Psychiatry Grand Rounds: November 13, 2020

Transcript

• 00:00Good morning everyone.
• 00:01It's great to see so many friendly faces
• 00:05in that bar on the right of my screen.
• 00:08I have the pleasure and honor of
• 00:11introducing Craig Brian this morning.
• 00:13I first actually got to know Craig
• 00:15in 2015 after reading a paper in
• 00:18the American Journal of Psychiatry,
• 00:20randomized controlled trial that
• 00:22he had done with David Rudd.
• 00:25That actually showed a 60%
• 00:27reduction in suicide attempts among
• 00:28active duty military personnel,
• 00:30which really caught my interest.
• 00:32So I reached out to Craig and David
• 00:35and started talking with them,
• 00:37learning more about their work
• 00:39and ultimately became a friend
• 00:42and admirer of his work.
• 00:43I looking very much looking forward
• 00:45to his talk today because I think
• 00:47there's data in it I'm not aware of.
• 00:50Craig is a clinical psychologist.
• 00:52He's the star professor of Psychiatry
• 00:54and Behavioral Health at the Ohio
• 00:57State University Wexner Medical Center.
• 00:59And is the division director for
• 01:02Recovery and resilience there.
• 01:03He did his doctorate in Clinical
• 01:05Psychiatry at Baylor University,
• 01:07completed clinical residency at
• 01:08the Wilford Hall Medical Center,
• 01:10Lackland Air Force Base,
• 01:11and was deployed in Iraq in 2009 and
• 01:14was the director of the traumatic
• 01:17Brain Injuries Clinic at the Air
• 01:19Force Theater Hospital there.
• 01:21He left active duty service and started
• 01:23researching post traumatic stress disorder,
• 01:25suicidal behaviors and
• 01:26suicide prevention strategies,
• 01:27as well as psychological health
• 01:29and resiliency.
• 01:30He's had faculty appointments at
• 01:31the University of Texas Health,
• 01:33same health system in San Antonio,
• 01:35the University of Utah,
• 01:37and now at the Ohio State University.
• 01:39He's had 10s of millions of dollars
• 01:42in grants focused on these topics.
• 01:44Suicide prevention,
• 01:45PTS di trauma and resilience.
• 01:47One has won many awards.
• 01:48I won't go into all of them.
• 01:51And I'll just let him dive in and
• 01:54get right to it.
• 01:55So Craig, thanks so much for coming.
• 01:57Thanks for everyone for attending.
• 01:58If you have questions,
• 02:00please enter them into the chat and
• 02:02when Craig is done we can go through
• 02:04them and try to get to them all.
• 02:06So thanks.
• 02:09Alright well thanks for the introduction.
• 02:11It's really great to be able to join all
• 02:14of you this morning and seeing a little
• 02:16bit of windows and in my boxes here,
• 02:19it looks like you guys are having a nice day.
• 02:23Up there in the Connecticut area and a
• 02:26beautiful day here in central Ohio as well.
• 02:28So today what I thought
• 02:30I would talk about is.
• 02:32More of some conceptual work around
• 02:35suicide that we've really been focusing
• 02:37on over the past several years now.
• 02:40Certainly present some of the
• 02:42empirical findings at our lab and
• 02:45other labs have published in recent
• 02:47years that really kind of backup.
• 02:50This sort of approach to understanding
• 02:52suicide and suicidal behavior.
• 02:54And then, after we kind of start with
• 02:57more kind of conceptual ideas about
• 03:00thinking of suicide in new ways.
• 03:03You know,
• 03:04I'll conclude with some of the work
• 03:06that we've done from an intervention
• 03:08and prevention standpoint,
• 03:10and how I think this alternative approach
• 03:12to understanding the emergence of
• 03:14suicidal behavior and might lead us to
• 03:17think about a range of interventions
• 03:19in a number of strategic ways.
• 03:22So the you know the basis for
• 03:24this line of research really was,
• 03:27you know, founded in.
• 03:28I think some traditional ideas
• 03:30about suicide prevention.
• 03:31I think some of the traditional
• 03:33ways that we have approached
• 03:35our understanding of suicide and
• 03:37approached prevention efforts are
• 03:39encapsulated in many ways by a lot
• 03:41of the catch phrases and sayings
• 03:43that I have on the screen right now,
• 03:46that I'm sure many of you are familiar with.
• 03:49You probably heard this plenty often.
• 03:52I'm going to maybe you as well as
• 03:54mental health professionals and
• 03:56researchers have said one or more of
• 03:58these things as well when talking
• 04:01about suicide and I start here because
• 04:03I kind of think that where we're
• 04:06heading now in in our research on
• 04:08suicide prevention is really a mini
• 04:10wasted on the cusp of a paradigm shift.
• 04:13And I think it's critical for us
• 04:16to understand where we come from
• 04:18as we talk about and think about
• 04:21where we're heading.
• 04:22So one of the challenges, of course,
• 04:25when it comes to suicide,
• 04:27is that despite decades of fairly consistent,
• 04:30I would argue fairly uniform thinking
• 04:32about suicide over the past two
• 04:35decades since the turn of the century,
• 04:37we've seen a steady rise in
• 04:40suicides in the US,
• 04:41and which is not really mirrored
• 04:43anywhere else in the world.
• 04:46About 3/4 of nations around the
• 04:48world during the same time frame.
• 04:50If you look at statistics reported too.
• 04:53Mental Health Organization.
• 04:55See that most nations are actually
• 04:58experiencing a decline in suicide
• 04:59rates in the United States is one of
• 05:02the minority that is actually seeing
• 05:04the reverse pattern occur ING in.
• 05:07On top of that, if we were to rank order.
• 05:11You know all the world's nations in
• 05:14terms of change in suicide rates since 2000,
• 05:17we see that the United States is
• 05:19actually #3 in terms of the fastest
• 05:21and the largest increase in suicides.
• 05:24And so there's something very unique I think
• 05:26about the United States that's different
• 05:29that Bing's very clearly are not working,
• 05:31at least not in the way that we would like,
• 05:35which has really stimulated
• 05:36me and others to second guess.
• 05:38A lot of those initial assumptions
• 05:41that I shared with you.
• 05:42Maybe there's something about the way
• 05:45that we've approached suicide and.
• 05:46Thought about it.
• 05:48That isn't really accurate.
• 05:49Or maybe we're missing the mark in some way.
• 05:52'cause if we were if we were correct
• 05:55about how we understand suicide in
• 05:58our interventions were as effective
• 06:00as we think they should be,
• 06:02then probably we would not be seeing
• 06:05this upward steady rise in suicide.
• 06:07We should be seeing either things
• 06:10holding steady or even a decrease.
• 06:13So as I think about what are some of those
• 06:16classic ways that we conceptualize suicide,
• 06:19I really think that many of the theories
• 06:21that have been proposed and there are
• 06:24a multitude of theories about suicide.
• 06:27In many ways,
• 06:28conform to this general framework.
• 06:30We could certainly probably quibble
• 06:32and argue about differences and
• 06:34the different theoretical models,
• 06:36but I think that this this sort of
• 06:40sequence of events, so to speak,
• 06:42kind of serves as a sort of a
• 06:45foundational way of capturing how
• 06:48we have thought about suicide.
• 06:50Where you know someone experiences
• 06:52an intensification or elevation of
• 06:55emotional distress and then some things
• 06:57happen, certain mechanisms occur.
• 06:59Different theories would posit
• 07:01different mechanisms,
• 07:02but something is happening whereby
• 07:04emotional distress evolves into
• 07:06suicidal ideations and then at that
• 07:09point individuals were thinking
• 07:11about suicide again.
• 07:12Some kind of mechanisms or processes occur.
• 07:16Different theories, again,
• 07:17are proposing different mechanisms,
• 07:19and then individuals transition
• 07:21to suicidal behaviors.
• 07:22So in essence,
• 07:24they act upon the suicidal thoughts
• 07:26and then another sort of typical
• 07:30assumption about.
• 07:31The sort of progression from ideations
• 07:34to action is that relatively soon
• 07:37before a suicide attempt occurs,
• 07:39there are theoretically some observable
• 07:42changes are observable variables that
• 07:45we refer to as warning signs that should,
• 07:48in theory signal that a suicide
• 07:51attempt is about to occur.
• 07:54So I would say despite in many ways
• 07:58this serving is kind of a foundational.
• 08:02Assumption about how we think about suicide.
• 08:05As I noted on the previous slide,
• 08:08we're not really good at preventing suicide,
• 08:10at least not in the United States,
• 08:13and so it's really prompted myself
• 08:15and a number of other researchers
• 08:18to question a lot of our long held
• 08:21assumptions about suicide and to start
• 08:24thinking about this in different ways.
• 08:26And so I just want to kind of
• 08:30talk today about some of my own.
• 08:32Halfway my own thinking over the
• 08:35past several years and kind of talk
• 08:38about how different experiences
• 08:40different studies have led us to to
• 08:43really approach our understanding
• 08:44of this emergent process of moving
• 08:46twords suicidal behavior that we
• 08:48think can potentially advance our
• 08:51ability to prevent suicide.
• 08:52And in many ways,
• 08:54sort of.
• 08:55I would kind of tie the
• 08:57origins of my thinking
• 08:59about this to some conversations
• 09:01that I had with a colleague of mine.
• 09:04In Utah. You know the first arrived.
• 09:08There's faculty, especially the
• 09:10conversations I had with John Buckner,
• 09:13whose social quantitative psychologist his
• 09:15expertise in complex dynamical systems.
• 09:17And we were having coffee one day on campus,
• 09:21just kind of chatting about our respective
• 09:24programs, getting to know each other,
• 09:27and it so happened that each of us had
• 09:30been funded by NASA to do some work,
• 09:34and in his project in his NASA.
• 09:37On the project is kind of interesting.
• 09:40He was being asked to model budget
• 09:42overages and what NASA's concern was,
• 09:45hey, you know we have all these grants.
• 09:48We've all these contracts and
• 09:50projects that we're funding.
• 09:51You know, the vast majority of them.
• 09:54They spend money as expected.
• 09:56They perform on track,
• 09:57so they're not really issues for us.
• 10:00We obviously don't worry about
• 10:02projects that achieve their
• 10:04objectives on time and on budget.
• 10:06And then there's another pool of
• 10:08projects where the projects are
• 10:10fairly consistently over budget.
• 10:12They're not dramatically over budget,
• 10:14but they are always spending more
• 10:16than what was initially projected,
• 10:19but there still achieving their
• 10:21milestones and an absolute told him,
• 10:23you know,
• 10:24we're not really worried
• 10:25about those projects either,
• 10:27because they're predictably over budget.
• 10:29We know exactly how much
• 10:31they're going to spend.
• 10:33And probably it's just we we
• 10:35probably just under estimated the
• 10:37resources needed for those projects,
• 10:39but they're doing OK,
• 10:40But there's this third group of projects.
• 10:43That are really, really problematic for NASA,
• 10:47and it's because.
• 10:48They initially start on track there,
• 10:51performing well they're spending as expected,
• 10:54and then all of a sudden
• 10:56they're massively over budget,
• 10:58and then once they seem to
• 11:00depart from being on track,
• 11:02they have one period where their massively
• 11:05over budget things just get worse and worse.
• 11:08They keep coming up over budget,
• 11:11over budget over budget dramatically,
• 11:13and so this creates this massive
• 11:15problem for the organization because it
• 11:18shuts down other projects it creates.
• 11:20Budget strain for everybody else.
• 11:22All of the projects depending
• 11:24on that one particular project,
• 11:25it slowed down and so NASA
• 11:28had actually a term for it.
• 11:30They called it,
• 11:31performance escapes,
• 11:31and so they wanted John to help
• 11:34figure out how do we know which of
• 11:37our seemingly on track OK projects
• 11:39are going to suddenly out of the blue
• 11:42turn into project escapes and and so
• 11:44John was modeling this mathematically.
• 11:46This is one of the simulations
• 11:48that he ran using their data and
• 11:51you see here there's just like 1
• 11:53project that sort of taking off.
• 11:56This is a performance escape as compared
• 11:58to all the other projects that are sort
• 12:00of clustered on their spending there.
• 12:02They're kind of on track,
• 12:03and as we're talking about
• 12:05these performance escapes.
