Measuring migrant deaths and disappearances

There are many challenges that hinder documentation of migrant deaths and disappearances but also much that can be done to improve the coverage and completeness of such data.

People trying to migrate to other countries drown in sewage canals; they freeze to death in remote forests; they die of dehydration and starvation after their vehicles break down in the middle of the desert. Documenting migrant deaths and disappearances helps to counter the invisibility of these tragedies, while collection, analysis and dissemination of such data are also valuable for advocacy campaigns, to inform evidence-based migration policies, and to improve operational planning. 

However, collecting information on fatalities during migration is challenging for several reasons. The lack of safe and legal options for migration forces people to travel on irregular migration routes to evade detection. As a result, deaths either go unreported or without witnesses. Remote, harsh topography also makes it difficult to find people’s remains; the remains of countless people who die on irregular migration journeys – especially those who go missing at sea – are never recovered.

There are additional, though more surmountable, issues involved: firstly, the difficulties of agreeing an operational definition of a ‘missing migrant’ and, secondly, the lack of reliable, accessible data sources.  In the context of the COVID-19 pandemic, data collection has become increasingly challenging, with additional obstacles posed by reduced media coverage of migrant deaths, inability to reach migrants to collect testimonies, and border closures pushing migrants to more invisible routes.

Who is a missing migrant?

The International Organization for Migration’s Missing Migrants Project (MMP) is a global, open-access dataset tracking deaths of people along mixed migration routes worldwide.[1] Since it was created in 2014, the project has recorded 40,505 deaths and disappearances during migration.[2] There is no agreed definition of ‘missing migrant’ but clearly an operational definition is needed in order to produce an evidence base.

MMP’s definition refers to a person who died (or who went missing and is presumed to be dead) in the process of migration towards an international destination, regardless of the person’s legal status. Migrants who die or go missing outside the process of migration – for example, when a migrant is residing in a foreign country – are not included in the database. MMP also excludes deaths that occur in immigration detention and settlements such as refugee camps, and deaths after deportation. This definition was designed specifically to identify the risks linked to irregular migration journeys.

A category of missing migrant that is absent from MMP’s definition relates to the many migrants who lose contact with their loved ones. These people may be alive but in situations where they are not able to contact their families, or they may choose not to for fear of deportation or other consequences.

Though adopting a definition is necessary for starting data collection on any topic, it can also have negative consequences. For example, the exclusion of deaths that occur in migrant shelters, refugee camps or in detention can lead to the perception that migrants in such spaces are safer than those on the move. It can also lead to underestimating the scope and scale of migrant deaths by those who are not aware of MMP’s methodology. This is especially problematic when MMP’s figures are used in the media with limited background information and when death rates are shown without an explanation about how they were calculated.[3]

In addition, establishing when a death happened ‘in the process of migration’ is particularly problematic. In general, the notion of transit implies temporariness but it says nothing about the length, direction or continuity of the migration journey. In the MMP methodology, it is assumed that transit is generally continuous; in reality, migrants often pause or re-route their journeys to regain strength and financial resources, and their completed journeys may take months or years. If a migrant dies or goes missing during one of their stopovers, their death would not be counted by MMP. The result is that many incidents are excluded from the database because it could not be established that the migrant was in transit. 

Likewise, migrants are often forced to pause their journeys because of government policies that restrain their movement. In 2020 mobility restrictions imposed to contain the spread of COVID-19 have been of particular concern, leaving thousands of migrants stranded in dangerous situations where humanitarian support, including access to health care, may be unavailable.

An Ivorian teenager died on board a quarantine ship in Italy one month after being rescued from the Mediterranean. Reportedly, the lack of access to adequate health care on the ship caused his health to deteriorate and eventually his death.[4] Had he had access to better health care immediately after his sea journey, his life might have been saved.

Sources of data

Few official sources collect and publish data on migrant deaths, and the available data are sometimes inaccurate or incomplete. The few official data that are available are typically collected by border agencies and forensic examiners. To address the issue that many missing migrants are not captured in government data, MMP relies on other sources, such as social and news media, NGOs, civil society, international organisations, and migrants themselves.

