The Diagnostic and Problem Area data offers a more detailed and concrete look into the specific impacts of therapy upon the relatedness group. The Diagnostic Areas include:  anxiety; depression; and behavior. The category of relatedness itself is not included. The Problem Areas include: family relationships; school performance; social relationships; and community activity.

The definitions and data collection process were discussed in Post 21. As a quick review, the presence of any of these seven problems was usually determined during the three session assessment process. Infrequently, an additional problem(s) could and did surface during the course of the therapy. An interconnectedness of Diagnostic and Problem areas usually existed in terms of any given symptom, e.g. depressive symptoms connected to declining school performance, or a flagging school performance leading to defiance at home. The existence and impacts of relatedness traits specifically were based more on clinical observations during intake and therapy, and were usually not included as explicit elements of the family’s problem and clinical goals list.

The determination that a diagnostic or problem areas was no longer an issue was based primarily on client and family reports. Collateral observations from community professionals, mostly school counselors, helped refine the overall clinical picture of a case. 

As a reminder when evaluating the above data, all the relatedness clients were males. How that fact skews the results is not clear, but some kind of skew can be assumed.

Consider dividing the seven DA/PA categories into three distinct sub-groups, including: anxiety and depression; behavior, school, and community; and family and social relationships. Anxiety and depression are internalizing symptoms.  Behavior, school, and community problems tend to be the result of externalizing processes. Family and social relationships involve both.

The comparative results indicate that the non-relatedness groups tended to have internalizing problems more frequently, 73% compared to 50% for the non-relatedness group. The relatedness group similarly tended to have externalizing problems (76% to 50%). While relational issues are presumably driven by both internalizing and externalizing factors, the respective rates of their relational issues are almost identical to the externalizing data, perhaps suggesting, that the two relational status categories are more driven by externalizing processes.  If so, this result may be more driven by the all-male cohort.

The relatedness group’s strongest area of improvement was anxiety. Their rate of resolution was above that of the non-relatedness group, as were the results in family relationships. The lessening of anxiety would appear to be related to the improvement of family relationships.

In contrast, the lowest improvement areas for the relatedness group were depression, school, and social relationships. This data is amplified further by the fact that all of the unresolved area problems within the relatedness group were those of the unresolved relatedness cases. The relationship between depression on the one hand, and school performance and social relationships on the other is probably one of mutual reinforcement. Again, these were the lingering problems of the unresolved relatedness group of thirteen at their terminations.

DA/PA Average Gain X 17/39, 13/43, 4/13

Note: In this chart, ‘Pre” refers to average number of the eight DA/PA categories each group has intake.  ‘Post’ is the number remaining at termination, and ‘Gain’ is the difference. In Post 22, one resolved category was calculated to be the approximate equivalent of a 5 point CGAS gain.


The average unresolved relatedness case improved 2.0 DA/PA during therapy, evidence that youth having the misfortune of relatively intractable relatedness traits can nevertheless benefit from effective therapy. Being in the CGAS 50’s or 60’s is eminently more tolerable than being in the 40’s and 30’s, and offers a better opportunity for continued change and growth.

The vignette ending the previous post, #27, suggests that at least a few of the clients with relatedness problems that were unresolved at termination can continue to improve after termination. Two factors are almost certainly involved in determining which cases continue improvement and those that substantially do not. 

First, as the number of traits in an individual rises, the possibility of full recovery appears to diminish. In the aftermath of asserting this, though, while the outpatient client may never resolve their relatedness tangles, they are still to be considered competent and capable of making improvements in at least some of the diagnostic and problem areas. The work could take a long while without much in the way of guaranteed gains. This clinical reality does pose a social policy conundrum.

Second, family support, particularly that of the parents individually and relationally, is of crucial importance. Part of the clinical task is to support and facilitate their roles. Capacity, tenacity, resilience, and reservoirs of energy are helpful attributes.

The statistical comparison above of the resolved group of four with the thirteen unresolved group highlights both the suppressive nature of a problematic trait system, and the almost kinetic recuperation when the traits fortunately dissipate.

The initial CGAS and DA/PA average scores of the resolved and unresolved relatedness cases are fairly close to each other: 49.8 CGAS for the resolved, and 44.8 for the unresolved; and 5.3 initial DA/PA for the resolved, and 5.8 for the unresolved. Both the clients who ultimately resolved their relatedness traits and those who did not began therapy functioning a decile or more lower than the non-relatedness group of thirty-nine. 

