United Nations warns Gaza is “one step away from famine.”

The Humanitarian Catastrophe
in Gaza Can Only Get Worse
Isaac Chotiner / The New Yorker

(February 29, 2024) — This week, researchers from the Johns Hopkins University Center for Humanitarian Health and the London School of Hygiene and Tropical Medicine released a report that attempts to project how many people will die in Gaza in the next six months. The authors estimated what are called excess deaths, which includes deaths owing to Israel’s war campaign directly and also those caused indirectly, owing to factors such as disease and a lack of access to medical care.

They modelled for three possibilities: if the next six months of the war are similar to the first three months, if the war escalates, or if there’s a ceasefire. If the war stays its course without escalation until early August—with Israel bombing densely populated areas, and blockading food and medicine—the researchers project somewhere between 58,260 and 66,720 excess deaths in addition to the more than twenty-eight thousand deaths that the Gaza Ministry of Health reported in mid-February. (That number is currently more than twenty-nine thousand.)

If the war escalates, the authors project that that death toll could rise to between 74,290 and 85,750 excess deaths over the next six months. (Their escalation scenario makes projections based on the highest single month of casualties.)

Even if a ceasefire begins immediately, the researchers project between 6,550 and 11,580 more people will still die over the next six months, than if there had been no war. (For each scenario, the higher number includes deaths from possible epidemics.)

“For now, this looks like a serious attempt to quantify the important issue of deaths and especially nonviolent deaths in Gaza,” Michael Spagat, an economics professor at Royal Holloway, University of London, and an expert on civilian death tolls, told me. (The study has not yet been peer-reviewed because of time constraints.)

To talk about how the report was compiled, and the dire humanitarian situation in Gaza, I recently spoke by phone with Paul B. Spiegel, one of the report’s authors and the director of the Center of Humanitarian Health at the Johns Hopkins Bloomberg School of Public Health. During our conversation, which has been edited for length and clarity, we discussed the different factors that shaped the authors’ projections, why even a ceasefire scenario could see more than ten thousand additional excess deaths, and what makes Gaza unique among humanitarian tragedies.

How did you decide to focus on three scenarios, and can you talk about what those scenarios consist of?

We wanted to look at the potential extremes as well as the current situation, so because of that, and from that, we developed three scenarios to give the range. These are not predictions; they’re projections. And there is a difference. Predictions have much more confidence. Projections are what could happen. These scenarios are lasting over six months, beginning from February 7th. So the first-case scenario is a ceasefire, the middle scenario is what we call status quo, which is what is happening, and what continues to be happening, now. The worst-case scenario would be an escalation.

So when you say “best” case and “worst” case, do you mean in terms of casualties, or are you saying something beyond that?

That’s right. We are just talking about casualties. If you read the report very carefully, we are coming at it from an academic perspective. We are apolitical and we’re not even making any sort of causal inferences. What we’re really trying to do is document excess mortality according to these three different scenarios.

Can you talk about what projections are? Should we be thinking of this the way we think of polls, with a margin of error? Should we be thinking of this as sort of back-of-the envelope calculations—

They’re definitely not back-of-the-envelope; we wanted to do this in a much more rigorous way. It’s more like thinking about the various projections of climate change. Predictions are usually for shorter periods of time. These are longer-term projections, over a six-month period, and they’re scenario-based. So with climate change, people asked what would happen if you see a 1.5-degrees-Celsius rise, etc. It’s that sort of thinking that we want to get across.

Under the immediate-ceasefire scenario, you still see excess deaths ranging from 6,550 to 11,580. Can you explain why that would be the case? And, moreover, can you explain exactly what you mean by excess deaths?

Excess deaths are deaths that would not have occurred if there wasn’t this conflict. So these are deaths that would be projected to occur because of the current conflict. For example, there were no deaths due to trauma when there was not the conflict, before October 7th. The trauma deaths are all excess deaths, but other deaths are due to diabetes, etc. We’ve had to look at how the health system is functioning now, the access to services as well as the types of services. Then we made certain assumptions. For example, that there would be a limitation for people who have diabetes in terms of accessing insulin by a certain percentage, and therefore the deaths will increase. Similarly, we know that every year there are endemic infectious diseases. We know how many people died of COVID and how many people died of influenza in the past. We now increase those numbers due to the current situation of overcrowding, the lack of water and sanitation, and the lack of treatment.

In the ceasefire scenario, one of the most important points, and perhaps surprising to some, is that there still is going to be a lot of excess death over that six-month period. The reason for that is because the situation is so severe. The population itself is in a very poor state. There are a lot of traumatic injuries, and, while there may not be a lot of new trauma—although there may be some because of unexploded ordnance—there are going to be traumatic injuries of people that are infected, and who will die. On top of that, there are going to be continued infectious diseases that are going to occur, and potentially also some epidemics. And it’s not going to be immediate, necessarily. You have a population that is malnourished. You have a health system that is hardly functioning, and you’ve got roads and infrastructure that are destroyed. To be able to rebuild that quickly enough to even provide basic care is going to take some time.

