Climate Change and Health Seminar: “Mental Health after Weather-Related Disasters: State of the Research and Future Directions”

Transcript

• 00:00- Hi all and welcome to the
• 00:03Yale Center on Climate Change and Health seminar series.
• 00:07So today is our first spring seminar series
• 00:11and we are very fortunate to have
• 00:14Dr. Sarah Lowe joining us today.
• 00:17So Dr. Sarah Lowe is assistant professor
• 00:20at the Yale School of Public Health
• 00:22the Department of Social and Behavioral Sciences.
• 00:26So her talk today will be mental health
• 00:28after natural disasters,
• 00:30state of the research and a future directions.
• 00:33So I was told that this seminar
• 00:36was one of the most popular seminar series we had.
• 00:41There were more than 80 participants registered
• 00:44and we have another roughly 10 students.
• 00:47So hopefully we can have a large audience today.
• 00:52And before handing over to Sarah,
• 00:56I want to mention that we will have our Q&A section
• 01:01at the end of this seminar.
• 01:04So if you have any questions please type
• 01:07in the chat box and I will raise the questions in the end.
• 01:12So without further ado, Sarah the stages is yours.
• 01:17- All right, thank you very much for that Kai
• 01:20for that nice introduction.
• 01:21I'm going to share my screen and get to it.
• 01:26Okay, so you can see my slides, yes?
• 01:29- [Kai] Yes, yes.
• 01:30- Okay, awesome.
• 01:32So as Kai said, I'm going to be talking about
• 01:35the research on mental health
• 01:36after weather related disasters,
• 01:39the state of the research and future directions.
• 01:41And you'll know what that I actually changed the name
• 01:43of my talk because the field is really moving away
• 01:46from referring to weather related
• 01:50or climate related disasters as natural disasters
• 01:53and acknowledgement of increasing findings showing that
• 01:59human beings are contributing to climate change
• 02:01and in turn increasing the frequency
• 02:03and severity of these types of events.
• 02:05And also that these disasters affects human-made structures
• 02:09and systems and people.
• 02:10So it's really an interaction
• 02:12between the environment and humanity.
• 02:16So I wanted to start off by giving...
• 02:19Hold on just a second.
• 02:20An overview of my talk today.
• 02:22I'm going to be first introducing myself
• 02:25and discussing my program of research.
• 02:27Then talking about the state of the literature
• 02:30on mental health after disasters,
• 02:32as well as some of the limitations they're in.
• 02:35And then give some examples of recent trends
• 02:37in the literature.
• 02:39I'm gonna end by discussing some of my current
• 02:41and hopefully future work.
• 02:45So starting off with my program of research.
• 02:47So I am a clinical psychologist by training.
• 02:51I received my doctorate
• 02:52at the University of Massachusetts Boston
• 02:55which has I think, unprecedented attention
• 02:58to social justice and multiculturalism.
• 03:02After getting my PhD, I did a post-doctoral fellowship
• 03:05in psych Epi at Columbia Mailman School of Public Health.
• 03:09And I stayed on there for a year
• 03:10as an associate research scientist.
• 03:12And that's where I really caught the public health bug
• 03:15and discovered that this would be a good home for me.
• 03:18I then actually spent four years
• 03:19in the department of psychology
• 03:20at Montclair State university in New Jersey
• 03:23before coming to Yale.
• 03:24This is my second year at the school of public health.
• 03:27And I've had a really great experience so far
• 03:30and I'm happy to be here today and to be affiliated
• 03:32with the center for climate change and health.
• 03:36My research program focuses on the long-term impacts
• 03:39of a range of potentially traumatic events.
• 03:42So much of it has focused on climate related
• 03:45and weather related disasters.
• 03:47But I've also been involved in research projects
• 03:49after the deep water horizon oil spill,
• 03:51projects focusing on the impact of gun violence,
• 03:53sexual assaults, child maltreatment, and community violence.
• 03:58Work-related potentially trauma exposures
• 04:01among first responders,
• 04:03the impact of discrimination
• 04:05on the mental health and Muslim youth.
• 04:07And most recently I've been involved in studies
• 04:10of the intergenerational impact of the 1994 genocide
• 04:14against the Tutsi in Rwanda
• 04:16and the impact of the COVID-19 pandemic on vulnerable groups
• 04:19including healthcare workers and persons with disabilities.
• 04:22And I'd be happy to talk
• 04:23about any of this research in the Q&A.
• 04:26So that is me in a nutshell, and I'm gonna move on
• 04:29to discussing some of the work on mental health
• 04:31after disasters and giving an overview of the literature.
• 04:36So in 2018, my colleagues and I were asked
• 04:40to do a review of a year of research
• 04:43on the mental health impact of environmental disasters.
• 04:46So climate related disasters,
• 04:47as well as disasters like oil spills and nuclear explosions.
• 04:52And when agreeing to do this I thought back
• 04:55to the Seminole Review by Fran Norris
• 04:57and colleagues in 2002, that reviewed all of the literature
• 05:00at the time on the psychosocial impacts of disaster.
• 05:04And that review had included a total of 160 papers
• 05:10on mental health and disaster.
• 05:12So I said to myself this is one year
• 05:14it's probably gonna be less than that,
• 05:16I can definitely handle that.
• 05:18But then when my colleagues and I looked at the literature
• 05:20in that single year, we found an enormous number.
• 05:23We actually ended up narrowing our inclusion criteria
• 05:26to those focused on PTSD and depression
• 05:29as our two key outcomes
• 05:30and only including quantitative research
• 05:32just to manage our workload.
• 05:35So I think this reflects the burgeoning interest
• 05:38in mental health after disasters, which is very exciting.
• 05:42Nonetheless, what we saw in the literature
• 05:44was consistent with prior research
• 05:48in that most studies were cross sectional
• 05:50and some included representative samples, and some did not.
• 05:57So I just wanted to review
• 05:58some of the mental health conditions that have been found
• 06:01across studies of mental health after disasters.
• 06:04So our review specifically focused
• 06:06on post-traumatic stress disorder and major depression
• 06:09but we know that these events are associated
• 06:11with increases in a range of mental health conditions
• 06:14such as acute stress disorder
• 06:17which is sort of a precursor to PTSD,
• 06:20other conditions like generalized anxiety disorder
• 06:22and substance use and other clinical phenomenon.