The post scoring was dramatically different: 76.8 CGAS for the resolved, and 53.2 for the unresolved, and 5.3 DA/PA improvement for the resolved cases compared to 2.0 for the unresolved. The unresolved relatedness termination average was the lowest of any sub-group, and that of the resolved relatedness group was the highest.

The inference of the low Pre score for the entire relatedness group, from those with three traits to those with eight, may be having traits of any number is enough to disrupt one’s ability to relate, adapt, be accepted, and be trusted. How many traits may be a less significant factor than their mere presence. 

The resolution of a relatedness problem, though, is definitely related to the number of traits existing at the outset of treatment. The resolved four cases had three and four traits while the unresolved cases had all the other scores, including the remaining four-trait cases and all else from fiver to eight.

The inferences of this data are that the presence of relatedness traits inhibit the development of normal relationships and mire the young person in situations of being distanced if not outright rejected as peers and others in their lives; and that once the traits begin to dissipate and disappear, the young person can climb into the normal strata of family, school, and social relationships.

The presence of traits are suppressive. Families, classes, and peer groups can be forgiving when the client starts to make changes. Speculatively, the combination of the conjoint identification of family problems early in the therapy, family relationship work using a socio-cognitive perspective exercise, and behavior management information for the parents helped lessen client anxiety. A good process and therapeutic relationship can provide modeling. Reinforcement both planned and spontaneous, anticipated and random can provide conditioning. The ability of children and adolescents to adjust and welcome the functionally improved into the social fold becomes an important feature of the client’s growth.

At least anecdotally, the experience with these four resolved cases suggests that trait resolution tended to follow improvements in other symptoms and problem areas. Notably, these traits were small in number. Perhaps, as suggested earlier, these low number trait systems are more the result of modeling and conditioning. Another viewpoint is that higher numbers of traits quickly develop a life of their own, mutually reinforcing each others’ existences and become much more difficult to resolve. Research on identical twins raised in separate environments suggests the genetics alone cannot explain why lower levels of traits tend to be more easily resolved, but genetics may account for the higher levels of traits that one have.

 As the number of traits rose, the number of DA/PA at treatment’s outset rose as well. While some improvements, such as a lessening of anxiety and an improvement in family relationships, could be achieved, the higher numbers of the traits tended to make other gains difficult. Another factor with increased traits is a tendency to have other family dysfunctions that complicate the case and limit the capacity of the change process. Personality disorder-type defense mechanisms are potent treatment inhibitors as well.

The sad part of this schema is that many child and adolescent relatedness problems do not resolve. Positive changes can be made, and functionality can be improved, but existence at CGAS levels less than CGAS 71 continues. We do keep working our best until the case terminates. 

The last point is a reiteration. These listed relatedness traits used here are culled from the eighty total traits listed in the personality disorder section of the DSM. Many of the cases had traits from more than one of the personality disorders. The same can be said for the DA/PA section in general. Most had two diagnostic area problems, a few had three. a couple had four. Most had more than one problem area, a few had three, and a couple had four. No one had all eight, but in a larger cohort, a small few could be reasonably anticipated.

A proposed dictum: the functional issue for therapy is not the diagnosis per se, but rather the number of extant DA/PA problem areas and relatedness traits.

Lastly, a system that identifies diagnostic and problem areas is graphically more indicative of change realities than a system based on diagnoses. This is said in some part because symptoms and problem areas are the prism through which line therapists have historically seen clients. A problem list is developed and a therapy proceeds from that point. An active mutual termination generally includes a review of what changed and what did not, for everyone’s reference. 

Parenthetically, that style of treatment may be in the process of changing.


Comparative CGAS Data

The following chart lists CGAS averages to compare, including:  the seventeen relatedness cases with the other thirty nine in the study; the thirteen unresolved relatedness cases with the other forty three; and the thirteen unresolved cases with the four resolved.


The average relatedness-troubled youth begins treatment with a CGAS score more than a full decile lower than the non-relatedness cases. The average unresolved relatedness client begins almost fifteen points below those non-relatedness cases, who generally appear in the 50’s and 60’. These are the moderate – mild disturbance deciles. At that level, the client’s problems are usually noticed only in limited areas of functioning where the traits and associated problems are manifest. In other areas of their lives, the client does not stand out as troubled.

