Psychotherapy Brown Bag

Emotions and suicide: The link is not as straight forward as we might think

by Michael D. Anestis, M.S.

I always feel a little weird writing about my own work on PBB, but, in all honesty, the online portal to the library is down today and I have time to write a post, so my options are either this or nothing for the moment, so here we go!

A few months ago, I wrote about a study that Thomas Joiner and I have in press at Journal of Affective Disorders examining the relationship between negative urgency, the components of the interpersonal-psychological theory of suicidal behavior (IPTS; Joiner, 2005) and lifetime number of suicide attempts (click here to read that PBB post).  In that paper, we found that individuals with high levels of all of the theory components (thwarted belongingness, perceived burdensomeness, acquired capability for suicide) as well as high levels of negative urgency reported the greatest number of past suicide attempts.  This made intuitive sense to us, as it seems that folks who are highly motivated to diminish the strength of negative emotions regardless of the long term cost would, in fact, be more likely to desire suicide and to develop the capacity to enact lethal self-harm through their repeated use of painful and provocative emotion regulation strategies (e.g., non-suicidal self-injury [NSSI]).

Some recent data we have been collecting in other populations, however, left us wanting to look at this from another angle.  Specifically, in another study we have in press at Archives of Suicide Research (Anestis, Bender, Selby, Ribeiro, & Joiner, in press), we found that high levels of distress tolerance predicted higher levels of the acquired capability for suicide. Now, for those of you unfamiliar with those variables, let me explain why this is a suprising and potentially important finding:

Folks with low levels of distress tolerance - who find the experiene of negative emotions to be highly aversive and overwhelming - have been shown in a number of studies to engage in a wide variety of painful and provocative behaviors (e.g., Nock & Mendes, 2008).  Given that, you might expect these folks to have higher levels of the acquired capability for suicide, a component of the IPTS that involves habituation to physiological pain and the fear of death through repeated exposures to painful and provocative stimuli.  Put more simply, folks with low distress tolerance often engage in behaviors like self-injury in an effort to reduce their negative emotions (emotion dysregulation).  Overtime, by repeatedly engaging in these behaviors, the pain associated with them seems to decrease (e.g., Nock & Prinstein, 2005; Orbach et al., 1997).  As such, for these folks, inflicting (potentially) lethal injury upon themselves becomes less physiologically and emotionally aversive.  The thing is...the data from our study in Archives of Suicide Research seemed to indicate something very different.  It was folks who COULD tolerate negative emotions who demonstrated elevations in the acquired capability. 

Given that this finding was somewhat surprising, we wanted to see if it would be replicated in an independent sample, just to make sure that it was not a random error or simply reflective of the particular group of folks we examined in the earlier study.  As a follow-up, I included relevant measures in my dissertation project and, during the two years we recruited for that study, we collected data that would help shine further light on this topic and ultimately became a paper we now have in press at Journal of Psychiatric Research (Anestis, Bagge, Tull, & Joiner, in press).  Overall, we recruited just under 300 undergraduate participants and had them fill out a series of measures, take part in a structured interview, and complete both card sorting and physiological pain tolerance tasks.  We had two main hypotheses:

1. High levels of negative urgency and low levels of distress tolerance (e.g., greater levels of emotion dysregulation) would predict elevations in the components of the IPTS that represent the desire for suicide (thwarted belongingness and perceived burdensomeness).
2. Low levels of negative urgency and high levels of distress tolerance (e.g., lower levels of emotion dysregulation) would predicted elevatiosn in the component of the IPTS that represents individuals' ability to engage in serious or lethal suicidal behavior (acquired capability for suicide).

The easiest way to summarize those hypotheses is that we expected individuals with greater difficulty regulating and tolerating their emotions to have a higher level of suicidal desire, but less capacity to act on such desire.  The results were highly consistent with this idea.

Both negative urgency and distress tolerance were correlated with burdensomeness and belongingness in the expected direction (e.g., high negative urgency/low distress tolerance -> greater levels of suicidal desire).  When we ran a series of hierarchical multiple regressions and controlled for sex, depression, and various components of impulsivity, distress tolerance, but not negative urgency significantly predicted burdensomeness and belongingness in the expected direction (negative urgency did as well if distress tolerance, which was highly correlated with it, was removed from the equation).