• 06:25And these are symptoms that are concerning
• 06:28but don't necessarily map nearly
• 06:30on to psychiatric diagnoses.
• 06:32Such as non-specific psychological distress
• 06:34internalizing symptoms, such as mood
• 06:37and anxiety symptoms in children.
• 06:39Externalizing symptoms including attention
• 06:41and conduct symptoms in children and adolescents,
• 06:44suicidality and adverse health behaviors
• 06:47including disruptions in sleep, eating and exercise.
• 06:50And what I would say is that
• 06:52across all of the studies today,
• 06:53there's been considerable variation
• 06:56in the prevalence estimates of these conditions.
• 06:59And this is likely due to divergences across the studies
• 07:02for example, in the timing of assessment
• 07:04relative to the disaster, the exposure severity
• 07:08of the sample included as well as the disaster itself
• 07:11as well as other characteristics samples.
• 07:16However, across this literature
• 07:18something that has been consistent
• 07:20is that we've identified individual level risk factors
• 07:22at least at the cross-sectional level
• 07:24of adverse mental health outcomes.
• 07:26And here I've organized them by timing relative
• 07:29to the disaster, starting with a predict disaster factor.
• 07:33So what we know about people
• 07:34going into these types of events.
• 07:36So studies have pretty consistently showed that women,
• 07:39those of low socioeconomic status,
• 07:42those who have preexisting health conditions
• 07:44and in particular mental health conditions
• 07:47who are socially isolated,
• 07:49who have experienced previous exposure
• 07:51not only to disasters but other events
• 07:54are at increased risk for mental health adversity.
• 07:56Whereas findings regarding race and ethnicity
• 07:59and age have been mixed.
• 08:02Turning to the peri-disaster period.
• 08:04So this is the period of the disaster itself
• 08:07and its immediate aftermath,
• 08:08we know that a range of experiences
• 08:10are associated with adversity
• 08:12including the perception that one's life was in danger
• 08:15experiences of physical injuries and bereavement and so on.
• 08:19We also know increasingly that media exposure,
• 08:21so exposure to versus details and images of disasters
• 08:26in their aftermath are associated with increased severity
• 08:30of psychiatric symptoms.
• 08:32Reflecting the potentially broader impact
• 08:35of these types of events.
• 08:37And then post disaster we know that
• 08:40when the storm clouds have cleared
• 08:42and the earth has stopped shaking
• 08:44disaster related stressors tend to persist.
• 08:46And those who experienced financial strain, unemployment,
• 08:50continue disruptions in their work and school lives,
• 08:53stressors in their relationships
• 08:55tend to be at increased risk.
• 08:57And that other stressful and traumatic life events,
• 08:59whether or not they're related to the disaster
• 09:01tend to exacerbate
• 09:02disaster related mental health conditions.
• 09:05So that in a nutshell is the research to date.
• 09:08And I think what we've seen in the past five years or so
• 09:11are some exciting ways in which researchers
• 09:14are trying to push the boundaries
• 09:16of disaster mental health research.
• 09:19So I have here some examples of recent trends.
• 09:22I know for those of you who read the review
• 09:24as part of the seminar
• 09:26you've seen examples of these already.
• 09:28But I'm gonna be focusing on on four trends
• 09:31and how my colleagues, students and I
• 09:33have in our work tried to push the field.
• 09:38So first focusing on long-term responses
• 09:40both in the general population and among vulnerable groups.
• 09:44Pathways to adversity, characteristics of communities
• 09:47and their impacts on mental health and treatment.
• 09:52So first off long-term responses.
• 09:54So what happens in terms of effective populations
• 09:58mental health not just in the immediate aftermath
• 10:00of disasters but in the longer term.
• 10:04And in this work, my colleagues and I
• 10:06have been very much influenced
• 10:07by the work of clinical psychologists and other scholars
• 10:10such as George Bonanno at Columbia
• 10:13and their theories about resilience
• 10:15and other potential trajectories of mental health symptoms
• 10:19after exposure to a potentially traumatic event or PTE.
• 10:23And what Bonanno and colleagues have said
• 10:24is that most people when exposed to trauma
• 10:27will experience what has been termed resilience.
• 10:30And resilience here means a trajectory
• 10:32of chronically low symptoms of distress and well being.
• 10:38So across studies, more than 50%
• 10:40tend to fall into this trajectory.
• 10:42However, other trajectories are common.
• 10:45About 25% on average experience
• 10:47what has been termed recovery.
• 10:49So short term elevations and symptoms
• 10:51and then smaller percentages have exhibited directories
• 10:55of chronic elevations and distress
• 10:57as well as delayed onset distress.
• 11:00So my colleagues and I have worked within this area
• 11:03while also trying to push its boundaries
• 11:05and question some of the key tenants of this theory.
• 11:10So as a first example, I'm going to be presenting data
• 11:13from the Galveston Bay Recovery Study.
• 11:16This was a study of...
• 11:17And I would say it's probably the gold standard
• 11:21of disaster mental health studies that Sandra Golia
• 11:24and Fran Norris led where they were able to gather data
• 11:27from a representative sample of areas that were
• 11:31most severely affected by Hurricane Ike.
• 11:33And they collected three waves of data
• 11:36within the first two years.
• 11:38So it's a really fantastic dataset.
• 11:42So what we did is we ran a trajectory analysis
• 11:44not just of PTSD, but also of depression,
• 11:48functional impairment and days of poor health.
• 11:52So I have our trajectory results here
• 11:53but they're very small and with good reason,
• 11:55which is that I want to put across the takeaway message.
• 12:00Which is that when we looked
• 12:01within each of these four domains
• 12:03resilience was indeed the modal outcome
• 12:05ranging from 45.1% to around 75% for PTSD.
• 12:10However, when we looked across all of these domains,
• 12:12we found that only 25% of our participants
• 12:15thereabouts had resilience across all four.
• 12:20Suggesting that a focus exclusively on PTSD
• 12:23or one other symptom domain might outscore
• 12:26the suffering and impacts of disasters
• 12:29on affected populations.
• 12:31Now, something I would know here is that all the data
• 12:35for the study were collected prior to Hurricane Ike.
• 12:38So we don't know how the participants were doing beforehand.