• 12:07He said,
• 12:07you know,
• 12:08that's interesting that what you're
• 12:10doing there because you know I'm a
• 12:13suicide researcher and one of our
• 12:15big problems that we run into him.
• 12:17Clinicians. We have some family members,
• 12:19lots of people is that
• 12:21suicide seems unpredictable.
• 12:22Patients will seem to be doing OK in
• 12:24treatment and then suddenly they're dead.
• 12:27Family members say everything
• 12:28seemed to be OK.
• 12:30They seem to be fine.
• 12:31And then all of a sudden
• 12:33they had attempted suicide,
• 12:35and so there is this sort of subgroup.
• 12:38Of suicide cases that seem to align with
• 12:41this notion of a performance escape,
• 12:44and we also have known for
• 12:46several decades if there's a,
• 12:48there's a significant percentage in some
• 12:50studies up to half of those who attempt
• 12:53suicide who deny suicidal planning.
• 12:56Severe suicidal ideations,
• 12:57where they seem to like skip these
• 13:00intermediate steps that we have long held
• 13:02to be a necessary precondition for suicide.
• 13:05The assumption that many of us have.
• 13:08Worked on for many years is that
• 13:10will you have to think about
• 13:12suicide in order to attempt suicide.
• 13:14But for whatever reason,
• 13:16there's a significant percentage of suicide
• 13:18attempt survivors who are saying no.
• 13:20Actually,
• 13:20I didn't do that.
• 13:21I skipped that step and
• 13:23so we were really there.
• 13:25Seems to be a lot of parallels
• 13:27between these sort of budget overages.
• 13:30And suicide.
• 13:30And so this is in many ways that
• 13:33sort of the pre genitor of a lot of
• 13:36our collaboration together to start
• 13:38thinking about suicide in unique ways.
• 13:42And this is where I started to become
• 13:45familiar with dynamical systems theory
• 13:47and one of the key concepts within
• 13:50that mathematical sort of branches.
• 13:52This notion of emergence and emergence
• 13:55just in general kind of refers to the
• 13:58process of coming to be how something.
• 14:01That was not there before
• 14:03eventually comes into being,
• 14:04and this sort of mathematical
• 14:06concept of emergence is largely
• 14:08founded upon four key assumptions,
• 14:10the first of which is that change
• 14:13within a system is constant.
• 14:15The second is that the whole is
• 14:17greater than the sum of its parts.
• 14:20The third is that each component
• 14:22of a system depends upon the other
• 14:25components of that system to function,
• 14:27and then 4th that complex systems
• 14:30behave in non proportional ways.
• 14:32And as John and I started
• 14:34working together over the years,
• 14:36we started to apply these principles
• 14:38of emergence to our study of
• 14:40suicide and what we're finding
• 14:42is that it seems to be a very,
• 14:44very useful way to conceptualize and
• 14:46think about suicide risk in general.
• 14:49And as we'll talk about later
• 14:51on in the presentation,
• 14:52and I think it is a lot of patients for
• 14:55clinical practice as well as suicide
• 14:57prevention strategies in general.
• 14:59And so I'd like to do is just.
• 15:02Spent some time talking about how
• 15:05these four concepts seem to apply to
• 15:08suicide and how this might change,
• 15:10or at least advance or upgrade our
• 15:13thinking about preventing suicide.
• 15:15And so first,
• 15:16let's talk about the first
• 15:18assumption that change is constant.
• 15:20So this is this notion that
• 15:22suicide risk is a dynamic concept
• 15:25that it's always ever changing
• 15:28really isn't necessarily
• 15:29a new idea. I mean this.
• 15:32This is certainly been written
• 15:34about for many, many decades.
• 15:36Anyone who's a clinician knows this to be
• 15:39true that patients are low risk sometimes,
• 15:41but then they in turn to a high risk state
• 15:44and then they return to a little risk stage.
• 15:48But it's only been within the past
• 15:50decade or so that I think technological
• 15:52advances and data analytic methods have
• 15:55gotten to a point where we're able to
• 15:57quantify it and describe these change
• 15:59processes with sufficient detail.
• 16:01And so, for instance, this is a.
• 16:03This is some of the results from.
• 16:06Study that Evan Kleiman at Ruckers did
• 16:09using ecological momentary assessment or MA
• 16:11where they were basically assessing suicidal
• 16:14individuals four to six times per day.
• 16:16They would assess multiple variables,
• 16:18but this one on the screen right now.
• 16:21Ask them about several aspects of suicidal
• 16:24ideations and so they were able to map
• 16:27out fluctuations in suicidal ideations
• 16:29over the course of several weeks,
• 16:31and what you see here.
• 16:33This is all of the participants
• 16:35in that study.
• 16:37Aggregated together in a few of the
• 16:39cases are kind of drawn out and put into
• 16:43color to help demonstrate the concepts.
• 16:45What you see is that variability.
• 16:48Is the rule not the exception an one of the
• 16:52more striking findings from climate studies,
• 16:55at about 90% of participants who were
• 16:58suicidal experience to 1 standard deviation
• 17:00or larger shift in their suicidal thinking
• 17:03within the span of just a few hours.
• 17:06And so you can see that here in some
• 17:09places where there are these sudden
• 17:11dramatic increases in suicide risk,
• 17:14and then sometimes it's the reverse.
• 17:16There's a sudden and dramatic
• 17:18drop in suicide risk.
• 17:20And every so often there's even these,
• 17:22like very brief peaks that seem to
• 17:25correspond with an acute suicidal crisis.
• 17:27And so this this we're seeing
• 17:29this pattern over and over again,
• 17:32not only with the MA studies with some of
• 17:35the work that I've done modeling change
• 17:38in suicidal ideations on a weekly basis.
• 17:40And now a daily basis.
• 17:42You know,
• 17:43an intensive outpatient programs,
• 17:45inpatient settings,
• 17:46or even outpatient therapy
• 17:47settings that these fluctuations
• 17:49and suicide risk are the rule.
• 17:51Now why this is sort of important
• 17:53for us to think about from a clinical
• 17:56perspective is that it provides an
• 17:59explanation for why suicide risk screening
• 18:02and assessment is so notoriously unreliable,
• 18:04and so here on the screen.
• 18:07Now I have two hypothetical patients,
• 18:09one depicted in Black,
• 18:11one depicted in red.
• 18:12You see that they are experiencing
• 18:15fluctuations in suicide risk
• 18:17that are comparable to what we
• 18:19saw in the previous screen.
• 18:21Now, if these two patients were to say,
• 18:24come into an emergency Department
• 18:26on the same day,
• 18:27and that day correspond to the arrow
• 18:29on the left, and we assess them,
• 18:32perhaps whatever scale that you use with pH,
• 18:34D 9, the ask the easy save the Columbia,
• 18:37whatever it is,
• 18:38what we would conclude on that day
• 18:41is that the person in black
• 18:43is higher risk for suicide.
• 18:45Been the person in red, and so we might.
• 18:48Perhaps we would consider
• 18:49hospitalizing the person in black,
• 18:51but may be discharged.
• 18:52Person in red to home.
• 18:54Now if we slide to the right of
• 18:56this graph to that second arrow,
• 18:58if those same two individuals had come
• 19:00into the emergency Department at a
• 19:02different point in their time series,
• 19:04we would conclude the exact reverse
• 19:06person in red is the higher
• 19:08risk than the person in black.
• 19:10And so maybe we would hospital as a
• 19:12person in red instead of the person in black.
• 19:15Now if we look at the entire
• 19:17T of the time series, however,
• 19:19what you would see is that
• 19:21they actually do not really.
• 19:22Differentiate from each other.
• 19:24There's not a whole way a real
• 19:26way to kind of say that one person
• 19:28is higher risk than the other.
• 19:30They're both experiencing these
• 19:32Upson Downs and this is,
• 19:33I think a critical new.
• 19:35Idea that many of us are starting
• 19:38to find in our studies over and
• 19:40over again is that suicide risk,
• 19:43especially suicidal ideations,
• 19:44isn't really all that useful for
• 19:47distinguishing between patients
• 19:48who are higher and lower risk.
• 19:50And that's largely because
• 19:51suicidal ideations suicide risk
• 19:52is a within person phenomenon.
• 19:54People have good days and bad days,
• 19:57and so we perhaps need to start shifting.
• 20:00I'm thinking away from some of
• 20:02our research models of clinical
• 20:04decision rubrics that are trying
• 20:07to distinguish which patients are
• 20:09going to attempt suicide in which
• 20:11are not and move towards a thinking
• 20:13of when is a given individual
• 20:16vulnerable or in high probability
• 20:18state to engage in suicidal behaviors
• 20:21and so along this line of sort of
• 20:23within person change processes.
• 20:25We started to now conduct several
• 20:28studies to really map out and
• 20:31mathematically model these
• 20:32fluctuations in suicide risk overtime,
• 20:35under the assumption that maybe
• 20:37there are certain patterns of change
• 20:40within individuals that might signal
• 20:42the near term emergence of suicidal
• 20:45behavior such that the concept
• 20:47of a warning sign, for instance,
• 20:50isn't necessarily the occurrence or
• 20:53the existence of a given variable or a given.
• 20:57Behavior,
• 20:57but perhaps the warning sign is a
• 20:59change process that departs from in
• 21:02individual standard baseline change process.
• 21:04So on the screen.
• 21:06Here I give an example.
• 21:08This on the left hand side of this time
• 21:11series we see fluctuations and suicide risk.
• 21:14Then at about the midpoint
• 21:16of the time series,
• 21:18what you see is those fluctuations in
• 21:21suicide risk become more and more prominent.
• 21:24The amplitude of the Upson Downs increase.
• 21:27In magnitude,
• 21:27this is actually a pattern that
• 21:30we've seen down to separate
• 21:32studies that have been published,
• 21:34that it serves as an indicator that
• 21:37suicidal behavior is coming where in essence,
• 21:40from a mathematical perspective.
• 21:42What this signals is a loss
• 21:44of self regulation.
• 21:45The system is breaking down in the
• 21:48system is no longer able to contain
• 21:51and preserve its homeostatic balance,
• 21:53and so it's getting
• 21:55agitated which signals and.
• 21:57Upcoming emergent shift in states we've
• 21:59also found in some of our research
• 22:02that a different but related pattern
• 22:04relates to the frequency of fluctuation.
• 22:07So again, here on the left hand side
• 22:09of the time series we see these
• 22:12periodic spikes that occur in suicide
• 22:15risk and then at the midpoint those
• 22:18spikes in suicide risk occur much,
• 22:20much more often.
• 22:22This is a second dimension of
• 22:24dysregulation such that perhaps
• 22:26sometimes we see an increase in.
• 22:28Amplitude of fluctuations.
• 22:29Sometimes we'll see an increasing
• 22:31frequency of amplitudes,
• 22:33and we might also see a combination
• 22:36of the two.
• 22:37So if we were to then apply
• 22:40this to clinical practice,
• 22:42what this would indicate is
• 22:44that repeated assessments of
• 22:46suicide risk overtime state,
• 22:48like every single therapy session,
• 22:50every single Med monitoring
• 22:52session might be able to detect
• 22:54within person change processes,
• 22:56which could then signal hey,
• 22:59this is a departure from the norm.
• 23:02This is not a typical change pattern
• 23:04for this particular individual.
• 23:06They're becoming more dysregulated and so
• 23:08perhaps that would prompt an intervention.
• 23:11We're actually working now.
• 23:12We have a narrow one.
• 23:14This is kind of the key idea behind
• 23:17the work that we're currently
• 23:19doing on a grant to look at.
• 23:22Could you potentially develop novel
• 23:24detection and alert systems that
• 23:26could be implemented in clinical
• 23:28practice that now takes into
• 23:30account multiple time points in
• 23:32applies new mathematical models?
• 23:34To look at within person, change to signal.
• 23:36Hey,
• 23:36this patient is now at a higher
• 23:39probability risk state and we
• 23:41should probably intervene with them.
• 23:44Now the 2nd two concepts of emergency.
• 23:46I want to talk about here is the
• 23:48whole is greater than the sum of its
• 23:51parts in each component of a system.
• 23:54Depends on the other components to function,
• 23:56and so this is.
• 23:57These are sort of two related ideas
• 23:59that I think is also highly relevant
• 24:01to our work and suicide prevention.
