Andrew Skodol: Personality disorders in DSM-5

well thank you professor woofer and Sophie steyr also for inviting me to address your group today it's a fairly long way from Phoenix Arizona to Stockholm it takes about 19 hours each way but it was a comfortable trip just the same and I've spent the my previous day here doing a little sightseeing you have a very beautiful city and setting to to live in so you're to be congratulated so I'm here to talk to you about personality disorders in dsm-5 and beyond so this is a very sort of future-oriented presentation and I'm going to talk about classification but I really want to underscore that the real purpose of classification you know is to try to increase our understanding of patients with disorders and so the better our classification can reflect true nature of particular psychopathology the better our patients will be understood and theoretically treated so this is just a disclosure in case anybody here cares they care a lot about this in America who pays you et cetera and whether you're going to talk about drugs I'm not going to talk about drugs at all if I mention a drug it's a mistake since I don't think they're actually that useful for people with personality disorders now dr. Brooke professor mentioned that dsm-5 is is unusual in a couple of respects in the section to the main section of the manual personality disorders the ten personality disorders as they were defined identical correct criteria dsm-4 appear so unchanged really unchanged an unedited the criteria for the disorders so you'll find the the three clusters the cluster a B and C that you may be familiar with the ten categorical disorders and personality disorder not otherwise specified the residual class there and the only really major change from DSM 4 to section to dsm-5 is that there's no multiaxial system of diagnosis anymore in dsm-5 and that's that's a long story in and of itself but so there's no access to anybody who's used to referring to personality disorders as access to disorders is going to have to retrain their brains to delete access to the other interesting thing about the the giving up on the multiaxial system is it happened in the process of dsm-5 without almost any discussion at all and certainly no no scientific justification was offered or anything like that it was just decided that they didn't want a multiaxial system and they would eliminate it but it remains to be seen whether that's a smart decision or not a smart decision one of the things that well I think that one of the beneficial aspects of multiaxial system is that since personality disorders were first defined by criteria and put on a separate axis which is now 35 years ago goes back to DSM 3 in 1980 we you know we've really learned a lot about personality pathology and its clinical significance first we've you know found out that personality disorders are very prevalent in clinical populations and in the community 10 to 12 percent in the community depending upon you know whether patients are in patients or outpatients minimum of 20% would have a person a disorders maybe even up to 50% among patient populations they're associated with high rates of social and occupational impairment now and this is both with respect to role functioning and quality at worlds status and quality of functioning so that means that people with personality disorders are less likely to be married less likely to be in long-term relationships you know less likely to be gainfully employed and so on and so forth but if they are in one of those roles they're functioning in that role be it social or occupational is lower and this has been established you know versus you know a major axis one so called axis one disorder major depressive disorder which is got its own functioning problems but personality disorders particularly the severe ones like borderline and schizotypal are significantly more impaired than patients with major depression who don't have a personality disorder we've also learned again in both clinical and epidemiologic samples that comorbid if you will personality disorders result in a slower recovery more likely relapse and greater chronicity over time for a variety of other mental disorders including major depression alcohol and other substance use disorders and a variety of anxiety disorders so these patients who are comorbid will you know will not respond to typical treatments for those so-called axis one disorders in the fashion that you would think and the problem is that oftentimes clinicians treating people with symptom disorders like major depression or anxiety sores don't actually pay attention to whether the person has a personality disorder and if they don't then when the person doesn't respond to treatment they may for example if it's a psychopharmacology change the drug or add the drug or something add another drug or do something like that when when in fact what the person has is an underlying personality disorder that is getting in the way of their full response and recovery and unless and until that personality disorder is given some specific treatment you know the course of the symptom disorder will not will not really improve and finally we found out that personality disorders are very costly to society it's a high association with various kinds of criminal behavior with substance abuse with child abuse at spouse abuse violence so on and so forth so personality disorders are a significant clinical and public health problem in and of themselves so when we set out to revise Personality Disorder section in dsm-5 we obviously needed to pay attention to the existing classification and approach so that we were not doing what we call throwing out the baby with the bathwater in other words we weren't getting rid of things that had shown to be shown themselves to be clinically that valuable just you know just in order to pursue a new system but there were also many many problems with the DSM 4 3 R + 3 categorical classification of personality disorders there are some general criteria in DSM 4 and section to dsm-5 for personality disorders that require that the person have cognitive affective impulse control and interpersonal problems these general criteria very nonspecific for personality disorders many many other symptom disorders could meet these criteria including schizophrenia forms of affective disorder and so on and there was no empirical basis for these in fact they were they're kind of made up because the group at the end of DSM 4 decided