• 12:42And it's fairly likely
• 12:43that those who were experiencing elevated symptoms
• 12:46that this had something to do with their wellbeing
• 12:49and health beforehand.
• 12:51So in another study, I've been a part of
• 12:54the Resilience in Survivors of Katrina Project,
• 12:57we've been able to address this limitation.
• 12:59And so what the RISK project is,
• 13:01is a longitudinal study of about 1000 women.
• 13:04Most of them are single low-income African-American mothers
• 13:08who all experienced Hurricane Katrina.
• 13:11What's very interesting about this study
• 13:12was that all of the participants
• 13:14were part of a study that was already going on
• 13:18prior to the hurricane called the Opening Door Study.
• 13:21But the Opening Door Study was a multi-site RCT
• 13:24of a community college intervention
• 13:27that sought to increase retention and graduation rates
• 13:30from community colleges throughout the country.
• 13:32And two of those colleges happened to be in New Orleans.
• 13:36So the hurricane hit in August of 2005
• 13:40and both of those colleges were closed
• 13:41for the fall 2005 semester.
• 13:44But my colleagues, Jean Rhodes and Mary Waters
• 13:47were able to secure funding to launch a new study
• 13:50of resilience among those participants.
• 13:53And we've not collected data three times after the hurricane
• 13:56at approximately one, four and 12 years after Katrina.
• 13:59And we just got back in the field last week
• 14:02to do an additional assessment
• 14:03of how they're fairing amidst the pandemic.
• 14:07So I'm gonna be talking about two analysis
• 14:09we did with these data, looking at trajectories over time.
• 14:14The first was actually my dissertation.
• 14:16And for this project, we looked at patterns
• 14:20of non-specific psychological distress
• 14:23from prior to the hurricane
• 14:25to four years after the hurricane.
• 14:27So at the time, and actually I would say probably still
• 14:31it's one of the few trajectory studies that had access
• 14:33to pre trauma data.
• 14:35So we were really able to look at how the patterns
• 14:38of symptoms over time might have been influenced
• 14:41by how people were doing before.
• 14:44And in a nutshell, we found a six trajectory solution
• 14:47and I know that this is a lot to look at.
• 14:49So I'm gonna try to break it down a little bit.
• 14:53So consistent with prior research,
• 14:55the modal trajectory was what we called resilience
• 14:59exhibited by over 60% of our participants.
• 15:02But what we can see is that those participants
• 15:03actually were doing well in terms of having low distress
• 15:07prior to the hurricane.
• 15:09Similarly, other common trajectories in our sample
• 15:13were marked by consistency from pre to post disaster.
• 15:17So we had a coping trajectory
• 15:19which may have looked like recovery
• 15:21and an increased trajectory
• 15:22which may have looked like chronically elevated symptoms.
• 15:25But again, here we see that prior to the hurricane
• 15:27they had significantly higher psychological distress
• 15:30than those who were resilient.
• 15:32Despite this consistency, we saw evidence
• 15:35for meaningful changes in distress.
• 15:40So we actually had two trajectories that were marked
• 15:43by decreasing symptoms.
• 15:45The first which we turned simply decreased
• 15:47had severe distress prior to the storm
• 15:49that decreased pretty consistently thereafter.
• 15:52Another trajectory that we termed improved
• 15:54also had a severe distress prior to the storm.
• 15:58And post disaster distress that was indistinguishable
• 16:01from those in the resilience trajectory.
• 16:03So had we only had post-disaster data
• 16:05we would have assumed resilience.
• 16:07And then we had a delayed trajectory
• 16:09consistent with prior research.
• 16:13In a more recent analysis, we used our latest data
• 16:17to run a trajectory analysis
• 16:19this time specifically of PTSD symptoms.
• 16:22So because their PTSD symptoms had ties to the disaster,
• 16:25we only have them after the disaster.
• 16:27And here we did a trajectory analysis
• 16:29and examined pre trauma predictors of our trajectories.
• 16:34What was notable here is that we did not find
• 16:37what would typically be termed a resilience trajectory.
• 16:40That is a trajectory of consistently low symptoms.
• 16:43The healthiest trajectory in the sample
• 16:45had actually moderate PTSD symptoms
• 16:47that consistently decreased over time.
• 16:50So in my more recent work,
• 16:52I have been trying actually not to use the term resilience
• 16:56although I hate to muddy the waters.
• 16:58I think that resilience as a trajectory
• 17:02of consistently low symptoms maybe does not capture
• 17:06what it means to be resilience
• 17:08in terms of people's lived experiences.
• 17:13So that's one thing.
• 17:14The other thing we found in this analysis
• 17:17that I think is notable is that the most robust predictor
• 17:21of trajectory membership
• 17:23was having probable pre disaster mental illness.
• 17:29Disaster related exposures, including bereavement,
• 17:32lack of vital resources like food, water and medical care
• 17:35and property damage were also predictive.
• 17:38Whereas other pre trauma factors seem to be mediated
• 17:43by either pre trauma mental illness or disaster exposure.
• 17:47So for example, we looked at pre disaster social support
• 17:52and at the university level
• 17:53this was associated with trajectory membership
• 17:55but not when we controlled for pre trauma mental illness.
• 18:00Similarly, we had access to data
• 18:02on pre disaster physical health conditions.
• 18:06And we found that its association with trajectory membership
• 18:09reduced to non-significant
• 18:11once we control for disaster exposure.
• 18:13Suggesting that there might be some mediational pathways
• 18:16from these risk factors to outcomes.
• 18:19Which brings me to the second area
• 18:23that I'm gonna be talking about today
• 18:24that I've observed in the disaster mental health literature
• 18:27which is an increasing focus on pathways.
• 18:30So pathways to both disaster exposure
• 18:33and even more so to post disaster mental health problems.
• 18:39Here my colleagues and I used what's called
• 18:42a pre peri post disaster framework
• 18:45thinking about how risk factors
• 18:48at these different time periods
• 18:49work together to shape disaster mental health.
• 18:53So for example, we would think that pre disaster factors
• 18:56not only increase post disaster mental health directly
• 18:59but they also increase adversity
• 19:01by influencing the extent to which people are exposed
• 19:05as well as the stressors they experience
• 19:07in the aftermath of disasters.