Right now it doesn’t seem like a ceasefire is about to occur, but if one were to be negotiated, what would be the important things that would need to happen immediately to insure that the death toll remained as small as it could be?

There needs to be—and very, very quickly—a huge amount of water as well as sanitation facilities. So water, food, and fuel need to get in, in large amounts. And it’s not just going to be typical food. It has to be some nutritious foods, particularly for malnourished kids. So that’s No. 1.

No. 2 is there are many, many emergency medical teams that are waiting to be able to get in, and they are semi-autonomous or autonomous. They have surgeons, they have medical tents, and they need to be allowed in and to be able to be set up to support the existing system—not to replace it but to support it. There’s going to be a need for a lot of personnel that have some of the specialties for trauma because the trauma cases are so severe. It’s not just going to be generalists but there’s going to have to be a lot of surgeons and rehab people. So there will be a massive amount of logistics, and I don’t know if whoever’s going to be controlling the borders are going to let all of this happen, even if there is a ceasefire.

In the middle-course scenario, which you call the “status quo” scenario, are you looking at current deaths and projecting them out?

We are doing it slightly differently. This was probably our most challenging scenario because it’s not as if these past four months have been the same. There have been a lot of ups and downs, so we used the average of the first three months of the conflict to say that this is the status quo. For the escalation, we used the highest amount of deaths that there have been over a one-month period, so that was how we differentiated between status quo and escalation.

What’s being reported are mostly deaths due to trauma. When you hear of twenty-eight thousand deaths, that’s primarily traumatic deaths—but it may not include certain deaths that are under the rubble that haven’t been reported. And it may not be including a bunch of other deaths due to infectious disease or non-communicable disease.

Sorry, what do you mean by “under the rubble”? You mean literally under the rubble?

Yeah, so there is an underestimation of the current deaths because not all the deaths are being reported because there are certain families, or certain areas, that are still under the rubble and they haven’t been recovered. Some of them are being reported, but we know that not all of them have been.

In terms of deaths that are either under the rubble, as you say, or from disease, how are you all getting access to trustworthy figures in either of those two cases?

For deaths under the rubble, we used another independent source, which is the number of deaths among employees and staff at the U.N. Relief and Works Agency for Palestine Refugees in the Near East (UNRWA). They’ve had over a hundred and fifty deaths. They’re very exact because UNRWA is a U.N. agency, and they have thirteen thousand workers in Gaza and they have a system in place. So we use that as sort of the gold standard, and then we use the deaths that have been reported early on by the Gaza Ministry of Health. We looked at the difference between those two and from there we projected the number of deaths that may be missing, and it was probably up to about ten to fifteen per cent more.

So by looking at these U.N. employees, of whom there are around thirteen thousand in Gaza, and how many of them have been recorded to have died by UNRWA versus how many of them have died in the official count, you’re using that to project for the population at large?

Yes, exactly.

That would go for both people under the rubble, as you say, and for people that are dying from disease or something else?

Not necessarily, because that latter aspect of disease is a bit more complicated. So, for example, we have trauma, we have infectious diseases, noncommunicable diseases, and maternal and neonatal health. My colleague Francesco Checchi, who has been doing this work for years, did one of the hardest parts, which is to understand when epidemics may occur, and that gets more complicated. We had to look historically at when there were epidemics, and we looked historically at the past conflicts in Gaza, we looked at the vaccination rates, and then what Francesco did is make certain assumptions.

For example, when was the last cholera outbreak? It’s actually been a long time. Luckily, there hasn’t been a lot of cholera in Gaza. But because of the water and sanitation conditions, and because we know that there has been cholera in Syria and then Lebanon, if cholera is introduced, then we ran models to ask, What is the percentage of people who may die from that?

Another example: we looked at how many people had chronic kidney disease, how many people were on dialysis, and then we were able to work to get reports of how many dialysis machines were working in which areas. So then we would come up with estimates of people who may die because of the current health situation, and that would be the status quo. When we move toward an exacerbation, we have to make the assumption that there’ll be fewer dialysis machines and therefore more people die.

How do we know that the population of UNRWA workers is representative of the population over all?

The question was: Might UNRWA have had a higher level of death rates because they were specifically targeted? And the answer was no, because at the time when we did this, only very few died while they were at work. The rest of them are spread out throughout Gaza and/or displaced and they were killed in apartment buildings, or at homes, for example.