that they needed some general criteria would it be nice to have some so let's sit down spend an afternoon putting them together but they're in the DSM for 5 another problem with categorical approaches is excessive co-occurrence or the comorbidity problem if you get a diagnosis of a personality disorder you are more than likely more likely than not to have at least one more personality disorder diagnosis and if you give a semi-structured interview like I have in some of my studies of personality disorders that actually ask all the questions for addressing all the criteria for the ten disorders in a severely impaired personality disorder population you might find patients who get six or seven or eight research based based diagnosis of personality disorder so it doesn't really make a lot of sense to make seven different diagnoses in a patient who you would consider to just have a really severe personality disorder they don't have seven independent conditions there's also a lot of heterogeneity in any this any of the disorders that are identified with so called poly thetic criteria sets we're like five out of nine or required but no single one is required so there are lots of combination combinations of permutations of five out of nine in fact there are 256 ways that you could meet the criteria for borderline personality disorder using the poly thetic system that's true of other disorders too when they're five of nine for major depression the new substance use disorder criteria set in dsm-5 requires only two of eleven criteria to make a diagnosis so that gives you over 2,000 possible combinations so the idea of within category heterogeneity plagues much of DSM and the personality disorders are problematic in that regard inconsistency of criteria content the criteria for personality disorders in DSM are mixtures some there are some criteria that are very trait like there are other criteria that are specific behaviors and you know we've shown in a longitudinal study that the trait part of personality pathology is much more stable than the behaviors borderline patients may cut themselves or do in a various sorts of self-mutilating acts but they don't do it on a daily basis or a weekly basis or a monthly basis or sometimes even on a yearly basis so infrequent behaviors have the same kind of weight in DSM for as more persistent and prevailing personality traits there's instability and personality disorders PD's one of the reasons for putting them on axis - originally was the thought that they were more stable types of psychopathology as opposed to episodic depressive or anxiety disorders well it turns out that both of those assumptions are wrong personality disorders can change and do have fluctuating levels of pathology over time we did a ten-year longitudinal study of patients with personalities orders and not only our study but other clinical studies and studies of non patients and studies in the general population of all shown the same thing and that is that personality disorders can change and really improve over time with some with some relapse and then it turns out that some of the anxiety disorders are much more chronic than than they were thought and same with depressive so the diagnostic thresholds this whole idea of five out of nine or four out of eight those were also pretty arbitrary the only two diagnoses in DSM for now dsm-5 section two that had any attempt to make their cut points scientific or empirical we're borderline and schizotypal and that was done in a study by Robert Spitzer who was in charge of DSM three in DSM three are the United States in 1979 he did a study to map criteria on two clinicians diagnoses of borderline personality disorder and borderline schizophrenia that eventually became schizotypal and he established these cut points but since then the criteria have changed the cut points have never changed and there are also these other eight diagnoses and the basic rule of thumb I guess among the committee's that preceded dsm-5 were that you know if you were going to get a diagnosis you should have at least half the criteria so that's that's basically the whole rationale for the numbers and I think clinicians know that the person who has four out of nine borderline criteria is not that different than somebody who has five out of nine there's nothing magical about you know making the threshold and it's really a continuum of pathology that you're talking about this is just one more evidence about diagnostic thresholds not not necessarily being appropriate poor coverage of pathology even though there are ten specific disorders were and are in the official classification the most common diagnosis is personality disorder not otherwise specified for people who don't meet the criteria for one of those ten poor convergent validity means that you know instruments designed to measure these disorders don't often show much correspondence so that it means that it's they're kind of difficult concepts for people to operationalize and so the instruments measure different things you could look at two studies of borderline personalities or if they were diagnosed by different instruments the overlap between the disorders might be fairly minimal and finally compared to trait models categorical models really have limited predictive validity and clinical utility so there are a lot of issues with the categorical approach and that's really why we started out trying to change DSM into a more dimensional model and the consequences of these problems sort of speak for themselves in that often the diagnoses were not used patients where people would write diagnosis deferred on access to in the United States where the multi actual system was required they would under use the diagnosis so that they we use that PD NOS most frequently or they were erroneously used sometimes the diagnoses were applied when the patient had you know one or two criteria so you'd have a patient who had mood instability and an impulsive suicide attempt and you'd call them borderline but that's not shouldn't be the case that's only to the potential criteria but no one actually took the time to do the full assessments in in clinical settings that would be required so the dsm-5 personality personality disorders workgroup set out to reformulate the ten DSM for PD diagnostic categories and we wanted to come up with a clearer conceptual basis of what pd pathology is what what really is a personality disorder as opposed to another kind of mental disorder we wanted to provide a