Accessibility menu

Content for the offcanvas goes here. You can place just about any Bootstrap component or custom elements here.


Disaggregating disaster deaths data

Ilan confirms the importance of knowing who dies in disasters, how, and why.

10 January 2022

To guide disaster-related life- and livelihood-saving measures, it helps to know who dies in disasters, how, and why. Doing so requires disaggregating the data on fatalities to explore the reasons for people being in particular situations in particular places at particular times. We need to understand these patterns across characteristics including sex, gender, age, and disability, among other traits.

These disaggregated data could then be used for evidence-based decision-making that is responsive to people’s different needs. Science can assist, yet also needs assisting. Anyone seeking to disaggregate disaster deaths data for policy and practice would gain by cooperating with scientists while ensuring that scientists understand non-research needs as well as communicating to a wide range of people involved, notably disaster-affected people.

In terms of involving scientists more, we offer the tools and techniques. Our job is often to understand data, design, sampling, collection, recording, and then analysis and interpretation—especially regarding limitations, sources of error, and areas of improvement. We typically have competent and eager students willing to contribute as part of the mandates of their university degrees. Never hesitate to ask us and to involve us in the work you need, especially since we are not prone to arguing against more research!

Our own limitation, though, can frequently be communication. We do not always express ourselves succinctly and diplomatically enough for policy and decision makers who continually suffer from limited time and requiring a direct way forward.

We also have intense bureaucracy to navigate, some of which is frustrating—such as internal systems for submitting contracts for review and approval—and some of which is legitimate such as ethics. After all, when dealing with disaster-affected people, no work (research or otherwise) should contribute to disaster problems. Who would approach a bereaved parent and ask, “Um, I am curious, if you do not mind, with what gender did your dead child identify and did they have disabilities?”

Yet if done carefully and respectfully, in many instances such questions ought to be asked and are welcomed. Relatives and friends may wish to tell the stories of their loved ones, seeking it as part of grieving, remembering, and memorialising. It is not always the case, making it tricky to navigate, especially since only some scientists are trained therapists, social workers, or medical practitioners. The baseline is nonetheless that all research, including on disaster fatalities, takes time and must be completed ethically.

Disaggregating disaster deaths data does not need to be completed directly at a disaster site with the people who are immediately impacted. Indirect and remote approaches can be successful where the data are available, publicly or formally acquired, for instance through anonymised databases or death certificates.

The processes of obtaining and understanding disaster deaths data for disaggregation reveal their own research pathways! We identified thirteen factors inhibiting not just disaggregated data collection, but merely consistent and verifiable data collection on who dies in disasters in order to determine the what and the why.

We need to develop techniques to overcome or bypass these difficulties in order to fully support evidence-based decision-making which addresses disasters in a gender-responsive (among others) manner. We can, at least, identify, analyse, and attempt to resolve the difficulties in the data, also entailing open and honest communication regarding what can and cannot be said about the numbers. If a scientist states that the data are perfect giving an indisputable answer, then consider finding another scientist.

More commonly, we are too hesitant, using “might”, “could be”, “likely”, and “a possible trend under some particular circumstances which need to be investigated more thoroughly”. In comparison, those using the data might just need an answer, preferably yesterday. Working together by taking the time to understand each other’s needs, to articulate what we each do and do not seek, and to collaborate for ways forward within what we can and cannot achieve, helps us all.

It especially helps those affected by disasters, and trying to prevent them, to involve everyone irrespective of sex, gender, age, disability, and other traits.

About the Author:

Ilan Kelman http://www.ilankelman.org and Twitter/Instagram @ILANKELMAN is Professor of Disasters and Health at University College London, England and a Professor II at the University of Agder, Kristiansand, Norway. His overall research interest is linking disasters and health, including the integration of climate change into disaster research and health research. That covers three main areas: (i) disaster diplomacy and health diplomacy http://www.disasterdiplomacy.org; (ii) island sustainability involving safe and healthy communities in isolated locations http://www.islandvulnerability.org; and (iii) risk education for health and disasters http://www.riskred.org

Risk Gender data