Hurricane Katrina survivors struggle with mental health years later

The study's lead author, Christina Paxson of Princeton University, said that the results were a departure from other surveys both in the design and the results. The researchers were able to collect data on the participants before Katrina and nearly five years after the August 2005 storm, finding a persistence of poor mental health and gaining insights into how different types of hurricane-related stressors affect mental health.

"On average, people were not back to baseline mental health and they were showing pretty high levels of post-traumatic stress symptoms. There aren't many studies that trace people for this long, but the very few that there are suggest faster recovery than what we're finding here," said Paxson, who is Princeton's Hughes-Rogers Professor of Economics and Public Affairs and dean of the Woodrow Wilson School of Public and International Affairs. "I think the lesson for treatment of mental health conditions is don't think it's over after a year. It isn't."

In addition to helping mental health professionals aid survivors of Hurricane Katrina and other disasters, the research may guide policymakers in addressing areas that had a significant impact on the women in the study, such as home damage and rebuilding.

The paper appears in the January issue of the journal Social Science and Medicine. Paxson's co-authors were Elizabeth Fussell, an associate professor of sociology at Washington State University; Jean Rhodes, a professor of psychology at the University of Massachusetts Boston; and Mary Waters, the M.E. Zukerman Professor of Sociology at Harvard University.

The project began in 2003 as a study of low-income adults enrolled in community college around the country, including three sites in New Orleans. The initial questionnaire contained questions about education, income, families and health. After Hurricane Katrina, some of the researchers decided to try to continue to track the New Orleans-based participants.

"I realized right away that the kinds of information we had on mental and physical health were very rare in disaster studies," Waters said. "Researchers never know if people are suffering because of the disaster or if they had underlying conditions that would have led to depression or poor health even before the disaster hit."

The sample size in the study was made up of 532 low-income mothers, most of whom were African American and whose average age was 26. They were interviewed in two follow-up surveys -- tracked down largely through their unchanged cellphone numbers, though they were spread across 23 states -- about 11 months and nearly five years after the storm.

Due to the makeup of the sample, Paxson cautioned that the study's results cannot be assumed to apply to the population as a whole, but they shed light on natural disasters' effects on a particularly vulnerable group.

The surveys helped rate the women on two signs of poor mental health: psychological distress and post-traumatic stress symptoms (PTSS). Researchers measured psychological distress using a series of questions (also in the initial questionnaire) typically used to screen for anxiety and mood disorders, asking about feelings such as sadness, hopelessness and nervousness experienced over the last 30 days. They measured PTSS using a test used to identify individuals at a high risk of meeting the criteria for post-traumatic stress disorder; for example, the women in the study were asked how often they thought about the hurricane in the last seven days and whether they had thoughts about the storm that they could not suppress.

The researchers found that even after four years, about 33 percent of the participants still had PTSS, and 30 percent had psychological distress. Though levels for both conditions had declined from the first follow-up 11 months after the hurricane, they were not back to pre-hurricane levels.

The researchers had also interviewed the study participants about the types of stressors they had experienced during the storm: home damage, traumatic experiences the week of the hurricane (such as being in danger or lacking food, water or necessary medical care), or death of a friend or relative.

Paxson and her collaborators found that these stressors played a role in whether the participants suffered from psychological distress or PTSS, or both. For the most part, the hurricane stressors, especially home damage, were associated with the risk of chronic, long-term PTSS alone or in combination with psychological distress.

"I think Katrina might be different from a lot of natural disasters in the sense that it completely upended most people's lives," Paxson said. "About two-thirds of the sample is back in the New Orleans area, but almost nobody lives in their old home. So they're living in new communities. They've been disrupted from their friends and their families. The whole fabric of their lives has really been changed."

Demographer Narayan Sastry, a research professor in the Population Studies Center and Survey Research Center at the University of Michigan, said that the study makes an important contribution toward understanding the medium- to long-term effects of Katrina on mental health.

"The most significant aspect was its study of long-term outcomes that were assessed at multiple points in time, including prior to the hurricane -- a unique feature of this study," said Sastry, who is familiar with the study but was not involved in it. "The results are important not just for designing policies and programs to help address any ongoing mental health problems being experienced by survivors of Hurricane Katrina, but also in devising the best response to help people who are affected by natural disasters in the future."

Sastry added that the study found very similar levels of psychological distress as his work on a broader, representative sample of the pre-hurricane population of New Orleans.

Paxson and her collaborators plan to further examine the data from their surveys, and to continue tracking the women in the sample. Some avenues for further study include looking at how genetics may have a played a role in the mental health responses, examining the hurricane's effects on physical health, and tracking the educational and mental health outcomes of the children of the women in the sample. They also plan to publish a book that combines the survey results with in-depth interviews with some of the women.

This research was supported by the National Institutes of Health, the National Science Foundation, the MacArthur Foundation and Princeton's Griswold Center for Economic Policy Studies.