Similarly, both negtive urgency and distress tolerance were correlated with our two measures of the acquired capability (self-report and a physiolgical pain tolerance task) in the expected direction (e.g., low negative urgency/high distress tolerance -> greater levels of the acquired capability).  We again ran a series of hierarchical multiple regressions and this time controlled for sex, painful and provocative events, and various components of impulsivity.  In this case, negative urgency but not distress tolernace significantly predicted self-reported acquired capability and distress tolerance but not negative urgency significantly predicted physiological pain tolernace (and again, if you take significant predictor out, the non-significant predictor became significant).

So...what does all of that mean?  Put simply, when folks find negative emotions highly aversive and are motivated to do whatever they can to immediately reduce the intensity of such emotions, they may be more likely than others to desire suicide, but less inherently capable than others to follow through on such desire.  Suicidal behavior is an inately frightening and physically uncomfortable event.  As such, people who have a difficult time tolerating and remaining in contact with emotional experiences may be less naturally capable of following through on a suicide attempt, choosing instead to opt for faster behaviors over which they may have more control with respect to the duration and intensity of the event (e.g., NSSI, binge eating). 

As you consider these results, however, I want you to keep a couple things in mind: 

First of all, these results do not indicate that emotion dysregulation makes it impossible or even less likely than an individual will die by suicide.  Instead, they simply indicate that such individuals may need to engage in more painful and provocative experiences in order to acquired the capability for suicide than would an individual for whom intense emotional and physical sensations are less aversive.  Given that emotionally dysregulated individuals are more likely to desire suicide, they may be more likely to engage in such behaviors anyway so, even though they have more to acquire with respect to the capacity to enact lethal self-harm, they still might be more likely to acquire it. The point, however, is that suicide would then not be due to their emotion dysregulation.

Secondly, these results do not mean that it is bad to tolerate distress and that clinicians should teach their clients to have low distress tolerance.  Instead, they simply point out that, what is often an adaptive feature (high distress tolerance) can become a liability when an individual is motivated to engage in a problematic behavior (suicide attempt) that requires him or her to experience significant distress, potentially for a prolonged period of time, in order to accomplish their desired outcome.

I'm very excited about this line of research and have a number of other projects at various stages of development.  In the meantime, I welcome your thoughts on these points and look forward to sharing more information with you as it becomes available to discuss.

************

If you would like to learn more about this or other topics discussed on PBB, we recommend that you consult with our online store for scientifically-based psychological resources.

Mike Anestis is a psychology resident at the University of Mississippi Medical Center and a doctoral candidate in the clinical psychology department at Florida State University

Narcissistic personality disorder and DSM-V: Confusing alterations with elimination

by Michael D. Anestis, M.S.

I came across a fairly bizarre article in the Health section of the New York Times recently and feel compelled to comment on it here (click here to read the actual Times article).  Written by Charles Zanor, a psychologist in Massachusetts, the article discusses the pending alterations to the personality disorder section of the DSM in the upcoming DSM-V. 

As we detailed in an earlier post, the proposed changes in the DSM-V include a shift away from classic personality disorder diagnoses and towards an evidence-based profile approach in which individuals are scored on various levels of a variety of variables (click here to view the DSM-V page describing all changes to personality disorders).  Think of it this way: 

• Everybody has a certain level of certain variables (e.g., height, intelligence, age).  Rather than report a yes-no answer on whether or not we meet a certain level of one of these things (generally speaking), we tend to report specific levels (e.g., five feet tall, 110 IQ, 30 years old).  The new personality system will do the same, recording individuals' levels of a number of traits so as to provide an overall view of their global personality.

Why is this important here?  Well, first of all, people are making quite a fuss about the DSM "eliminating" diagnoses.  In this case, the uproar is about narcissistic personality disorder (NPD).  The thing is, it's not being eliminated.  If you look at the list of personality traits that will be assessed in clients, you'll see one entitled antagonism, which is defined by the committe as:

"Exhibits diverse manifestations of antipathy towards others, and a correspondingly exaggerated sense of self-importance."

Taking it a step further, each personality trait includes a list of trait facets to be considered.  Listed amongst those for antipathy is...wait for it.....NARCISSISM.

That's right, every client should, in theory, be rated on their level of narcissism.  In this sense, not only is narcissism not being eliminated from our assessment of personality, it is being included in a profile system that, in theory, will be utilized in assessments of all clients, thereby INCREASING the likelihood clinicians will assess this construct.