• 19:09Similarly, we think disaster related experiences
• 19:12are important for post disaster mental health
• 19:14both directly and in so far
• 19:16as they increase risk for further stressors downstream.
• 19:20And then finally we see the relationship
• 19:23between post-disaster stressors
• 19:25and mental health as being bi-directional
• 19:28in that post disaster stressors likely increased risk
• 19:31for mental health symptoms, but mental health symptoms
• 19:34in turn, make it more difficult to cope
• 19:36with post disaster stressors and actually can lead
• 19:39to more stressors in the post disaster environment.
• 19:44My colleagues and I recently published a paper
• 19:47testing such a model using data from the risk project.
• 19:50And we were specifically interested
• 19:51in the pathway from pre disaster trauma.
• 19:54So we assessed trauma exposures separate from disasters
• 19:58including assaulted violence,
• 20:00bereavements, physical assaults, that sort of thing.
• 20:05And then we looked at both PTSD symptoms
• 20:07and generalized psychological distress symptoms.
• 20:10And today I'm just gonna be presenting
• 20:12the results from PTSD.
• 20:15So what we hypothesized was a bit of a complex model
• 20:18at least to look at.
• 20:19But we essentially thought that pre disaster trauma exposure
• 20:24would be directly associated
• 20:25with long-term post-disaster PTSD symptoms.
• 20:29So PTSD symptoms directly tied to one's experience
• 20:32of Hurricane Katrina assessed at around 12 years
• 20:36after the hurricane.
• 20:37But we thought even more so there would be indirect pathways
• 20:40to variables downstream.
• 20:42Among them pre disaster psychological distress
• 20:46that these would work together and the likelihood
• 20:48of exposure to disaster related trauma,
• 20:51to short term post disaster PTSD symptoms
• 20:54and then also to post disaster trauma experiences.
• 20:59- And in a nutshell,
• 21:00we found support for this type of model.
• 21:03The model had good fit with the data
• 21:05and most of our pathways were significant
• 21:07and they expect a direction.
• 21:09Although notably in this model
• 21:11the path from pre disaster trauma
• 21:13to long-term symptoms was non-significant.
• 21:17- However, it had a significant indirect effect
• 21:20on long-term PTSD through other variables downstream
• 21:24and in particular by increasing risk for disaster related
• 21:27and post disaster trauma.
• 21:29Suggesting that people might have factors that increase
• 21:32their vulnerability to trauma across the board
• 21:36disaster related trauma and other types of trauma.
• 21:40Which brings me to the third area of research
• 21:42that my colleagues and I have been focusing on,
• 21:44which is attention to community level factors
• 21:47and characteristics and exposures of communities
• 21:50that could increase or mitigate the impact of disasters
• 21:55on mental health.
• 21:58So much of this research has been using data
• 22:01from the community resilience after hurricane Sandy study.
• 22:04Which is a study we launched in New York city
• 22:07after the hurricane in 2012.
• 22:11And what we did is a serial cross-sectional approach
• 22:13where we sampled two representative sub samples of survivors
• 22:18from highly effective neighborhoods within New York City.
• 22:22We gathered data from around 500 participants
• 22:25a year after the storm and 500 participants
• 22:28two years after the storm.
• 22:30We would have loved for the study to have been longitudinal
• 22:32but we did not have the funding to run that type of study
• 22:34so we took this approach instead.
• 22:38And we also gathered data on where our participants
• 22:41were living and community characteristics
• 22:44including property damage within the communities
• 22:47as well as demographic data
• 22:50from the American Community Survey.
• 22:53We were fortunate to have a health geographer on the team
• 22:56Oliver Grooner who did geospatial analysis
• 23:00including spatial autocorrelation analysis.
• 23:03In which we were able to identify clusters
• 23:05of low and high PTSD that were related to exposure
• 23:09but not entirely so.
• 23:11Suggesting that there might be unique characteristics
• 23:14of these different neighborhoods that could have increased
• 23:17or mitigate risk.
• 23:20In another study, we looked at the interaction
• 23:23between exposures experience at the individual level.
• 23:25These included stressors like financial losses,
• 23:28displacement, and bereavement.
• 23:31Participants in communities that either experienced
• 23:34high or low levels of damages.
• 23:37And what we found was perhaps not surprisingly
• 23:39that individual and community level exposure
• 23:42had a synergistic effect on the likelihood
• 23:46of perceived need for mental health services.
• 23:49And that it was those who experienced both stressors
• 23:51themselves and who lived in communities
• 23:53that were highly damaged
• 23:55who had the greatest mental health needs.
• 23:58We've also using the serial cross-sectional data
• 24:02been able to look at interactions
• 24:03between individual and community level factors
• 24:06in shaping mental health risks over time.
• 24:09So there's one example we looked at the interaction
• 24:12between again individual level disaster related stressors
• 24:17in participants who are living in communities
• 24:19with either high or low unemployment.
• 24:22And what we found was that a year after the hurricane
• 24:25it didn't matter whether our participants lived
• 24:27in higher or low unemployment areas
• 24:30at least for their PTSD symptoms.
• 24:32Across the board, hurricane related stressors
• 24:34were associated with elevated risk for PTSD symptoms.
• 24:40However, two years after the storm,
• 24:41the picture dramatically changed.
• 24:43And at this point, a disaster related stressors
• 24:46experienced at the individual level,
• 24:48their impact on post-traumatic stress disorder symptom
• 24:51severity was grossly exacerbated among our participants
• 24:54who were living in a high unemployment neighborhoods.
• 24:57And what this suggests is that the impact
• 25:00of community vulnerability might not manifest
• 25:03until the longer aftermath of disasters.
• 25:07And this is problematic because oftentimes the resources
• 25:10that are funneled to vulnerable communities
• 25:13are cut off at about the one-year anniversary.
• 25:14So this suggests greater needs over time.
• 25:19Which brings me to my fourth area
• 25:21that I've been seeing Burgeon in the research,
• 25:25which is a focus on treatment approaches.
• 25:28And I should say, I have not been involved
• 25:30in this research as much as I would like.
• 25:33But there are many different treatment approaches
• 25:35that I have received empirical support,
• 25:37including Psychological First Aid,
• 25:40Skills for Psychological Recovery,
• 25:42Project Hope in New York City,
• 25:44Bounce Back Now which is a smartphone-based app
• 25:49that focuses on a variety of mental health symptoms
• 25:51that could be experienced after disasters.
