A fundamental fact of wars is chaos: even basic information is often kaleidoscopic, contested and fleeting.
This is true in Gaza as well, despite the territory’s small size, heavily documented population, relatively robust pre-war medical system given its development level and a massive international and media presence.
The unprecedented flow of images, videos and data out of the Strip provides an illusion of precision conveyed through metrics such as daily casualty updates, aid truck counts, targets struck or munitions used. Much more is known about the death toll in Gaza – which the Hamas-run Gaza Health Ministry (MoH) claims is now more than 54,000 – than in many other war zones.
Take Sudan, where aid shipments are vanishingly rare and the official casualty tally is considered a massive undercount. Or Ukraine and Russia, where both sides are broadly understood to be dishonest about military losses and where the United Nation’s Ukrainian civilian death toll does not include the multitudes killed in areas under Russian control. Or the Ethiopian region of Tigray, where the UN said only 66 children had died during the civil war but where outside analysts estimate between 300,000 and 800,000 people died – many of famine.
However, that does not mean that the death toll in Gaza is known precisely. Due both to this and the unique level of politicisation of the death toll in this conflict, careful and rigorous study and reporting remains essential.
While what is known about the death toll appears mostly substantiated, large, unanswered questions remain about the MoH’s methodology, other Hamas-run bodies continue to spread disinformation, and there are key unknowns. These include the scope and structure of the uncounted death toll; whether Israel, Hamas, or other armed groups are responsible for specific deaths; and to what extent natural (deaths that would have occurred regardless) and indirect (deaths that occur because of the war but are non-violent) deaths are included in the GHM lists.
Delving into these issues requires a much longer format, and my recent report for the Washington Institute for Near East Policy, “Assessing the Gaza Death Toll After 18 Months of War”, does just that, examining the most recent MoH list published last month and outlining remaining areas of uncertainty. Simply collecting and cleaning the available fatality data, determining which body is reporting what, and conducting basic analysis has already escaped most analysts, reporters, and even governments, despite the prominent position of such statistics in nearly every story about the war between Israel and Hamas.
To cut through noise, opaque or shoddy reporting, and outright falsehoods that often accompany reporting and analysis on the war, readers should keep several questions in mind:
Who is the source of the data?
There is only one institution authorised to or capable of reporting the Gaza death toll: the Gaza MoH. Others are either relaying MoH figures, as the UN does, or are pushing false figures, as does the Hamas-run Government Media Office (GMO). Those relaying the information may do so improperly (as the UN did repeatedly early in the war) or attribute it incorrectly (either through circular citation or masking GMO claims as coming from the MoH). This includes the Palestinian Authority, major newspapers, and sitting US Senators.
Is the claim about a specific incident or the overall toll?
Reporting on specific incidents may come from eyewitnesses, journalists, medical workers from the Gaza Civil Defence, hospital spokesmen, the MoH, or the GMO. Many of these may be unreliable or lack the full picture, and it is unlikely that death tolls produced immediately after events are fully accurate.
What is known and unknown about the death toll?
The MoH list has become significantly more complete over time and has very few duplicated entries or invalid IDs. Many prior problems with the dataset have been minimised, though major questions remain about the insufficiently explained “media reports” and mass removals of some 4,000 entries over the last year.
However, completeness – having information for each entry – does not imply validity (that this information is accurate) or accuracy (that the entry belongs on the list). For this, information outside the list is needed, either from open sources or the population registry.
The stated death toll does not capture everything. However, the undercount is most likely of moderate size and will likely skew towards militants, who more often die in remote or inaccessible areas and whose families are less likely to report them. This is independent of any potential mass omission of militants from the MoH list. The MoH has admitted to removing some natural deaths from the lists, and without outside information, it is difficult to know if more are to follow. However, it is unlikely that natural deaths are included en masse.
What are the demographics of the death toll? The death toll, of the 52,958 deaths reported in mid-May 2025, was 49.3 per cent men (18-59), 31.2 per cent children (0-17), and 19.5 per cent women (18-59) – a major over-representation of men relative to their share of the population and under-representation of women and children. Men aged 18 to 45 were 2.84 times more likely than women to have been killed in that age range. While civilian men die at higher rates than civilian women during war, the imbalance is striking and indicates militant presence in the count (although many militants are likely excluded).
What is the civilian-combatant ratio?
The MoH does not distinguish between civilians and combatants, and it is likely that many combatants are excluded from their death toll list. The IDF has provided some assessments of how many militants it has killed but has not released data to back this up. Moreover, the IDF and third-party observers all have different methods and evidence thresholds for determining combatant status. Accordingly, MoH and IDF figures should not be combined – they overlap only imperfectly, and doing so will obscure more than it illuminates.
Comparisons to other conflicts are also challenging. It is difficult to find another war that took place in similarly dense urban conditions with full militant exploitation of civilian infrastructure, massive tunnel networks, and a population not allowed to flee the fighting into Egypt. Later assessments of a civilian-combatant ratio will have to take this into account.
The above are all technical points that only scratch the surface. Just as important, however, is the moral component. Lists and datasets are abstract, but each name reflects a person and a life cut short, many of them trapped in a war not of their making started by Hamas’s brutal assault on Israel on October 7, 2023. The terrible cost of this war must not be minimised or dismissed.
Responsibly analysing and reporting the Gaza death toll means accurately identifying who is making a claim, what is included and excluded from that claim, and caveats if the information cannot be verified. Many news organisations, and certainly the public discourse, have almost completely failed on these basic points for 19 months. There is no better time to start than now.
Gabriel Epstein is a research assistant in The Washington Institute’s Koret Project on Arab-Israel Relations
