The current argument over precisely how many people died in Puerto Rico when Hurricane Maria came barreling through has gotten ugly. The original death toll was put in double digits, but recently a new study placed the “official” count at just short of 3,000. The President has disputed this, leading to all sorts of acrimony, including the recent decision by House Hispanic Caucus to snub an invitation to the White House. But where did that gargantuan number come from and was it “politically motivated,” as Donald Trump has suggested? Perhaps a bit of calm discussion can lead to some sanity here.

Karen Tumulty of the Washington Post tossed out a link to an article from her newspaper on Twitter this morning purporting to settle the issue.

The article at the WaPo turns out to be from Lynn R. Goldman of the Milken Institute School of Public Health at George Washington University, which compiled the death toll of 2,975 “excess deaths” in Puerto Rico as a result of Hurricane Maria. Bear with me here, because the phrase “excess deaths” will become critical later. She took to the pages of the Washington Post to assure everyone that their study “was carried out with no interference whatsoever from any political party or institution.” (Emphasis added)

Make no mistake: The death toll did continue to rise in the months after Maria. In September 2017, when Puerto Rico recorded a total of 2,906 deaths, we found there was an excess of 574 deaths above what would have been expected in a year without the storm. The death toll continued to mount every day, with an excess of 697 deaths in October, 347 in November, 479 in December, 558 in January, and 320 in February, for a total of 2,975.

Throughout that time, researchers produced other estimates. The government of Puerto Rico came up with a figure of 68 excess deaths through October 2017; the New York Times, 1,052 through October 2017; a team at Harvard University, 4,645 through December 2017; and researchers at Penn State University, 1,139 through December. Of these, only the estimate by the government of Puerto Rico involved the examination of individual deaths to determine if the hurricane had caused them.

Here’s where we run into trouble. Look at the description of the methodology they used. It’s based on “a careful examination of all of the deaths officially reported to the government of Puerto Rico between September 2017 and February 2018… using state-of-the-art mathematical modeling to compare the total number of deaths during that time to the expected number of deaths, based on historical patterns as well as age, sex, socioeconomic status and migration from the island.”

They’re counting deaths which took place five months after the storm had passed.

Pardon my incredulousness over such a serious matter, but when did we start defining hurricane deaths in this fashion? When we think of people dying in a hurricane, we picture those who are washed out to sea, struck by flying debris, trapped under collapsed buildings, drowned in rising storm surge waters or even expiring from exposure while trapped on their roofs awaiting rescue. But at some point the storm is over and the immediate rescue and recovery operations are complete. People no doubt suffer ill effects from a deadly storm for some time to come, but that’s largely a matter of how good the medical and social support services available to them happen to be.

And just picture that number. 2,975 dead from Hurricane Maria. It’s not that hurricanes can’t kill that many people or more. It’s certainly happened. The 1900 Galveston hurricane is credited with killing anywhere from six to twelve thousand people (nobody is really sure) because it washed an entire town out into the Gulf of Mexico. But other big, damaging hurricanes in the modern era haven’t been “officially listed” as coming close to that. By the end of September 2005, the death toll of Hurricane Katrina was pegged at 1,033. A comprehensive study done almost ten years later raised the number, but only to 1,170. They were studying people who actually died in or directly from the hurricane. And they were looking at autopsy reports for conclusive causes of death.

This new study of Maria, as indicated above, was looking at the number of people who probably would have died on average in each month following the storm going out to February fo the following year. Then they looked at the number of actual death certificates they could find (no matter what they died of) and any in excess of the projection were attributed to the storm. Five. Months. Later.

I’m not saying that there was or wasn’t any political bias involved in the compilation of the Milken Institute study, but that’s really not the point. The question is whether or not this is actual science. If you’re suggesting that every person who gets sick and dies months after a storm comes through because the infrastructure is inadequate, or medical supplies aren’t available, or because they were bitten by a dog who might have been too hungry because he lost his owners in the storm, are directly attributable to the hurricane, then… please. If that’s the case, the death tolls of every hurricane, tornado, wildfire, earthquake or drought in the history of the country should probably be multiplied by a factor of anywhere from 10 to 100.