In the middle of an ongoing conflict, it's difficult — if not impossible — to understand the true human cost of war.
Casualties go uncounted and injuries untreated as fighting rages on, leaving an imperfect tally of lives lost or damaged. Indirect effects of conflict, from missed cancer treatments due to hospital bombings or disease outbreaks caused by poor sanitation and overcrowding in refugee camps, can take months or years to become apparent.
The chaos of war usually means that researchers wait until a conflict is over to pore over the data and reconstruct as complete a picture as they can of how many lives were lost and what took them.
But soon after Israel began its military response to the October 7 Hamas-led attack that killed more than 1,200 Israelis and kidnapped about 250, a team of researchers from the Johns Hopkins University Center for Humanitarian Health and the London School of Hygiene and Tropical Medicine tried something that's never been done before as a conflict rages. Instead of trying to calculate the toll of war in the moment, they looked forward.
By combining on-the-ground data from Gaza with knowledge from past wars, the researchers projected how many additional lives would be lost over the next six months under three conflict scenarios: immediate and permanent ceasefire, status quo and escalation.
Their results, which the researchers caveat with a hefty dose of uncertainty, are sobering.
Here are their projections
Even if ongoing ceasefire talks are successful, the humanitarian crisis could claim an additional 6,550 to 11,580 lives by August 6, on top of the 29,000 deaths documented by the Gaza Ministry of Health when the study was published last week. Now, the death toll exceeds 30,000.
If a ceasefire isn't reached and fighting continues apace over the same timeframe, the researchers project between 58,260 and 66,720 excess deaths — fatalities caused directly by Israel's military or indirectly by factors like disease and restricted access to medical care or sanitation.
If the war escalates, excess deaths could climb to between 74,290 and 85,750. (The higher number in each range includes deaths from possible outbreaks of infectious disease).
The researchers caution that they're not trying to predict the future. "These are projections, not predictions," says Paul Spiegel, the director of the Center of Humanitarian Health at Johns Hopkins University and study co-author. Rather, they're projecting what would happen given a slew of assumptions about the human costs of war, he says, similar to how climate scientists project future warming under different carbon emissions scenarios.
"There's no perfect, pre-established methodology for this kind of projecting," says Yara Asi, a public health expert who studies the health impacts of war at the University of Central Florida and wasn't involved in the analysis. But she says it's an innovative and valuable effort.
"It shows that even if the bombing stops tomorrow, people will continue to die, not simply from the destruction of the health-care system but [loss of] access to food, water, vaccinations and shelter," she says. "Even if the numbers aren't perfect, putting this all together forces us to confront the true toll of what this means for the population there. We cannot say that we did not anticipate this."
Other experts agree. "It's a rigorous way of talking about the human cost of human decisions," says Patrick Ball, director of research for the Human Rights Data Analysis Group, a nonprofit organization. While he stresses that the projections are speculative, that kind of speculation can be "immensely useful" in clarifying the potential costs of military action, which could both hold actors to account and help guide humanitarian action, he said.
"I've spent my career looking backward and trying to calculate the cost of war that's wrapped up," says Ball. "I'm never projecting tomorrow's deaths, but I can imagine a whole new field coming from this [analysis]. We're always going to be doing this from here forward."
How they came up with their numbers
The researchers decided to scope out three scenarios for how the Israel-Gaza war might unfold over the next six months, from best- to worst-case. Given those scenarios, they projected how many people would die directly from trauma or indirectly from infectious disease, maternal and neonatal causes and non-communicable diseases, like cancer or diabetes.
"We wanted to define scenarios that were realistic and then based on those scenarios project what might happen, giving us some [upper and lower] bounds to work with," says Tak Igusa, a civil engineer at Johns Hopkins University.
Constructing those two extreme scenarios — ceasefire and escalation — as well as a status quo middle ground, required making a lot of assumptions. To inform these assumptions, the researchers gathered up all the data they could from the conflict so far, filling in the gaps with information from past conflicts, as well as consultation with trauma doctors in Gaza.
That data is messy and imperfect but better than data from other conflicts, says Spiegel. "It's rare in these situations to have such strong baseline data," he says, ranging from immunization levels that can inform estimates of how deadly a disease outbreak might be, to how many people died from landmines in past Gaza conflicts. Still, the researchers include a range of possible values for each number they project to reflect uncertainty in their estimates.
In the ceasefire scenario, the researchers assume fighting stops immediately. While no more people would die from traumatic injuries due to active warfare in that scenario, injuries sustained before the ceasefire would continue to kill some. And civilians would continue to inadvertently set off unexploded ordnance as they try to rebuild Gaza. Estimating those deaths alone is quite complicated.
The researchers analyzed data from Israel's 7-week-long 2014 war on Gaza to project deaths from unexploded munitions and used existing injury data from Gaza to estimate the number and severity of wounds across the population at the time of ceasefire. Whether a wounded person dies depends in part on their access to health care.
"We have to think about how the health-care system is changing over time," says Spiegel. The situation right now is dire, according to many of the trauma doctors the researchers consulted for the study. "If hospitals are functioning, a person with a head or chest wound might survive. But in the current situation, it's likely that he or she will not."
Given those assumptions, the researchers project that an additional 3,250 people will die from traumatic injuries after fighting stops. The rest of the 6,550 to 11,580 deaths post-ceasefire come from nonviolent causes.
