It's not what you might think.
From drought-linked hunger crises in East Africa to floods in South Asia and hurricanes in the United States, disasters have repeatedly hit people all over the world in the past year.
About 1.5 million people have died in disasters in the last 20 years, and crises such as hurricanes and severe droughts cost $30 billion a year across 77 of the poorest countries in the world, according to Britain's Department for International Development (DFID).
Learning from such disasters - including crises such as pandemics and earthquakes - will be as crucial as how we respond to them if we are to prevent future loss of life, experts told a London event hosted by the Overseas Development Institute (ODI) on Thursday.
According to Jennifer Leaning, director of Harvard University's FXB Center for Health and Human Rights, "in the past 20 years we've learned a great deal about pandemics, for example, but not enough about how to build a robust enough system to respond to them."
During West Africa's Ebola outbreak, which started in 2014, for example, better use of technology could have contained the disease, she believes.
"Mapping people's whereabouts with mobile phone data, for example, could have allowed health workers to track social contacts through which the disease is spread," she said.
Similar geospatial data can also help direct search and rescue efforts in case of an earthquake, or in issuing early warnings to vulnerable people, she added.
WHAT PEOPLE NEED
Knowledge and research on disasters needs to be organised in a way that is attuned to affected communities, panelists said at the event.
Hugo Slim, head of policy and humanitarian diplomacy at the International Committee of the Red Cross, said that "rather than working from habit we need more empirical research about people's lives - what they experience and what they need".
Resources like the Resilience Exchange, an online report published this week by the DFID-funded Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED) programme, seek to gather "stories of change" on how vulnerable populations are becoming more resilient to climate shocks and stresses.
"We - the disaster community - need a detox from endless fascination with our own system," he added.
Yasemin Asyan, a consultant, agreed. "The humanitarian community tends to be inward-looking. We should engage more with other sectors like development to address populations' longer-term needs," she said.
Experts warned against solely blaming nature for disasters. Often shocks - such as from earthquakes - could be prevented from becoming disasters by ensuring housing is resilient, for instance, or that the right people are included in decision making.
Ilan Kelman, a disaster reduction expert from University College London, said that "language is powerful, and calling a disaster 'natural' risks deterring people and governments from action".
Those that respond to disasters need a deeper understanding of how communities recover from an extreme event to better target their interventions, experts said.
Charles Parrack, an architecture expert from Oxford Brookes University, said that about 80 percent of people around the world rebuild their homes themselves after a disaster rather than relying on aid.
"Perhaps that's where we should be focusing our efforts" to make recovery efforts more effective, he said.
Elenia Fiddian-Qasmiyeh, co-director of the migration research unit at University College London, said that assistance from developing world organisations is often overlooked by international aid agencies during emergencies "because they don't think their response will be fast enough."
But it's worth remembering that some agencies, beyond the usual developing-world actors, are increasingly working hard to cope with disasters, she said.
"The Organisation of Islamic Cooperation and others, for example, have greater access to Myanmar" where minority Muslim Rohingya people are fleeing a military offensive, she said. That means "the international community needs to explore different modes of response".
Any views expressed in this article are those of the author and not of Thomson Reuters Foundation.