After the hurricane, the media wanted to inflate Puerto Rico’s death toll from dozens to thousands. Their media star was on television wearing, “We Are All Dying” shirts. Something had to be done.
That something was a series of dubious studies that I have been writing about and disproving for a while now.
The previous, incomplete death toll showed that the actual hurricane seemed to have only killed 19 people. 2 suicides were also attributed to the hurricane. As were respiratory failures “indirectly linked” to the hurricane.
But the media has been working hard to inflate the death toll.
Then came the Harvard study that used dubious models to estimate population size and death tolls.
Now Harvard has delivered. “Hurricane Maria killed more than 4,600 in Puerto Rico, not 64,“ USA Today blares. “A New Study Says Nearly 6,000 Died In Puerto Rico,“ BuzzFeed shouts. “Puerto Rico’s Hurricane Maria Death Toll Could Exceed 4,000,” the New York Times reports slightly more cautiously.
4,000, 4,600 or 6,000. Which is it?
Except that the margin of error is “plus or minus 3,852.”
Instead they surveyed 3,299 random Puerto Rican households, used the survey results to generate a death rate of 14.3 per 1,000 between September 20, the date the hurricane hit, and December 31.
That’s over 100 days after the hurricane hit.
Then they compared those numbers to the same period in 2016 and produced 4,645 deaths.
Now the official GWU study conducted in cooperation with the Puerto Rican government is out. And the media is crowing even though the study has some of the same problems.
Here’s CBS News.
Hurricane Maria killed far more people in Puerto Rico than initially thought, accounting for an estimated 2,975 deaths on the island from September 2017 through February 2018, according to a new analysis.
No, it really didn’t.
The study uses estimates of the population, based on possible displacement, and then guesses how many people should have normally died based on a fairly short trend time frame, and then assumes that deaths in excess of this may have been caused by the hurricane.
Somehow.
Here’s the actual study.
We estimate that in mid-September 2017 there were 3,327,917 inhabitants and in mid-February 2018 there were 3,048,173 inhabitants of Puerto Rico, representing a population reduction by approximately 8%
If you’re estimating population size, then all you have are guesses.
Results from the preferred statistical model, shown below, estimate that excess mortality due to Hurricane María using the displacement scenario is estimated at 1,271 excess deaths in September and October (95% CI: 1,154-1,383), 2,098 excess deaths from September to December (95% CI: 1,872-2,315), and, 2,975 (95% CI: 2,658-3,290) excess deaths for the total study period of September 2017 through February 2018
Which would actually that there were a whole lot of excess deaths quite a few months after the hurricane.
There’s over 800 excess deaths from December to February. How could the hurricane manage to kill 1,271 people in the first months, but then consistently go on killing some 800 people in the next two periods?
Leave a Reply