Disasters and Public Health
Before beginning today’s note, an acknowledgement of the recent tragedy in Toronto. Last Monday, 10 people were killed, and more than a dozen were injured, when a man drove a van into pedestrians. While the investigation into what caused him to allegedly commit this crime is still ongoing, he appears to have been motivated by hatred against women. I have written previously about the public health consequences of hate; sadly, there has been ample cause to revisit these thoughts in recent years, from the mass shootings we have regularly seen, to bombings like last Sunday’s attack in Kabul, to the hate that has infused much of our political discourse, both in the US and abroad. In this context, it is all the more important that we continue working to reduce hate and promote the values of empathy and community that are the basis for a healthy world.
Just as we have seen health undermined by acts of hate in recent years, we have also seen it threatened by natural disasters, including last summer’s string of hurricanes. Tomorrow, we will welcome Anne Schuchat, principal deputy director of the Centers for Disease Control and Prevention, for a Public Health Forum titled “Crises, Calamities, and Chaos: How Public Health Can Improve Response to Emerging Threats Wherever They Arise.” At the event, she will address how we can best face acute public health challenges such as disasters, opioids, and infectious threats like Zika and Ebola. In advance of her talk, we today rerun a modified version of a Dean’s Note on disasters and the health of populations, to inform thinking about how we can better respond to such crises.
Disasters are unexpected, large-scale events that disrupt communities and cause death, destruction, and trauma. The number of disasters worldwide—both natural and human-made—is increasing, principally as a result of global environmental climate change and urbanization. In 2012 there were almost 360 natural disasters registered worldwide. China is the country most frequently affected by disasters, followed by the United States, the Philippines, India, and Indonesia. In the US, an estimated 13 percent to 19 percent of adults have reported having experienced a disaster in their lifetime.
Between 2003 and 2012, natural disasters killed an average of 106,654 people each year. Examples of some of the major storms in the past several years include the tsunami in Japan in 2011, which killed more than 15,000 and left more than 2,500 people missing; the 2010 Haiti earthquake, which had a death toll of more than 230,000; the 2008 earthquake in China’s Sichuan province, which killed about 70,000 with more than 18,000 missing; and the 2004 Indian Ocean tsunami, which killed about 230,000, displaced 1.7 million people, and injured more than half a million. Closer to home, in the last 20 years alone we have witnessed Hurricanes Katrina, Ike, Sandy, Harvey, and Maria, as well as droughts, floods, and wildfires—events that have resulted in loss of life and billions of dollars in damage. Aside from natural disasters, mass trauma can also occur from terrorist attacks, such as the September 11 attacks, or technological disasters.
The burden of disasters includes lost life, property loss, infrastructural damage, monetary loss, years lost to disability, interruption of services, and damage to individual and population health. Physical injury and death after a disaster tend to be immediate, happening within minutes of the event onset. They also tend to be very much the “tip of the iceberg”—a small fraction of the health burden following disasters that principally involves mental health, which can be both long-term and debilitating. By way of illustration, in reviewing the PTSD literature with colleagues, we previously found that 30 percent to 60 percent of direct victims of disasters experience PTSD. This prevalence was smaller among rescue workers (10 percent to 15 percent) and the general public (about 5 percent to 10 percent). PTSD prevalence varies greatly between studies due to different scales of disasters, degree of exposure, timing, and methods. Many studies have found PTSD symptoms to be associated with younger age, female gender, and a history of mental illness.
The risk of depression and substance use is also increased after disasters. Alcohol and substance use disorders are more prevalent among men and are often used as a coping mechanism. Many studies found increased use of alcohol, cigarettes, and drugs in New York City after the September 11 terrorist attacks. A study of Hurricane Sandy survivors in New Jersey found that high hurricane exposure, physical health limitations, and environmental health concerns were all associated with worse mental health outcomes. Other risk factors found to be associated with post-hurricane health include ongoing stressors, lower social support, and financial loss.
It is worth emphasizing that the majority of people who are faced with disaster tend to be resilient, or continue normal functioning relatively quickly after experiencing a traumatic event and potentially initial mental health symptoms. Longitudinal work allows us to see this phenomenon over time; for example, Pietrzak and colleagues followed Hurricane Ike survivors at three time points after the storm and found that the prevalence of any past-month mental disorder, and specifically hurricane-related PTSD, decreased over time. Most studies suggest that the course of mental health after these events is different for different groups of people, with a majority having few symptoms of mental disorder at any time point, and others demonstrating rapid resolution of symptoms, ongoing moderate symptoms, ongoing substantial burden of psychopathology, and, occasionally, increases in psychopathology over time. For example, a study of children affected by the 2008 Sichuan earthquake found that the proportion of participants who used mental health services dropped substantially between follow-up periods, although PTSD and depression prevalence remained stable.