more efficient and effective approach to assessment and we wanted to have a more empirically based formulation of diagnostic criteria for the the PDS and you know a long story made very short at the eleventh hour the Board of Trustees of the American Psychiatric Association decided that our proposal was too new and too different and therefore it should be placed in Section three for emerging measures and models although it was given this name of the alternative model for personality disorders suggesting that it could be used to you know assess and diagnose patients with personality pathology as a option to the standard approach so what is our alternative model what you know what basically does it look like and so we were told that it was too complicated and I've you know thought a lot about how to boil it down to a sort of a very simple kind of schematic but first I'm going to just review that the changes in the alternative model we've revised the general criteria for personality disorder we've put in a new measure of severity of personality Pathology called the level of personality functioning scale we've required an assessment of pathological personality traits to go along with the severity assessment and we've redefined personality disorders in terms of impairments and personality functioning and pathological person he trades and we have a new concept of what we call personality disorder trait specified to replace personality disorder Nos so that now if somebody doesn't meet the actual criteria you would have some information about what kind of personality type or traits the individual had rather than just the blanket and not that informative diagnosis of PD nos so first the general criteria for personality disorder what are they well we call this a hybrid dimensional categorical model and it's not that all that complicated although some people did think it was it starts out with this assessment of personality functioning and I'll be talking about each of these component parts in the lecture today but the model requires moderate impairment and personality functioning along with the presence of pathological personality traits and both of these assessments are dimensional so there's a scale for measuring personality functioning and each of the personality traits is also measured on a dimensional scale and you add those two things along with some other standard DSM criteria that have been changed a little bit but not that much relative stability and consistency and various developmental cultural substance use other disorder and medical exclusions I mean those are and I'll show you a set of criteria a little bit later but those are kind of the standard exclusion criteria from DSM so the main idea here is adding these personality functioning and pathological personality trait models and putting that all together you get the revised general criteria for personality disorder so first the level of personality functioning what is this the level of personality function is based on a review conducted by Donna bender who's among one of the workgroup members for dsm-5 she did a literature review of existing measures of assessing clinician administer measures to assess personality pathology what was out there already that was designed specifically to determine whether somebody had personality pathology or not and she reviewed these instruments looked for constructs that appeared in multiple instruments and constructs that could be shown to already have been reliably measured by clinicians and research studies and she created this continuum made up of five levels of personality functioning addressing key self and interpersonal elements now the elements addressing SEL what we call self functioning our identity and self direction an identity is defined as the experience of oneself as unique with clear boundaries between self and other stability of self esteem and accuracy of self appraisal and capacity for an ability to regulate a range of emotional experience and under self direction we have pursuit of coherent and meaningful short-term light and life goals utilization of constructive and pro-social internal standards of behavior and ability to self reflect productively under the interpersonal domain we have empathy and intimacy empathy being comprehension and appreciation of others experiences and motivations tolerance of differing perspectives understanding effects of own behavior on others and under intimacy we have depth and duration of positive connections with others desire and capacity for closeness and mutuality of regard reflected in interpersonal behavior so these four elements then are added together in the clinical mind the mind of the clinician and rated on a five-point scale ranging from little or no impairment through moderate serious and extreme impairment and in the scale if you have a DSM manual you can access the section three part of it you'll see you know there's a scale on the back and there are detailed definitions of what some impairment versus moderate impairment versus severe impairment would look like under four different domains identity self-direction empathy and intimacy so this is this is all spelled out and this is a single global dimensional rating so so Donner designed this scale based on this literature review and then another colleague les Mori from Texas A&M was also on our workgroup did a study in which over 300 clinicians 307 clinicians rated one of their patients on all DSM four criteria and all elements of the dsm-5 alternative model DSM four criteria by the way were presented randomly so that they couldn't you know they couldn't have biases about you know what kind of pathology the person had or didn't have and the dsm-5 elements were also presented in random order and in looking at this scale this five-point scale he was able to establish that level of to moderate impairment where greater had the best combination of sensitivity and specificity for a personality disorder so if you looked at some impairment you know obviously everybody with the personality disorder would have some impairments but so would a lot of other kinds of patients who didn't have personalized ORS so that would have high sensitivity and low specificity if you looked at severe impairment well very few people without a personality disorder have severe impairment but you don't pick up all the people who do have so you have lower sensitivity so that if you're looking for a cut point on a scale to make judgment about a diagnosis where you would