georges said in October. “But still I think we should be able to get sufficient information.”


Family members of Aubrey Levons believe that Irma-related stress played a role in the death of the Anegada tourism pioneer, who launched the Big Bamboo Restaurant and Beach Bar on the island’s north shore in the 1980s.

On the evening of Nov. 2, 2017, Mr. Levons attended a party with Puerto Rican boaters who were visiting the island to bring relief supplies.

Early the next morning, he died of a heart attack before medical help could arrive.

“By the time I get to him in the bathroom he was already dead,” his wife Diane recalled.

Though the 67-year-old was diabetic, she added, he had no known history of heart disease.

“He had hypertension,” she said, “but that never bothered him.”

Ms. Levons believes his death was likely related to the stress of rebuilding after the storm.

“The work was a lot,” she said. “He had a lot of work going. He worked tirelessly from the time the hurricane was finished until the night of his death.”

Their son Ceon Levons spoke similarly.

“The hurricane made him see things that he’d never seen in all his life here in terms of certain trees that been through many hurricanes and never broke, broke — all of his lime trees in the yard that used to bear fresh limes,” the son told the Beacon last year. “The after-stress.”

Aubrey Levons
On Oct. 27, 2017 (above), Anegada tourism pioneer Aubrey Levons gave the Beacon an energetic interview about his efforts to rebuild his restaurant and bar on the sister island. A week later, he died suddenly of a heart attack that family member believe may have been caused in part by Irma-related stress. (Photo: FREEMAN ROGERS)

Rosie Dawson, the office manager for the BVI Diabetes Association, said that greatly elevated stress levels persisted in the population at least into November 2017, when her organisation held a screening at Rite Way in Pasea.

Of more than 230 people screened, she said, about 40 percent had such high blood pressure that association officials urged them to rush to a doctor immediately.

“If they did not have a doctor, we had to send them to the hospital,” she explained, adding that during a pre-Irma screening offered in March 2017, only about four out of 40 people screened — 10 percent — had high blood pressure, a number she said was typical. “It seems after the hurricane, it’s stress that can carry up your blood pressure.”


Many residents also struggled to access health care after the storm. Though the five-year-old Peebles Hospital survived with only minor damage — a success story that provided a stark contrast to the USVI and Puerto Rico, where many hospitals were largely destroyed by Irma and Maria — it struggled to meet demand.

“Confusion, chaos,” Dr. Georges recalled of Peebles in the days after Irma. “Some order to the chaos, but a lot of people in a small space trying to do a lot of different things at the same time.”

For weeks, injured, sick and homeless residents competed for space with government officials who used the building as the National Emergency Operations Centre.

The situation elsewhere was worse: Some residents were unable even to reach the hospital, and community clinics across the territory were destroyed by the storm.

“The access of individuals who needed to access [health care and medicine was] severely impacted, particularly in Tortola, in the districts on Tortola,” explained Dr. Georges.

Ms. Dawson’s village of Carrot Bay was among the communities that were largely cut off from the rest of Tortola.

“I think you had persons who would go to the clinic, but some of those clinics were not functioning in some areas and some people weren’t driving,” she said. “Getting to health care facilities — that was big problems.”

In Carrot Bay, Ms. Dawson and some of her neighbours volunteered to help at-risk residents get the care they needed.

“I walked from house to house taking the prescription bottles to be filled and monitoring their vitals, and that was good,” she said. “I would bring them into town and refill and take back, especially my seniors.”

Insulin, she added, was a major concern for some diabetics, because it must be kept cold and many residents lacked access to electricity.

Alred Frett, the manager of B&F Medical Clinic in Road Town, said various types of medication were in short supply in Road Town as well.

“We had made a humongous efforts to stretch the medicine for persons,” he said. “Instead of giving them a four-month supply, they might have gotten enough to tide them over until the shipments would come in so we’d have enough for everybody.”

Dr. Georges blamed the broken supply lines.

“I think the main issue was a lot of the stuff comes in commercially, and because of the length of time the airport was closed and the ports were closed — the commercial traffic — it was difficult to replenish supplies,” he said. “A lot of things actually come in by FedEx, UPS, courier.”

Mr. Frett also complained that his clinic didn’t receive the support he felt it needed.

“I feel that the government wasn’t doing enough to be helpful,” he said. “In our case, after we worked so hard, they came to us and they promised they gonna help with this, that and the other. And zero.”

Still, Mr. Frett added that he didn’t know of any death that could be blamed directly on such issues.

“At my clinic, what you may have seen more of would have been persons stressed out, and you’d have had persons who would have been hypertensive, diabetic: These persons would have been a more significant risk, and I would think that could have contributed,” he explained. “I don’t know what can be traced directly to it. … My suspicion is that these persons were already stressed and they became more susceptible to disease.”

On Virgin Gorda, Vincent Wheatley, who helped operate the VG Recovery Operations Centre and now serves as the minister of natural resources, labour and immigration, acknowledged that health care posed a challenge on the sister island after Irma, but said the Nurse Iris O’Neal Medical Centre continued to function.

He added that it is difficult to determine whether any post-storm deaths on Virgin Gorda were caused by Irma.

“The hurricane might have aggravated certain things, but you have no way of knowing,” he said last year. “It’s only speculation.”


