A Natural Disaster's Imprint, 20 Years Later

In the aftermath of the 2004 Boxing Day Tsunami, which killed more than 227,000 people across 14 countries in the Indian Ocean, a research team faced a humbling challenge. What questions do you ask survivors that will shed light on the impact of the disaster years or even decades later?

“We had to begin with those long-term research goals in mind,” says Duncan Thomas, Ph.D., the Norb F. Schaefer Distinguished Professor of International Studies at Duke and an affiliate of the Duke Global Health Institute.

Thomas and Elizabeth Frankenberg, Ph.D., the Cary C. Boshamer Distinguished Professor of Sociology at the University of North Carolina-Chapel Hill, have conducted longitudinal research studies in Indonesia since the 1990s, long before the tsunami struck. In the early 2000s, they had been studying the economic impacts of the 2002 bombings on Bali, one of the islands in the Indian Ocean that forms the country.

But then came the tsunami, which on Dec. 26, 2004, struck the Indonesian island of Sumatra with waves reaching more than 100 feet high. Soon after, Frankenberg, Thomas and collaborators at the Indonesia-based nonprofit SurveyMETER began to discuss developing a project that would shed light on the immediate impacts of the tsunami on the health and economic wellbeing of survivors and, eventually, how those impacts would evolve over the long-term.

Working with Dr. Cecep Sumantri of SurveyMETER, they decided to focus on Aceh and North Sumatra, the two westernmost provinces of Indonesia, which had borne the brunt of the tsunami.

“We thought, how can we possibly do this?” says Thomas. “We all knew that it was a very high-risk project that was not going to be easy, but because it was important for the country and the people of Aceh and Indonesia, and because it was important for science, we should give it a crack, knowing full well that the chances were that it would be a dismal failure.”

Twenty years later, they are still there asking questions. The answers are painting an important picture for understanding the long-term effects of natural disasters, including trends in fertility and mortality, strategies for reconstructing livelihoods, risk aversion, and the evolution of psycho-social and physical health. The results have helped researchers to target who might need the most attention after future tragedies, and how to help them through challenges that may continue to emerge years, or even decades, later.

Less than a month after the tsunami struck, the research team leapt into action, applying for several large research grants. When they started putting together their plans, they had no idea if the money would come through, or how. To their surprise, it did -- from the World Bank, the National Science Foundation, the National Institutes of Health and others. It was enough to recruit and train 150 people to find and interview survivors and to establish what would become known as The Study of the Tsunami Aftermath and Recovery (STAR) program.

In the summer of 2005, Frankenberg traveled to Aceh to survey the damage and meet with the team. She had first visited Indonesia as an undergraduate student at UNC-Chapel Hill in the 1980s, and had returned regularly throughout her career. But on this trip, Aceh looked totally different. Many buildings had been completely destroyed or badly damaged, with piles of debris everywhere. Entire communities were washed away.

“It was so disturbing to see the immense destructive force that these communities had been subjected to,” she says. “The tsunami just changed so much. Nobody in the areas where the water came ashore was operating as though anything was normal.”

In a striking illustration of the tsunami’s power, waves carried a large ship, called Apung 1, three miles inland. It remains there today and now hosts a museum about the disaster. It’s an image that sticks with Thomas. “The Achenese do a very good job of not trying to scrape away the horror, but rather have it stand as a memorial to how successful the people have been in rebuilding their lives,” he says.

The researchers’ first goal was to understand the demographics of the tsunami survivors, including their age, sex, occupation, education, and where they were living, says Frankenberg. From there, they could develop questions to determine how their lives had changed through economic, sociological, behavioral and health-focused lenses.

The situation was unique in that, 10 months before the tsunami, more than 30,000 people living in Aceh had participated in a broad survey conducted by the country’s national statistical agency. The research team started with those questions so that they could compare change over time.

“A lot of the questions that we would ask would have to be consistent with the questions that had been asked pre-tsunami to measure change,” says Thomas. “We also had to develop new questions that  were relevant to the tsunami survivors about how their lives and livelihoods had changed.”

Collaborating with Statistics Indonesia, the team interviewed 28,000 people over the course of 14 months. They followed up with the same respondents annually for four more years and again at 10 and 15 years after the tsunami. Because family is central to well-being, they collected information on all household members and interviewed everyone who was at least 15 at the time of the tsunami.

Locating the respondents was an extraordinary challenge with all the displacement that had occurred, says Thomas, and Sumantri and his team have re-interviewed over 98 percent of the tsunami survivors at least once – which Thomas says is simply exceptional. Interviewers have tracked people all over Aceh and North Sumatra as well as to other provinces across Indonesia.

Over time, the researchers have developed an expansive dataset that they can analyze to better understand trends across many areas, including mortality, family disruption and relocation, fertility and community re-population,  social and economic well-being, and cognition and cognitive aging. They also evaluated physical and psycho-social health, including biological measures of cardio-metabolic risks.

Since their expertise and training lie in economics, demography and sociology, Frankenberg, Sumantri and Thomas have collaborated with experts in the health and cognitive sciences to ensure they are doing the highest quality science. This includes evaluating post-traumatic stress reactivity (PTSR) rather than post-traumatic stress disorder (PTSD), which requires a clinical assessment. The team has taken hair samples to measure cortisol, a hormone that helps regulate the body’s response to stress, and used novel approaches to assess memory and cognition.