• 24:04And to demonstrate these two concepts,
• 24:06I'm actually going to share with you
• 24:08a video that I found very sort of
• 24:11informative in my own thinking about
• 24:13systems theory and dynamical systems.
• 24:15So in this video I want to say maybe
• 24:18like a minute and a half long.
• 24:21It's sort of outlines the process
• 24:23of emergence for hurricane and so
• 24:25hurricanes go through several.
• 24:27We've sort of classified different stages,
• 24:29whether that's tropical depression to
• 24:30a tropical storm to a full blown hurricane,
• 24:33and then we have different
• 24:35categories of hurricanes.
• 24:36But even though there are these unique
• 24:39sort of stages that we have just sort
• 24:42of arbitrarily come up with as humans.
• 24:45There are these sort of continuous
• 24:47change processes that lead you know
• 24:49weather patterns to kind of move through
• 24:51these different arbitrary stages.
• 24:53So as we watch this video,
• 24:55the thing I wanted to just pay attention
• 24:58to is again this notion of complexity.
• 25:01How different variables are
• 25:02interacting with each other and
• 25:04also pay attention to it will
• 25:06show different arrows moving in
• 25:08different directions to to
• 25:10convey some these concepts.
• 25:14Depends upon the following conditions
• 25:17coinciding Alarge still and warm ocean
• 25:19area with the surface temperature
• 25:22that exceeds 26.5 degrees
• 25:23Celsius over an extended period.
• 25:25This allows a body of warm air to
• 25:30develop above the ocean surface.
• 25:33Low altitude winds are also
• 25:35needed to form a tropical cyclone.
• 25:38As air warms over the ocean,
• 25:40it expands, becomes lighter and rises.
• 25:43Other local winds blow in to
• 25:45replace the air that has risen.
• 25:47Then this air is also warmed and rises.
• 25:52The rising air contains huge amounts of
• 25:54moisture evaporated from the ocean surface.
• 25:57As it rises, it cools,
• 25:59condensing to form huge clouds about
• 26:0210 kilometers up in the troposphere.
• 26:05More warm air rushes in and
• 26:07rises drawn by the draft above.
• 26:10The amazing drafts are there carrying
• 26:13moisture high into the atmosphere
• 26:15so that these clouds eventually
• 26:17become very thick and heavy.
• 26:20Condensation then releases the latent
• 26:22heat energy stored in the water vapor,
• 26:25providing the cyclone with more power.
• 26:27This creates a self sustaining heat cycle.
• 26:31Drawn further upwards by the
• 26:33new release of Energy,
• 26:35the clouds can grow to 12
• 26:38to 15 kilometers high.
• 26:40The force created by the Earth's
• 26:42rotation on a tilted axis,
• 26:44the Coriolis effect,
• 26:46causes rising currents of air to spiral
• 26:49around the center of the tropical cyclone.
• 26:52It is at this stage that the cyclone matures
• 26:55and the eye of the storm is created.
• 26:58As the air rises and cools.
• 27:00Some of this dense air descends
• 27:02to form the clear.
• 27:04Still,
• 27:04I as the cyclone rages around it.
• 27:10So a couple of key ideas in here is that
• 27:15as the hurricane is a cyclone emerges.
• 27:19It's not simply that you have to
• 27:21have wind speed at a certain level
• 27:24temperature at a certain level have
• 27:27to be over a suitable body of water.
• 27:30There's a lot of conditional
• 27:31interdependence amongst these variables.
• 27:33It's the right wind speed at the
• 27:36right temperature in the right place,
• 27:38under the right conditions,
• 27:40and if any of these variables shift
• 27:43or are they off just a little bit,
• 27:45it can actually dissolve the
• 27:47entire system when aspect about.
• 27:49Hurricane formation that I learned
• 27:51about in the past few years doing
• 27:54this work is that wind speeds
• 27:56Tradewinds at very high altitude.
• 27:58Have to actually be in this
• 28:00very narrow band of velocity.
• 28:02If the trade winds are not fast enough,
• 28:05it does not provide enough momentum
• 28:07for the hurricane to actually move in.
• 28:10Things break apart.
• 28:11Conversely, if the wind,
• 28:13the trade wind speeds are too
• 28:15high of a velocity,
• 28:17it doesn't allow enough time for all
• 28:19of these other processes to congeal.
• 28:22And for man,
• 28:23it basically breaks up the system
• 28:25and So what we have here then.
• 28:28Is a system where lots of variables
• 28:31are interacting with each other
• 28:33in those variables are changing in
• 28:35very key ways that we start to see
• 28:38feedback loops and if those feedback
• 28:40loops in those change processes
• 28:42aren't in the right configuration,
• 28:44then in essence the hurricane doesn't
• 28:46emerge and so this seems to be
• 28:49something comperable to suicide that
• 28:51the conceptual theoretical model that
• 28:53we use in a lot of our research is
• 28:56referred to as the suicidal mode,
• 28:58and this is a graphic depiction of that.
• 29:02And so a lot of people have
• 29:04sometimes commented clinicians like,
• 29:06well,
• 29:06they're like all these sort of
• 29:08arrows pointing to different pieces.
• 29:10You know,
• 29:10it's sort of like a complicated
• 29:12model is compared to some of the more
• 29:15classic like cognitive models or
• 29:17the biomedical biobehavioral model,
• 29:18so I've seen.
• 29:19But what this really captures in
• 29:21the suicidal motives,
• 29:22this notion of a network that
• 29:24lots of different variables are
• 29:26influencing each other,
• 29:27and they have to influence each other
• 29:30in the right way at the right time.
• 29:33In order for the conditions to be
• 29:35right for suicidal crisis to emerge,
• 29:38and then critically for suicidal
• 29:40behaviors to emerge,
• 29:41and some of the newer research
• 29:43that's being done right now,
• 29:45particularly IMA researchers that are
• 29:47using network computational models,
• 29:49is showing us that there actually
• 29:51are multiple pathways suicide.
• 29:53One of the limitations of a lot
• 29:56of our traditional thinking about
• 29:58suicide is that we focus on.
• 30:00One or two risk variables or factors.
• 30:03Hopelessness, perceived,
• 30:04burdensomeness thwarted belonging.
• 30:06This fearlessness about death or
• 30:08some of the ones that are more
• 30:11popular that are more well known.
• 30:13But what we're seeing now using these
• 30:17more complex dynamic models is that,
• 30:19you know,
• 30:20hopeless hopelessness is a primary
• 30:22factor in firmly some suicidal people.
• 30:25Perceived Burdensomeness is an
• 30:27important variable for a different group
• 30:29of individuals, but in essence.
• 30:31The reason may be that we haven't
• 30:33been better at preventing suicide
• 30:35as we create these monolithic
• 30:37models that don't have sufficient
• 30:39complexity and therefore don't allow
• 30:41us to fully understand the different
• 30:43processes by which suicide might be
• 30:45a common endpoint of another way of
• 30:47saying this is sort of like there.
• 30:49There might be multiple roads to Rome,
• 30:52but because we assume there's
• 30:53only one Rd to Rome,
• 30:55we're not doing a good job setting
• 30:57up the barriers on all of the
• 31:00many different roads to block
• 31:01their pathway to get to room.
• 31:06I'll say one other point about
• 31:08that that I didn't mention is that
• 31:11this notion of interdependence,
• 31:13where all of these many different factors
• 31:15and variables influence each other,
• 31:17and so you know depression.
• 31:19Hopelessness, for instance,
• 31:21if you experience an increase
• 31:22in your hopelessness,
• 31:24we might reasonably expect
• 31:25an increase in depression.
• 31:27Conversely, an increase in depression
• 31:29might reasonably be expected
• 31:30to increase once hopelessness,
• 31:32and so we really can't sort
• 31:34of parse out and separate.
• 31:37The notion of depression and hopelessness
• 31:39and other risk variables because
• 31:41they all depend upon each other and
• 31:43we have a study under review right
• 31:45now or what we did is we use the
• 31:47best scale for suicidal ideations.
• 31:49And instead of modeling the scale for
• 31:51suicidal ideations as a single total score,
• 31:53which is how most of us are traditionally
• 31:56used it in looking at that change process.
• 31:58Overtime,
• 31:58what we've done is we actually
• 32:00took each of the individual items
• 32:02and we said all of the items.
• 32:04In a sense they can change
• 32:06in any way they want.
• 32:08Independence of each other and let's
• 32:10see what happens in what we found
• 32:13was that can sort of consistent with
• 32:15the notion of emergency complexity
• 32:18be different asset aspects or facets
• 32:20of suicidal ideations under some
• 32:23conditions were tightly correlated
• 32:25with each other such that an increase
• 32:27in the wish to die corresponded with
• 32:29an increase in suicidal intent.
• 32:31But there are other cases in
• 32:34other circumstances where there
• 32:35was no relationship between these
• 32:37two otherwise supposedly.
• 32:39Correlated facets of suicidal ideations,
• 32:41and So what we're starting to see,
• 32:43then,
• 32:44is that there might be these sort
• 32:46of different networks and Inter
• 32:48related webs of risk variables and
• 32:51processes that have implications.
• 32:53For prevention is from a network perspective.
• 32:56We can identify some of the key
• 32:58variables that are most tightly
• 33:00influential and interconnected
• 33:02with other aspects of the network.
• 33:05What we could start doing, then is targeting.
• 33:08This is the one thing that if we
• 33:11can change this one variable,
• 33:13it has this downstream effect that
• 33:16ripples out and starts to impact
• 33:18other risk factors for suicide,
• 33:20leading to a more efficient
• 33:22reduction in suicidal crises.
• 33:24Conversely,
• 33:24if we're not targeting the
• 33:26right piece of the network,
• 33:28we might have interventions that are really,
• 33:31really good at changing a
• 33:33particular risk variable,
• 33:34but that change isn't sufficiently
• 33:36connected to other members of the network,
• 33:39and so we don't see that downstream
• 33:43effect in reduction in suicide risk.
• 33:46So the 4th concept that I wanted to
• 33:49talk about is that complex systems
• 33:51behave in non proportional ways and
• 33:53what this refers to is that you can.
• 33:56You can sometimes put a whole lot
• 33:58of stuff into a system to change
• 34:01that system and in the system just
• 34:04sort of ignores all of the input.
• 34:06It doesn't really change how it behaves.
• 34:09Conversely you see the opposite
• 34:11where we don't really do a whole lot
• 34:14in terms of input or intervention.
• 34:16And then the the system will
• 34:18dramatically change in how
• 34:20it behaves, and so a little bit
• 34:23and sometimes paying big dividends.
• 34:26Is a whole lot of input doesn't
• 34:28really seem to have any effect.
• 34:30This is like one of those really
• 34:32frustrating aspects of clinical practice
• 34:33where some of our patients that come in.
• 34:35We just really hit them with a lot of
• 34:38interventions. We really are group,
• 34:39you know they do therapy.
• 34:41They do meds.
• 34:42Did you all sorts of stuff and
• 34:43then we don't really see a change,
• 34:46whereas other patients it
• 34:47seems like they come in.
• 34:48We have clinicians or like I didn't
• 34:50really even do anything and now all of
• 34:52a sudden there dramatically better and
• 34:54we really don't understand why there
• 34:56are these differential response patterns.
• 34:57In one of the possibilities of that well,
• 35:00complex systems behave
• 35:01in non proportional ways.
• 35:03There isn't any guarantee that what you
• 35:05put in will have a proportional output,
• 35:08and so I have another video to kind of
• 35:11display and highlight some of this.
• 35:13This is a gymnastics example
• 35:15before I introduces further.
• 35:16I'll encourage you might want to adjust
• 35:18your volume a little bit down on this one.
• 35:22The volume on this video tends to
• 35:24be a lot louder than the last one,
• 35:27and so perhaps we can avert some
• 35:29panic attacks here if we.
• 35:31Give you a little bit of warning.
• 35:33This is a 30 second video.
• 35:35This is Simone Biles.
• 35:37I'm sure everybody recognizes her
• 35:39and she's doing one of her routines
• 35:41and I found that as I was really
• 35:43trying to understand the concepts of
• 35:45non proportionality that gymnastics
• 35:47was like this.
• 35:48Really really great way of kind
• 35:50of capturing it.
• 35:51Embodying this notion of non proportionality.
