Why Are the Death Tolls in Puerto Rico From Hurricane Maria So Different?

Widely different estimates of Hurricane Maria’s death toll in Puerto Rico have led to confusion. Here is a guide to the tallies, what accounts for their differences and how a new study aims to provide a more definitive account:

In December, The New York Times analyzed vital statistics from the Puerto Rican government that showed that in the 42 days after Hurricane Maria made landfall on Sept. 20, 2017, 1,052 more people than usual died in Puerto Rico.

That figure was particularly striking because thousands of people had left the island, including many with chronic medical conditions. Based on the likelihood that the population there was smaller in the fall of 2017, we would have expected the number of deaths per day to decrease, not increase.

To obtain our figure of 1,052, we compared the number of deaths for each day in 2017 with the average of the number of deaths for the same days in 2015 and 2016. The figures came from the Puerto Rican government, which provided us with tables showing the number of deaths per day and deaths broken down by cause. The 2017 numbers were preliminary, so we limited our analysis to September and October.

We found that people died from certain causes more often in September 2017 than they had in the previous two years. Those included sepsis — a complication of severe infection — which jumped 50 percent over the previous year. Pneumonia and emphysema deaths also rose notably.

Using the same data and similar methods, Puerto Rico’s Center for Investigative Journalism reached a similar estimate, that 1,065 more people than usual had died in September and October. Likewise, demographer Alexis Santos of Pennsylvania State University, and Jeffrey Howard, an independent researcher, compared deaths in 2017 with figures derived from 2010 to 2016 and established an estimate of 1,085 excess deaths.

The study, which was published in The New England Journal of Medicine, one of the most highly regarded peer-reviewed medical journals, analyzed a longer period than we did. It also used completely different methods.

Researchers visited more than 3,000 residences across the island and interviewed their occupants, asking whether anyone in their households had died and whether the storm and its aftermath might have contributed. Residents reported that 38 people living in their households had died between Sept. 20, 2017, when Hurricane Maria struck, and the end of that year.

That toll, converted into a mortality rate, was extrapolated to the larger population and compared with official statistics from the same period in 2016. Researchers arrived at an estimate of roughly 4,600.

Because the number of households surveyed was relatively small in comparison to the population’s size, there was a large margin of error. The true number of deaths beyond what was expected could range from nearly 800 to close to 8,500 people, the researchers’ calculations showed. The widely reported figure of 4,645 was simply the midpoint of that statistical window, known as a 95 percent confidence interval. Including a midpoint figure in such a report is standard academic practice.

The study’s main finding was that residents of Puerto Rico died at a significantly higher rate during the three months after the hurricane than they did during the same period in the previous year, and that roughly a third of those deaths resulted from delayed medical care. The researchers said in the report that their conclusions were consistent with the analyses of The Times and others.

“All these methods complement each other,” said Dr. Satchit Balsari, one of the study’s senior authors. “By no means is the number we’re estimating the end-all.”

At the time our analysis was released, the official death toll was 64. That number included only people whose death certificates listed Hurricane Maria as a contributor, as certified by the Puerto Rico Forensic Sciences Institute in San Juan. But under pressure from a skeptical public, the Puerto Rican government announced in December that all deaths that had occurred in the months after Hurricane Maria would be reviewed and that people who died either directly or indirectly from the storm and its aftermath would be included in a revised tally.

In relation to hurricane deaths, the term “direct” means those that occurred from drowning or other effects of the storm itself. “Indirect” deaths include those in which related factors, such as difficulty reaching a hospital for care, or trouble refilling prescriptions, played a role. The Centers for Disease Control and Prevention recommends that medical examiners include both types of deaths in official storm tolls, and the government of Puerto Rico has agreed to do so.

The government acknowledged that its tally of 64 was likely to be a significant undercount. In the days after the storm, with widespread power outages and extreme difficulty moving around the island, it was likely that many storm-related deaths went uninvestigated by the island’s medical examiner.

In November, CNN compiled figures from half the island’s funeral homes to report that funeral directors believed that 499 more deaths than the official count were tied to the hurricane. In many instances, local doctors had not sent the cases to the forensics bureau for analysis.

Also in November, BuzzFeed reported that 911 bodies were cremated following Hurricane Maria without being examined, and that in many cases, funeral home and crematory directors believed that those people had died as a result of the hurricane.

The government of Puerto Rico has a contract with the Milken Institute School of Public Health at George Washington University to conduct a more thorough review. The initial phase of the study will use the island’s vital records and information from funeral homes, the medical system and the larger public. In a second phase, the researchers will conduct interviews with survivors of those who died. Dr. Lynn Goldman, the school’s dean, said she expected to deliver the initial review this summer, with the results from the second phase following perhaps nine months later.

In recent months, pending the completion of the review, the government of Puerto Rico had stopped issuing statistics on deaths. After Puerto Rico’s Institute of Statistics went to court to demand their release, the commonwealth’s Health Department on Friday released a table of overall deaths by month. Figures for September and October of 2017 were similar to those that had been released to The Times last year, having increased slightly. Total deaths in November and beyond did not appear elevated compared with previous years. However the population of Puerto Rico had likely declined, so an analysis of the rate of deaths would be needed to detect any possible increases, also accounting for changes in the makeup of the population remaining.