• 25:54And TF-CBT and cognitive behavioral interventions
• 25:57in schools have also been investigated in literature.
• 26:02So I've been involved, not in these treatment studies,
• 26:05but in studies using a system science approach
• 26:09to simulate populations
• 26:13or communities exposed to disasters
• 26:16and the potential impact of different ways of providing care
• 26:20on levels of PTSD, DK, Snus.
• 26:23So in this first study, we use data from
• 26:26our Hurricane Sandy study as well as studies
• 26:30of the effectiveness of different treatment approaches
• 26:33to create an agent-based model of New York City
• 26:35after Hurricane Sandy.
• 26:38And we tested two different approaches to providing care.
• 26:42First was termed care, which was skills
• 26:45for psychological recovery applied broadly
• 26:48irrespective of our agent's PTSD symptoms.
• 26:53We also then tried a step care approach
• 26:56where our agents were screened for their levels of PTSD.
• 27:00And those with lower moderate symptoms were given
• 27:03the skills for psychological recovery intervention.
• 27:05And those who had like the PTSD were given
• 27:08a more intensive treatment of cognitive behavioral therapy.
• 27:12And through the simulation study,
• 27:14we found that the step care approach
• 27:17had benefits in decreasing the prevalence of PTSD over time
• 27:22as well as lead to cost savings.
• 27:26We did a follow-up using the same data
• 27:28and adding on a social service case management approach.
• 27:32And what we found here was that this approach
• 27:34had even greater benefits and reducing PTSD
• 27:37and across our population of agents in our simulation.
• 27:41And in particular for those who experienced greater exposure
• 27:45to the hurricane characterizes having been displaced
• 27:48or losing income.
• 27:50So while this is not a direct test
• 27:52of these types of interventions
• 27:54it represents an approach to system science to simulate
• 27:58and test different possibilities in effected populations.
• 28:03So now I'm gonna turn to some of my current
• 28:06and hopefully future directions.
• 28:07And for these, I have three.
• 28:11The first is considering cumulative exposure
• 28:14which we think is important given that
• 28:16we know that there are some areas within the United States
• 28:19and beyond that are disaster prone
• 28:21and have unfortunately experienced
• 28:23more than one environmental disaster
• 28:26as well as other stressors.
• 28:29So one example of this is an analysis
• 28:31my colleagues and I did using data
• 28:33from the Gulf long-term follow-up study.
• 28:36And what we did is we looked at exposure
• 28:38amongst the sample to hurricane Katrina
• 28:41to clean up work after the deep water horizon oil spill.
• 28:45And then the combination of these two different exposures.
• 28:48And what we found was that participants who were exposed
• 28:51to both disasters, both oil spill cleanup
• 28:54and to hurricane Katrina tended to have
• 28:56higher mental health symptoms, including PTSD, depression
• 29:00and anxiety symptoms, as well as physical health symptoms,
• 29:03including headaches, back pain and digestive problems.
• 29:09In a future project, I mentioned that we're collecting data
• 29:12on the COVID-19 experiences of our risk sample.
• 29:16And what we're hoping here is to investigate
• 29:19the impact of the pandemic on this group
• 29:22that has already been exposed to a major disaster
• 29:25and their perceptions of whether having experienced
• 29:27hurricane Katrina exacerbated the impact of the pandemic
• 29:31or help them cope.
• 29:36Another future direction is that
• 29:37I've been increasingly interested
• 29:40in the broader impacts of climate change
• 29:45both on people living in areas that are affected
• 29:48by disasters and other climate change indicators,
• 29:51but more generally in the population
• 29:55even in less affected areas.
• 29:58So for this work, I have had the honor
• 30:01of working with Susan Clayton,
• 30:04who is an environmental psychologist
• 30:06at the college of Wooster.
• 30:07And she, this past year developed and validated
• 30:11a measure of climate change anxiety.
• 30:13So the two of us are working
• 30:15with a former classmate of mine, Sarah Schwartz,
• 30:17who's a psychologist at Suffolk University
• 30:20on a study looking at college and graduate students
• 30:23climate change anxiety, its relationship
• 30:26with mental health indicators
• 30:27such as depression and generalized anxiety disorder.
• 30:31And the protective role of constructs such as climate hope
• 30:35and climate activism, and mitigating this relationship.
• 30:39And some of you in the climate change and health seminar
• 30:41may have been invited to participate
• 30:44in this study last semester.
• 30:48And then finally, I've been increasingly interested
• 30:51in other climate change indicators beyond disasters
• 30:55including some of those that are more chronic and persistent
• 30:58as well as other environmental exposures
• 31:00that are likely to affect mental health.
• 31:04An example of this work I have had the honor
• 31:08of working with Kai Chen
• 31:09from the Yale Center for Climate Change and Health
• 31:13on a study looking at particulate matter, air pollution
• 31:16and its association with outpatient visits
• 31:19for mental health problems in Nanjing China.
• 31:22And what we found that was on days
• 31:24where there was greater levels of particulate matter
• 31:29the use of outpatient services increased.
• 31:32Suggesting that this environmental indicator
• 31:34could increase the demand for mental health services
• 31:37and also impact the likelihood of mental health symptoms.
• 31:42And then I've been collaborating on a systematic review
• 31:45trying to conceptualize climate change indicators
• 31:49and look at their impact on mental health.
• 31:51This has been sort of slow going.
• 31:53I think in our initial screening
• 31:55we looked at around 12,000 abstracts
• 31:59and in doing so recognize the challenges
• 32:02of measuring chronic climate change impacts
• 32:07and their potential influence on mental health.
• 32:09So, hopefully that will come out in the next few years.
• 32:13So that is actually all I've got for today.
• 32:16I think that was faster than I expected.
• 32:18But I have my email here and I would be happy
• 32:22to answer questions about this work both today and offline.
• 32:27So feel free to email me and reach out.
• 32:30I love connecting with people, hearing from students
• 32:33and so on.
• 32:35So, thank you very much.
• 32:38- Great, thank you Sarah for this wonderful presentation,
• 32:41giving the state or the knowledge regarding
• 32:44the mental health after all these weather related disasters.