• 35:53So let's let's watch your 32nd routine here.
• 36:19That's I warned a second Trisha
• 36:21that usually this video and
• 36:23practice runs works beautifully.
• 36:25And then during the live
• 36:27presentation this happens.
• 36:28Let me try an alternative approach.
• 36:30Going to switch to you.
• 36:33Another video here.
• 36:35Let's see if this works better.
• 36:45Share this instead.
• 36:49The same video I'm just going to use
• 36:52the raw video and maybe that will.
• 36:54Correct the problem, right? Let's try now.
• 37:32Right, so I shared that video because
• 37:36I think it really helps to capture
• 37:40that notion of that non proportionality
• 37:43where here we have this undisputed,
• 37:47very accomplished.
• 37:48Athlete who's very good at what she does.
• 37:52Ann is 1 lots of medals and is very very
• 37:56accomplished and is doing quite well on her.
• 38:00Routine for the first 29 1/2 seconds
• 38:02and then she does a split an she lands
• 38:05on the beam and up until that moment is
• 38:08sort of like we had absolutely no way of
• 38:11knowing that she was actually going to fall.
• 38:14And from the time that we see her
• 38:17legs shooting up for her trying to
• 38:19touch her balance to her feet being
• 38:21on the ground with the fall,
• 38:23it's about half of a second
• 38:25so it's very dramatic.
• 38:27Very rapid,
• 38:27seemed to come out of the blue despite.
• 38:30All of this good performance
• 38:32up until that point.
• 38:34And so this really sort of aligns
• 38:36with that notion of the projects
• 38:39that John was working on with NASA,
• 38:41as well as some of what we
• 38:44know to be true with suicide.
• 38:46That it seems to sometimes come out of
• 38:49the blue without much advanced notice.
• 38:52Well that in essence this fall
• 38:54here that we saw in gymnastics.
• 38:57It's sort of like we don't really
• 39:00know what contributed to it.
• 39:01We can probably infer that maybe she
• 39:04just landed in a way where her center
• 39:07of gravity was perhaps just a little
• 39:09bit off from where she wanted it to be.
• 39:12Maybe her,
• 39:13maybe her foot was just twisted in a slightly
• 39:16different way than what she was expecting.
• 39:18We really don't know, but a very,
• 39:21very slight shift in the mechanics of
• 39:23the move led to this dramatic outcome,
• 39:25and suicide seems to function
• 39:27in the same way.
• 39:29So mathematically,
• 39:29this is referred to as a catastrophic
• 39:33change where the system suddenly
• 39:35shifts from its prior behavior to a
• 39:38completely different behavioral pattern,
• 39:40and we've started to think about
• 39:43catastrophic change processes and apply
• 39:46it to our work with suicidal individuals.
• 39:49And we're actually finding it.
• 39:51It's a very useful model,
• 39:53and the reason we think this is
• 39:57the case is that.
• 39:59Again, from assistance.
• 40:00Perspective to have these sort of
• 40:02sudden catastrophic changes,
• 40:04there's another inherent assumption
• 40:06embedded within non proportionality,
• 40:07which is the notion of a tipping
• 40:10point where there exists two distinct
• 40:12distinct and discrete States and in
• 40:15between these two separate states of
• 40:18being is this tipping point such that
• 40:20if you fall just shy of the tipping point,
• 40:24you stay in one particular state.
• 40:27If however you just barely move
• 40:29over the cusp of that tipping point.
• 40:32You can shift into a new state,
• 40:35and indeed there's now an accumulation
• 40:37of evidence indicating that this is
• 40:39actually how suicide risk works.
• 40:41So this is a study here, but Tracy Witty.
• 40:44They've done.
• 40:45Taxa,
• 40:45metric analysis with large sample of
• 40:47suicidal individuals and found that
• 40:49there was a subgroup of suicidal
• 40:51individuals who
• 40:52had very, very high risk for suicide,
• 40:54and they were categorically
• 40:56distinct from everybody else.
• 40:57These results were actually replicated by
• 40:59team down at Baylor College of Medicine,
• 41:02with psychiatric inpatients,
• 41:03and found that.
• 41:04Among psychiatric inpatients,
• 41:05there's a small group that was really,
• 41:07really high risk for suicide and then
• 41:09everybody else was in a lower risk state.
• 41:12We've seen this as well.
• 41:13An epidemiological study.
• 41:14This is a. Project done better.
• 41:17Army Stars research team led by Ron Kessler.
• 41:20What they did was they took a large sample
• 41:22of soldiers who had been discharged
• 41:25from psychiatric inpatient care,
• 41:26used machine learning to split
• 41:28them up into vent tiles.
• 41:30And then they looked at they rank
• 41:32ordered the vent tiles in terms of
• 41:35overall risk level and then mapped
• 41:37onto who and where are the suicide
• 41:40deaths occur and what they found
• 41:42was that over half of the suicide
• 41:44deaths occur in the first subgroup.
• 41:46The 1st event tile.
• 41:48This is on the left hand side
• 41:50of the screen and you see the
• 41:52rest of the 19 vent tiles.
• 41:54There's not as much difference between
• 41:56them is a little bit of variability.
• 41:58Groups two and three or a little
• 42:00bit higher risk than the others,
• 42:02but not dramatically so.
• 42:04So this also suggests that there's
• 42:06this sort of tipping point.
• 42:07There's one small subgroup that accounts
• 42:09for more than half of the suicides,
• 42:11and then there's everybody else
• 42:13but the rest of the 95% of the
• 42:15population now on the right hand side.
• 42:17This is not hypothetical.
• 42:19Model where if suicide risk was a continuum,
• 42:22this sort of unidimensional gradual change
• 42:25process moving from lower to higher risk.
• 42:27The graph on the right hand side
• 42:30is what Kessler should have found.
• 42:32There should have been the sort of
• 42:35incremental changes in the probability
• 42:37of suicide deaths within each ventile,
• 42:39but that's of course not what they found.
• 42:43We've since mathematically modeled
• 42:44suicidal Ideations on a session by
• 42:47session basis within psychiatric
• 42:49outpatient psychiatric outpatients,
• 42:51and what we found was that patients
• 42:53who had a history of multiple suicide
• 42:57attempts at the start of treatment,
• 43:00they had a unique change process that
• 43:03was characterized by sudden dramatic
• 43:05Upson Downs in suicidal ideations
• 43:07on a session to session basis.
• 43:10It was like a sawtooth pattern.
• 43:13In that that sawtooth pattern
• 43:15is consistent with the existence
• 43:17of two discrete states of risk,
• 43:18where in essence the patients are
• 43:21bouncing back and forth between
• 43:22low and high risk that they're
• 43:24not doing it in a gross, gradual,
• 43:26slow manner because there's a tipping point,
• 43:28and so they can be.
• 43:30It's sort of like a binary or
• 43:32a one or a zero.
• 43:34There is no in between,
• 43:35and there is a distinct subgroup of patients
• 43:38who seemed to conform to this change process,
• 43:40and So what that means is,
• 43:42I think.
• 43:43Our traditional ways of thinking
• 43:45about suicide risk in this sort
• 43:48of unit dimensional hierarchy,
• 43:49where at the bottom we have
• 43:51things like I don't
• 43:53want to live anymore.
• 43:55I don't want to be alive,
• 43:57moving to active suicidal ideations,
• 43:59moving to planning,
• 44:00moving to preparatory behavior,
• 44:02and then at the highest level,
• 44:04suicidal behavior where there's at least
• 44:07a significant subgroup of individuals
• 44:09who do not conform to this pathway.
• 44:11In this hierarchical model,
• 44:13some patients do seem to follow
• 44:15this gradual change process.
• 44:17But there's another substantial minority.
• 44:19At the very least,
• 44:20that don't follow this,
• 44:22and So what this means is we
• 44:24probably need to change how we
• 44:26conceptualize suicide risk.
• 44:28And so we've now informed
• 44:30by the notion of emergence,
• 44:32move towards a 3 dimensional model of suicide
• 44:35risk and and this figure on the screen.
• 44:38With this sort of curved claim of
• 44:40suicide risk is what's referred
• 44:42to as a cusp catastrophe model.
• 44:45So this provides us with.
• 44:47A way to understand why is it
• 44:49that some individuals seem to
• 44:51have this gradual change process.
• 44:53They sort of slowly move through these
• 44:56incremental stages towards suicide,
• 44:57whereas other individuals seem to
• 44:59very suddenly shift from a low
• 45:02risk state to a high risk state,
• 45:04and so let me demonstrate several of
• 45:07the kind of a key change processes here.
• 45:10This first pathways pathway a.
• 45:12This is sort of the classic
• 45:14way of thinking about suicide,
• 45:16it's unidimensional.
• 45:16As you move up the surface plane that
• 45:19corresponds to increasing suicide risk,
• 45:21and these individuals then follow back down.
• 45:23They kind of move up and down Hill,
• 45:26and the way that this typically
• 45:28manifests is those fluctuations in
• 45:29suicide rates that have inclined in,
• 45:31as noted in that,
• 45:32like very first one of the very first
• 45:35slides that I've showed you today.
• 45:37And so we do see that people
• 45:39have good days and bad days that
• 45:42they have these fluctuations,
• 45:43but they tend to be somewhat
• 45:46smooth and gradual.
• 45:47These individuals often come
• 45:48in to mental health care.
• 45:50We typically diagnosed him with
• 45:52like mood disorders and this is,
• 45:54I think the sort of preponderance of
• 45:56the people that we have traditionally
• 45:58thought about when it comes to suicide.
• 46:01We didn't have a different group of
• 46:03individuals who are located in a
• 46:05different place on this behavior surface.
• 46:08They also move up and down Hill,
• 46:10but because they are located in a
• 46:12place of the this curved plane that
• 46:15corresponds to a steeper slope,
• 46:17they experience much more rapid
• 46:19increases in rapid decreases in suicide risk.
• 46:22It's no different than if we were
• 46:24to kind of like roll a ball up
• 46:27and down a really steep Hill.
• 46:29The ball would start to roll up very quickly,
• 46:32stop, and then come back down very rapidly.
• 46:35In the steeper the slope,
• 46:37the faster that return process would be.
• 46:40Now this this sort of sub group of
• 46:43individuals seems to characterize
• 46:45those who are diagnosed with
• 46:47borderline personality disorder and
• 46:49have other related conditions that
• 46:51are characterized by deficient self
• 46:53regulatory processes where we're
• 46:55more likely to see a sawtooth pattern
• 46:58and then one of the key aspects of
• 47:02this is that that up and down that
• 47:05dramatic up and down actually increases
• 47:07their vulnerability for suddenly
• 47:09shifting from a low risk state.
• 47:12To a very high risk state from the bottom
• 47:14of the surface to the top of the surface,
• 47:17and that was actually what David wrote,
• 47:19and I found in some of our initial
• 47:22research with multiple attempters
• 47:23and we now have a new study we just
• 47:26published in behaviour research and
• 47:27therapy in the past year where we
• 47:30took patient psychiatric out patients
• 47:31who had attempted 1 one time in their
• 47:33life and then we mapped out their
• 47:36fluctuations in suicidal ideations over
• 47:37therapy and what we found was that the
• 47:40patients who had larger fluctuations
• 47:41on a session by session basis.
• 47:44Or significantly more likely to
• 47:46attempt suicide during treatment
• 47:48or soon after treatment.
• 47:50And so these individuals also,
• 47:52I think,
• 47:53are more likely to come in for
• 47:55mental health treatment.
• 47:57And so this is where a lot of our
• 47:59work that Seth mentioned in his
• 48:02introduction on cognitive behavioral
• 48:04therapy for suicide prevention,
• 48:06I think has been really impactful.
• 48:08I think other treatments in addition
• 48:10to BTI but also dialectical behavior
• 48:12therapy has a lot of implications here.
• 48:15What we were finding is that these
• 48:17treatments seem to sort of calm the storm
• 48:20where the fluctuations in suicidal ideations.
• 48:23Over the course of treatment tend
• 48:25to settle down so that there is a
• 48:28decreased probability that a person
• 48:29will suddenly shift to a high risk
• 48:32state where suicidal behavior
• 48:34is more likely to occur.
