On December 13 the Henry Jackson Society published a report by Major (ret’d) Andrew Fox attacking the death-toll figures produced by the Gaza Ministry of Health (MoH) and differential media treatment of MOH and IDF claims. Here we review just the five “key findings” in the Fox’s report – “Questionable counting” – and conclude that they contain some truth but that the evidence that Fox presents does not support the dramatic charges he levels against the MoH and the media.
In public comment, Fox – a retired British army officer who has been the guest of the Israel Defence Forces (IDF) on trips to Gaza – has claimed that media outlets worldwide have irresponsibly relied on fatality figures from the Gaza Ministry of Health (MoH), which is controlled by Hamas, a designated terrorist organisation.
His report argues that the MoH inflates civilian casualty counts by misclassifying combatants and natural deaths as civilian war fatalities. He argues that media reporting of such death tolls has fuelled global antisemitism, legitimised anti-Israel narratives, and eroded trust in journalism. He calls for greater scrutiny of data sources by journalists to prevent the spread of disinformation, uphold ethical reporting standards, and avoid amplifying Hamas’ propaganda, which frames Israel as the primary aggressor while downplaying Hamas’ role in civilian harm.
Systematic Bias vs. a Small Number of Random Errors
In response, we here first address three of the key findings in Fox’s report, specifically that the MoH casualty accounts list:
- men as women “to inflate female fatalities”
- adults as children
- natural deaths as well as violent ones
Admittedly, Fox finds examples of the above errors that should be corrected but, for two principal reasons, we argue that no important implications flow from his findings.
First, Fox reports only evidence that supports a narrative of intentional data manipulation by the MoH, while more comprehensive analysis is more suggestive of random error.[1] Thus, while men are misclassified as women, women are also misclassified as men. A review of over 21,000 entries (53% of the total) from the most recent list published in October shows that of 116 entries with a likely misclassified sex (roughly 0.5% of the sample), 67 are men listed as women and 49 are women listed as men (see technical appendix).
Similarly, a comprehensive comparison of the April and June MoH lists of claimed dead reveals at least 13,389 age revisions, with over 95% subtracting 1 year.[2]
We found that downward revisions are spread widely across sex and age groups, including men moved into a “fighting-age male” category however defined.
We think the MoH probably calculated ages for many entries using a spreadsheet formula that measured times between dates of birth and the present until it realised that this device continuously “aged” the dead until they were “de-aged” by fixing the problem (see technical appendix for details). It would hardly make sense for the MoH to try and inflate child numbers by decreasing the age of thousands of entries far from the adult-child boundary.
The second problem with Fox’s critique is that the errors identified are too few to discernibly affect the percentages of women and children among the dead. For example, transferring Fox’s 103 cases from female to male only decreases the percentage of female entries on the last MoH list from 35.3% to 35.1%.
Similarly, Fox identifies only 2 cases of adults registered as children and 3 cases of “inclusion of natural deaths”, based on their appearance in medical transfer lists after being listed in a death list. We reviewed 5,004 cases of patients and companions listed for transfer out of the Strip and found 12 who appeared on a prior MOH death list – a few more than Fox found but still a tiny fraction of all listed deaths (see technical appendix). Fox would need to find far more natural deaths to support the claim of deliberate fraud.
Fighting Age Males
A fourth key finding of Fox’s report is concerning the “disproportionate deaths of fighting-age men.” It is true and well known that the proportion of males aged 15-45 on MoH death lists (38.7%) is substantially higher than their population share (~25%), and current data show they are 2.4 times as likely to have been killed as women and girls in that age bracket. Yet, once again, Fox’s argument finds itself running up to greater complexities than he gives credence to.
Recent MoH data releases distinguish between deaths recorded by the MoH’s “Central Collection System” and deaths reported to the MoH through “Family Self Reports”.
Fox trumpets the fact that the percentages of young males are much higher for family-reported deaths than for central-system deaths and argues that the family-based system more accurately measures the true demographics of the dead than the central system does.
But Fox works with the penultimate MoH data release covering the war through August, where this argument was more plausible, not the newest one released on October 23 and covering the first year of the war.
(The MoH released the list covering up to October 7, 2024, on October 23 through its public WhatsApp group, rather than Telegram, where all its previous lists have been distributed. Many analysts likely missed it for this reason, so we have posted the original list (PDF), a spreadsheet conversion true to the original, and a cleaned-up version with duplicates removed and basic demographic processing.)
The differing demographics across the two casualty recording systems evaporates in the October list and is even slightly reversed. For example, males aged 15-45 constitute, respectively, 38.8% and 38.1% of the deaths in the central and family systems.[3] The reason for this shift is unclear but its existence calls into question the idea that comparing the demographics of deaths captured by the two methodologies can illuminate the true death toll of the war.[4]
Some Media Criticism is Warranted
There are numerous examples of high-quality media coverage of the Gaza death toll. However, we agree that the media has largely failed to apply proper scrutiny to some high-profile MOH claims, such as the allegation that Israel was responsible for the al-Ahli hospital explosion in October 2023, which was almost surely caused by a misfired Palestinian Islamic Jihad rocket.