• 32:48And thank you very much for sharing your future
• 32:52and the current directions in this field.
• 32:54It's all, it's very fantastic.
• 32:56And I'm sure the audience will have a lot of questions.
• 32:59So while the audience is preparing the question
• 33:02and typing in the chat box,
• 33:03we do have already clacking a question from the students.
• 33:07So there are a lot of student questions.
• 33:10But the first question the student is wondering is
• 33:15you have shown different types of disasters
• 33:20especially in your review paper.
• 33:22Several students are kind of wondering
• 33:25is there a way to compare the mental health matters
• 33:29across different types of disasters?
• 33:32Like when you compare the different types of disasters,
• 33:35does this matter?
• 33:37Is a particular type of disaster has a strong effect
• 33:41on a particular mental health outcome?
• 33:45- That is a really good question.
• 33:47So I know that it used to be said
• 33:50that disasters that were clearly human made
• 33:54such as oil spills and terrorism
• 33:58we're likely to trigger more severe impacts on mental health
• 34:02because there was someone to blame
• 34:04and they seemed less fateful.
• 34:08However, I don't think that has been shown empirically
• 34:11although perhaps someone else in this seminar
• 34:14knows more than I do.
• 34:18And I do think that it is again worth emphasizing that
• 34:21what we've typically seen as natural disasters
• 34:24do have a clear tie to climate change and human impacts
• 34:29and affects human made systems.
• 34:31And I think that that can lead to feelings of anger
• 34:36and blame and neglect that can exacerbate risks
• 34:39sort of in the same way that would happen
• 34:42after a technological disaster or terrorism.
• 34:46So I think it's difficult to really make the comparison.
• 34:50But my sense is that both have the potential
• 34:53to trigger symptoms across the board.
• 34:58- Thanks, so another type of question follows
• 35:01the interventions you mentioned.
• 35:04So the students are wondering,
• 35:06you mentioned give some examples
• 35:09more from the clinical science clinical based interventions.
• 35:13And you have also mentioned your own research
• 35:17and other papers has shown some individual level
• 35:22or community level characteristics
• 35:25such as the employment rate
• 35:28that it can kind of modify the risk.
• 35:31So is there wave, can you talk about
• 35:34more this nonclinical intervention strategies?
• 35:38And are there community-based programs are happening
• 35:42or are there any further readings for the students?
• 35:47- Yeah, so that is a really good question.
• 35:49So yeah, so as a clinical psychologist, I'm most well-versed
• 35:54in trauma-focused CBT and those types of treatments
• 35:59for people who have moderate or severe symptoms.
• 36:03But I think that there are public health approaches
• 36:06to treating mental health across the board
• 36:08including psychological first aid.
• 36:11And I think a key here is that psychological first aid
• 36:14acknowledges that most people are going to be resilient
• 36:18in terms of their mental health.
• 36:20And so aren't going to benefit from more intensive services.
• 36:25And in fact, you know, therapeutic approaches
• 36:28might actually impede their coping processes
• 36:30and increase their risk.
• 36:32So psychological first aid as I understand,
• 36:34I have not been trained in it
• 36:36and I would love to at some point,
• 36:38focuses on assessing how people are doing,
• 36:41providing them information and then referring them
• 36:45to resources that help them
• 36:46either with their mental health problems
• 36:48or other social service needs.
• 36:52I think a social service approach that integrates
• 36:54both psychological first aid and that assesses
• 36:58the broader range of post disaster needs
• 37:00and provides some case management
• 37:02in navigating the various systems
• 37:04that disaster survivors come into contact with
• 37:07is very important.
• 37:09And I know that in our Katrina study
• 37:12so that was a mixed methods project
• 37:15a lot of our, not a lot, some of our survivors
• 37:18talked about how their encounters with social services
• 37:22after Katrina was actually their first touch point
• 37:26to getting mental health services for preexisting problems.
• 37:30So I think the post disaster period
• 37:32could actually be in some cases, an opportunity
• 37:35for people to get help that they needed all along.
• 37:38And it's unfortunate that it takes a disaster to do that
• 37:40but could actually facilitate not just psychological growth
• 37:45but access to social and economic resources
• 37:48that foster their wellbeing across the board.
• 37:52- Oh, thanks, Sarah.
• 37:53I think there's a question from the audience relate to this
• 37:56from Pat Haney.
• 37:58Just thank you, Sarah, can you give an explanation
• 38:01of the step heard care you discuss in your model?
• 38:05- Yeah, so that was a really interesting project
• 38:08to be a part of.
• 38:09So we use what's called agent-based modeling
• 38:12which you actually put in, you create a population
• 38:16within a computer programming software.
• 38:19We use Python and then you put in various inputs.
• 38:23So you distribute disaster exposure,
• 38:26you distribute risk factors for psychopathology
• 38:31and then you can apply an intervention to that population.
• 38:34So intercept care approach, what I believe we did
• 38:37is we screened our participants
• 38:39meaning that we assign them different levels of PTSD.
• 38:44And then those who met a certain level
• 38:45I think we said seven PTSD symptoms
• 38:48who likely had the disorder were then given
• 38:51in the simulation cognitive behavioral therapy for PTSD.
• 38:56And that others who had non-zero
• 38:58but less than seven symptoms of PTSD were given
• 39:02quote unquote skills for psychological recovery.
• 39:06And based on the findings of prior research
• 39:10on the effectiveness
• 39:12of these two different intervention approaches
• 39:15our agents within the model, their symptoms declined
• 39:18in a way we would expect
• 39:19based on their socioeconomic demographics.
• 39:22So again, it was a simulation, it was not a test
• 39:25of an approach, but more of a demonstration
• 39:28that screening participants and providing services
• 39:30that meet their mental health needs
• 39:33could more effectively lead to decreases in PTSD over time.
• 39:39- Oh, great, I think another, it's not maybe a question
• 39:43but a comment from Massey asking
• 39:46as a clinician and a public health practitioner
• 39:49how best to translate this information
• 39:52to first advocate clinician to be aware now of these issues.
• 39:57So I think it's first within the interaction question.
• 40:02There has been other questions from students as well.
• 40:07So while the students is asking
• 40:10like we study the association between disaster
• 40:13and the mental health, is that a case that is
• 40:16some solution, will there be some underestimation
• 40:21of their mental health status due to the stigma
• 40:25of the mental illness
• 40:26especially in a lot of surveys you have performed?