• 48:35The key aspect about some of our
• 48:38findings with BTI is that we see
• 48:40treatment effects very quickly,
• 48:42typically within three months
• 48:43of starting treatment,
• 48:44and so not only is it a long lasting
• 48:47effect up to two years post baseline,
• 48:50but we tend to see pretty early
• 48:53separation between treatment groups.
• 48:55Another related intervention
• 48:56that that we've worked on a lot
• 48:59as crisis response planning,
• 49:00which serves as sort of like the kind
• 49:03of a precursor to safety planning,
• 49:06which a lot of people are familiar with.
• 49:09Crisis response planning was,
• 49:10apart of BTE,
• 49:11and then we subsequently extracted
• 49:13this procedure and tested it as a
• 49:16standalone intervention and emergency
• 49:17department's as well as walk in clinics
• 49:20for UNE scheduled crisis evaluations.
• 49:22Whereas what we do is we hand
• 49:25write on an index card what?
• 49:27Ultimately,
• 49:28amounts to sort of a self regulation plan.
• 49:30How to know when you're getting upset,
• 49:33how to know when you're heading uphill.
• 49:35And then.
• 49:36Here's a list of things that you can
• 49:39do to stop yourself from progressing
• 49:41towards a high risk state into self
• 49:44regulate to kind of calm yourself back down.
• 49:46We completed a randomized controlled
• 49:48trial that we published in 2017
• 49:51comparing crisis response planning
• 49:52to treatment as usual and found 76%
• 49:54reduction in suicide attempts over
• 49:56the six month follow up period.
• 49:58And so this.
• 50:00Very simple.
• 50:00Ateji that definitely takes 30 or 60 minutes.
• 50:04We found this actually quite potent
• 50:06for averting suicidal behaviors.
• 50:08But then there's one last group that my
• 50:11research programs increasingly focused on.
• 50:13This is this pathway.
• 50:14See, this is the.
• 50:16This is the performance escape,
• 50:18the catastrophic change group,
• 50:19for whom suicidal behaviors
• 50:21seemed to come out of the blue.
• 50:23And you can see their position
• 50:25on the behavior surface here.
• 50:27What happens with this group is when they
• 50:30experience an increase in suicide risk.
• 50:32It's very gradual.
• 50:33It's not very pronounced,
• 50:35so maybe they only get to a point where
• 50:38they're wishing that they weren't.
• 50:40Around anymore they would.
• 50:42They think they'd be better off dead,
• 50:44but in some cases they're not
• 50:46manifesting active forms of suicidal
• 50:48ideations or even planning.
• 50:50But then what happens is they hit
• 50:52this curve in the behavior surface
• 50:55and they sort of like teleport up
• 50:58to the top where they see this very
• 51:00sudden shift in the probability
• 51:02for suicidal behavior in these
• 51:04individuals are much less likely
• 51:06to come to mental health care,
• 51:08and it's because for the most part
• 51:11they're existing in a low risk state.
• 51:14And so when we say get help,
• 51:16watch out for these warning signs
• 51:18of suicide that largely focus
• 51:19on active suicidal ideations.
• 51:21It doesn't apply to these individuals.
• 51:23This isn't the nature of their
• 51:25subjective experience of suicide risk,
• 51:26and so we end up missing this group,
• 51:29and we keep encouraging.
• 51:30Go get help to get help to get help,
• 51:33but this is a group for whom that is
• 51:36a message that just isn't hitting the
• 51:38target in the way that we needed to.
• 51:41So we know that about 40 to 50% of those
• 51:44who attempt suicide sort of skipped.
• 51:47The planning stage and we have
• 51:49some newer data coming out.
• 51:51Showing that manifestations of suicide
• 51:53risk do not necessarily involve active
• 51:56thinking about suicide where we are
• 51:58now able to capture using a scale
• 52:01called the Suicide cognition scale.
• 52:03Things like I can't take this anymore.
• 52:06I deserve to die.
• 52:07No one can help me solve my problems.
• 52:11These are perhaps thought processes
• 52:13that signal this is someone who's
• 52:16at elevated risk for suicide,
• 52:18but they aren't necessarily experiencing
• 52:20suicidal thoughts in the way that we
• 52:23have traditionally conceptualized them.
• 52:25Traditionally asked about that,
• 52:27and so we're effectively missing
• 52:30a large group of people.
• 52:33And so this brings me into the kind of
• 52:35the last idea of what we're working on,
• 52:38which is the importance of means restriction,
• 52:41especially with firearm owners.
• 52:42So now a big portion of my research
• 52:45is really focused on understanding.
• 52:47Suicide risk amongst firearm owners
• 52:49and what we're finding is that there
• 52:52are really high risk subgroups of
• 52:54gun owners who they typically own.
• 52:57Guns for self protection purposes,
• 52:59but they are much,
• 53:00much more likely to engage in suicidal
• 53:03behaviors because we're starting to
• 53:05see now is that they have a number of
• 53:09biobehavioral vulnerabilities to suicide.
• 53:11They view.
• 53:12The world is a dangerous place.
• 53:14They perceive hostility and others
• 53:17actions even with no hostility.
• 53:19Exists in this seems to degrade
• 53:21cognitive control processes,
• 53:22which increases the propensity for them to
• 53:24engage in these dysregulated behaviors,
• 53:26and if they have a loaded weapon during
• 53:28moments of acute distress when they suddenly
• 53:31shift to a high probability high risk state,
• 53:34it weaponizes their emotional States
• 53:36and they die very quickly and we do
• 53:39not have a way to intervene with them,
• 53:41and so we do have now a new study.
• 53:44It's impressed with the American
• 53:46Journal of Public Health,
• 53:47that's hopefully will be coming out
• 53:49in the next few months, showing that.
• 53:51And we brought gun owners into our lab
• 53:55and we sat down with them and we talked
• 53:59about safe gun storage and using gun safes,
• 54:02trigger locks,
• 54:03things like that and what we found
• 54:05was that very simple conversations
• 54:08guided by motivational interviewing
• 54:10principles significantly increased
• 54:11the likelihood of gun owners actually
• 54:14using safe storage methods which we
• 54:16know from epidemiological data are
• 54:18correlated with dramatic reductions
• 54:20in suicide mortality.
• 54:22We now have an R61 that we're about
• 54:24to start enrollment on to further
• 54:26understand a lot of these processes,
• 54:29but I think this is in many ways,
• 54:31one of the most important directions in
• 54:34suicide prevention is to move outside
• 54:36of the healthcare system because there
• 54:38is a subgroup who don't have access,
• 54:40are not going to seek out mental
• 54:43health treatment,
• 54:44and during those moments of a
• 54:45sudden onset of acute despair,
• 54:47we just don't have time to find
• 54:49them in to intervene with them.
• 54:52So we need to change the environment.
• 54:55So that we suicide proof their lives in,
• 54:58decrease the probability of a fatal outcome.
• 55:03It's all,
• 55:04so that's that's kind of where we're at.
• 55:06With the research a lot more
• 55:09to be done understandably,
• 55:10but I'm happy to take any questions that
• 55:13anyone might have in the remaining time,
• 55:15and likewise,
• 55:16if you have to drop off,
• 55:18totally understandable and by all
• 55:20means feel free to shoot me an email
• 55:23afterwards with any questions or
• 55:24follow up thoughts that you might have.
• 55:27So thank you.
• 55:30Thanks Greg, if there any questions
• 55:33folks wanna put them into the chat
• 55:36chat room that would be great.
• 55:38Two things I'd never say.
• 55:39Seeing you give this talk before
• 55:41Craig so you know two things I
• 55:43thought were really interesting.
• 55:44One was on tipping points. You know,
• 55:46when you think about the medical model of.
• 55:50Of of that, you presented around suicidality.
• 55:52This is true, I think,
• 55:54across many different medical diseases.
• 55:56So kidney function,
• 55:57for instance, right?
• 55:58We learn in medical school that
• 56:00your kidney function declines
• 56:01gradually throughout your lifetime,
• 56:03and then boom,
• 56:04all of a sudden something happens
• 56:06and you cross a tipping point,
• 56:08and then all of a sudden you're
• 56:11creating starts going up.
• 56:12You know you're you're essentially
• 56:14entering kidney failure even
• 56:15though you've been in decline.
• 56:17There's been these issues all along.
• 56:19There's a tipping point.
• 56:20And same with coronary artery disease
• 56:23where your things are going on in the
• 56:25background for an extended period,
• 56:26sometimes something can happen early.
• 56:28Maybe you're taking a drug abuse.
• 56:30Maybe there's another disease
• 56:31process going on,
• 56:32but then boom,
• 56:33all of a sudden there's a myocardial
• 56:35infarction because something
• 56:36is occurring in the vessel that
• 56:38was either an it's similar to
• 56:39the 3D model you presented,
• 56:41where it can be gradual or all
• 56:43of a sudden something can happen
• 56:45and you just jump up the curve.
• 56:47So I really like that,
• 56:48right?
• 56:50Yeah, it's the notion of tipping
• 56:52points is yeah, it's widely used.
• 56:55In other words not widely used,
• 56:58but it's sort of applicable
• 57:00and well understood,
• 57:01and other branches of medicine but
• 57:04also other Sciences within biology,
• 57:06chemistry, physics.
• 57:07All of these processes are quite
• 57:10widely used in even the notion of
• 57:13catastrophic change processes.
• 57:14And so I think there's a lot of potential
• 57:18value and thinking about change.
• 57:21In these more complicated ways and
• 57:22the good news is that we do have a
• 57:25lot of precedent and a lot of examples
• 57:27from other scientific disciplines,
• 57:29and so it's really now just sort of
• 57:31a matter of applying the methods
• 57:32in the computational concepts
• 57:34to the work that we do.
• 57:37Then there's a question from Tammy.
• 57:41Say I'll skip over the very nice words
• 57:43at the beginning of the question,
• 57:45although everyone can read it,
• 57:47any data you're aware of Crag that's
• 57:49been collected with micro longitudinal
• 57:51methods like EMA or experience
• 57:52sampling with firearm owners.
• 57:56So not yet, but that's actually so.
• 57:58There are 61 that I just received.
• 58:00We are going to do that this year,
• 58:03so January is when we're
• 58:04hoping to start enrollment.
• 58:05What we'll do is the design of our study will
• 58:08be bringing gun owners and non gun owners.
• 58:11And then we're looking at two
• 58:13different subgroups of gun owners,
• 58:14those who carry a firearm on a
• 58:17regular basis and those who do not.
• 58:19Maybe they have a gun in their house,
• 58:21but they don't necessarily carry
• 58:23it with them on a regular basis.
• 58:26And yeah, we'll be hanging them six
• 58:29times a day for four weeks to see.
• 58:32Like what are they doing?
• 58:33Who are they with will be asked about.
• 58:36Mood will also be think
• 58:38about suicidal ideations.
• 58:39So Needless to say we have a pretty
• 58:41robust risk management protocol in
• 58:43place because consultation with
• 58:45other suicide researchers using MA.
• 58:47Some of the comments I've heard is,
• 58:50like you know,
• 58:51we all know intuitively that probably some
• 58:53of our participants are carrying a weapon,
• 58:56but.
• 58:56This is the first time that you actually
• 58:59know they'll be carrying a weapon when
• 59:01they're suicidal and that sort of
• 59:03changes the scariness of the project,
• 59:06so we have a pretty robust risk management
• 59:08protocol in one of the hypothesis we have it.
• 59:11We will be asking the mini may are
• 59:14you carrying a firearm right now,
• 59:16or is there a firearm within?
• 59:19I can't remember the wording.
• 59:21Basically,
• 59:21is there something firearm
• 59:23within reach something like that?
• 59:26And one of our hypothesis is
• 59:28that when someone is in physical
• 59:30possession of a firearm,
• 59:32that might increase their
• 59:35emotional liability.
• 59:36Their reactivity to stressful events,
• 59:37so my hope is that maybe maybe
• 59:40in a couple of years, yeah,
• 59:42we'll be able to have some of
• 59:44those those data available.
• 59:46You know on the group this
• 59:48is just a question for me.
• 59:50It's not in the chat box.
• 59:51And please if anyone has questions,
• 59:53feel free to jot them down in the box.
• 59:55But I'm thinking about that
• 59:57third group on your 3D model.
• 59:59I've done some work with health.
