Talk:Hurricane Maria death toll controversy

Usefulness of a Timeline?
One reason that the controversy is difficult to untangle is that there is no timeline that places the events in a coherent order. Furthermore, the various studies and estimates were all based on different datasets, released and/or collected at different times. I am not a heavy wikipedia editor, so I prefer to propose this here rather than implementing a timeline and having it reverted/rejected. Thoughts from others who follow this page? OtianNgocnion (talk) 15:35, 5 October 2018 (UTC)
 * September 6th, 2017 - Irma passes the island and affects the electrical grid:
 * September 20th, 2017 - Maria makes landfall:
 * November 21st, 2017 - Santos-Howard paper in pre-print reports excess deaths by end of November at 1,085 using partially released vital statistics
 * December 9th, 2017 - Puerto Rican Government sets death toll at 64 - NYT reports deaths may be as high at 1,052
 * December 18th, 2017 - Puerto Rican Government orders a recount
 * January / February 2018 - Harvard study data collection is completed
 * February 22nd, 2018 - Puerto Rican Government announces that GW will conduct recount, no official government data is released
 * May 29th, 2018 - Harvard paper released reporting a midpoint of 4,645 deaths till the end of December 2018
 * June 4th, 2018 - court orders the release of official government death records
 * August 27th, 2018 - GW releases report with a midpoint of 2,975 excess deaths through February 2018

Some comments about the controversy
I'm having a hard time following this article. The GWU study cites somewhere around 3000 deaths during a nearly 6 month period. Trump's comment was "3000 people did not die in the two hurricanes that hit Puerto Rico." This would actually appear to be true, since the hurricanes did not last nearly 6 months. I think we need to separate into the deaths that occurred during the actual storms, and the deaths that occurred in the aftermath and follow up periods. As it stands right now it is very difficult to understand the facts. Mr Ernie (talk) 19:20, 13 September 2018 (UTC)
 * not sure what you're looking for here. There are multiple mentions of the deaths covering the six months during and after the hurricane. It's clearly stated thrice in the lede and implied a fourth time that large numbers of people died in the months after the hurricane. As for the differentiation, we have almost no clue of the split. We do know of 64 direct deaths from the storm based on early autopsies/death certificates; however, hundreds of bodies were cremated without examination. The report by GWU is a statistical assessment not a direct tabulation of deaths, hence the range of 2,658–3,290 mentioned at the end of the lede. ~ Cyclonebiskit (chat) 19:27, 13 September 2018 (UTC)
 * I guess my comment was more in line with Trump's response and the response to Trump's response. I would understand Trump's comment to refer to the actual hurricane, and how many were killed in the days the storm was active, not in the weeks and months afterwards. But it seems the RS have taken his comment to refer to the entirety of the events spanning the 6 month period. It makes me wonder how they could attribute an elderly person's death to the hurricane, and not on some other factors that normally end the lives of elderly people, although I would assume the study tried to account for this. Mr Ernie (talk) 19:36, 13 September 2018 (UTC)
 * I didn't add in the section regarding Trump's recent tweets, but it's a touchy area to analyze. In one sense he is correct in saying that 3,000 people didn't die in the storm; however, it was the inadequate response from his administration, FEMA, and Puerto Rico's government that drastically worsened things. The GWU study covers "excess deaths", meaning the number of deaths exceeding what would normally be expected given mortality trends and accounting for the exodus of 300,000 people in the hurricane's wake. Elderly deaths attributable would include people in need of dialysis unable to receive such treatment, lack of food/water compounding pre-existing illnesses leading to their death, and otherwise treatable injuries that led to death because of no medical care. Him mentioning the very early numbers of 6–18 deaths did not help at all. ~ Cyclonebiskit (chat) 19:46, 13 September 2018 (UTC)
 * Thanks for the response. I think the section about the study should be expanded a bit to show the methodology for how they arrived at the number. As I understand it there were many assumptions and statistical methods and computer models, but the PR government has accepted the result of the study as the official death toll. Mr Ernie (talk) 15:45, 14 September 2018 (UTC)
 * It May need further clarification of how other death counts may differ.... where others are direct deaths by wind or drowning. Cheers Markbassett (talk) 14:23, 17 September 2018 (UTC)

