Displacement used to be described as an event. It is closer to a condition. Last year, storms, floods and other hazards triggered 29.9 million internal displacements, according to the Internal Displacement Monitoring Centre's 2026 report. That was a sharp fall from 2024's extreme figure and still 13 per cent above the average of the past decade. Some of those people went home within weeks. Others have not gone home at all.
Most coverage of climate mobility stops at the moment of leaving: the flooded street, the packed bus, the tent. What happens next is harder to photograph and matters more. It is mostly about water, medicine and whether anyone knows you are there.
WHO published its global research prioritisation and action plan on health, migration and displacement in the context of climate change this year, launching it at a webinar on 22 July. The plan is a map of what we don't know, drawn up by 59 specialists from academia, policy and practice. Reading it is a useful corrective to the assumption that the health consequences of climate displacement are already understood and merely unaddressed. They are not well understood at all.
The health effects of displacement are less exotic than the phrase suggests. People lose the clinic that held their records. A course of treatment stops halfway. Vaccination schedules slip. A pregnancy that was being monitored stops being monitored. Someone managing diabetes or hypertension loses the prescription and the pharmacy in the same week.
Then there is the accommodation. Communal shelters, informal settlements and camps concentrate people in places built for fewer of them, with drainage and waste systems that were adequate for nobody in particular. The European Environment Agency's assessment of flooding and health puts the mechanism plainly: people housed in temporary shelters face higher exposure to infectious pathogens in communal accommodation, on top of the disruption to their usual care. Sewage overflow during floods raises infection risk, particularly for children, and the organisms involved are named: norovirus, hepatitis A, rotavirus, Cryptosporidium, Giardia, Campylobacter.
That list is worth pausing on, because it is the same list whether the flood happened in Accra or in a European river basin. The pathogens do not consult the income statistics.
WHO's specialists ranked the gaps they identified by how much difference closing them would make, whether the research is feasible within five years, and whether the findings would actually reach policy. Under the first of the plan's themes, the item that came out top was the least fashionable one on the list: work out which social determinants matter most, with food security, nutrition, access to safe drinking water and sanitation, and psychological trauma given as the examples.
Water and a working toilet, in other words, before anything else.
UNHCR's 2025 report on water, sanitation and hygiene shows why. In Luakdong settlement in Ethiopia, built on low-lying flood-prone land with no drainage outlet, one of the heaviest rainy seasons in years pushed the river over its banks. Water points stopped working. Latrines collapsed as the soil gave way. A cholera outbreak followed, and a protective dyke was built afterwards. The sequence took weeks.
Slow-onset failure is quieter. In Bangladesh's Teknaf settlements, dependent on surface water, average supply fell from 33 to 23 litres per person per day between 2024and 2025. Drought forced rationing down to 10 litres, against a post-emergency minimum of 20.
The scale behind those examples is unglamorous and large. WHO's fact sheet on diarrhoeal disease still records it as the third leading cause of death in children under five, killing roughly443,832 a year across some 1.7 billion childhood episodes. Cholera is the visible edge of that; most of it is caused by organisms nobody ever identifies.
The plan does not treat mental health as a soft addition. It identifies the mental health needs of displaced people as a research gap in its own right, noting that displacement brings trauma, loss and instability, and that climate stressors such as extreme weather, lost livelihoods and broken social networks feed into depression, anxiety and post-traumatic stress. What it asks for is evidence on which psychosocial support actually works, and works in a way that fits the culture it is offered in.
It also flags something rarely discussed in the same breath: the risk that displacement raises exposure to gender-based violence, including child marriage and sexual violence in post-disaster settings. That is listed as a gap because the research to size it has not been done.
The plan's most repeated complaint is administrative rather than clinical. Displaced and migrant people are often missing from the national systems that generate health statistics, particularly when undocumented or in transit. If you are not in the register, you are not in the plan, the budget or the emergency response.
That is the practical route from research to fairness, and it is more mundane than a declaration of rights. It means agreeing what the terms mean, since there is still no settled definition of a climate migrant. It means indicators built into national health information systems that can see mobile populations at all. It means integrating displaced people into primary care and emergency preparedness rather than running a parallel service that ends when the funding does.
It also means a safeguard the plan is careful to name. Health data on undocumented people becomes a surveillance risk the moment an immigration authority can read it. Any system built without that separation will teach people to avoid the clinic, which produces both worse data and worse health.
Santino Severoni, who heads WHO's health and migration work, said at the launch that "Closing research gaps is a matter of equity”. The version of that sentence which survives contact with a flooded settlement is narrower and harder. Somebody has to count, somebody has to pay for the latrine, and somebody has to make sure the counting cannot be used against the person counted.
Sources