In its efforts to help the families of missing migrants, the Pima County Office of the Medical Examiner in Arizona examines unidentified remains found in its jurisdiction and determines which ones could be coded as ‘undocumented border crossers’. The criteria they use include whether the body was recovered in a known migration corridor, whether the person’s belongings match the items typically carried by migrants, and whether those supplies came from Mexico. This simple classification translates into an invaluable record of migrant deaths along a border with one of the highest fatality rates. Since 2001, they have recorded over 3,398 migrant deaths.[5]

Media reports are one of the most frequently used sources by MMP. However, very often the information that the media provide about migrant fatalities is inaccurate, incomplete or misinterpreted and exaggerated.[6] In addition, where deaths occur in large numbers and on well-known routes, such as the Mediterranean route, these are more likely to be reported by the media and this can bias MMP data. This problem is further exacerbated when a large part of the media coverage focuses on just a few issues. In 2015, when there was wide media coverage of the Bay of Bengal crisis, our data saw a substantial increase. MMP documented 577 Rohingya migrants’ deaths at sea that year. In contrast, in 2020, MMP has only recorded 165 such deaths, even though NGOs have declared that well over a thousand Rohingya have been stranded at sea for months, indicating that the death toll might well be higher.

MMP also uses other sources of information such as migrant surveys and survivors’ testimonies. The more recent crackdown on NGOs that carry out search and rescue operations and offer lifesaving support to migrants – such as those active in the Mediterranean and at the southern US border – not only puts migrant lives at risk but also hampers our ability to collect evidence. Lastly, MMP uses maritime departure data to identify possible incidents involving migrant deaths or disappearances on overseas routes.

In 2020, the sea passage from the northwestern coast of Africa to Spain’s Canary Islands saw a significant increase in migration flows compared with 2019. Information about the sex and age of people who go missing on this route and the precise location of their death or disappearance is scarce. In 2020, at least 849 people lost their lives on this route. Several incidents potentially involving hundreds of deaths are still to be verified and recorded by MMP but evidence is difficult to obtain. MMP has consequently started to collect data on departures from the Senegalese coast; this information can then be compared with that on arrivals in the Canary Islands to account for potential shipwrecks. 


Though there are many challenges that hinder documentation, much can be done to improve the coverage and completeness of data on missing migrants, as the following recommendations suggest.

Data must represent the lived experiences of people taking migration journeys. Actors using operational definitions, such as MMP’s, should develop guidelines on how to adapt their methodologies to the evolving realities of migration journeys.

All actors working in spaces where migrant deaths and disappearances occur should collect and share data. Also, actors collecting missing persons reports should disaggregate these data so that disappearances linked to migration can be identified.

Data on missing migrants should be shared among actors, regardless of definition. Data on missing migrants tend to be scattered and fragmented, and there is great value in bringing disparate sources of data together for comparison and verification.

States must urgently start collecting data on migrant deaths within their territories. Official actors are likely already to collect data about deaths within their jurisdiction. Local, national and regional authorities that collect such data should disaggregate it by migratory status and publish it in accordance with data protection standards. This includes collecting information – in a sensitive way – from migrants about deaths or disappearances of fellow migrants that they have witnessed.

NGOs, international NGOs and civil society should collaborate to resolve cases and generate evidence. Data collection efforts led by non-governmental actors working directly with people on the move can provide new information and increase the likelihood that data can be verified, particularly in cases where migrants’ bodies are not recovered. Collaboration with and between families of the missing should also be encouraged, as their involvement can both provide crucial information and help families cope with the uncertainty of the loss that they face.

The media should use existing guidelines for reporting on migrant deaths. Media reports can be the first indication that a migrant death or disappearance has occurred, and as such can play a vital role in data collection. Details on the people involved in such an event should be included whenever possible, while giving due consideration to the deceased and their families’ interests. Beyond this, news reports should mention the limitations of any data published, so as to avoid inaccuracies or misinterpretations, and should endeavour to portray the people behind the numbers.


Andrea Garcia Borja @agarciaborja

Data Analyst


Julia Black @_Julia_Black

Project Officer


IOM Missing Migrants Project



[2] Figure covers period from 1 January 2014 to 30 January 2021.

[3] Dearden K, Sánchez Dionis M, Black J and Laczko F (2019) ‘Calculating “Death Rates” in the Context of Migration Journeys’

[4] Agenzia Italia (2020) (Italian only)

[5] Map of Migrant Mortality

[6] White A and Singleton A ‘Mixed messages: Media coverage of migration and fatalities’, chapter 3 in IOM (2017) Fatal Journeys, Volume 3, Part 1: Improving Data on Missing Migrants


Opinions in FMR do not necessarily reflect the views of the Editors, the Refugee Studies Centre or the University of Oxford.
FMR is an Open Access publication. Users are free to read, download, copy, distribute, print or link to the full texts of articles published in FMR and on the FMR website, as long as the use is for non-commercial purposes and the author and FMR are attributed. Unless otherwise indicated, all articles published in FMR in print and online, and FMR itself, are licensed under a Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. Details at