New Study Questions Reliability of Widely Cited Gaza Mortality Survey in The Lancet Global Health

A recently published correspondence in The Lancet Global Health casts serious doubts on the reliability of a highly cited Gaza Mortality Survey, which had estimated approximately 75,200 violent deaths during the recent Israel–Hamas conflict. The reassessment, conducted by Professor Emeritus Sergio DellaPergola, a distinguished demographer affiliated with the Hebrew University of Jerusalem, alongside independent researcher Mark Zlochin, reveals significant methodological flaws that undermine the survey’s claims of representativeness and the validity of its mortality estimates.

The Gaza Mortality Survey was initially lauded for providing one of the few systematic attempts to quantify conflict-related fatalities within the Gaza Strip during a period of intense violence. However, DellaPergola and Zlochin’s critical reanalysis of the publicly accessible microdata and related documentation reveals substantial anomalies at the level of individual interviewers. These anomalies raise questions about the fidelity of data collection to the declared probabilistic sampling methodology and potentially bias the extrapolated mortality figures applied to the wider population.

One of the most striking findings from the reanalysis concerns the distribution of reported deaths among interviewer teams. Contrary to expectations derived from the survey’s design, a single interviewer team was responsible for recording nearly one-quarter of all violent deaths reported in the study, despite surveying only a small fraction of the total sampled households. This disproportionality hints at possible sampling bias or data reporting distortions that compromise the assumption of equal probability sampling and population representativeness fundamental to producing reliable mortality estimates.

Further scrutiny uncovered demographic irregularities within the data collected by certain interviewers which did not align with externally available demographic profiles of Gaza’s population. These irregularities include inconsistent household structures and reporting patterns that deviate markedly from expected norms. Additionally, despite the survey’s stated quality-control protocols, documented evidence suggests these measures failed to identify or mitigate such interviewer-level heterogeneity during data collection, raising concerns about the integrity of the dataset as a basis for mortality estimation.

Supplementing their analysis, the authors employed GPS-based evidence to examine adherence to the survey’s outlined sampling strategy. Such spatial data indicated problematic overlaps between sampling areas that were intended to be mutually exclusive. Interviewers reportedly concentrated their efforts within only small portions of their assigned geographic regions, which likely failed to capture the full spatial diversity of the Gaza population. Consequently, this introduces systematic sampling errors due to undercoverage and spatial clustering, further calling into question the generalizability of the mortality data.

An additional methodological flaw highlighted pertains to the potential double counting of conflict-related deaths. The extreme conditions faced by families during wartime—such as displacement, communication breakdowns, and fragmented household units—may have led to multiple reports of the same fatalities among different households included in the survey. Since the survey lacked robust mechanisms to identify and reconcile duplicated reports, this issue threatens to inflate reported death counts erroneously.

The critique also notes a conspicuous omission in the published survey results: while causes of death were reportedly collected during fieldwork, these data were not disclosed or analyzed. This lack of transparency impedes independent validation of violent death classifications and obscures critical epidemiological insights necessary for interpreting mortality patterns within the conflict context.

Moreover, the correspondents question the reliability of the survey’s estimates related to imprisonment during the conflict, noting substantial inconsistencies with independently reported detention figures. The discordance between the survey data and external administrative or monitoring sources introduces further doubts about data accuracy and the robustness of conclusions drawn from the survey in portraying the full scope of conflict impacts on Gaza’s population.

DellaPergola has a well-established reputation in demography, with extensive contributions to population studies in Israel, the Jewish diaspora, and the broader Middle East. His expertise in analyzing poorly defined populations and designing rigorous data collection and evaluation methods underpins the current reassessment. Together with Zlochin, their correspondence embodies a crucial scientific contribution to an ongoing debate on the appropriate methodologies for casualty estimation in conflict zones, where data are inherently difficult to acquire and verify.

Their critical review underscores the principle that population-level mortality estimates rely fundamentally on representative sampling frameworks. Any substantial deviation from sampling protocols, interviewer biases, or data irregularities must prompt cautious interpretation. The Gaza Mortality Survey’s claims to national representativeness appear compromised, limiting the utility of its casualty figures for policy formulation, humanitarian response, or scholarly understanding.

This published correspondence not only highlights the specific methodological deficiencies in a widely referenced survey but also has broader ramifications for the field of conflict epidemiology. It calls for enhanced transparency, rigorous independent validation, and the integration of alternative data sources—such as satellite imagery, administrative records, and crowd-sourced information—to triangulate mortality estimates more reliably.

In conclusion, while the Gaza Mortality Survey sought to fill a critical knowledge gap in quantifying the human toll of conflict, the reanalysis by DellaPergola and Zlochin makes a compelling case that the survey’s findings should be treated with considerable skepticism. Their work serves as a reminder of the persistent challenges in generating accurate population health metrics amidst warfare and the imperative for methodological rigor and critical appraisal in the science of conflict mortality estimation.