Most Severe Disaster in Modern History
The first sentence of this article states that this is "the island's most severe natural disaster in modern history." While I agree with this, and I am confident (though I have not confirmed so) that this is backed up in the rest of the article, it is a somewhat controversial statement and deserves an inline citation. I would add  to it, but I am not about to dispute that sentence, since that is what I understand to be the case and I do not wish to accidentally mislead readers by implying that the statement is controversial. I do think though that an inline citation on that sentence would give readers much more confidence as to what the reality of the situation is. Thanks for listening to my feedback. 2601:140:6:8047:89C4:F77D:3D92:515E (talk) 04:02, 14 September 2018 (UTC)
 * Seems a reasonable need for the source of somewhat flamboyant language, an editorial side remark not directly part of the controversy. I found no good match by google so will try a citation needed and see if it works.   Cheers Markbassett (talk) 14:16, 17 September 2018 (UTC)
 * I've provided a citation for the line in question; the statement comes from Governor Rosselló. ~ Cyclonebiskit (chat) 16:57, 17 September 2018 (UTC)
 * User:Cyclonebiskit thanks, that makes sense. Markbassett (talk) 02:22, 18 September 2018 (UTC)

Table asterisk
The table of deaths cites a 2010 study... I am going to asterisk the Maria line as that insert is not part of the cite. Please discuss whether to do a further footnote in tablet, to remove the line from table to text, or what. And if anyone can clarify why it says 2982 instead of 2975 .... Cheers Markbassett (talk) 02:29, 18 September 2018 (UTC)
 * The extra deaths include those that occurred in the US Virgin Islands and along the US East Coast. ~ Cyclonebiskit (chat) 03:01, 18 September 2018 (UTC)
 * Thanks — now see that in the template editing page ... 2,975 PR, 3 USVI, 4 CONUS Markbassett (talk) 03:26, 18 September 2018 (UTC)

Bias?
Although This was a real controversy with real events, I still feal that there is a bias in this article. Such as parts like "Trump made these accusations with out evidence." A tag should at the least be added. — Preceding unsigned comment added by 63.142.60.10 (talk) 20:51, 11 December 2018 (UTC)

An error in the displacement data....
Under the George Washington University study section, there is a statement "An error in the displacement data used by the GWU team led to an overestimate of Hurricane Maria death toll by about 300[54]". The citation associated with this edit does not show that a an error led to an overestimation of Hurricane Maria death toll. This statement should be removed or the reference needs to be improved to support the statement.

Additionally, the reference is to "Supplement to: Santos-Burgoa C, Sandberg J, Suárez E, et al. Differential and persistent risk of excess mortality from Hurricane Maria in Puerto Rico: a time-series analysis". Lancet. October 10, 2019. [emphasis added]. I am writing this in August of 2019. — Preceding unsigned comment added by 107.217.159.40 (talk) 11:31, 5 September 2019 (UTC)

A controversy is not a "conspiracy theory"
This article is in the category "conspiracy theory promoted by Donald Trump". The hurricane Maria death toll tweets are a Trump controversy but not even close to being a conspiracy theory by usual definitions. 73.149.246.232 (talk) 12:10, 10 December 2019 (UTC)
 * Google it and you'll find it. He did. https://www.vox.com/2018/9/13/17854408/trump-hurricane-maria-death-toll-tweets . He thought the death count was just to make him look bad... --The Eloquent Peasant (talk) 12:37, 10 December 2019 (UTC)
 * Trump said, correctly, that the "death" toll estimates in the thousands are not counting hurricane "deaths" in the usual and historical sense of the term, and that death tolls in the ordinary sense are 2 orders of magnitude lower. Indeed, the table of hurricane death tolls in the article has a big fat asterisk only next to one number, the one for hurricane Maria.

Trump also said, correctly, that people dying of old age (i.e., chronic conditions such as emphysema and diabetes listed in this artice) He talked, possibly correctly, of REPORTS (that is, media accounts) of the death estimates, not the estimations themselves being used to make him look bad. He did not say the study itself was done to make him look but, but he did tweet that "GWU research" was "hired" (from context I guess he meant by the Puerto Rico government) to do the estimate. I don't have any evidence for or against that, but the study authors mostly have Spanish surnames and the articles you cite state that GWU did the study in collaboration with the University of Puerto Rico. This was at a time the PR government was under fire for underplaying the death toll, and wanted as much money as possible for hurricane recovery efforts. Ultimately they simply took the GWU "excess deaths" concept as their (re)definition of the number of deaths; instead of getting a more accurate death count by including missing later found, surveying funeral homes, etc they just took a new concept and decided to call it the "official death toll".