As the survivors have navigated the new post-tsunami landscape, the questionnaires have been adapted to reflect that reality. But much of what the researchers have been able to pinpoint about the population came from their foresight 20 years ago.

“All of the research we've done, all of the pieces, was basically in mind when we designed the first follow-up survey,” says Thomas.

A massive data trove collected over 20 years has shed light on how the people of Aceh suffered and how they have rebuilt their lives, their families and their livelihoods, providing important insights into the long-term impacts of natural disasters.

Because the tsunami’s devastating effects were essentially random, depending on the location of the earthquake that caused the tsunami and variations in the topography of the sea floor and coastline, researchers  compared coastal communities that sustained different levels of damage to measure the impacts of tsunami exposure. The first post-tsunami survey revealed that physical strength was a key factor in survival. It also showed that the mortality rate in the hardest hit communities, which soared at the time of the tsunami, had dipped lower than communities that were not directly affected fifteen years later, indicating a robustness among the surviving population. The fertility rate was higher in communities that sustained the greatest damage, which the researchers say was attributable to young women having children earlier in hopes of rebuilding their communities. .

The researchers have also uncovered interesting social and economic trends among survivors. In the near term after the tsunami, for example, survivors were more willing to take economic risks such as starting a business, and they launched new ventures at unprecedented rates. But even 15 years after the disaster, they still lagged economically behind peers living in areas not affected by the tsunami. The willingness to take economic risks was also temporary, returning to pre-tsunami levels after a few years.

Some of the most interesting findings have been on the long-term impacts of stress. In the immediate tsunami aftermath, the researchers not surprisingly found high levels of PTSR among individuals from directly affected communities, but people in adjacent communities also experienced PTSR, suggesting that counseling services – which were used by very few survivors - would likely have been impactful across the entire region.

Their longer-term work shows that stress in some cases has not gone away. For example, female tsunami survivors still showed signs of extreme stress in the form of cortisol burnout fourteen years after the event. When hair samples were taken from 615 survivors over the age of 45, women exposed to the tsunami’s direct impacts had levels of hair cortisol thirty percent lower than those spared by the tsunami, says Thomas, and those with the lowest levels of cortisol had presented with the worst PTSR symptoms that persisted for several years post-tsunami.

Extremely low levels of cortisol occur when the body’s stress response system, known as the hypothalamic-pituitary-adrenal (HPA) axis, is flooded with so much cortisol for so long that it essentially burns out, similar to the way an amp can be blown out by too much voltage.

People living with HPA axis burnout may feel lethargic and less resilient to daily stressors, according to Ralph Lawton ’20, who started working with Thomas and Frankenberg on the study when he was an undergraduate student at Duke. They are also at increased risk for other health conditions, including immune system dysfunction, mental health disorders, and cardiovascular and metabolic diseases.

“While there has been remarkable resilience among survivors of the tsunami, the fact that the HPA-axis remains different after such a long time is surprising,” says Lawton, who is now a medical student at Harvard University jointly pursuing a Ph.D. in health policy and economics. “There is a lot of emphasis on stress during critical periods of childhood, but these long-term findings among people who were exposed as adults are noteworthy.”

This is just a sampling of published research findings from the longitudinal study; for a full list, see the STAR website.

Now almost 20 years after the study began, the researchers have no intention of stopping. When asked why he devoted so much of his career to studying the survivors of the Boxing Day Tsunami, Thomas has a counter-question: Who wouldn’t?

“The value of the study just grows,” says Thomas. “Survivors who were impacted in their thirties and forties are now in their fifties and sixties. We want to know how they are doing as they’re getting older.”

Frankenberg, Lawton, Sumantri and Thomas have studies underway to measure survivors’ cognition and memory through their ability to recall certain smells. Animal studies have shown that extreme stress is connected to permanent deficits in the function of the hippocampus, the region of the brain that controls memory, says Thomas. Working with Margaret Sheridan, at UNC, the researchers are measuring the long-term impacts of exposure to the stresses of the tsunami on multiple domains of cognitive function, and whether those exposed to stress are more prone to develop Alzheimer’s disease and related dementias, as they get older.

Because the study follows all household members, the researchers are also investigating the impacts of exposure to the tsunami at an early age on health and well-being over the long term. Fifteen years after the tsunami, they documented high rates of hypertension among those who were age 10 or younger at the time of the disaster, particularly among the exposed. This likely portends high levels of cardiovascular disease as these people age, Thomas notes.

“Now those kids who were five are 25, so they're starting their adult lives, and we need to better understand what the consequences are for these people,” says Thomas. “We should be trying to follow them for as long as we can while being respectful of their time, of course, so that we can know what to expect from other disasters, including those brought on by climate change.”

Rising global temperatures will make extreme weather events more likely, according to the U.S. Environmental Protection Agency. “Learning about these long-term effects is  critical because climate change and other natural disasters are likely to leave fingerprints that last for a very long time,” says Thomas. “It’s important to think about what can be done early on to mitigate persistent effects on health and well-being, rather than wait for 20 years to find out.”