On Dec. 8, 2017, the New York Times published a sobering report on Puerto Rico’s death toll from Hurricane Maria: mortality on the island spiked abruptly in the months following the storm, leading to over 1,000 more deaths than during the same period last year. This number dramatically exceeds the official death toll of 64 — which only counted people who were directly killed by immediate storm damage.
After major natural disasters, emergency responders often warn of the risk of a wave of secondary deaths unless immediate relief is delivered. Puerto Rico — where federal relief efforts have been widely criticized as inadequate — presents a grim test case.
Is it fair to blame these extra deaths on relief failures? Let’s explore these four takeaways:
1. The Times used an extremely reputable technique to arrive at the higher death toll.
The Times’s analysis renders the official death toll irrelevant. The report applies a methodology that humanitarian researchers often use to approximate the casualties from a major hazard. Since it is rarely possible to track down and individually attribute every death after a shock, humanitarians look at how many deaths occur beyond normal levels, and for how long those elevated mortality levels persist. This makes it possible to estimate a famine death toll, for example, without having to individually attribute each death due to cholera, malnutrition or unknown causes.
This technique gives a reliable approximation of an event’s impact — and has been used to estimate the death tolls from events like the war in the Congo or the 2011 Somalia famine. In Puerto Rico, this technique gives us a considerably more valid picture of the actual human impact than simply counting immediate storm-damage deaths.
2. Secondary mortality is not inevitable.
While humanitarians often warn of possible secondary mortality after a disaster, in practice these deaths are often averted. Super Typhoon Haiyan, a Category 5 storm that stands among the most powerful ever to make landfall, ravaged the Philippines in November 2013 and triggered a massive relief effort. Like Maria in Puerto Rico, Typhoon Haiyan hit a low-income area with poor infrastructure and all the logistical complications that come with island settings. The immediate storm itself killed 100 times more people than Hurricane Maria: 6,300 versus 64.
But there is no indication that Haiyan set off a wave of secondary mortality in the following months, despite widespread damage to shelter, water infrastructure and health services. National mortality statistics compiled by the Philippines Statistics Authority show that the three regions (Eastern, Central, and Western Visayas) in the storm’s path had 6,572 more deaths in 2013 than in 2012. This figure mirrors the formal death toll of 6,300 from the immediate storm impact. It does not suggest additional excess mortality in the final months of 2013.
But if the kind of secondary mortality we’re seeing in Puerto Rico is not inevitable, it is also not unheard of. Look no further than Hurricane Katrina. Like Maria, it devastated already impoverished parts of the United States and was followed by a federal relief effort widely critiqued as lackluster. A mortality survey found that the death rate in New Orleans in the months following the disaster was approximately 50 percent higher than it had been before the storm, and that death rates remained elevated for months. While these deaths — like those in Puerto Rico — may not be directly tied to the immediate storm damage, their timing indicates a clear correlation to the longer-term fallout from the disaster.
3. Relief effectiveness matters — a lot.
There are good reasons that the Philippines saw no surge of secondary mortality after Haiyan. Its government offered effective leadership and simultaneously sought international help. The international humanitarian system responded with a relief mobilization that was massive, swift, well-coordinated and robustly financed. Foreign governments, the United Nations, the U.S. military and hundreds of NGOs deployed in force following the storm, resolving logistical obstacles and delivering immediate lifesaving aid. Numerous post-storm evaluations praised the effectiveness of the response — a classic example of what the international relief system does well.
The international relief architecture was built for settings where international actors must often do much of the heavy lifting. However, the Federal Emergency Management Agency was not built to tackle this kind of challenge: a major disaster in a setting with widespread poverty, weak local response capacity and extreme logistical obstacles. FEMA is designed, under the government’s National Response Framework, to support relatively capable state-level disaster managers. But disaster management capacity in Puerto Rico is weaker than in Texas or Florida, meaning that FEMA had take a much stronger lead role than it is accustomed to.
And so FEMA struggled to adapt, falling short in a number of ways identified in the Times’s report. Puerto Rico’s excess deaths have come mainly from sepsis and respiratory problems, classic post-disaster health problems when there is not enough clean water, safe shelter and adequate health care. Available clean water was so inadequate that as excess deaths were spiking, Puerto Ricans were reportedly turning to sewage-contaminated rivers, condemned wells and Superfund cleanup sites for water. Health-care coverage was so weak that the Navy took the rare step of deploying one of its hospital ships to Puerto Rico. But the ship was poorly suited to Puerto Ricans’ actual health needs — reliant on a bureaucratic referral process that proved difficult to navigate — and ultimately saw very few patients. And it took so long to deliver emergency roofing kits that families stayed in unlivable homes for weeks and months, exposed to the rainy season and creeping mold. Even three months into the response, shelter remains woefully inadequate: FEMA reports that it has sheltered only 28 percent of those who need it, with just over 20,000 emergency shelter kits installed — leaving more than 50,000 households still in need.
4. The U.S. government had options to prevent many of these deaths — but failed to use them.
The picture that emerges is fairly damning: a federal response effort that has failed to address basic needs that the international response system would — and does — address as a matter of course. This is all the more staggering given that the international disaster response system was largely built and funded by … the U.S. government.
With FEMA understandably overmatched by this fall’s series of megastorms and faced with a uniquely challenging context in Puerto Rico, the Trump administration could have gotten creative: deploying its own international responders at scale, or seeking help from international partners. Inexplicably, it did neither. And the response suffered as a result.
Take shelter as an example. The federal government has managed to cover the emergency shelter needs of only 20,000 households in the first three months, while international agencies reached over 474,000 households with emergency shelter support in the same period after Haiyan. On the health front, 151 foreign medical teams deployed to the Philippines following that storm, enabling direct delivery of care in front line communities, rather than floating in a harbor.
It is impossible to say how many of the thousand extra victims could been saved if the administration had swallowed its pride and sought international assistance from the United Nations and other partners. But it is clear that many died — and may still be dying — from causes that could and should have been addressed months earlier.
Jeremy Konyndyk is a senior policy fellow at the Center for Global Development and teaches humanitarian field operations at Georgetown University. He served from 2013 to 2017 as the Obama administration’s director for foreign disaster assistance at USAID, overseeing the U.S. government’s Ebola response in West Africa and other emergencies.