The Media’s Shameless Politicization of the Death Toll in Puerto Rico

By Kent R. Kroeger (Source:, November 1, 2017)

{Feel free to send any comments about this essay to: or}

How many people died in Puerto Rico as a result of Hurricane Maria? It is an important question that needs a serious, non-partisan answer.

An answer we will not get from CNN or any of the major news outlets covering the aftermath of Hurricane Maria.

Why? Because they are not really in the news business. They are in the Trump-bashing business. News stories are not pursued on their merit, they are selected based upon how well they serve the current popular narrative — and that narrative since November 2016 is: “Trump-is-a-liar-and-an-incompetent-Russia-colluding-stooge.”

Why focus on the narrative over objective facts? Because strong narratives build audiences, much like presidential candidates with the strongest narrative attract the most voters. Humans prefer narratives over hard, cold facts. The research supporting this conclusion is long, varied, and convincing.

The major news outlets’ coverage of the Hurricane Maria aftermath in Puerto Rico is an exemplar of this ‘feed-the-narrative’ journalism and its hurting their credibility and the people of Puerto Rico.

Hurricane Maria devastated Puerto Rico, but by all objective accounts, the immediate death toll was relatively small

The official death toll in Puerto Rico from Hurricane Maria stands today at 54. These are deaths directly related to the storm — mostly caused by drowning, blunt force objects, and stress-caused physical traumas, such as strokes and heart attacks. This official number includes deaths in the more remote sections of Puerto Rico, according to the Puerto Rican governor’s office.

It is not a perfect number and probably an under-count, given the realities of Puerto Rico’s terrain and socio-economic conditions.

But the news media have decided to manipulate the suffering of Puerto Ricans for a political purpose. They have discovered local and state Puerto Rican officials willing to suggest over 900 people in Puerto Rico died due to Hurricane Maria.

Here is just a sampling of recent headlines suggesting this death toll:

ABC News: 900-plus cremations since Maria, but hurricane death toll still 51

Newsweek: Puerto Rico says more than 900 people were cremated after Hurricane Maria

The Hill: Puerto Rico says over 900 people died of ‘natural causes’ after hurricane: report

Is The Hill‘s convenient use of quotations around ‘natural causes‘ meant to suggest someone is doctoring the death toll? Perhaps Donald Trump is doing it himself between 3 a.m. tweets? As if Donald Trump or anyone in his administration would be connected or knowledgeable enough to manipulate a death toll count generated by Puerto Rico’s state bureaucracy. [See, even I am willing to hitch a brief ride on the ‘Trump-is-incompetent’ narrative]

How did the news media get to this 900+ number? Its a bit murky and unsystematic, but, generally, it is coming from body and cremation counts from local morgues and funeral homes across Puerto Rico.

On a superficial level, that approach may make sense to a journalist or the general public; but, in practice, it yields an inaccurate and biased number.

CNN reporter John Sutter’s approach to covering the death toll has been particularly creepy and dishonest. It is revealed in the first personal story he offers in his Oct. 27th article on titled: Puerto Rico’s uncounted hurricane deaths — CNN visits every funeral home in one town to test the government’s count.

He writes: “Isabel Rivera González was 80. She loved to dance, and was known in this hilly enclave of Puerto Rico for her Saturday-night merengue moves…On October 15, three weeks after the storm, Rivera died awaiting a procedure at a hospital that had lost power in the hurricane and whose backup generator failed, according to several of her family members.”

Rivera was 80 and in poor health (prior to the hurricane). This is a sad death and possibly an indirect (not direct) result of Hurricane Maria. The direct versus indirect distinction may seem hardhearted, but it is an important distinction to those who study natural disasters and help prepare local, state and national governments for the next natural disaster.

Indeed, Sutter could have saved himself a lot of effort trolling Puerto Rico’s funeral homes had he first talked to an epidemiologist or public policy expert specializing in natural disasters. His methodology, which he painfully details in his Oct. 27th article, is not appropriate for measuring direct or indirect fatalities related to Hurricane Maria.

Here is why: people die in Puerto Rico every day. They were dying long before Hurricane Maria and they will die in Puerto Rico long after the visible effects of Maria have vanished. In fact, prior to Maria, around 80 Puerto Ricans were dying every day, according to the U.S. Census Bureau.

This is why epidemiologists and policy experts establish mortality rates prior to an event (e.g., hurricane) and re-measure that rate during a post-event period of time when assessing the impact of natural disasters.

Establishing a baseline mortality rate is a normative benchmark that can be compared to the post-Maria mortality rate. This methodology in its simplest form is called a Pre-Post measurement design.