Israel's military operations have destroyed many hospitals, interrupting crucial medical care for cancer patients, those with diabetes and other noncommunicable diseases. The researchers combined pre-war estimates of the burdens of those diseases with their estimates of health system capacity to project that 1,680 people would die from lack of medical care. Bombing also cuts off pregnant people from necessary care, which the researchers project would cause 100 to 330 maternal and neonatal deaths, depending on the scale of conflict.
Infectious disease outbreaks, like cholera or COVID, are common in conflict zones, as overcrowding in shelters and poor sanitation can stoke the spread of pathogens. But "it's very difficult to estimate when such outbreaks might occur," says Spiegel, which is why the team included projections with and without epidemics.
To estimate the possible toll of an outbreak, the team combined existing data on baseline health status, malnutrition, sanitation and vaccination rates with models of infectious disease spread. Such outbreaks would be especially hard on children, the researchers say, who are more vulnerable to infections, especially when they're malnourished. More than half a million people are "facing catastrophic levels of deprivation and starvation," according to the United Nations.
Altogether, several conflict data experts who weren't involved in the research applauded the authors. "This is a very serious effort, they're trying their best to get it as right as they possibly can," says Michael Spagat, an economist at Royal Holloway, University of London and chair of Every Casualty Counts, a nongovernment group that quantifies the human cost of war. "But it's extremely complicated, and rests on assumption after assumption after assumption," he says. "Though it can't be otherwise, given what they're trying to accomplish."
Caveats aside, Spagat says the analysis "focuses attention on the fact that deaths will continue even under the most optimistic scenarios."
The projection does not take into account Israel's concern that a ceasefire would allow Hamas to regroup and launch more bloody attacks against Israel — resulting in additional Israeli deaths beyond the October 7 toll and the potential killing of those still held hostage.
Projecting what might happen if fighting continues, or gets worse, proved somewhat tricker.
The status quo and escalation scenarios
The ultimate toll of the other two scenarios — status quo and escalation — depends heavily on the number of new traumatic injuries.
"We had to be very careful in defining these scenarios," says Igusa. The team decided to base the scenarios on existing casualty data. But the current death toll, largely documented by the Gaza Ministry of Health, is likely an underestimate, as not all deaths are reported.
For six months of war comparable to the early stages, they took the average casualties from October 15 through January 15 and spun them forward over the next six months. For escalation, they assumed fighting from now through August would be as intense as October 11 through November 10, the worst month of the conflict so far that claimed more than 11,000 lives.
To Ball, those are plausible assumptions, but he stresses there's a lot of uncertainty.
"Escalation might involve a whole lot of indiscriminate bombing in densely populated areas, or Israeli forces could decide to flood the tunnels with seawater," he says. "We don't actually know what any of those scenarios will mean in terms of the armed groups' actions."
There's also uncertainty in the underlying numbers. As Israel's military assault has gone on, collecting good numbers has been harder, says Spagat.
"I've got less trust in what the [Ministry of Health] is releasing now, not that I think they're making stuff up, just that it's worse quality," says Spagat. In fact, he suspects that more deaths are likely going uncounted now, compared with the beginning of the war. Still, he says these numbers "are reasonable to use. Honestly there's no other alternative."
The researchers tried to account for those missing deaths with an independent data source. The U.N. Relief and Works Agency for Palestine Refugees in the Near East closely tracks deaths among their staff of over 13,000 people in Gaza. By comparing death rates from the agency and the Ministry of Health, the team arrived at a better estimate to make their projections, says Igusa.
Non-violent deaths increase in the status quo and escalation scenarios too, as there's greater chance for interrupted medical care and epidemics. But given the authors' assumptions, ongoing violence would account for the bulk of those excess deaths, which could reach 66,720 under the status quo scenario and 85,750 under escalation.
While those specific numbers come with lots of caveats, says Spagat, "the broad brush picture that they're painted should be taken seriously."
The scope of these projections can be seen by looking at estimated trauma deaths from two of history's most devastating bombing campaigns during World War II, with Allied bombs killing some 25,000 over two days in Dresden and Nazi attacks taking approximately 40,000 lives in London over eight months.
Will these numbers make a difference?
This initial report is very much a first draft. Spiegel, Igusa and colleagues plan to release updated projections over the next several months, refining their assumptions based on experience and incorporating new data, including measures of mental health.
Until then, the authors hope these projections of the future might spur action to avoid the worst outcomes.
"I hope we've made clear that there's still going to be a lot of death if there's a ceasefire," says Spiegel, underlining the importance of getting adequate food, water and medical attention to where it's needed as soon as possible. Right now, blockades, continued assaults and damaged roads are preventing that aid from reaching those who need it.
"These very quantitative, evidence-based efforts are valuable," in forcing politicians and humanitarian agencies to confront the human cost of continued fighting, says Asi. "But that's just the first step. They have to be coupled with advocacy and political action."
Since the conflict began, wrenching images and videos of the suffering in Gaza have made the conflict real to many around the world, spurring some advocacy and political action. But the fighting continues. Whether these numbers push political actors towards a lasting ceasefire "remains to be seen," says Asi.
Jonathan Lambert is a Washington, D.C.-based freelance journalist who covers science, health and policy. He's been a staff writer at Grid and Science News and has contributed to NPR, Nature News, Quanta Magazine and the Dallas Morning News. He holds a Master's degree in evolutionary biology from Cornell University. Follow him on X @evolambert or on bluesky @jonlambert.bsky.social.
Copyright 2024 NPR. To see more, visit https://www.npr.org.