Now, I realize that this is still a marked change from the current diagnostic system, but the thing is, there is essentially no empirical evidence demonstrating that the diagnosis of NPD is a valid and reliable idea (which is the reason it is being massively altered) and there is not a single evidence-based treatment for NPD.  Massive changes to an idea that has no evidence supporting that it exists in exactly the way it is described (e.g., number of symptoms required) and for which we have no effective treatments does not seem like a horrific idea to me.

In the article, Dr.Gunderson refers to these changes as "draconian," citing the fact that such a massive amount of change within one diagnostic area has never before been done.  The fact that it is being done for the first time, however, does not qualify it as draconian.  If it flew in the face of empirical evidence, that would make more sense to me, but it does nothing of the sort.  Using this same logic, the decision to make DSM-III based upon observable behaviors instead of psychodynamic principles would also be viewed as draconian and there aren't many people trumpeting that idea.  The issue here is the accuracy of the changes, not the scope of the changes.

The final aspect of this article that caught my attention and to which I feel compelled to reply is Dr.Shedler's comment that the committe is made up predominantly of researchers who do little clinical work. 

• Question #1: on what basis is this claim made (e.g., how many patient hours were accumulated by each committe member?)?  
• Question #2: what evidence is there that somebody with a greater level of clinical activity (should his assumption be correct) make more valid decisions regarding diagnostic criteria?

I completely understand folks having reservations about these changes.  I certainly have some and I have voiced them in a number of forums (e.g., the DSM-V website, the online forum of Personality Disorders: Theory, Research, and Treatment).  That being said, it is important that, when we voice these concerns, we ensure that they actually reflect reality.  If not, things like this happen: misinformation spreads in an internationally renown forum and folks argue about something that is not even happening.

************

If you want to learn more about this or other topics discussed on PBB, we recommend that you consult our online store for scientifically-based psychological resources.

Mike Anestis is a psychology resident at the University of Mississippi Medical Center and a doctoral candidate in the clinical psychology department at Florida State University

What do clients value in psychotherapy: Empirical support versus common factors

by Michael D. Anestis, M.S.

This morning, I came across an interesting article published in the most recent issue of the Journal of Clinical Psychology.  The article was written by Joshua Swift of the University of Alaska Anchorage and Jennifer Callahan of the University of North Texas and it's central purpose was to determine the extent to which a client coming in for therapy values common factors (e.g., therapeutic alliance) more or less than the scientific evidence supporting the effectiveness and efficacy of a particular form of therapy (e.g., what percentage of clients with a particular diagnosis tend to recover as a result of this specific form of therapy?).

A total of 57 clients from a university-based psychology clinic filled out a series of questionnaires prior to the onset of treatment indicating the degree to which they value particular aspects of therapy.  Specifically, they were asked to indicate whether they would prefer a form of treatment with scientific evidence supporting it's efficacy (70% recovery rate) that lacks a particular common factor (e.g., strong therapeutic alliance) or a treatment with varying levels of support (ranging form 10% to 70% recovery rate) that promises high levels of that same common factor.  As an example, a client may have been asked to indicate their preference for each of the following options:

****

70% recovery rate/weak therapeutic alliance vs 10% recovery rate/strong therapeutic alliance
70% recovery rate/weak therapeutic alliance vs 20% recovery rate/strong therapeutic alliance
70% recovery rate/weak therapeutic alliance vs 30% recovery rate/strong therapeutic alliance
70% recovery rate/weak therapeutic alliance vs 40% recovery rate/strong therapeutic alliance
70% recovery rate/weak therapeutic alliance vs 50% recovery rate/strong therapeutic alliance
70% recovery rate/weak therapeutic alliance vs 60% recovery rate/strong therapeutic alliance
70% recovery rate/weak therapeutic alliance vs 70% recovery rate/strong therapeutic alliance

****

The idea here is to determine to what extent evidence is valued by a client prior to treatment relative to other factors that they might believe are more important.  In this particular study, the authors examined the importance of four common factors:

• Therapeutic alliance
• Interpersonal responses of the therapist
• Therapist level of experience
• The degree to which the client determined the direction of therapy sessions

What they found was that the client would accept a treatment with a 38.14% lower recovery rate in order to have a stronger therapeutic alliance; a treatment with a 48.54% lower recovery rate to ensure that the therapist is warm, empathic, and accepting; a treatment with a 25.77% lower recovery rate to ensure a therapist with greater amounts of education and professional experience; and a treatment with a 34.71% lower recovery rate in order to ensure that they would do more of the talking during therapy sessions. In other words, clients were willing to accept a treatment that does not work nearly as often in order to attain other things that are important to them.