• 01:00:00And where they're looking for signals
• 01:00:02and in non behavioral health claims
• 01:00:04data so people are not feeling well,
• 01:00:05they're going to primary care.
• 01:00:06They are finding flags that seem to be
• 01:00:08predictive of eventual suicide attempts,
• 01:00:10even though the people are actually
• 01:00:11not in any sort of mental health
• 01:00:13care based on their claims data.
• 01:00:14Do you have any sense of whether or
• 01:00:16not I mean have you looked at that?
• 01:00:18Is that something that's come up
• 01:00:20as to are there going to be signals
• 01:00:22that we can acquire to reach out?
• 01:00:25The so the the one project we did
• 01:00:28that was not tide to like medical
• 01:00:31record data was social media project
• 01:00:34is actually funded by Department
• 01:00:37of Defense several years ago and we
• 01:00:41basically content coded a years worth of.
• 01:00:44It was mostly Facebook posts.
• 01:00:47Amongst users personnel who died by
• 01:00:49suicide and then a control group,
• 01:00:52they died of other non suicide causes
• 01:00:54and we looked at we kind of compared
• 01:00:57what was on their social media
• 01:01:00accounts and one of the findings was,
• 01:01:02you know we did find a statistically
• 01:01:05significant higher rate of,
• 01:01:07you know post that were hopeless
• 01:01:09and despondent and depressed.
• 01:01:10And had those types of themes.
• 01:01:13They were particularly huge
• 01:01:15effect sizes once we applied.
• 01:01:17These sort of dynamical systems
• 01:01:19models what we found was that yeah,
• 01:01:21that was like a game changer in one of
• 01:01:24the key findings was that we that was
• 01:01:27where we started to see the increase
• 01:01:30in instability in certain variables
• 01:01:32was a very very clear indicator.
• 01:01:34And what was interesting is as we got
• 01:01:37closer and closer to the date of death,
• 01:01:40those change processes got
• 01:01:42more and more pronounced.
• 01:01:43The second key finding was that we
• 01:01:46found that it was certain sequences.
• 01:01:48Of variables, so if you had,
• 01:01:51say variable in variable,
• 01:01:52feed if variable a consistently
• 01:01:54happened right before variable
• 01:01:56B that signaled suicide.
• 01:01:57But if variable B came
• 01:01:59right before variable A,
• 01:02:01it did not signal suicide.
• 01:02:03So that sort of notion of
• 01:02:05sequencing ended up being really,
• 01:02:07really key and then an independent
• 01:02:09team led by Glenn Coppersmith and
• 01:02:11they actually used the technique
• 01:02:13faceoffs presented at a conference.
• 01:02:15They went back to use it on Twitter data,
• 01:02:19and they called it Micro patterns,
• 01:02:21and they found that.
• 01:02:22These micro patterns of sequencing in
• 01:02:25certain orders doubled the amount of
• 01:02:27information and predictive ability
• 01:02:29to identify not only suicide but
• 01:02:31also like schizophrenia, anxiety,
• 01:02:33depression.
• 01:02:33So we have
• 01:02:35two questions. Actually,
• 01:02:36I think related to your EPI scenario,
• 01:02:38so I'll jump to those real quick.
• 01:02:42Alec was asking about the dependent variable.
• 01:02:45Do any of the conclusions you have
• 01:02:47presented applied differently according
• 01:02:48to whether we're looking at attempts
• 01:02:50that could have ended in death?
• 01:02:54Yeah, so we've not gotten to
• 01:02:56that point yet because our
• 01:02:58samples haven't been big enough.
• 01:03:00We do have some studies
• 01:03:02where death was the outcome.
• 01:03:04We have studies where attempts were
• 01:03:07the outcomes and so the general
• 01:03:09concept seemed to be the same,
• 01:03:12but I think the question is well
• 01:03:14placed in it is something that I've
• 01:03:16wondered about is are are there
• 01:03:19differences between highly lethal and
• 01:03:21less lethal suicidal behaviors that?
• 01:03:24Yeah, sort of those.
• 01:03:26Those sudden shifts to high risk
• 01:03:28states we see at least are thinking
• 01:03:31right now is perhaps more relevant
• 01:03:34for things like firearm suicide
• 01:03:37and other highly lethal methods.
• 01:03:39Whereas pathways A&B,
• 01:03:41those are individuals who tend to select
• 01:03:44less lethal methods like medications,
• 01:03:46and so there may actually be
• 01:03:49different change processes that signal
• 01:03:51lethal versus non lethal outcomes.
• 01:03:53I would love to.
• 01:03:55Hopefully get a data set
• 01:03:57large enough where we can
• 01:03:59test that out more.
• 01:04:00Yeah, one of the things I loved about
• 01:04:03your work is to focus on attempts,
• 01:04:06which is like heart attacks, right?
• 01:04:08It potentially lethal,
• 01:04:09and it relates to Alex,
• 01:04:10question around that.
• 01:04:11And then Sally asked about alcohol
• 01:04:14consumption and whether or not that
• 01:04:16teleports someone to severe risk.
• 01:04:19Yeah, yeah, that's that's a great question,
• 01:04:22and that's that's something
• 01:04:24that we wondered about as well.
• 01:04:27We don't know for sure, but you know,
• 01:04:31so there's some evidence outside of
• 01:04:33the suicide field that acute alcohol
• 01:04:37intoxication perhaps attenuates.
• 01:04:38Fluctuations in suicide risk owns a
• 01:04:40person down and oftentimes a sort of
• 01:04:42the motive for alcohol consumption.
• 01:04:43As I'm up and down,
• 01:04:45I kind of feel out of control.
• 01:04:48So I drink alcohol and it numbs
• 01:04:50and sort of pegs these Upson
• 01:04:52downs that they're less extreme.
• 01:04:54But then there are other individuals
• 01:04:56for whom alcohol consumption sort of.
• 01:04:58Yeah, sort of ramps them up and make some
• 01:05:01more vulnerable to these sudden shifts.
• 01:05:03So we will be we.
• 01:05:05That's one of the other things
• 01:05:07we are going to look at.
• 01:05:09Some of our PMA studies is how does
• 01:05:11acute alcohol consumption intoxication
• 01:05:13perhaps alter change processes?
• 01:05:15My suspicion what I really think
• 01:05:17we'll probably find is that
• 01:05:19there is going to be sort of a.
• 01:05:22A conditional finding where it
• 01:05:24will be some for some people.
• 01:05:28Alcohol consumption and those sort of
• 01:05:30calm them down and then for others
• 01:05:33it increases their reactivity,
• 01:05:35which I think would be really helpful
• 01:05:38because that might help us to more
• 01:05:41accurately target different subgroups
• 01:05:42of patients and individuals for alcohol
• 01:05:45consumption within their treatment plans.
• 01:05:50There's another question that
• 01:05:51came in between those two,
• 01:05:53but it wasn't as related to what
• 01:05:55you had said, so I hold it back,
• 01:05:58which is the international
• 01:06:00variability safe to countries.
• 01:06:01I think she meant the country Australia.
• 01:06:04Where they ban access to firearms
• 01:06:06in a more stringent way?
• 01:06:08Can you comment on that?
• 01:06:10Yeah, so there certainly are.
• 01:06:13You know, lower suicide rates and.
• 01:06:17In countries with more
• 01:06:18stringent access to firearms,
• 01:06:20what one of the things that related
• 01:06:22to this when we look at just sort of
• 01:06:25means restriction in general is that
• 01:06:27there are sort of international and
• 01:06:29cultural variability regarding this,
• 01:06:30and so one of the key assumptions about
• 01:06:33means restriction in general that we
• 01:06:35talk about is that the method that
• 01:06:37you're going to restrict needs to be
• 01:06:40sufficiently lethal to have an effect,
• 01:06:42so it has it has to kill people with
• 01:06:44a high enough rate that restricting
• 01:06:46access will actually make a difference.
• 01:06:49And then Secondly,
• 01:06:50the method has to be common enough
• 01:06:53within the population that again
• 01:06:55restricting it would potentially
• 01:06:57reduce an observable notable,
• 01:07:00meaningful proportion of the population.
• 01:07:02And so yeah,
• 01:07:03in many international spaces
• 01:07:05in other nations,
• 01:07:06firearms isn't really sort of initiative.
• 01:07:09Are no firearm suicides are
• 01:07:11very little because of their
• 01:07:14limited access to that method,
• 01:07:16but in those nations they typically look at.
• 01:07:20Other methods in from a means
• 01:07:22restriction perspective.
• 01:07:23They would target something else so.
• 01:07:26The most classic examples of this
• 01:07:29historically was in Sri Lanka,
• 01:07:31Indonesia or maybe it was Indonesia
• 01:07:34and some other Pacific island nations.
• 01:07:37It was poisoning with fertilizer.
• 01:07:40Someo farmers would drink there.
• 01:07:42Basically fertilizer to kill themselves in.
• 01:07:45So in these nations they restricted
• 01:07:48basically the sale of pesticides and
• 01:07:51fertilizers that that removed some of
• 01:07:54the most toxic chemicals and agents and so.
• 01:07:58In essence, former still had the tools
• 01:08:01that they need to do their jobs,
• 01:08:04but now they were less dangerous and
• 01:08:07so if they drink pesticide or whatnot,
• 01:08:10they were less likely to die as a result.
• 01:08:14In UK it was carbon monoxide
• 01:08:17being sent to homes to fuel ovens,
• 01:08:20things like that,
• 01:08:21and some of the Scandinavian
• 01:08:23nations is restricting paracetamol.
• 01:08:26And so,
• 01:08:27so in the United States,
• 01:08:28it's guns and I don't think will ever
• 01:08:31been the curve on suicide until we
• 01:08:34start really thinking about firearms as
• 01:08:36a key central aspect of a conference.
• 01:08:39Is suicide prevention plan.
• 01:08:43Um, we have a very long statement
• 01:08:45in question from Shelly.
• 01:08:47I don't know if she wants to
• 01:08:50actually go on audio and say it.
• 01:08:53I mean, I think.
• 01:08:55Well, that's thought about just
• 01:08:57sort of a question comment crag
• 01:08:59because when I think about some of
• 01:09:01the questions in the talking about
• 01:09:03because 'cause this shell is question
• 01:09:05comment was really around a bunch
• 01:09:07of linear variables and then some
• 01:09:08unpredictable variable or exponential.
• 01:09:10I think you know comes in from
• 01:09:12some direction and then pushes
• 01:09:14everything over the tipping point.
• 01:09:15But I think on the treatment side
• 01:09:17or intervention side one of the
• 01:09:19things that I find interesting.
• 01:09:21If you take time to really think about.
• 01:09:24Crisis response planning.
• 01:09:26CBT in your work.
• 01:09:28In some ways,
• 01:09:29it helps neutralize the unpredictability
• 01:09:31of what those things are.
• 01:09:34Because you're identifying
• 01:09:35specific things that will either
• 01:09:38distract or soothe the storm.
• 01:09:40You know that are not 100% likely,
• 01:09:42but at least are highly likely,
• 01:09:44and so some of that is in the crisis
• 01:09:46response plan some more of it is in
• 01:09:48the CBT in terms of better identifying
• 01:09:50when those storms are brewing.
• 01:09:52But can you comment on that?
• 01:09:54Like why is it so effective?
• 01:09:57Yeah, I think it's you.
• 01:10:01So it doesn't sort of like provides
• 01:10:03this sort of buffer zone between these
• 01:10:05two distinct States and so in essence,
• 01:10:08the first piece of it is how to
• 01:10:10know when you're heading towards
• 01:10:12a higher state and being able to
• 01:10:15recognize your own personal indicators
• 01:10:17or warning signs of that process,
• 01:10:19and then so now that you're aware of,
• 01:10:22like Now you know that you're
• 01:10:24heading towards that space,
• 01:10:25then the rest of the plan is all about.
• 01:10:29Here's a whole bunch of
• 01:10:30different things that you could.
• 01:10:32Do to report back up to reverse the
• 01:10:36process to take an offering up to.
• 01:10:38In essence,
• 01:10:39lots of lots of different ways to avert
• 01:10:42getting into that high risk space,
• 01:10:45and so in some cases,
• 01:10:47and it really is sort of riding the wave,
• 01:10:51but it does.