Nevertheless, one Virgin Gorda resident blames Irma on the early October 2017 death of a relative who she said was unable to access the medical care and medicine he needed in part because he lived on the sister island.

The relative, who was in his 60s, had suffered a haemorrhage and aneurism weeks before the storm, she explained, and his health had been improving steadily. Then Irma hit.

“He resided in Virgin Gorda, and there was limited ferry service to get to Tortola,” she said, speaking on condition of anonymity. “And I guess everything just came on at one time. Because he was pretty okay and doing well prior to the storm, but after that everything just escalated with the storm. He was a taxi driver, so with the economy being hit so hard and having to be backed up with different bills, he took all those into consideration. So that deteriorated his health.”

The days after the storm, she recalled, were harrowing.

“It was like trying to balance both conditions to stabilise it, and finding the right medication at the time was a challenge,” she said, adding, “Getting him overseas was not feasible at the time.”

Other survivors are reluctant to point fingers at Irma as the cause of a loved one’s death. Police officer Francia James-Liburd died of a heart attack on Oct. 6, 2017 at her Fort Hill home while awaiting an ambulance. Her husband Lincoln Liburd doesn’t blame the hurricane, but he does blame the emergency response time.

“It took the ambulance nearly 30 minutes to get from the hospital to Fort Hill, which is a five-minute drive,” he said. “I believe that was the main factor there, because you’ve got someone with a heart attack and they can’t find the home.”

Though the medical system was still facing severe challenges at the time, Mr. Liburd believes the ambulance would have been delayed anyway, and he doesn’t think his 50-year-old wife’s stress levels were dramatically elevated: The two were away when Irma hit, and they returned to find that their home and vehicles had not sustained major damage.

Other survivors aren’t sure what to blame. Joseph Frett’s 68-year-old wife Nedia died on Sept. 25, 2017 at Peebles Hospital after an infection developed in her foot, which she injured in her backyard following the storm. Mr. Frett, who has hung a portrait of his wife in front of his Little Dix Hill home, was hesitant to speak in detail, but he said that Ms. Frett suffered from existing medical conditions including diabetes.

Asked if the storm was to blame for her death, he shrugged his shoulders.

Nedia Frett died in Peebles Hospital on Sept. 25, 2017 with an infection in her foot, which she injured in her backyard after Irma. Her husband has hung her portrait (above) in front of his Little Dix Hill home. (Photo: FREEMAN ROGERS)


Drs. Redlener and Asgary stressed the importance of assessing post-storm mortality, particularly with another hurricane season now under way.

“The more we know about what causes the long-term deaths, the more we can do to support programmes, initiatives, access that would mitigate those factors,” Dr. Redlener said.

The death count, he added, also affects the international response.

“There’s a political issue involved here, which is that if you — quote, unquote — only had four fatalities reported but the fact is there were 40 or 50 or 60, [that helps] disincentivise media coverage of the severity of the storm, and that’s related to getting enough resources to mount an appropriate recovery,” he said. “If it looks like it was a minor event — because people will say, ‘Oh, it’s only a small handful that died’ — that’s much less likely to incentivise significant recovery money.”

This effect was evident after the George Washington University researchers estimated that nearly 3,000 people had died due to Maria in Puerto Rico. US President Donald Trump hotly disputed the number without providing contrary evidence.

“3000 people did not die in the two hurricanes that hit Puerto Rico,” Mr. Trump tweeted last September after the release of the study, which was commissioned by the Puerto Rico government. “When I left the Island, AFTER the storm had hit, they had anywhere from 6 to 18 deaths. As time went by it did not go up by much. Then, a long time later, they started to report really large numbers, like 3000 [sic].”

The president added that the new number was created “by the Democrats in order to make me look as bad as possible when I was successfully raising Billions of Dollars to help rebuild Puerto Rico.”

The GWU researchers stood by their study, and Mr. Trump has been accused of underfunding the relief effort and exacerbating Puerto Rico’s plight with his angry words.

After other disasters, including Hurricane Katrina, similar controversies have been hotly debated. Dr. Redlener said that such conversations increased in frequency following the 2001 terror attacks in New York City.

“Since 9/11, there’s been, I think, generally a greater interest in the issue of disasters, whether they’re natural or manmade,” he said. “And then we have the Ebola crisis and lots of other things everyone’s concerned about; an avian flu epidemic. And then we started getting increased frequency of weather types of disasters, especially hurricanes and tropical storms.”

He added that climate change likely will make such questions increasingly relevant in the future.

“Years ago people said, ‘Well, this is just a cycle that occurs naturally,’ but I think there’s increasing evidence that unchecked climate change is now responsible for increased frequency and increased intensity of storms,” he said.


For now, the VI likely will have to continue waiting for a study on post-Irma deaths. Though Dr. Georges agreed that such research would be valuable, he predicted that other issues would continue to get priority given the limitations imposed by the recovery process.

“What happened in the territory impacted everybody’s health status,” he said. “Some people, we didn’t get sick, or we did; others died.”

The general reasons for post-storm deaths, he added, are well known.

“We actually don’t need a study to figure that out and to say, ‘Well, these are the things that we have to strengthen,’” he explained. “And I guess that’s positive. You kind of focus on some other things that we need to strengthen rather than trying to figure out, ‘Yeah, well, okay: Ten people died because of this and 15 of, you know.’”