• 40:31- Yeah, so the question is whether
• 40:35mental consequences will be exacerbated
• 40:37if there's stigma experienced?
• 40:39- Or maybe underestimated in the service.
• 40:42Some people would maybe reclined
• 40:46to answer these questions, so.
• 40:49- That is a good question.
• 40:52I don't think I have a good answer for you.
• 40:54I think it's certainly possible
• 40:56that people who experienced mental health stigma
• 41:00might be less likely to report symptoms.
• 41:04That being said in these studies
• 41:06we use validated scales that ask about specific behaviors
• 41:11and experiences, not disorders.
• 41:14So for example someone who experienced mental health stigma
• 41:17might be more likely to say I haven't had good sleep
• 41:21over the past two weeks, or I've been feeling
• 41:23like a lack of pleasure.
• 41:26Than saying that they experienced depression per se.
• 41:31So they are sort of behaviorally anchored questions.
• 41:35And it's interesting 'cause I think people are more likely
• 41:40to report symptoms if they're doing so anonymously,
• 41:44such as via an online survey or something like that.
• 41:47But a lot of, especially the epidemiologic studies
• 41:49are done over the phone, at least historically.
• 41:52And that having that personal contact could potentially
• 41:55be a barrier to reporting.
• 41:58And then absolutely stigma is a barrier to service seeking
• 42:02but you know, there are other barriers too.
• 42:05So in one study we looked at the frequency
• 42:07of different barriers
• 42:08and a major one was a lack of resources.
• 42:12So not knowing where services were,
• 42:14not having time, needing childcare,
• 42:18not having transportation.
• 42:19And I think those can get in the way as well.
• 42:23- Yes, another question kind of related
• 42:27to the respondents characteristics is,
• 42:29there's one question from Peter asking,
• 42:31has any of the current research considered the difference
• 42:35in PTSD among first responders and long-term
• 42:39community responders versus those who are impacted
• 42:44but did not assist them with the response?
• 42:48- Yeah, that is a really good question.
• 42:49So from the research that I've seen,
• 42:54epidemiologic studies have shown that people
• 42:56who are involved in the response
• 42:58tend to be at increased risk for mental health problems
• 43:01relative to the general population.
• 43:03However, there is substantial variability
• 43:06amongst first responders.
• 43:08So those who are exposed to atrocities, such as,
• 43:14dead bodies, people who are harmed
• 43:16really severe property damage,
• 43:18who are exposed to environmental toxins,
• 43:20like mold and things of that nature
• 43:23and who have not received adequate training.
• 43:27So I know for example, I think there was a study
• 43:30after the Deepwater horizon oil spill, or maybe not,
• 43:33I'm trying to think.
• 43:34This may have been a disaster in one of the ones in Japan
• 43:37that was conducted that showed that people who were
• 43:41police officers or who had previously been involved
• 43:44in response work tended to have fewer
• 43:48adverse mental health impacts
• 43:49relative to those who volunteered.
• 43:52Which suggests the benefits and importance
• 43:55of resilience training prior to these exposures,
• 43:59which is really hard to do, right?
• 44:00Because these events by their very nature are unexpected
• 44:03and people are going to volunteer
• 44:05which is great to help out.
• 44:07There might not be adequate time to really prepare them,
• 44:10but probably at least some.
• 44:15- Great, so there's a couple of other questions
• 44:19relating to the study.
• 44:23Actually to the review paper you presented.
• 44:25One of them is actually asking
• 44:28about not weather related disaster, but
• 44:32a question from the audience asking,
• 44:34have you worked or research interests such as
• 44:38with manmade disaster, such as armed conflict?
• 44:42And looking into the displacement
• 44:44and how these may impacted them in the house?
• 44:48- Absolutely, that's a very good question.
• 44:51So I have been involved in studies of human made disasters,
• 44:55namely the study, I mentioned with the workers
• 44:57after the deep water horizon oil spill
• 44:59but that seems very different
• 45:00than what the student is asking about
• 45:02which is armed conflict and displacement.
• 45:05I would love to get involved in this type of work.
• 45:07I haven't yet had the opportunities.
• 45:10But what I can say is that there are some clear parallels
• 45:16to weather related disasters
• 45:19as well as some clear distinctions.
• 45:22So a parallel is that being displaced from your community
• 45:26not by choice can be really stressful
• 45:29and potentially traumatic.
• 45:31And that we found in our Katrina study,
• 45:34that those who relocated which was a good percentage
• 45:37of our sample tended to be at increased risk
• 45:39for mental health problems.
• 45:41Both those who like stably relocated
• 45:44who found a new place to live
• 45:45in a different state and settled there
• 45:46and those who had unstable housing trajectories.
• 45:50I think another commonality is that
• 45:55both types of community level trauma
• 45:59involve exposure to death and destruction.
• 46:02But I think the particulars of it are very distinctive
• 46:08and the level of violence who is perpetrating it,
• 46:15the extent of displacement could be very different
• 46:18in ways that could exacerbate mental health risks.
• 46:22So I think that there are some ways are similar
• 46:23and some ways they're very different.
• 46:26- Yeah, I wanted a follow up
• 46:28on the like displacement request.
• 46:30And we know what you also mentioned
• 46:32the kind of anxiety conscience is your future direction.
• 46:36So we know there's issue on the counter refugees
• 46:41especially considering even the whiteflies in the West.
• 46:45A lot of people just were displaced due to the whiteflies.
• 46:48So when talking about to the mental health burden
• 46:52of these kind of refugees,
• 46:56can you give more like an explanation
• 46:58on the state of the science on that?
• 47:00And are there any new directions that you want to ask?
• 47:06- Yeah, that is a really good question.
• 47:09In terms of the state of the science,
• 47:10I don't know a lot of good literature
• 47:13on climate refugees and displacement aside from
• 47:17like domestic displacement after hurricane Katrina.
• 47:21That doesn't mean that there's not good research going on,
• 47:22I just might not know about it.
• 47:25But my overall sense is there's probably not a lot of it
• 47:28going on and that this is to be a major issue
• 47:31'cause being displaced from one's home community
• 47:34either because your community has been destroyed
• 47:36or that it's at great risk is incredibly stressful.