want to make a diagnosis then you want to do the best balancing between sensitivity and specificity that you can and that's why we empirically decided that moderate impairment or greater is the cut point for having a potential to have a personality disorder this this also means that in dsm-5 section three approach to PDS that if you if you mastered this one scale you have a very good screening instrument for personality pathology in your patients so if you you know with it with this kind of sensitivity of specificity you could do an assessment of this scale this one item basically and and be in pretty good shape about whether the person was at high or low risk for having a personality disorder in high risk then you could go on and do more extensive evaluation so the second part of the model then is the pathological personality traits now there are five trait domains some of you may be familiar or have heard of the five factor model of personality disorder while these five trait domains in the dsm-5 model represent if you will the pathological ends of the five factor spectrum so negative affectivity corresponds to five factor neuroticism that is generally pathological in its own right detachment is is the opposite of extraversion if you will antagonism is the opposite of agreeableness disinhibition is the opposite of conscientiousness and psychoticism isn't really the opposite of openness but it's needed in the in the model to account for disorders such as schizophrenia in a general population that involved over a thousand contributors to a set of 25 reliably assessed and overlapping personality traits so what if some of these trait facets look like within the trait domains well under negative affectivity we have traits such as emotional lability anxiousness separation and security hostility depressive 'ti submissiveness perseveration under another domain disinhibition we have impulsivity irresponsibility recklessness distractibility and ridge the absence of rigid perfectionism so rigid perfectionism counts sort of negatively towards the disinhibition domain so each of the domains then is defined so relatively short pithy definition of what the domain reflects so negative affectivity domain reflects frequent and intense experiences of high levels of a wide range of negative emotions such as anxiety depression guilt shame worry anger etc and their behavioral like self-harm and interpersonal like dependency manifestations so that's a definition for negative activity another example of definition of a domain is disinhibition orientation towards immediate gratification leading to impulsive behavior driven by current thoughts feelings and external stimuli without regard to pass learning or consideration of future consequences so each of the domains has you know one of these definitions and these you know these things I think ring relatively true to people who work with patients with personality pathology as things that that characterize the various personality disorders that we now deal with in addition all 25 of the trait facets come with a short definition so emotional liability something you would see in borderline patients is identified as unstable emotional experiences and frequent mood changes emotions that are easily aroused intense and end or out of proportion to circumstances parallel definition for impulsivity something else you might see and patience or anti-social patience acting on the spur of the moment in response to immediate stimuli acting on a momentary basis without a plan or consideration of outcomes and difficulty establishing and following plans so all 25 of the traits are arrayed in these five domains and then given these brief in these brief definitions and then they're rated on again dimensional scales from very little or not at all descriptive of the patient through mildly moderately and extremely descriptive and generally speaking a trait would count if you will as describing a patient if it was moderately or extremely descriptive then that would be something that would apply to a given patient so what about the disorders okay so section three personality disorders one of the things we decided fairly early on is that we weren't going to describe with criteria all ten of the disorders now talk a minute a little bit more about the rationale for that but we focused on six disorders antisocial avoidant borderline narcissistic obsessive-compulsive and schizotypal and then as I mentioned before we have the four we have the trait specified PD trait specified that could represent the other four personality disorders as well as any other presentations of personality disorder that one might find so the the the reason that we focused on the sixth was which to ask ourselves and we did look at the literature you know which of these disorders had a reasonable amount of data about either their validity or clinical utility and we looking at the publication's from just a number of publications from 1994 which is when DSM 4 was published to 2007 or 8 it was clear that the most information existed for three of the disorders and those were you probably would guess antisocial borderline and schizotypal and then there was hardly any data at the other end of the spectrum on disorders such as schizophrenia Birchley no empirical data at all to support you know those entities and then in the middle there was kind of a modest amount of information about clinical utility or validity so within it within that middle group there was a little bit of picking and choosing but you know taking a diagnosis such as avoidant personality disorder surprising to me to find out that it was it's actually one of them might not be surprising to you but it was one of the most commonly made diagnoses and clinical populations and also one of the most empowering in fact second to borderline avoidant was as you know in the literature is associated with more impairment than you know most of the other PDS obsessive compulsive doesn't have nearly as much impairment but it is also a very commonly made diagnosis in clinical populations and has been shown to be associated with very high a cost of treatment utilization so it has its own you know importance borderline of course nobody would particularly agree disagree with some of the other diagnoses paranoid you know I said there wasn't a whole lot of information literature about paranoid there were virtually no studies of paranoid person eyes or that didn't have some other diagnosis interestingly either had a an axis