It is becoming increasingly clear that context plays a role in shaping the health of populations after disasters. Community-wide destruction has been shown to worsen PTSD in post-tsunami Indonesia, even when considering a range of individual exposures and loss. On the other hand, community social capital can promote resilience in individuals. Many post-disaster events can also affect psychopathology. For example, displacement and lack of order in a community increases the likelihood of violence, especially violence toward women and children. At a more macro level, disasters affecting places with lower incomes and those with more overall destruction are associated with worse health outcomes. This was particularly the case after Hurricane Katrina, where many of those affected were economically vulnerable. There has also been some evidence that human-made disasters and mass violence may have a bigger psychological impact on survivors compared to natural or technological disasters, although this finding is debatable, and complicated by a paucity of longitudinal data in disaster studies.
There is also much interest in population behavior after disasters—that is, what we can expect to be population behavior in the aftermath of these events? Although after Hurricane Katrina there was significant media attention on extreme behavior prompted by unprecedented circumstances, in our work we articulated a model of population behavior following large-scale disasters—based on a dataset of 339 disasters from 1950 to 2005, and grounded in theory—suggesting replicable and consistent patterns of behavior after large-scale events. This model posits five overlapping behavioral stages (see figure below). Stage 1, group preservation, involves directly affected people acting to preserve life and secure safety, which involves both information-seeking (from friends, employers, media, government etc.) and action (targeting the source of the hazard, evacuating etc.). Stage 2, population preservation, is similar to Stage 1, but happens among the larger population, not just those directly affected by the hazard. This can include disseminating information to people who may be at risk of experiencing a hazard’s effect, and leaders taking charge to determine what action should be taken. Also included in this stage are volunteers and formal response agencies assisting or donating to the directly affected group.
Stage 3 begins after the initial danger of the hazard has passed, and involves the mourning of loss of people and property, memorialization, recognizing a new set of norms, and creating a narrative. Stage 4, externalizing, involves seeking redress and addressing vulnerabilities. Formal investigations or criminal charges are examples of ways a society might seek redress after a terrorist attack. We have also seen externalization in terms of anger and criticism of government preparedness after a natural disaster. Addressing vulnerabilities can include plans for preventive measures to avoid similar effects of future disasters. The final stage, renormalization, includes cultural adaption to post-disaster circumstances, normalization of vulnerability, and new modes of behavior. This phase ends when these new post-disaster modes of behavior become dominant. For example, this could be when new technology is adopted or new security policies become regular.
Each of these stages can affect health; for example, volunteerism as part of Stage 2 may improve the mental health of those outside the directly affected circle who volunteer. There are also other suggested models of behavior after disasters, including those that focus on community response, as well as evidence-based post-disaster policy recommendations.
One final, perhaps more personal, note about disasters and their consequences. I have had the privilege of being involved in scholarship about the consequences of disasters for the past 17 years. I have learned much from this work, and hope that some of our work has made a contribution to our understanding of these events and how we can mitigate their consequences. At core, though, I have repeatedly impressed that disasters do happen, and will continue to happen, and that the difference between disasters with devastating consequences—such as Hurricane Katrina—and those with much milder consequences—such as Hurricane Sandy—is frequently the fundamental underlying conditions that produce health, i.e. the fundamental salutary conditions that promote population health about which I have written frequently in Dean’s Notes. This suggests that our work towards creating healthier societies and healthier populations is probably the most important work we can do together to mitigate the consequences of disasters. When disasters happen, they are evocative and compel much attention and an urgent—and very human—desire to help. Unfortunately, these events capture our attention fleetingly, before the news cycle moves on. It is, sadly, during that period that much of the damage wrought by disasters comes about, when our attention has moved on. After a disaster has occurred, I am frequently asked what one can do to help from a distance. It is a difficult question, but I often recommend providing support to organizations that are committed to helping societies rebuild in the medium- and long-term aftermath of these events, and that are committed to creating the conditions that make people healthy at all times.
I hope everyone has a terrific week. Until next week.
Sandro Galea, MD, DrPH
Dean and Professor, Boston University School of Public Health
Acknowledgement: I am grateful for the contributions of Laura Sampson to this Dean’s Note.
Previous Dean’s Notes are archived at: https://www.bu.edu/sph/category/news/deans-notes/