The following research study on the measurement of indirect deaths related to the 2011 earthquake and tsunami in Japan is an excellent example of a high-quality scientific study measuring natural disaster-related mortality rates. A more basic analytic approach was employed here to measure the impact of Hurricane Katrina on the mortality rate in the Greater New Orleans area in Louisiana.

Yet, even a quick, back-of-the-envelope attempt at understanding the deaths related to Maria reveals the inherent flaw in the news media’s unscientific assertion that 900 people died in its aftermath.

Hurricane Maria hit Puerto Rico on September 20th. The 900+ death toll estimate being promoted by the media emerged around October 26th. Lets assume those estimates were derived in the week prior to the news stories, that puts us around October 20th. Rounding, that is 30 days between landfall and the 900+ death count.

Based on U.S. Census Bureau data, in just a normal 30-day period Puerto Rico would have seen around 2,400 deaths. That is the number of deaths without a hurricane at the start of the period. This makes the 900+ fatalities number a little suspect, I would say. At a minimum, it demands more information before we can take it seriously.

In fact, I wonder if Puerto Ricans didn’t become even more attentive to their sick and elderly in the hurricane’s direct aftermath, thereby decreasing (if only temporarily) the mortality rate in Puerto Rico.

No, I’m not going to go that far. That would make me no better than CNN. And it may be true that the post-Maria mortality rate is higher in Puerto Rico. Counting cremations and dead bodies in morgues however is not a reliable way of getting to that true number.

This is why the news media’s ‘feed-the-narrative’ motivation is so important to understanding what it reports as news. The primary concern of CNN or The New York Times or MSNBC or The Washington Post is not understanding the impact of Hurricane Maria on Puerto Rico. Their primary motivation is finding ways to make Donald Trump look bad.

‘So what if the news media is reporting a sketchy death count, does it really matter?’

Unbiased mortality, morbidity, and financial loss estimates due to natural disasters are critical to understanding trends and long-term disaster planning. When these numbers are manipulated for political and economic reasons, public policy suffers.

“This poses a problem for any attempt to characterize trends in disaster impact and – maybe more importantly – to use those trends to identify optimal policy choices,” according to Dr. Llan Noy, from the Victoria Business School (Canada). “Trends in disaster losses are crucial because the distribution of losses across regions – and across countries at various levels of wealth and development – informs the discussion of climate change mitigation policies.”

Inflating the Puerto Rican death count may cultivate the news media’s anti-Trump narrative, but it harms those trying to prepare the world for the possible impact of climate change and natural disasters in general.

There is not going to be a quick answer to the “How many people died in Puerto Rico due to Hurricane Maria?” question. In the meantime, exploiting this period of scientific uncertainty to bash Donald Trump should be beneath the ideals of the news media. Unfortunately, it is not.


About the author:  Kent Kroeger is a writer and statistical consultant with over 30 -years experience measuring and analyzing public opinion for public and private sector clients. He also spent ten years working for the U.S. Department of Defense’s Office of the Under Secretary of Defense for Personnel and Readiness and the Defense Intelligence Agency. He holds a B.S. degree in Journalism/Political Science from The University of Iowa, and an M.A. in Quantitative Methods from Columbia University (New York, NY).  He lives in Ewing, New Jersey with his wife and son.

A post-essay addendum:

I used these two imaginary scenarios to explain to my son how someone might measure the impact of Hurricane Maria:

In the first scenario, a hurricane hits an island and, sadly, a boat with 200 people from the island capsizes and everyone dies. However, on the island, nobody dies and power and normalcy return quickly. Using the Pre-Post design, the researchers would see a spike of 200 people (above the normal mortality rate) but the rate would promptly return to its historical norm. Deaths directly attributable to the hurricane would stand at 200 and indirect deaths would be around zero (assuming the 200 people that died on the boat weren’t all doctors and first-responders from the island).

In the second scenario, a hurricane hits the same island but nobody dies on the day of the hurricane. Instead, the island’s power and transportation infrastructure is destroyed and is not repaired for weeks, even months. In this case, the researchers would see no spike on the day of the hurricane but may see a steady — maybe even abrupt — increase in the island’s post-hurricane mortality rate. This mortality rate change is the estimate of the hurricane’s impact.

Of course, reality is more complicated than offered in these simple scenarios. For example, life-expectancies can change due to natural disasters and would not be easily discernible a simple Pre-Post measurement design. For this reason, researchers employ much more sophisticated (but analogous) statistical techniques to assess the impact of natural disasters.