The authors looked at these numbers and concluded that it might not make sense to always start with the treatment with the greatest level of empirical support, as clients appear not to weigh effectiveness very highly and, as such, might drop out of treatment prematurely given the absence of other factors.  That is certainly one way to look at these numbers, but I, perhaps not surprisingly given the focus on this site, have a very different view:

First of all, implicit in the question asked in this study is the idea that scientifically supported treatments are lacking with respect to common factors.  In other words, it makes it sound as though choosing the evidence-based treatment means choosing a less experienced, less empathic therapist you won't enjoy working with and who will talk at you the entire time you are there.  Who would want that?  Actually...I might if that therpist is the one most likely to solve the issue that is causing me such distress and impairment, but I suspect I might be in the minority on that one.  The problem with all of this is that there is absolutely no evidence whatsoever that evidence-based treatments maintain lower levels of any of these variables.  In other words, the choice presented to the clients is one that there is no reason they should ever have to make. 

My second issue with this is the weight given to client preference over outcome.  The assumption in this paper is that, if a client ends up in a treatment they prefer less, they are more likely to drop out and less likely to have a strong outcome.  This makes intuitive sense.  The thing is, the evidence does not back that up either, at least not with any degree of consistency.  A little over a year ago, I wrote about results on a study that examined this very question (click here to read that article) and which found that clients who received their treatment of choice were no less likely to drop out of treatment and no more likely to exhibit greater improvement post-treatment.

My third issue with this is that clients first presenting for treatment are often lacking experience with respect to the notion of empirical evidence.  In other words, most folks out there are not a nerd like myself.  They do not read scientific journals in their free time.  They have not taken a boat load of graduate level stats classes and developed a strange affinity for numbers.  They do not think about things like regression to the mean, mechanisms of change, and actuarial data when considering the value of therapy.  As such, when they are given the opportunity to express their preference for numbers or common sense ideas that we know from day-to-day life (e.g., talking to somebody who seems to understand you), the question in inherently weighted in favor of the second option because the client has not been given any opportunity to fully understand the first one and they're in a hurry to get some help for something troubling enough to prompt them to come to a clinic in the first place.  Afterall, common factors are what most people think about when they think about psychotherapy.  What makes the work of folks like Paul Meehl, who so clearly and repeatedly demonstrated the superiority of actuarial methods to clinical intuition, so remarkable is that it runs in direct contrast to our natural expectations.  Sometimes the facts are not obvious.  Sometimes things do not work as we expect that they will, even though we make a compelling case for why we intuitively believe that they will.  That's why we need to rely on numbers.  That's why sometimes, when we are upset and we have a preferred path, it is important for an imformed health care provider to encourage us to consider an alternative.

My point here is not that clients should be forced to take part in a particular form of treatment (although I do think that only once evidence-based treatments have been attempted should non-evidence-based treatments be covered by insurance...this as a means to motivate clinicians to provide treatments that work rather than simply treatments they can sell).  Instead, I think that, when a client expresses an interest in non-evidence-based treatments due to a preference for common factors that, for one reason or another, they fear the evidence-based approach will lack, the therapist should engage in motivational interviewing and psychoeducation, thereby allowing the client the opportunity to arrive at their own conclusion from a more educated stance.

Swift and Callahan (2010) put together an interesting study here and raised an extremely important topic of conversation, but I ultimately disagree with the general conclusions of their paper.  That being said, I do agree with their conclusions that therapists and clients should make decisions collaboratively and that therapists should not assume that they know their clients' preferences.  I simply think that client preferences could, at times (based on these data), be due to a lack of information and, regardless, be unlikely to steer the client towards a better outcome.  What do you think about all of this?

************

If you would like to learn more about this or other topics discussed on PBB, we recommend that you consult our online store of scientifically-based psychological resources.

Mike Anestis is a psychology resident at the University of Mississippi Medical Center and a doctoral candidate in the clinical psychology department at Florida State University