As Trump said, correctly, this newfangled death-counting method was never the official count for prior hurricances. Obviously we could also get 1-2 orders of magnitude higher for Katrina and most other storm using the same method.

Trump's statements are not super-careful or full of citations, but this is not the stuff of conspiracy theory, and he is right about the death tolls. — Preceding unsigned comment added by 73.149.246.232 (talk) 17:44, 11 December 2019 (UTC)
 * According to media reports and newspapers he was theorizing; his arguments that it was to make him look bad, arguing against "reality" is a conspiracy theory. (BTW, I actually have nothing against Trump, voted for him and like how he's managed the US, mostly - but that's irrelevant to this). --The Eloquent Peasant (talk) 01:46, 12 December 2019 (UTC)
 * Trump only mentioned "reports" as being done to make him look bad. That is a controversial claim, but not a conspiracy theory, as a statement about media reports.  The severely POV article you just cited is an example; it is not a report of Trump's errors but a screed against his person.  It's possible that Trump meant the Harvard or GWU "reports" were done to make him look bad, which would be a claim of conspiracy against him, but this seems less likely and his comments on GWU are either correct ("old age" deaths counted) or nowhere disputed (ie, that PR "hired" them).
 * For everything Trump says there is a media report calling it a conspiracy theory, the question here (especially for the designation of article as "conspiracy theory promoted by Trump") is whether the information in those reports actually supports that rather than it being controversial.

73.149.246.232 (talk) 03:30, 12 December 2019 (UTC)
 * Your mention of old age is odd and people "correctly' dieing of old age is very odd. People who are old sometimes depend on machines, machines which were not functional due to no electrical power for more than 8 months, in some places. So yes even old people died due to the hurricane, in fact it was mostly old people and sick people who died as a result of the hurricane. People didn't have what they needed to live; machines, dialysis, emergency rooms, hospitals didn't have electricity, etc. Are you saying they were going to die anyway 'cause they were old so they should not have been counted? Everyone will die sooner or later but many old people's death were expedited by Maria and that didn't happen to make Trump look bad and that is what Trump was saying - that the death count was to make him look bad. Also, you should visit the island so you can understand what you're talking about better. The Eloquent Peasant (talk) 03:44, 12 December 2019 (UTC)
 * For many reasons, ranging from having a definite standard to having a number that doesn't keep changing over time to the need to arbitrate insurance claims, official "death tolls" are essentially corpse counts done immediately after the disaster with some minor adjustments (presumed dead, severely injured who die shortly after, etc). Exceptions are events like genocides where data and bodies are often unavailable, and some numbers have to be estimated indirectly, but that was not the case in Puerto Rico.  As Trump implied, they made a political decision to use a different standard, which isn't even a standard (notice the huge difference between Harvard and GWU estimates), but that doesn't mean Trump or anyone else have to accept the redefined number as a "death toll" in the ordinary sense.


 * Trump is right that excess mortality studies include people who die of old age. This WP article lists estimated extra deaths percentages from emphysema, diabetes, and other pre-existing chronic conditions.  If someone smokes (overconsumes sugar) all his life, and dies at age 70.5 instead of 71 from emphysema (diabetes) because two hurricanes forced half the nurses off the island and put stress on the medical care system, then Maria is indirectly a contributing factor to his death but not what we would normally call a "cause of death".  The proximate cause of death is the illness, and the ultimate cause is old age, or bad habits, or natural causes, or Puerto Rico having bad government leading to more fragile infrastructure and delayed recover, or a million other things having nothing to do with Maria, FEMA, Trump, Democrats or what have you.  Assigning causation is largely an exercise in storytelling which is why the death tolls apply the principle of Habeas Corpus: find the body.  — Preceding unsigned comment added by 73.149.246.232 (talk) 05:07, 12 December 2019 (UTC)


