Megatrends met migration patterns as we launched the all-new PopTalks conversation series, hosted by Columbia’s own Thoai Ngo and featuring Brian Thiede of the Population Research Institute (PRI) at Penn State. On February 17, 2025, nearly 130 in-person and virtual attendees gathered for a spirited discussion. The focus: Dr. Thiede’s recent work on the nuanced—and sometimes contradictory—health and demographic outcomes of climate shocks.
“We face a rapidly changing world,” said Dr. Ngo in his opening remarks. “We need new frameworks and new thinking to confront climate crisis, aging societies, and the needs of the rising generation. Most of all we need new solutions, and that means changing how we do research on population health.”
The PopTalks series aims to engage just those sorts of challenges, and this kickoff event wasted no time diving into the big questions.
“Big Microdata” Captures Nuance and Exposes Research Gaps
Our most reliable research methods have proven less than adequate for grappling with overlaps between climate, health, and migration. How do you nail down the effects of extreme weather events—droughts, wildfires, and heavy rainfall—on millions of people? Researchers like Dr. Thiede start by mapping out correlations in data on climate and health.
His recent work with collaborators at PRI showcases both the value of “big microdata”—massive collections of individual-level data from surveys and censuses—and the need for interdisciplinary creativity to help us find solutions. His talk highlighted three recent examples.
1. Who’s most at risk — migrants, or those who stay behind?
The PRI team built a dataset tracking migration for some 55 million people in six Asian countries, all between 15 and 49—the prime age for migration. Then they modeled migration to other provinces as a function of unusual spikes in heat and rainfall(link is external and opens in a new window).
Previous research(link is external and opens in a new window) supports the idea that people tend to migrate away from drought-stricken areas, but Dr. Thiede’s team found the opposite: across all six countries, migration went down after sharp drops in rainfall. But when they zoomed in, the drop in migration was specific to China. In the Philippines, drought increased migration. In Indonesia, Malaysia, Nepal, and Vietnam, it had no effect.
“These results complicate common narratives around displacement and ‘climate refugees,’” Dr. Thiede explained. “We usually frame this issue in terms of national security or anti-immigration policies, but what we see here is that and households adapt in ways that can be hard to predict, and vulnerability is distributed in complex patterns. Sometimes, the people most at risk are the ones who aren’t able to leave despite the danger of drought or floods.”
2. Heat shocks and malnutrition: A hidden survivorship bias
Using Demographic and Health Surveys (DHS) data from a 36-year period, Dr. Thiede and co-author Clark Gray dug into the connection between climate stress and nutrition(link is external and opens in a new window). They modeled both short- and long-term malnutrition as a function of neighborhood-level exposure to heat and precipitation shocks.
What they found surprised them: results were different for infants under two years old than for children aged three to five. The infants were less well nourished, on average, if their neighborhood or village had higher temperatures. For older children, the reverse was true.
Untangling the mystery required a second set of models that uncovered a strong survivorship bias. Heat shocks correlated with higher mortality at all ages, meaning that only the healthiest children were reaching age three and showing up in the long-term data.
“We need to be open to counterintuitive findings and pay close attention to possible biases,” said Dr. Thiede. “Population level health indicators are intertwined with mortality, migration, and fertility.”
3. Complex causality in flooding and fertility
Working with an interdisciplinary team, Dr. Thiede explored the effects of climate-related flooding on the fertility of women in Bangladesh(link is external and opens in a new window). They found patterns that were clear, but complex. When more of the land around a village flooded during monsoon season, birth rates went up among women who already had two or more children, or had less than a primary school education. At the same time, fertility declined among women who had never given birth before and women who lived in cities.
That kind of pattern was a key theme of the event. To understand how the stresses of climate change lead to particular health outcomes or demographic shifts, we need to think about how climate affects everyday realities. How does flooding alter the decision to have children? Does it change the availability of contraception, or damage the local economy?
Effective interventions—policies that really build resilience to climate shocks—target those kinds of practical, mundane issues. And to design, study, and implement solutions that work, we must train students of global health to understand and assess them.
“Big, Crazy Ideas” and Bold Responses
Dr. Thiede closed his discussion with a call to action. Engineers have proposed wild, hugely ambitious solutions to rising sea levels, like buttressing ice shelves or draining the beds of glaciers(link is external and opens in a new window). Sociologists and demographers, he said, need to do the same. We bear a responsibility to think in bold, big-picture terms about responses to the climate crisis.
The session ended with an engaging back-and-forth about finding solutions and the future of the field. In response to questions from Dr. Ngo and audience members, Dr. Thiede offered a list of practical implications and take-home ideas based on his work.
Key Takeaways:
- We may have reached the limits of DHS data; global health and population science need a major investment in high-frequency data collection.
- New research needs to focus more directly on solutions, testing scalable interventions and moving beyond descriptive methods.
- “Climate is never the only stressor”—and our research needs to grapple with the mutually aggravating effects of armed conflict, social instability, and economic pressures.
- We must change how we teach. The next generation of population scholars should be better prepared for the distinctively interdisciplinary complexities the field now faces.
To hear more about exciting new research, keep an eye out for upcoming PopTalks events by signing up here.