Despite the fact that we’ve been talking a lot about second-order effects of the war, is it still expected that for the foreseeable future the vast majority of deaths will come directly from the war itself?

In some other long wars, like in the Democratic Republic of Congo, we know that the indirect deaths are higher than the trauma deaths, but that’s because of the poor health-care system and because infectious diseases are so much higher in those settings, and because you don’t have the kind of massive aerial bombardments in a small concentrated area like Gaza. We expect, over time, that in the status-quo scenario, there will still be a tremendous amount of bombardment, but the infectious-disease rates will increase, particularly if there’s an epidemic.

An escalation of the war could mean different things. Could you talk more about how you projected that possibility?

We took the deaths over a three-month period, but we chose the month where there was the highest death rate in a given month, which was really more at the beginning—because, if you remember, Israel started by bombing in Gaza City. These are projections, so we have to make assumptions, but what we said was we are going to project this for the next six months. Also, in the ceasefire scenario, most of the deaths occur because of infectious diseases, but there are deaths, as I mentioned, due to trauma, and there are maternal deaths. In the status quo, and then even more so in the escalation, you understandably see a significant increase in deaths due to trauma, because more and more deaths will be occurring because of the bombing.

Several times you have mentioned the status-quo scenario, which you define in the study as the first three months of the conflict, and thus includes the first month, when deaths were higher. If you had defined the status quo as only the last couple of months, would the status-quo projection be less?

Yes, and we had to choose a starting date and this included the ups and downs. In the escalation scenario, we originally did this with the most extreme two weeks, but we felt it was far too high and would be too intense. So we made it a four-week scenario. We think these are reasonable. The point about projections is that assumptions have to be made.

So you’re saying for the status-quo scenario, you settled on the first three-month period, even though that would show a higher toll than the most recent two-month period, but for the escalation period, you looked at a whole month rather than looking at the very, very, very most extreme two weeks.

Yes, exactly. That’s exactly right.

I was struck that the gap between ceasefire and the current course is way bigger than the gap between the current course and escalation, which is to say that the current course the war is on, even without an escalation, is one with an incredibly high amount of projected deaths.

Yes, that’s right. But we were surprised to a degree that you’re still going to have so many deaths in the ceasefire scenario—because many people think that after a ceasefire then everything should go back to normal rather quickly, but that’s not the case. Then there is also the epidemic modelling. What hasn’t happened is a significant increase in the infectious diseases, but the longer and longer this goes on, the worse and worse the water and sanitation, which is already horrific, and the shelter get. Therefore, when you look at our infectious diseases that are non-epidemics, it’s going to be a lot of respiratory-tract infections, a lot of COVID and influenza. The one difference is that some of the infectious diseases go down in that second phase because they’re seasonal. Right now, we’re in a bad season for COVID and influenza.

What role do starvation and hunger play here? Is the lack of access to food something that you’re viewing as having the potential to directly lead to more deaths, or is it more a matter of the declining general health of Gazans, which is going to lead to secondary effects?

Yes, it’s not an immediate cause, but it’s an underlying cause. Our models include the effects of acute malnutrition on death rates. The most vulnerable are children between the ages of six and fifty-nine months. Looking at the numbers of trucks going in for food, and reports by UNRWA and others, we used all of this to come up with acute-malnutrition rates.

How is this conflict different or similar to other conflicts you have examined?

I’m talking less here about the report per se, and more about my past, but before coming to Hopkins, seven years ago, I was with the United Nations High Commissioner for Refugees for fourteen years. I’ve been throughout Africa and the Middle East and parts of Asia. And the closest thing that it would be like would be urban warfare in Mosul or in Fallujah. This is different for several reasons. One is that there’s not what I call an escape valve. In most situations, people are able to move and they’re able to escape either to a different part of the place they are. But here people are not able to keep safe despite being told, Go south, go south, go south. There is no safe zone. There is no secure space. There is a horrible situation right now in Sudan and there’s not a lot of safety, but people are able to leave Sudan or they’re going to certain parts of Sudan that are secure.

And No. 2 is that this is a very, very small piece of land, the Gaza Strip. The intensity of the bombardment and the extensiveness of the bombardment is massive on a proportional level. There have been bombardments we’ve seen in the past similar to this, but perhaps not in the same intensity. And the high density of the population, the urban population, makes it extremely difficult.

And then finally, the last thing that makes it different has been that Gaza’s been under siege for a very, very long time. People do not have access to basic water, food, and fuel, and that’s exacerbating the health conditions. Many of these health facilities and health systems have been attacked and therefore they’re not functioning, either. This makes it extremely precarious and a crisis situation of the sort that has not really been seen for a while.

Posted in accordance with Title 17, Section 107, US Code, for noncommercial, educational purposes.