 * Maybe this will make it easier to understand: (1) excess mortality is more for measuring the effect of things like pollution, unemployment or divorce, where (almost) nobody dies of it in a definite direct way but in a diffuse statistical sense it could change the number of deaths by a lot; because (2) under the excess mortality standard you can never point to a specific individual and determine whether his death is due to that cause. You only know that there are more deaths than some expected number in an econometric model, and (to the extent you trust the model, which is usually debatable) declare that the difference will henceforth be called the excess mortality or "official PR  death toll" or whatever.  In contrast, for death tolls from natural disasters, one usually requires a clear standard for deciding, for every death in some time and place, whether or not it was due to the disaster, and then the deathtoll is the total number of individuals for whom the answer was Yes.

73.149.246.232 (talk) 05:44, 12 December 2019 (UTC)


 * It is a conspiracy theory created by Trump to make people disbelieve that many people died in P.R. as a result of his inefficient handling of the event. "A Harvard-led team conducted an extensive survey of 3299 households throughout Puerto Rico and produced an all-cause mortality estimate of 14.3 deaths per 1000 between September 20, 2017 and December 31, 2017. This value was 62% higher than the well-documented rate from the corresponding period in 2016, suggesting that the true number of excess deaths from hurricane-related debris, drowning, illness, injury, or lack of access to medical facilities was 4645 (>70 times the official value of 64)." People who support Trump's conspiracy theories should be ashamed of themselves. The researchers from Harvard weren't hired by P.R. Also, why did you mention researchers having Spanish surnames. Did you want to say that because researchers had Spanish surnames that means something? Because your mention of Spanish surnames proves that Donald's conspiracy theory worked.The Eloquent Peasant (talk) 14:10, 12 December 2019 (UTC)


 * "It is a conspiracy theory created by Trump" The question is whether it actually meets the definition of a conspiracy theory or is just another political controversy. That you prefer to repetitively label it a conspiracy theory does not make it one.


 * Trump in his 4-5 controversial tweets made about a dozen factual assertions and arguments that were apparently correct or reasonable, plus one (much more disputable) claim that "reports" of 3000 deaths were done with Democrats to harm him politically. The last is the sole basis of the term "conspiracy theory". Under the most-probable interpretation that Trump meant media "reports" of the massively revised death tool, it is not hard to imagine that he could be right, which is not the mark of a conspiracy theory.  Conspiracy theories generally involve very improbable claims requiring complex chains of causation, stated without evidence and resistant to correction.  Trump's claims are nothing like that: it is extremely easy to imagine the media seizing upon Democratic Party talking points, which in turn seized on the huge revised death toll, to embarrass Trump.
 * to make people disbelieve that many people died in P.R. as a result of his inefficient handling of the event -- Trump's reasons for making his statements are unknown and are completely irrelevant to determining whether the statements were "conspiracy theory".


 * Trump's (correct) comments about GWU study also apply to the Harvard study. The excess mortality numbers simply are not what anyone, including the Puerto Rican government, had ever previously defined as the "death toll" of natural disasters.
 * I looked up the GWU study, which says it was "commissioned" by the Puerto Rico government. So Trump was apparent right, and not inventing a conspiracy, when he said GWU research was "hired" to do the study. The remark you (dishonestly) jumped on about Spanish surnames of the researchers was, as should be obvious from reading it, an observation that the study was carried out largely by people in PR where there was pressure to raise the official death toll.  To what degree the needs of people in PR influenced the choice of methodology and ultimately the numbers estimated is a level of detail I am not interested in researching, since I don't think Trump was even talking about the studies when he used the word "reports". — Preceding unsigned comment added by 73.149.246.232 (talk • contribs)

Remove infobox on nonexistent Senate act?
There is a large infobox on a proposed US Senate bill to study methods for counting disaster deaths, but the bill died in committee and shows no sign of being revived. Remove? 73.149.246.232 (talk) 16:52, 13 December 2019 (UTC)

New study
There is a new study in the American Journal of Public Health (AJPH) published in June 2019 - "Causes of Excess Deaths in Puerto Rico After Hurricane Maria: A Time-Series Estimation"

https://ajph.aphapublications.org/doi/10.2105/AJPH.2019.305015