• 01:10:52It's something that I think really hits
• 01:10:54that notion of non proportionality
• 01:10:56is something very simple,
• 01:10:58like I was shocked to be perfectly
• 01:11:01honest when we did that CRP study.
• 01:11:05I was like, you know, this.
• 01:11:06We're writing a bunch of stuff
• 01:11:08done on an index card.
• 01:11:09In about 30 minutes,
• 01:11:10you know this is going to help in
• 01:11:13the next few weeks, but you know,
• 01:11:15we really need to get someone into
• 01:11:17like treatment like the CBT or
• 01:11:19something like that to have a longer
• 01:11:21term effect and what we ended up
• 01:11:23finding was that the effects tended to
• 01:11:25get bigger and bigger as more time passed,
• 01:11:28and so I was like wow,
• 01:11:29there's something happening here and
• 01:11:31it appears to help individuals to
• 01:11:33know where their tipping point is.
• 01:11:35Into in essence,
• 01:11:37a stay away from that,
• 01:11:39and it seems to work very,
• 01:11:41very effectively as result.
• 01:11:44So I think we're how we have 60 seconds left,
• 01:11:47so just given all this data,
• 01:11:49what do you think? Why?
• 01:11:51Why do we still have these
• 01:11:53issues in the military,
• 01:11:54the VA and in our country?
• 01:11:56What's preventing these interventions from?
• 01:11:59And be part of it is.
• 01:12:01I think we still buy and larger than the
• 01:12:04middle health disciplines conceptualize
• 01:12:06suicide as a symptom of mental illness.
• 01:12:09And so by and large treatment, as usual,
• 01:12:12is largely treat the depression.
• 01:12:14Treat the PTS di treat anxiety,
• 01:12:16whatever the you know,
• 01:12:18whatever the diagnosis is,
• 01:12:20and then theoretically suicide
• 01:12:21risk should resolve after that.
• 01:12:23But now I think we increasingly recognize
• 01:12:26that suicide is transdiagnostic
• 01:12:27that it's better to look at.
• 01:12:30Suicide risk is.
• 01:12:31Independent of psychiatric diagnosis
• 01:12:34and so treatments like bteen crisis
• 01:12:37response planning are going directly at.
• 01:12:40Several mechanisms that seem
• 01:12:42to give rise to suicide risk,
• 01:12:44and so it's a more direct targeting
• 01:12:46of the problem as opposed to an
• 01:12:49indirect targeting through these
• 01:12:51other reported mental health.
• 01:12:53Channels the last thing that I think
• 01:12:55is really key that another aspect of
• 01:12:58my thinking is really moving out of
• 01:13:01this sort of mental health model of suicide,
• 01:13:04and we're now looking more and
• 01:13:06more at more like environmental.
• 01:13:09Injury prevention models where
• 01:13:10you know we don't reduce traffic
• 01:13:12fatalities by saying everyone needs to.
• 01:13:14Lookout for the warning signs of
• 01:13:16a car accident and then if you see
• 01:13:19those warning signs you should go
• 01:13:21and get enroll in a drivers Ed class.
• 01:13:24I mean,
• 01:13:25that's basically how we approach
• 01:13:26suicide prevention right now.
• 01:13:28What we do is we build safer
• 01:13:30roads and we build safer cars and
• 01:13:33we require people to buckle up.
• 01:13:35And we criminalize conditions like
• 01:13:37drinking and driving that increase
• 01:13:39the probability of fatal outcomes.
• 01:13:40And so,
• 01:13:41how do we now take a similar
• 01:13:44thought process with suicide,
• 01:13:46where we change the environment in a
• 01:13:49way that reduces the likelihood of a
• 01:13:52person if they do shift to a high risk state,
• 01:13:56they have a seat belt now and they
• 01:13:59are significantly more likely
• 01:14:01to survive that unforeseen,
• 01:14:03unpredictable conditions.
• 01:14:04So
• 01:14:05can you one quick thing on that
• 01:14:07that issue of of the systemic?
• 01:14:10Approach and associating
• 01:14:11it with mental illness.
• 01:14:13There's varying degrees in different
• 01:14:15studies about how many people actually
• 01:14:17even have an independent mental health
• 01:14:19condition from suicidality, right?
• 01:14:21Just like heart attacks like some
• 01:14:23people just don't have another
• 01:14:25cardiac condition that we're aware of.
• 01:14:28So what is your best guestimate of the
• 01:14:30people who attempt suicide and actually
• 01:14:33don't have some other standalone
• 01:14:35health condition?
• 01:14:36Yeah, I would say so.
• 01:14:38It's a bigger percentage than we think.
• 01:14:41Or at least that we've traditionally assumed.
• 01:14:43So if you look at CDC data,
• 01:14:4555% of suicide decedents in the US
• 01:14:47do not have a known mental illness.
• 01:14:49Now the typical worst retort to that is,
• 01:14:51well, they didn't go in to meet
• 01:14:53with the therapist or psychiatrist,
• 01:14:55and so they were never diagnosed,
• 01:14:57so we just didn't know that they had
• 01:14:59the condition that we assume is there.
• 01:15:02But if you look at some
• 01:15:03other lines of data and ones,
• 01:15:05I think that is really informative.
• 01:15:07This Joe Franklin's meta
• 01:15:08analysis from a few years ago.
• 01:15:10Looking at 50 years of
• 01:15:12suicide risk factor research,
• 01:15:13if you kind of do some reverse engineering
• 01:15:16of some of the statistics he reported,
• 01:15:18in essence,
• 01:15:19what his data would suggest is that.
• 01:15:21Somewhere around 40% of those
• 01:15:24who attempt suicide or die by
• 01:15:26suicide have a mental health
• 01:15:29diagnosis or elevated symptoms.
• 01:15:30Things like that.
• 01:15:34I think the third line of evidence
• 01:15:36comes from psych autopsy studies,
• 01:15:38which are often used to support that.
• 01:15:40This is where the 90% status
• 01:15:42obtained from 90% of suicide
• 01:15:44decisions have a mental illness.
• 01:15:46It comes from cycle autopsy studies.
• 01:15:50The second party method is
• 01:15:52highly vulnerable to bias,
• 01:15:54and if you look at case controlled
• 01:15:56cycle top studies where you can see
• 01:16:00is that they generally overestimate
• 01:16:02rates of psychiatric illness in
• 01:16:04control cases by double and so you
• 01:16:07can reasonably infer that maybe
• 01:16:09psych autopsies are doubling their
• 01:16:12biased towards over diagnosing mental
• 01:16:14health conditions by a factor of two,
• 01:16:17in which case 90% of.
• 01:16:20Which is their finding would convert to
• 01:16:22about a 45% and estimate the correction,
• 01:16:25so to speak.
• 01:16:26So this is sort of like these several
• 01:16:29lines of evidence that I would say
• 01:16:32maybe around 40 to 50% of suicide.
• 01:16:35Decedents Oregon suicide
• 01:16:37attempters probably do not have.
• 01:16:39A diagnosable mental condition.
• 01:16:41They might be really stressed out.
• 01:16:43It might be really upset
• 01:16:44under pressure for sure,
• 01:16:46but come on like being upset
• 01:16:48when your partner tells you that
• 01:16:50they're going to leave you there.
• 01:16:52Having an affair is not a mental illness.
• 01:16:54That is a normative emotional
• 01:16:56reaction to a life stressor,
• 01:16:58and so I think we've blurred
• 01:17:00the lines between.
• 01:17:01Between these experiences and as a result,
• 01:17:04we put all of our eggs into
• 01:17:06the Git mental health.
• 01:17:08Basket mental health treatment basket
• 01:17:10and so we're ignoring a lot of other
• 01:17:14potentially useful and impactful
• 01:17:17suicide prevention strategies.
• 01:17:21Some very positive feedback in the chat.
• 01:17:22I'll take some screenshots so
• 01:17:24that you can show it to your loved
• 01:17:27ones and family so that they know
• 01:17:29what a great job you did crag.
• 01:17:31Any other questions or comments for Crag?
• 01:17:35Before we let him get on
• 01:17:37to his weekend I guess.
• 01:17:41Hi this is Shelly I guess I I apologize
• 01:17:43for the length the set up there so I've
• 01:17:45read Malcolm Gladwell's tipping point.
• 01:17:47Understand I've done a lot of work in the
• 01:17:49chemistry lab with pH buffering and it's
• 01:17:51very very simple to overshoot because
• 01:17:52you're on an exponential pathway right?
• 01:17:54So what I was thinking is it sounds like
• 01:17:56you have a lot of linear factors and then
• 01:17:59you have an exponential factor that may not
• 01:18:01be the same exponential factor every time.
• 01:18:03So very very quick to overshoot
• 01:18:05at that moment of suicidality.
• 01:18:06Most of the work I do is
• 01:18:08with veterans at this point.
• 01:18:10And most of the veterans are very,
• 01:18:12very familiar with weapons.
• 01:18:13The conversations I start out early I
• 01:18:15start out often talking about guns.
• 01:18:16I talked extensively about it
• 01:18:17in the interview,
• 01:18:18find out how they're stored,
• 01:18:20what they do with somehow they carry it.
• 01:18:22You know,
• 01:18:22there's a fairly lengthy conversation
• 01:18:24and a lot of times sometimes they
• 01:18:25back off and sometimes it it goes to.
• 01:18:27You know, I just say listen,
• 01:18:29you're probably not anything
• 01:18:30to worry about right now,
• 01:18:31but if the time ever comes in the future,
• 01:18:34I want to, you know,
• 01:18:35I have a comfort level talking
• 01:18:37with you about this,
• 01:18:38and we've laid out the groundwork for that.
• 01:18:40Now I don't have any data.
• 01:18:42I don't know if it helps us.
• 01:18:44Don't know if it doesn't,
• 01:18:45but one of the frustrations I have
• 01:18:47is with my fellow clinician who
• 01:18:49sometimes in all best efforts.
• 01:18:51Seem like that conversation
• 01:18:52is incredibly minimal,
• 01:18:54is sort of the richness that
• 01:18:56it could be done.
• 01:18:57Yeah,
• 01:18:58so I guess I'm wondering what we
• 01:19:00can do to train the clinician to
• 01:19:03inquire in a way that's useful
• 01:19:05to the patient in those times.
• 01:19:08So we yeah,
• 01:19:09so we trained clinicians pretty
• 01:19:10often in mean safety counseling,
• 01:19:13but now the training workshop that I do,
• 01:19:16I just call firearm safety counseling.
• 01:19:18I'm like, yeah,
• 01:19:19'cause I find clinicians are usually
• 01:19:21reasonably comfortable talking about like.
• 01:19:23Limiting access to medications
• 01:19:25and things like that,
• 01:19:26but it is a firearm issue that often is much,
• 01:19:30much more challenging, and so we do.
• 01:19:33So we do have data.
• 01:19:35Their forthcoming that if you do this,
• 01:19:38if you have these conversations in
• 01:19:40a very sort of non judgmental way.
• 01:19:43But again we used the motivational
• 01:19:45interviewing based approach we found.
• 01:19:47There was incredibly highly
• 01:19:49acceptable in our study with military
• 01:19:52personnel out of 200 year old 130.
• 01:19:54200% of them said I would recommend
• 01:19:56this conversation for a friend,
• 01:19:58which is sort of like OK then.
• 01:20:01That's some St cred there.
• 01:20:02This is evidence of acceptability.
• 01:20:06Reached their likelihood of locking up their
• 01:20:08guns and we found that the vast majority,
• 01:20:11even though there was there,
• 01:20:13was a small number who kind of still had,
• 01:20:15you know, screw you guys at
• 01:20:17the end of the conversation,
• 01:20:19all of those during the follow up.
• 01:20:21They said, you know,
• 01:20:22I felt respected and listened to you guys
• 01:20:25didn't push me and so we had a few who
• 01:20:28at least we're left with a good impression,
• 01:20:30even though perhaps we hadn't yet convinced
• 01:20:32them that it was in their best interest.
• 01:20:35But one of those things that we
• 01:20:37have found really helpful working.
• 01:20:39Gun owners in general.
• 01:20:40Is using the metaphor of a designated
• 01:20:42drivers in wearing seatbelts and
• 01:20:44we say do you when you drive?
• 01:20:46Do you only wear seat belt on the day that
• 01:20:49you expect to get into a car accident?
