Conflict-attributable mortality in Tigray Region, Ethiopia: Evidence from a survey of the Tigrayan diaspora.

Background: The war in Tigray Region, Ethiopia (November 2020 to November 2022) ended with formal commitments to accountability, but these have yet to produce publicly available accounts of the harms caused by the conflict.

Methods: We carried out an online survey of the Tigrayan diaspora to estimate mortality amongst children, adults, and older adults during, and prior to, the war-period. We collected retrospective demographic information on respondents', and their spouses', extended family inside Tigray. To mitigate selection bias, we standardised mortality estimates by rural-urban residence and wealth index.

Results: Of 1011 participant-reported decedents, 810 died within Tigray, and 310 of these individuals died during the war-period. Of the 310 deaths in Tigray during the war-period, 224 (72.3%) died from intentional injuries. The standardised mortality rate for adults (15-49 years) was 21.3 per 1000 person-years (29.4 for men, 14.8 for women) during the war, and 1.0 in the preceding period (2010-2020). The mortality rate amongst older adults (≥ 60 years) was 45.1 per 1000 person-years during the war-period, compared to 22.8 in 2010-2020, and is higher than the period encompassing the Ethiopian Civil War (1974-1991) and Tigray/Wollo Famine (1984-1985). The mortality ratio (men to women) was approximately 2:1 in both adults, and older adults. The mortality rate amongst adults and older adults had been declining across the pre-war periods. Few deaths were reported amongst children. We estimate that the conflict has resulted in more than 102,000 deaths amongst those aged ≥ 15 years.

Conclusions: Our study suggests a significant elevation in all-cause mortality, largely driven by intentional injuries. Although our pre-war-period data are likely under-reported, comparisons with other pre-war estimates corroborate these plausible elevations, particularly amongst adults. The mortality ratio, when compared to those from other settings, does not support assertions that the military strategy primarily involved the targeting of adult males, and instead suggests widespread killing of women and older adults who would not have posed a combat threat.