Who's counting?: Tigray's hidden mortality data crisis

This is the sixth post in the Mortality Estimation in Crises blog series accompanying the Mortality Estimation Systems Innovation Partnership, supported by UKHIH–Elrha, which brings together diverse partners to strengthen how mortality data is collected, understood, and used to inform humanitarian decision-making.

Between 2020 and 2022, intense conflict in the Tigray region of Ethiopia created a perfect storm of political and operational challenges that disrupted the production of mortality evidence – information governments and their humanitarian partners need to inform crisis responses. Occupation by Eritrean forces, an information blockade, political tensions between the federal and regional authorities and bureaucratic power suppressing critique meant that little robust, timely data was produced.

Today, with some areas still experiencing displacement and low-level conflict, the health system is slowly recovering but livelihoods and the food supply remains insecure, and there are new national restrictions on humanitarian surveys. Throughout, researchers at Ethiopian public universities have been among the only actors able to keep working.

This post draws insights from research conducted by the London School of Hygiene & Tropical Medicine for the Mortality Estimation Systems Innovation Partnership, including interviews on how local university teams adapted to Ethiopia's shifting constraints on mortality estimation. Speaking openly about these constraints carries real risk for those still working inside them, which is why this account is written from outside the country, drawing on what colleagues there and non-academic stakeholders were willing to share. University actors in Ethiopia are trusted, independent, government-adjacent and community-rooted, and they deserve the humanitarian community’s recognition and support.

When the usual systems broke

The Tigray war disrupted the usual ways the government’s emergency department and humanitarian agencies track deaths. SMART surveys – which measure the crude death rate, under-five death rate, and malnutrition – and the systems around them, are a legacy of Ethiopia’s 1999–2000 famine. Securing government approvals for these surveys became impossible during the Tigray conflict.*

Likewise, Ethiopia’s long-term demographic surveillance system became crippled in conflict regions. This system collects regular, high-quality community-level information on deaths from a network of villages to monitor the long-term impact of development initiatives and policy changes. Population scientists from Mekelle University had run a set of surveillance sites since 2009. The 2021 government banking blockade cut university funding, conflict displaced field teams, and internet outages blocked data sharing.  

Filling the vacuum

With state-run data collection paralysed, different actors found various ways to re-establish the supply of mortality information – and use this data to speak out.  

Outside Ethiopia, online civil society-led projects compiled lists of civilians killed in the violence and surveyed mortality through family networks.

Inside the region, Mekelle University drew on its standing with the Regional Health Bureau and local communities to run rapid field surveys. One assessment of household food insecurity and child malnutrition led the team to publish a stark warning in The Lancet, calling for urgent action on the unfolding crisis. Later, with UNICEF and UNFPA support, the same group produced Tigray's first estimates of maternal, neonatal and under-five mortality. The estimates showed under-five deaths had doubled since before the war, crossing the threshold used to define an 'emergency' and reversing decades of progress on child survival.

Beyond this, no full-scale mortality estimation work was possible.

A fragile, fragmented recovery

Following the peace agreement in November 2022, data collection resumed. UN agencies and nine other international NGO partners worked with the government nutrition department to conduct coordinated SMART surveys in 2023 and 2024. However, following the introduction of new national guidance and laws in January 2025, mortality and nutrition assessments were restricted to government institutions – stalling evidence collection again.  

Public universities, as government entities, remain eligible to carry out mortality estimation but sit awkwardly in a fragmented ecosystem. As described by one NGO informant, universities and government departments tend not to be “emergency oriented”, having few direct connections with the humanitarian actors who have historically been the ones with funding, systems and mandates to commission and act on mortality evidence. Yet, as we have argued before, universities like Mekelle are exactly who communities and regional government trust to provide independent technical guidance on mortality measurement.

Why academics belong at the centre of this work

Universities will undoubtedly be central to strengthening Ethiopia’s mortality data ecosystem, but effective collaborations with government and UN agencies must connect better around them. Public health initiatives should be designed to bridge the gap between currently disconnected UN/NGO and government ecosystems so university-led evidence reaches decision-makers faster and university programmes can be sustained rather than rebuilt from scratch each time.  

More than that, academics bring something the humanitarian system urgently needs: a long view. In a recent webinar from our project, collaborators from Somalia, Democratic Republic of Congo and Ethiopia spoke about community data systems as the next frontier in mortality estimation. In Tigray this might include churches, Edir platforms (which help support families during death), vital events registration offices, health facilities and other organisations. University teams are uniquely placed to build and sustain grassroots networks and to help local public health actors analyse and act on what these community data systems reveal, in crisis and beyond it.

Realising this potential means a shift in how humanitarian actors invest: away from one-off emergency surveys, and toward sustained support for the local institutions equipped to build evidence systems that last – in Ethiopia, and wherever crises strike.

Further reading

Results from a new mortality survey conducted by Mekelle University among rural, urban and displacement camp populations in 6 out of the 7 zones in Tigray region in 2026 are forthcoming and will be available here.  

*One international NGO (anonymised to protect identity) attempting to work outside the government system completed a household survey to inform their programme, however they did not share results for fear authorities would cancel their operating permit.