one diagnosis and had paranoid symptomatology or had another axis - like schizotypal that had you know shared some paranoid cinematography there's virtually no study about paranoid personalities or in the absence of some other kind of pathology other problems with something like dependent personality disorder clinical populations that you know what people have done review studies some patients some studies would find that 30 percent or 25 percent of the patients had dependent personality disorder in another clinic that same percentage would be 2 percent so there's if something has that variable prevalence in two kind of similar types of clinical settings you wonder about how difficult or it is to you know assess when somebody really has a dependent precise so there are various reasons that we eliminated some and focused on others but these are the six that we chose and I'm going to now just walk you through the criteria for the one that probably interests most clinicians which is borderline personality disorder so thinking back to that model of the general criteria impairment in personality functioning pathological personality traits when we get to the disorders criterion a becomes the impairment and personality functioning and criterion B becomes the pathological personality traits and in criterion a each of those parts of the personality functioning model identity soft Direction empathy and intimacy have a disorder specific description of the type of impairment that they typically show the patient typically shows so patients with borderline personality disorder in the identity domain typically show markedly impoverished poorly developed or unstable self-image often associated with excessive self-criticism chronic feelings of emptiness and dissociative States under stress in the self-direction domain these patients would have instability and goals aspirations values or career plans and borderline is by definition the personality disorder of instability so it definitely affects how they ouch you know life's goals under the empathy domain borderline patients would have compromised ability to recognize the feelings and needs of others associated with interpersonal hypersensitivity ie prone to feel slighted or insulted as well as perceptions of others selectively biased towards negative attributes or inch or vulnerabilities and under intimacy we would find the typical intense unstable and conflicted close relationships marked by mistrust neediness anxious preoccupation with real or imagined abandonment close relationships often views in extremes of idealization and devaluation or alternating between over involvement and withdrawals so this is a typical interpersonal behavior of borderline patient that would be captured by the a criteria in the dsm-5 section 3 model under the criterion B pathological personality traits we have a somewhat complicated algorithm here but I'll explain it in a second we have four or more of the following seven pathological traits including at least one of the following which is number 5 impulsivity number 6 risk-taking or number 7 hostility so first of all from a theoretical point of view it within our group some a few people thought that borderline patient out personality disorder could adequately be represented by emotional dysregulation or in our terms negative affectivity but other folks in our group felt that no that really wasn't a true representation of borderline that there had to be some elements of impulsivity or some elements of antagonism some additional personality traits that were beyond emotional dysregulation and so I'll come back to this in a sec we did a I think I'll explain as I I'll take it in order and then I'll come back up and tell you how we came to the algorithm so then the rest of the criteria are reflected in these C through G which are the this again the standard criteria that measure whether the person's impairments and personality functioning and personality traits are relatively inflexible and pervasive whether they're relatively stable across time on sets back to adolescence or early adulthood whether the they're not better explained by another mental disorder and so on with the various exclusions for substance and medical conditions and also cultural and developmental processes so these again are not major changes from DSM for the big changes are in criteria a and criterion B so how did we arrive at any two of the four a criteria for a diagnosis well again we use that survey of questions who rated their patients on both the DSM for the dsm-5 and at first we were actually going to require one self functioning criterion one interpersonal criterion but when we actually looked at the data we found that the best combination of sensitivity and specificity was any any two so for each of the disorders one self and one interpersonal had you know reasonable sensitivity and specificity but we actually wanted the criteria to have higher sensitivities and it turned out that any two was really preferable to one of either self or interpersonal and so we ended up based on this you know this data set deciding that the a that the algorithm for the a criteria should be of any two and not what we originally set out thinking it would be now how do we assign the traits to the disorders again this is a busy slide but I'll point out a few highlights and try to illustrate what how the traits end representing the disorders again using the same sample of 337 who patients were rated on all the disorders and all the traits we were able to look at the correlations between each and every trait and each and every disorder and we were able to look at them both at the domain level so the red bold numbers in bold are the significant correlations at the domain level and then the black bold numbers are the correlations at the at the personality facet level so you can see for example that antisocial personality disorder is made up of primarily traits in the antagonism and dissing abyssion domains with much lower correlations with the other three in contrast to that borderline is has a very high correlation with negative affectivity that should be and lesser correlations with antagonism it's still still significant and disinhibition but not much in psychoticism and of course not much at all in detachment and schizotypal it looks like it's largely a combination of detachment and psychoticism and so this is the way that the traits kind of are clustered within the disorders by these large domains and then the individual