• 47:41And not only can impact mental health
• 47:42but it can impact the things that foster mental health.
• 47:46Such as social connections, employment,
• 47:49community attachment, things of that nature.
• 47:54So, you know, what I would say is that we need to be mindful
• 47:57that this is going to happen
• 48:00and trying to create communities that are accepting
• 48:05and supportive of people who are displaced.
• 48:08You know, I know for our Katrina sample
• 48:10one of the things qualitatively that was very difficult
• 48:12for them was moving to places where they were not welcome,
• 48:16where they were stigmatized,
• 48:17where they had difficulty getting jobs,
• 48:19because they were from New Orleans.
• 48:22Or heard people say things about people
• 48:25from New Orleans and the culture of New Orleans
• 48:27and this is within the same country.
• 48:29So I could only imagine, you know, when we're talking about
• 48:31people crossing international borders
• 48:32that these types of issues within communities
• 48:35are gonna be heightened.
• 48:40- Yeah, another question kind of related
• 48:43to the culture inference, one of the students is asking
• 48:46among these community level characteristics,
• 48:50do you expect these different characteristics
• 48:54such as the culture inference can be a factor
• 48:58influencing the substantial variability
• 49:01you observed in the review paper
• 49:04on the premise of the PTSD and the depression?
• 49:07- Yeah, that is a really good question.
• 49:08And I don't know, offhand I'd have to actually look closely
• 49:11at the review paper that you all read
• 49:13to see what literature was came out at that particular year.
• 49:16What I would say having been involved in this research
• 49:19you know, we try to get community level data
• 49:21from the Census Bureau and the American Community Survey.
• 49:26And oftentimes when you run these analysis
• 49:29they explained very little variability in outcomes.
• 49:33And I think part of the reason is because
• 49:36census tracks and census blocks don't necessarily
• 49:39map onto what people perceive as their communities.
• 49:43Like I know in after Hurricane Sandy
• 49:45like I technically I was eligible for the study that we did.
• 49:48I have no idea what my census track was.
• 49:52And it would be hard to imagine
• 49:54that it really mapped onto what I saw as my community
• 49:57given that the people that I interacted with
• 50:00on a day-to-day basis didn't necessarily even live
• 50:02in that particular census track.
• 50:04So I think it's tricky.
• 50:06And then an alternative source that people sometimes use
• 50:09is they ask people about their perceptions
• 50:11of their own community and that's going to be biased
• 50:13by their mental health and functioning.
• 50:16So I think, you know, there are advantages
• 50:17and drawbacks to different approaches
• 50:20and very likely community level characteristics
• 50:23do shape mental health after disasters.
• 50:26But I don't think we've been able to
• 50:28very precisely estimate that.
• 50:32- Great, so due to the time limitation
• 50:35we will have the last two questions.
• 50:36So the one is from Diane,
• 50:41excuse me, if I pronounce it wrong
• 50:45from the audience, what might the considerations be
• 50:48for substance misuse services pre and post disaster?
• 50:53And what has to be ensured to help these populations most?
• 50:58- That is a really good question.
• 51:01So I am not super well versed in substance abuse services.
• 51:08I can say that there have been studies
• 51:09that have shown increases in alcohol use
• 51:13and use of other substances including non-medical use
• 51:18of prescription drugs after disasters
• 51:20and often they're endorsed as a means of coping with stress.
• 51:24And I think certainly we've seen that
• 51:25with the COVID-19 pandemic as well.
• 51:28So I think in general a population-based approach
• 51:30could be to acknowledge that that is something
• 51:33that people do to cope
• 51:34as well as the potential negative consequences of that
• 51:37and alternative ways of coping if people feel like using.
• 51:41I do know anecdotally I have
• 51:46colleagues, not super close colleagues
• 51:48but contacts who have done some work
• 51:50with opioid and methadone maintenance after hurricanes.
• 51:54And I think it's really challenging
• 51:56because the people who run these clinics are also impacted.
• 52:00And when people are displaced, they have disruptions in care
• 52:04that can be really devastating for their recovery.
• 52:08So I think it is a major issue
• 52:10both in terms of people using substances to cope
• 52:12and then people in recovery not only experiencing
• 52:16an additional stressor
• 52:17that can exacerbate their risk of abusing
• 52:20but also major disruptions in their care.
• 52:26- Okay, so last question is actually from the student
• 52:30is asking one of your future director
• 52:32is the community disaster exploring.
• 52:35So the students are wondering, do you know any study
• 52:39exploring the potential interaction facts
• 52:42from these individual characteristics you observed
• 52:45and also the community characteristics
• 52:47including some of the pre disaster finding?
• 52:51- Yeah, so I'm trying to think if there are good examples
• 52:54other than the one that I presented today
• 52:56which looked at individual and community level exposures.
• 53:01I don't know of any offhand
• 53:05that have looked at community level factors,
• 53:08such as indicators of socioeconomic status
• 53:12and individual level impacts.
• 53:14There is some work that has been done
• 53:16by Elizabeth Frankenberg and colleagues
• 53:18after the Nepal earthquake and tsunami
• 53:22that I believe found something in that effect.
• 53:26But I can't remember offhand what exactly they found.
• 53:30And then there's another study that was conducted
• 53:32after flooding in England by Compro,
• 53:35is the author Winden and Compro,
• 53:36I know are their last names
• 53:38that found interactions I believe between exposure
• 53:41and social cohesion.
• 53:43But social cohesion in that case was measured
• 53:46based on the participant's own perceptions
• 53:50of social cohesion across the area.
• 53:52So, yeah, those are two examples
• 53:55but I don't know a ton of literature in that area.
• 53:58And I think that is an open area
• 53:59for further explanation or examination.
• 54:03- Great, thank you Sarah.
• 54:05And I think there's a lot of excitement to conduct research
• 54:09in this field and thank you all for listening today.
• 54:12And just a reminder that this seminar is recorded
• 54:15and will be posted online
• 54:18on the Yale Center for Climate Change and Health
• 54:20so check out later.
• 54:22With that, thank you Sarah.
• 54:24- Yeah, feel free to be in touch.
• 54:26- Thanks Sarah.
• 54:28- [Sarah] Thanks Rob.
• 54:29- Bye everyone.
• 54:30- [Sarah] Bye everyone.