• 01:20:51And they're like, well, no,
• 01:20:52you can't predict it is like right?
• 01:20:54And we never know when we're going to
• 01:20:56have those really, really bad days.
• 01:20:58And so that's why we wear a
• 01:21:00seat belt just in case.
• 01:21:02And so what's going to be
• 01:21:03the seat belt for your gun?
• 01:21:05And they find that we were finding
• 01:21:07that oftentimes helps him to think.
• 01:21:09OK,
• 01:21:09maybe a gun safe and unlock something
• 01:21:11like that is sort of like a seat belt.
• 01:21:14And that might be the difference
• 01:21:15between life and death in the same way
• 01:21:18that wearing your seat belt can be.
• 01:21:19Those difference between life and death
• 01:21:21in a car accident and then the designated
• 01:21:23driver model is really resonated a lot.
• 01:21:25Where,
• 01:21:26like you know,
• 01:21:26if you're worried about a friend,
• 01:21:28we we target this sort of community rather
• 01:21:30than this like a high risk individual,
• 01:21:32but will say you know when
• 01:21:34your friend has been drinking,
• 01:21:35you take away their parties because
• 01:21:37it's dangerous and it doesn't mean they
• 01:21:39never get to drive again in their life.
• 01:21:41It's just that you have to wait
• 01:21:43until they're sober so it's
• 01:21:45safer and then they can.
• 01:21:46You know,
• 01:21:47have a car keys back again and so
• 01:21:49it works the same with the gun,
• 01:21:51shouldn't it will ask you should
• 01:21:53suicidal people have access to
• 01:21:54loaded weapons and everyone's like,
• 01:21:56no, that's a terrible idea,
• 01:21:57like right,
• 01:21:58and so if you're if you know we have a friend
• 01:22:01is going through a hard time like that.
• 01:22:04Let's think about this like a
• 01:22:05designated driver and let's maybe
• 01:22:07offer to temporarily hold onto
• 01:22:08their weapons until they're in
• 01:22:10a different state of mountain.
• 01:22:11And then perhaps it would be
• 01:22:13safer for them to have access.
• 01:22:15And we're finding those messages
• 01:22:17are very positively received.
• 01:22:18But the hard part is,
• 01:22:20yeah,
• 01:22:21training clinicians.
• 01:22:21My sense is that clinicians are often
• 01:22:25biased towards complete removal.
• 01:22:27Of the fire room.
• 01:22:28From the households which makes
• 01:22:30a lot of sense,
• 01:22:31that is definitely a probably
• 01:22:33the safest option,
• 01:22:34but we're really taking more of a
• 01:22:37harm reduction approach and saying,
• 01:22:39all things being equal,
• 01:22:40if someone has multiple loaded weapons
• 01:22:43that are readily available in the home
• 01:22:45and we move them to a state where now
• 01:22:48those weapons are all in a gun safe,
• 01:22:51or they're all locked up with a trigger lock,
• 01:22:54even though they're still in the home.
• 01:22:58They are in a safer space,
• 01:23:00relatively relatively speaking
• 01:23:01than they were before,
• 01:23:03and I see that as a valuable
• 01:23:06movement that's worth us talking
• 01:23:08with this clinicians.
• 01:23:09While I see Mike and Alec and Howard here,
• 01:23:13you have three, you know,
• 01:23:15preeminent forensic psychiatrists.
• 01:23:16And I heard what you said about
• 01:23:19absolute removal of the gun,
• 01:23:21which my suspicion is done out of fear of
• 01:23:25liability for not insisting on it versus.
• 01:23:28Google clinical evidence 'cause
• 01:23:30your middle ground is so
• 01:23:31intuitive and makes so much sense.
• 01:23:33You know my suspicion,
• 01:23:34and I am always have a little
• 01:23:36little touch of cynicism,
• 01:23:37and I guess my perspective on things
• 01:23:39is that it's because we don't go for
• 01:23:41the absolute removal were afraid.
• 01:23:42If something happens,
• 01:23:43there's a liability risk.
• 01:23:44It's black or white.
• 01:23:46And I remember the first time
• 01:23:48I I saw you speak live,
• 01:23:50you put up a little table of like the disk,
• 01:23:53disjoint between the misalignment
• 01:23:55between what does the clinician want
• 01:23:57when they meet with the suicidal patient?
• 01:23:59And what does the patient want?
• 01:24:01Do you remember that slide?
• 01:24:03That was quite awhile ago?
• 01:24:05Comment on that because I thought it
• 01:24:07was really incredibly elegant, yeah?
• 01:24:11Yeah, it's sort of a key point in a
• 01:24:13lot of the trainings I do is yeah
• 01:24:15this that the clinicians goals are
• 01:24:18to prevent the patient from dying,
• 01:24:20and then I also tongue in cheek,
• 01:24:22say and to avoid getting sued.
• 01:24:24And then I like to say well in
• 01:24:26most cases let's be honest,
• 01:24:28we don't want to get sued and we know
• 01:24:30the way to not get sued is to not have
• 01:24:33a patient die something like that.
• 01:24:36Whereas the patient's goals are
• 01:24:37to alleviate their suffering
• 01:24:38and solve their problem,
• 01:24:40and So what ends up happening is we.
• 01:24:43Sometimes gets so focused on preventing
• 01:24:45death that we forget that in another
• 01:24:48important part of suicide prevention is
• 01:24:51creating lives that are worth living,
• 01:24:53and so we may succeed in the short term in
• 01:24:56keeping someone breathing biologically alive.
• 01:24:59But if we don't do something
• 01:25:02that changes their environment,
• 01:25:04their psychology,
• 01:25:04things like that,
• 01:25:06and probably they're going to get
• 01:25:08into the state again,
• 01:25:10and so we need to look at quality
• 01:25:13of life in addition to simply just.
• 01:25:17Like restraining people from dying,
• 01:25:19and so when it comes to the issues
• 01:25:21surrounding like means restriction,
• 01:25:23I think of you know I've written
• 01:25:26about this a few times.
• 01:25:27I think one of the downsides we
• 01:25:29recommended as a suicide prevention
• 01:25:31intervention for clinicians.
• 01:25:33But as was Shelley,
• 01:25:34was alluding to we don't really
• 01:25:37do any training at all on how
• 01:25:39to do this effectively,
• 01:25:40so we almost set ourselves up for failure,
• 01:25:43and we think I think in terms of
• 01:25:46these sort of binary extremes.
• 01:25:49Where I don't know.
• 01:25:50We'll see how things emerge overtime.
• 01:25:52You know I do some consulting on
• 01:25:54legal cases like negligence cases,
• 01:25:56things like this and it's a topic
• 01:25:58that is very key to a lot of those
• 01:26:01is at what point do you say that a
• 01:26:04clinician have been reasonable and
• 01:26:06done enough 'cause we can always
• 01:26:08find something more that a person
• 01:26:11could have theoretically done?
• 01:26:13And it I found it tough with the gun issue,
• 01:26:16but they give some cases now.
• 01:26:18Clinicians that actually did
• 01:26:20needs restriction.
• 01:26:21Yeah,
• 01:26:21they didn't need description and
• 01:26:23then the patient didn't tell them
• 01:26:25they still had a secret gun.
• 01:26:27I mean it's good news.
• 01:26:29Craig is that unlike drinking and driving
• 01:26:31gun ownership is not not all politicized.
• 01:26:34So since drinking and driving and
• 01:26:36designated driver that just simple,
• 01:26:38it's just simple problem.
• 01:26:40The gun issue I'm sure will
• 01:26:43be even easier to tackle.
• 01:26:45Yeah, well thank you, anything else?
• 01:26:48I just wonder what you think
• 01:26:50the APA is considering having
• 01:26:53suicidal behaviors, a diagnosis.
• 01:26:56And given the fact that yeah don't have,
• 01:26:59I just wonder what you think about that.
• 01:27:03Yeah, I'm sort of ambivalent about it.
• 01:27:06On the one hand.
• 01:27:08I think some of the arguments being
• 01:27:11made for why we want to do it,
• 01:27:14I think makes sense.
• 01:27:16You know to recognize it as being discrete
• 01:27:19from other psychiatric conditions diagnosis,
• 01:27:22I think probably there might be what?
• 01:27:27Joyner is called the closet with an acute
• 01:27:30suicidal syndrome or something like that,
• 01:27:32where maybe there is kind of a
• 01:27:34unique mental state associated
• 01:27:36with near term risk for suicide.
• 01:27:38Now the downside kind of my hesitation
• 01:27:41about it as I kind of feel like begin
• 01:27:44were now now with a diagnosis as
• 01:27:47opposed to a manner of death. And.
• 01:27:50You know, I don't know that we have like,
• 01:27:53yeah, Heart Attack syndrome.
• 01:27:54We don't.
• 01:27:54You know, it's sort of.
• 01:27:56I kind of feel like it's our continuing trend
• 01:27:59to conceptualize and think about suicide.
• 01:28:02From the nearly exclusive
• 01:28:03lens of psychiatric illness.
• 01:28:05And so I think that will slow
• 01:28:07us down from in my opinion.
• 01:28:09I think we need to be a little bit
• 01:28:12more divergent and are thinking
• 01:28:14about suicide to consider it from
• 01:28:16these different perspectives and
• 01:28:18angles as opposed to continuing to
• 01:28:20kind of narrow us down more and
• 01:28:23more into the DSM and the domain of.
• 01:28:26You know one particular health
• 01:28:28care discipline so.
• 01:28:32And no one is asking me,
• 01:28:33but I happen to the heart attack
• 01:28:35thing to me or an arrhythmia is
• 01:28:37in fact a parallel in my mind.
• 01:28:39Anyone can have.
• 01:28:40I honestly believe anyone can
• 01:28:41become suicidal with the right.
• 01:28:43External variables,
• 01:28:44and certainly if other conditions
• 01:28:45occur as they go through life,
• 01:28:47just like in a heart attack,
• 01:28:49anyone can get Melanoma.
• 01:28:51They are like important,
• 01:28:52they had outcomes,
• 01:28:53and I wouldn't worry about it, Craig.
• 01:28:55Yeah, well and I think yeah,
• 01:28:57I've thought about this a lot in
• 01:28:59the past year when they were getting
• 01:29:02close to a year of a pandemic.
• 01:29:04And so I'm constantly asked
• 01:29:06about how the pandemic influence
• 01:29:08suicide risk and how do we improve
• 01:29:10access to mental health care.
• 01:29:11And I think the point that.
• 01:29:14We often overlook is like, yeah,
• 01:29:16you know, therapy medications might
• 01:29:17help to reduce a person's distress,
• 01:29:20but it doesn't inherently change the
• 01:29:22fact that the environment right now is
• 01:29:25placing a lot of strain on individual,
• 01:29:27so you can't therapy your way out of
• 01:29:30COVID-19 you know you can't therapy
• 01:29:32your way out of a minimum wage job.
• 01:29:35You can't therapy your way out of an abusive,
• 01:29:38toxic work environment.
• 01:29:40And and so I think this is.
• 01:29:43This is one of the main reasons why
• 01:29:46I think we're so bad that suicide
• 01:29:49prevention as there are things in
• 01:29:52life that contribute to suicide risk.
• 01:29:55But we we largely think about suicide
• 01:29:57as being within the individual.
• 01:29:59The thing that we.
• 01:30:00Just to sort of root out and get rid of
• 01:30:04when in reality there are conditions
• 01:30:06all around us that contribute to that,
• 01:30:09and we would probably actually
• 01:30:11see way better results if.
• 01:30:13If we increase the minimum wage,
• 01:30:15if we expanded access to health care,
• 01:30:18if we targeted racism and discrimination
• 01:30:20against vulnerable populations,
• 01:30:21and there's not much that I'm going to
• 01:30:23be able to do as a health care provider.
• 01:30:26Yeah,
• 01:30:27to improve the financial stability
• 01:30:28of my patients that they're working
• 01:30:303 minimum wage jobs.
• 01:30:32It's it would be way more impactful to
• 01:30:34just increase their salary in resolve.
• 01:30:36A lot of the stress that they're coming in.
• 01:30:39Complaining to me about.
• 01:30:41But of course,
• 01:30:42it's really not coming to complain.
• 01:30:44About that because I can't afford it, right?
• 01:30:47So anyway.
• 01:30:51Craig, thank you.