traits within it I mean if you look at the black bold black for example for antisocial hostility manipulative nnessee fulness irresponsibility impulsivity and risk-taking those are the six criteria for antisocial personality disorder and those are the six criteria that have the highest correlations to the DSM for diagnosis of antisocial personality disorder for schizotypal the highest correlations are in the unusually sand experiences and eccentricities and cognitive and perceptual dysregulation as well as these three traits in the detachment domain and so that so again those are the two groups of traits that define - gets a type of personality disorder so in ever in in virtually all cases of those disorders the traits that had the highest correlations represented up representing the criteria now there are a couple of exceptions if you look at narcissistic personality disorder the to that criteria that are assigned to that diagnosis are grandiosity and attention-seeking but there are some other you know fairly significantly correlated traits particularly within the antagonism domain such as manipulative Ness deceitfulness callousness etc so those could have been added perhaps to the criteria for narcissistic personality disorder but you know what happens if you do that is that you're basically making the diagnosis of narcissistic personality disorder into the diagnosis of antisocial personality disorder that you would be increasing the the co-occurrence of those two disorders and making them almost identical some people believe that narcissistic personality disorder is a minor variant of antisocial but you wouldn't necessarily want to conflate these diagnoses and so we made the decision a conscious decision not you know to try to make these criteria sets as discriminating discriminating between themselves as we could we could be so there are some instances where a highly correlated trait you know we decided not to put into the into the criteria set and then in terms of the actual algorithm using the trait so we so we had decided that we would you know look at various combinations of traits and we would look at the DSM for diagnosis in this 337 person dataset and we look at the prevalence and then we would look at all various combinations of traits including complicated algorithms that had you know traits from different domains alternatives this being the one that we have eventually chose require one from negative affectivity and one from at least one from disinhibition antagonism and the reason that we picked four or more which is the rule for borderline and making one at least one be from disinhibition and antagonism is that this prevalence is almost the same as that which is the dsm-4 and this kappa of agreement is virtually is the highest this one is slightly higher but we were also looking at making the discrimination with other disorders as low as possible and maximizing the relationship to functional impairment in order to choose a rule and so we basically applied this complicated formula to all six of our diagnoses to be in order to be able to say that we had now some kind of empirical rationale for what the algorithm for making the diagnosis on the B criteria would be and so the rules we developed based on those kinds of analyses for antisocial we had there are seven criteria and a six of seven was the one that worked the best in terms of meeting that you know matching the prevalence for DSM 4 as well as the Kappa and decreasing that's the increasing the discrimination with other disorders and maximizing the relationship the functional impairment for avoidant it ended up being three or three or more for one of which is anxiousness we're borderline I just told you is four or more of seven one of which is from detachment or antagonism narcissistic which only has two criteria both a required obsessive-compulsive schizotypal again so the all of those again thresholds diagnostic thresholds for the six disorders were empirically determined a complicated system well you know the dsm-5 did a field trial in large academic centers and one of the one of the sets of disorders they were interested in was personality disorders and in this field trial this was take people who who did work at academic centers but weren't particularly trained in the new model or any of the criteria particularly in dsm-5 they were basically looking at the new criteria for the first time and you know filling out information on patients that that they were seeing in their clinics well it turned out that at least for borderline personality disorder the Kappa for the inner class Kappa for the disorder was 0.5 for considered good in the our terminology and it really was as good as the diagnosis of bipolar one slightly better than the diagnosis of schizophrenia a lot better than the diagnosis of major depression which didn't change at all and it really only exceeded by significantly by a major neurocognitive disorder which is dementia and and PTSD so you know even with these fairly complicated personality functioning and trade criteria experienced but relatively naive in terms of their exposure to the system clinicians these criteria could could be applied reliably in the academic field of trials so what about clinical utility so again one of the criticisms that we faced was you know were we being disruptive you know to practice and to research I mean there's all this evidence out there now about various treatments for borderline and so on and so forth and you know were we changing things so much that you know we would have lost the the concepts on which all the research and training had been based well as it turns out that correlations between our diagnosis and DSM for diagnosis are really quite high borderline was the highest point eight oh three and a social was high avoidant was high narcissistic was high schizotypal was okay and obsessive-compulsive was a little lower but these correlations between so the change from dsm-4 to dsm-5 based on these correlations would be less than the change that had occurred between DSM 3 and D s and 3 are in other words there were more they're more profound changes made in between DSM 3 and D s and 3 are that were really not protested by anybody they just happen but they were that was obviously a much more drastic change than than we were making so we felt based on this data reasonably confident that we had preserved the traditional constructs we had just recast them in terms of these personality functioning and personality trait