Further, as Fox points out, the MoH claim that all deaths on their detailed lists were directly caused by “Israeli aggression” is wrong and should not be repeated – some people are killed by Palestinian militant groups directly or by fallen rockets. The media has also frequently repeated a claim, incompatible with the MoH’s own data, that 70% of Gazans killed are women and children.
Deaths of Combatants
The final key finding, “Media Underreporting of Combatant Deaths”, is based on research showing that MOH numbers are frequently presented in the media without scrutiny and/or without mentioning IDF claims to have killed over 17,000 militants in Gaza. This is a good point – IDF claims are relevant and should be reported. They should also be taken seriously and subjected to scrutiny, which is, unfortunately, difficult to do at present because the IDF has not released detailed evidence to back its claims.
In fairness, many states e.g., Russia and Ukraine, commonly make poorly evidenced claims for opposition fighters killed, although Every Casualty Counts maintains that they have a legal obligation to account for wartime casualties. Israel ostensibly has some means to do so, despite battlefield difficulties. Reportedly it has fully confirmed the identity of some 10,000 militants killed by September. The IDF has seized extensive documentation from terror groups in Gaza, including personnel files, which it has excerpted in public releases. These can be cross-referenced with the MOH lists and the Palestinian population registry, which Israel has a mostly up-to-date copy of, to produce a partial list of militants killed. Even a partial accounting would allow for some analysis of IDF claims.
Trying to Reconcile the MoH Lists with IDF Claims
The overarching goal of the HJS report seems to be to shoehorn the IDF estimate of militants killed into MoH death totals to demonstrate that combatants constitute a large percentage of the dead. This simply does not work because the demographics of the MoH lists are incompatible with the IDF claims – there are not enough fight-age males on the MoH lists to make this argument work.
The case for MoH-IDF reconciliation would have to run along the following lines. Many militants are killed deep in combat zones, in tunnels, or in other places that make them unlikely to be recorded at hospitals or morgues. Families may not know, and therefore not report, militant family members killed. Moreover, Hamas actively discourages acknowledgement of its undoubtedly extensive casualties so family members may stay silent even when they know about deaths. Thus, the deaths that are uncounted by the MoH may include enough militants to sustain the IDF claims.
The above argument would require that the true death toll of the war is higher than the MoH number. Adding these militants would shift the demographics toward fighting-age males, although there will also be uncounted civilians. Substantially lowering the ratio of civilians to combatants would require that the demographics of uncounted deaths lean heavily toward militants. This is possible but unprovable given the current state of knowledge.
Conclusion
Criticism of MoH data, something that both of us have done, is necessary and can help to improve their quality. As such, the Henry Jackson Society’s report by Andrew Fox raises legitimate concerns about the reliability of Gaza MoH casualty figures and media reporting on them.
However, a closer analysis of the report’s key findings reveals significant flaws in its evidence and reasoning. While it identifies instances of misclassification and questionable claims by the MoH, these errors appear to be more indicative of random mistakes than systematic manipulation. Furthermore, the small number of errors identified is insufficient to meaningfully shift overall casualty demographics.
Fox’s argument that media outlets have uncritically echoed MoH claims is valid in part, particularly regarding misreported death tolls and the attribution of responsibility for certain attacks. His attempt to reconcile MoH death counts with IDF claims of militant fatalities is fraught with methodological issues and unsupported assumptions. Without greater transparency from both the MoH and IDF, such reconciliations remain speculative at best.
Ultimately, while the report underscores the importance of media scrutiny and the verification of war casualty figures, its own selective presentation of evidence undermines its most serious claims. Balanced reporting requires scepticism toward all parties in conflict, and while the media should critically assess MoH data, the same standard should apply to IDF casualty claims.
The truth, as is often the case in wartime, is far more complex than Fox’s report allows.
[1] The MOH has made a major effort to clean up its death toll list. In April 17% of the entries had problems such as missing information, duplicates, and invalid ID numbers while by October only 0.2% had such problems. However, the remainder may still contain errors such as sex misclassification, incorrect ages, or complex duplication, which are difficult to assess without access to the population registry.
[2] There were several significant revisions between April and June so this comparison focuses exclusively on 13,400 entries for which age is the only difference between the two lists.
[3] Fox claims that “most” deaths are in this “fighting-age men” category.
[4] The August list contained four methodological categories: the central collection system, family reports, “judicial committee”, and an unspecified “House” or “Local Committee”. The October list combined the latter three categories together, which may explain some of the large demographic shift, but the category also expanded by several thousand new entries.
Gabriel Epstein is an Israel Policy Forum policy and communications associate, based in Maryland. He previously worked as a research assistant in the Koret Project on Arab-Israel Relations at The Washington Institute for Near East Policy, and his analysis has appeared in The Washington Post and The Associated Press.
Michael Spagat, Professor of Economics at Royal Holloway, University of London, researches armed conflict, including civilian casualties and war death measurement and has been widely published in leading journals such as Nature and PLoS Medicine.
Did you find this story interesting? Please support AOAV's work and donate.
Donate