models that we felt were much more closely reflective of what we think about personality disorders then then the criteria of DSM 4 and apparently so did the people who participated in the field trials this is the personality disorder column and if you looked at the number of people who thought that DSM 5 was moderately very or extremely more useful than DSM 4 now remember these people had worked with deism 4 since 1994 and had been given the criteria for DSM 5 probably two days before they assessed their patient but over 80% of people who participated in those field trials and made a diagnosis of personality disorder thought that DSM 5 was moderately very or extremely more useful in DSM 4 and that was the third-highest of all the categories that were examined in the field in the academic field trial and the same basically was true in the so-called private practice field trial small solo and small group practices these were not academic clinicians more than 80% found the new criteria moderately very extremely more useful in comparison to DSM 4 we did a as part of the survey again of the 337 patients we asked those clinicians to rate the two system on questions like how easy was it how useful would it be to communicate with other professionals communicate to patients comprehensively describe your patients personality problems formulate effective interventions describe the patient's global personality not only in these first two the DSM for exceed some aspects of dsm-5 and you wouldn't be that surprised if somebody who had been using a system for you know 20 years told you that it was easier to use DSM for than it was to use dsm-5 but you know when it came to the trait model its kind of amazing that even with ease-of-use and communicating with other professionals and all the others the trait model was Rick was rated as you know more clinically useful by these clinicians than DSM 4 we also asked another question so it's one thing just to say to somebody well how useful is it it's another thing to find out whether they actually whether the information predicts some of the judgments that they make about the person that would be you know clinically important so remember they didn't know exactly what diagnoses they were making they were rating criteria randomly but we then looked we then put their diagnosis together based on their criteria ratings and we correlated and predicted and did regression analyses on the clinicians judgment about the patient's composite social functioning so social occupational and leisure functioning risk composite that's future risk for self-harm criminality or violence estimated prognosis from poor to you know good and level with treatment intensity from no treatment you know through outpatient inpatient and so on and so forth and basically if you looked at the DSM total dsm-5 total it's all aspects of the model put that put those together the predictions of these judgments was much more related to their dsm-5 assessment than it was to their DSM for assessment so that this is much closer to you know demonstrating that these constructs that were embodied in dsm-5 were used by you know the clinician in making these judgments or could be used by the collision in making these judgments and if you actually took the two systems DSM 4 and dsm-5 and you took control for all the information dsm-5 and then saw whether the information DSM 4 added anything to these predictions it rarely did slightly to the risk composite but if you did it the other way around dsm-5 information once you controlled for all the information and do support you've got these highly significant predictions above and beyond the information contained in those criteria so you again could argue that in terms of incremental validity with predictions such as these that the dsm-5 was much more related to the predictions than the DSM 4 so just a you know again looking to the future want to mention some other models of personality pathology you're going to guess here from Thomas in Seoul and from the NIH in a while and I think he'll probably be talking about the you know research domain criteria you know the NIH is the place where most American researchers go to try to get grant funding and that's that's you know you apply for grants to the NIH to fund your research and the NIH is basically said you know don't come to us with any grant about any diagnostic category be it schizophrenia major depression borderline personality disorder they really are not interested anymore categorical diagnosis because they believe as it says that there's no apps that there's no rigorously tested reproducible clinically actionable biomarker for any psychiatric disorder at this point and so they want to shift the emphasis of research towards underlying cognitive and motivational processes that go cut across diagnostic categories but are more basic to the kinds of things that where we might be able to find biomarkers or genetic patterns or so on and so forth and so this group of what they call research domain criteria consists of five domains one is negative valence that's aversive types of conditions involving fear anxiety loss and that could be anxiety disorders depressive disorders other disorders a positive valence involving reward seeking and habit influencing things like substance abuse and so on cognitive domain involving perception attention memory and control and then for our purposes of social processes which involve things like affiliation attachment and understanding and perception of self and others and that is almost identical to the personality functioning constructs that we put into the personality disorders and then there's a domain also about arousal and regulatory systems but you know Tom Insel believes and the nmh people believe that there's this need to deconstruct current diagnostic groupings into these more fundamental pieces in order to kind of move research forward and it's just not going to happen based on you know traditional DSM categorical diagnoses and if you look at how they're already the RDoc and the dsm-5 alternative model align you know we have a bridge here you know we have a bridge between traditional psychopathology and the RDoc in that the negative valence system is very much like our negative affectivity trait domain their positive valence system is the opposite of an tag our antagonism domain and under social processes we have a trait domain of detachment and then we have identity and self-direction and empathy and intimacy that align very well with understanding perception of the self and understanding perception of others so again I think you know we think that there is a big opportunity here to go through the alternative model a personality to tie the RDoc concepts in with more traditional diagnostic approaches and you'll undoubtedly be hearing soon about that if you haven't already about the icd-11 proposal for personality disorders again it's a little different from ours but if you're looking for similarities you can certainly find them there they're diagnosing personality disorder on a continuum of severity so they're having a one single personality disorder diagnosis that can be primarily DIF personally difficulty which is sub threshold for personality disorders somewhat like our some impairment and personality functioning and then they have mild moderate and severe personality disorder kind of like we have moderate severe and extreme impairment and personality functioning and then they have five trait broad trait domains and we have five broad trait domains and they align very closely in fact the only major difference is that they have an Anand caste ik domain that has sort of obsessive-compulsive components and we have the psychoticism domain so that we can make the diagnosis of schizophrenia sooner and they will have no PD types no borderline no antisocial no type of any kind if this goes through they will just have personality disorder and then characterized by combinations of these domains so it's very very similar in a way to you know what we proposed of some differences but you know I think that these two systems moving forward will provide you know much different kinds of information about personality disorders than what we get from current you know categorical approaches a word on measurement tools to assess personality traits there's something called the personality inventory for dsm-5 it's a two hundred twenty items self-report scale patient fills it out by themselves they rate the traits on four-point scales from very false or often false to very true or often true they're scoring for the twenty five trait facets as well as the five main domains and there's a informant version a brief version and a child version full and brief I might say that the child version is no different from the adult version I'm not exactly sure why the APA put that in there but it is it's basically the same but so there's this self report that's been out and there's a lot of research has been accumulating about the the trait model how it relates to the five factor and other trait models and how well it represents personality disorder pathology and then we have a new semi structured interview for the alternative model that we just sent to the American Psychiatric press and they're agreeing to publish it it's going to be a four module semi structured interview with one module addressing level of personality functioning one personality traits one personality disorders and one the complete alternative model this would be for clinicians and you know would require some experience with patients who have personaiiy pathology and and specific training and instruments but so this will be probably available within six months for you know people who wanted to actually have a semi structured interview that would address the various components of the alternative model general features of the skid a.m. PD we have a general overview that includes demographics education and work history currents or past psychopathology and treatment suicidal thoughts or actions and then we have a personality overview that has from two to eight questions depending upon the module about perceptions of self and others and interpersonal relationships so we ask the patient you know how would you describe yourself how would other people describe you who are the most important people in your life how do you get along with them how well do you think you understand yourself how well do you think you understand other people those kinds of overview questions to get the person to be talking about the kind of material that you need to rate but then we have multiple specific questions that reach that rate every element of personality functioning every trait and every criterion and then it can be scored on multiple levels and there are course detailed instructions for ratings and a user's guide so to summarize dsm-5 section three advantages over section two we think that actually the DSM 3 approach decreases clinician burden it increases clinician information you think about that level of personality functioning scale as a screening instrument you know you have this one item test that you could apply to your patients that would give you a high likelihood of knowing whether they had a personality disorder or not and if you went through all the criteria there there are 94 criteria for PDS and DSM 4 there are 54 criteria for PDS and in section 3 so we think it decreases clinician burden and increases the information the section 3 decision rules you know the diagnostic algorithms clearly more empirically derived than what's in DSM 4 or section 2 we think DSM 3 better addresses PD complexity and enhances understandings of patient's personality pathology if patients have traits that are in the same domain like antagonism it's no wonder that those disorders often get diagnosed in the same patient it's because they share these underlying fundamental features and maybe there shouldn't be any categories maybe it should be just based on the the fundamental underlying features but at least you can see based on the you know the trade assignments and representations of the personalities or is why certain kind of comorbidities frequently occur the section three approaches perceived as more clinically useful that was both in the field trials and in our own 337 clinician study and the section three approach is more strongly related to clinical decision making about functioning risk prognosis and treatment assignment so with that I will stop we have time for formal questions and then I guess the laughter will gather outside and have more informal discussion so thank you very much you

Channel: PsychiatryLectures

Views: 75,719

Rating: 4.81106 out of 5

Keywords: dsm-5, Psychiatry (Medical Specialty), Diagnostic And Statistical Manual Of Mental Disorders (Book), Personality Disorder (Disease Or Medical Condition), Personality (Quotation Subject), Health (Industry)

Id: 4mgifm3ftl8

Length: 65min 28sec (3928 seconds)

Published: Thu Apr 30 2015