For the serious or severe deciles, the 40’s and 30’s, the problems are noticed in most or all areas of functioning, meaning the young people who have relatedness issues receive more negative critique. Consciously aware or not, the client is experiencing greater degrees of stress in their day-to-day lives. To add to their dilemmas, the traits themselves are clinically more difficult to treat than common anxieties, depressions, and behavior disorders. 

On average, the CGAS gains of the relatedness group were within a point of the average non-relatedness gain. In spite of the relatedness deficits, the clients nevertheless made progresses. On its face, the result was surprisingly positive. The overall statistic, though, disguises a significant disparity between the results of the thirteen unresolved relatedness cases and the four resolved cases.

The thirteen unresolved relatedness cases collectively had among the lowest average gains of any sub-group in the study. Paradoxically, the four resolved relatedness cases had the highest average gain of any specific group. In a statistical sense, the major distinction between the two group is the numbers of traits for individuals in each group. The four resolved relatedness cases included the one with three traits and three of the five cases with four traits. The thirteen unresolved cases included the other two with four traits, and the remaining eleven constituted all the cases with five, six, seven, and eight traits.

While the small N of seventeen precludes any suppositions about why this particular division in the relatedness group might be the case, speculatively, an overarching concept does come to mind. 

The DSM lists a number of diagnostic traits for each personality disorder, usually eight to nine. As with virtually all mental health diagnoses, to qualify for a formal diagnosis the patient needs to display a minimum number of the listed traits for the personality disorder in mind, usually five or six. Four or lower would be viewed as sub-clinical.

A second consideration of any given trait is whether the genesis is genetic or experiential/environmental. Would it be possible that a greater proportion of traits in a sub-clinical case be more experientially based, and for those that meet the clinical criteria of, say, five traits or more, a greater proportion would be genetically based? 

The reason this makes sense is that the four “resolved” cases all gained 24 CGAS  points or more, which was quite a long way to go from the beginning to end a therapy. Because they had been demonstrating these traits in their various life arenas, they all had experienced negative and sometimes punitive responses from those effected, complicating their lives further. The resolution of the traits themselves within this group was clearly easier.

The four resolved cases were able to overcome the emotional, behavioral, and social issues including the traits themselves and rise into the functional areas of the 70’s and 80’s, or transient problem and normal deciles. In contrast, resolution of traits with the unresolved relatedness cases were far less noticeable, and in several cases, none at all. Problems that result from modeling and conditioning are, almost by definition, easier to resolve than the more genetically based, thus the speculation that the resolved cases particular traits may have been more the result of conditioning and modeling than those of the unresolved.

The overall point when looking at the levels of improvement for both groups is that CGAS gains can be made whether the trait problems get resolved as not.  At lower numbers of traits, for whatever reason, gains can be truly impressive. But gains can still be made by those youth whose trait problems do not change much. 

Improvement can and does occur without necessarily changing the traits themselves. The modest changes as a result of the clinical work can and sometimes do evolve into greater changes over time. Once the positive feedback loop is injected into their living systems, an ability to potentiate change can persist. As in other kinds of problems, family involvement can be crucial to creating a change process, reinforcing the changes that occur, maintaining the changes over time, and promoting further growth.

Comparative Results Using Other Variables

This chart compares the unresolved relatedness cases (13) to the remaining study (43) using six variables, including: age; number of sessions; length of treatment; parent configuration; primary treatment format; and termination type. Values of each as follows:as follows:


For all variables except format, the relatedness and non-relatedness case were closely correlated. In that sense, they are similar. The format distribution leaning toward some kind of individual work with the clients is an indicator of both severity and complexity both. Remember that most of the relatedness cases began in the CGAS 30’s and 40’s where, for the most part, the non-relatedness clients began in the 50’s and 60’s. 

A picture of how a relatedness case presents does begin to emerge. If the client seems to be in a serious or severe state of functional impairment at the beginning of counseling, and the history does not include serious trauma, spectrum disorder, significant learning disability,  developmental disorders specified and unspecified, or other adequate explanation, a strong likelihood of a relatedness problem exists. 

Cases of this complexity and severity generate excitement – not necessarily the positive kind. For the less experienced clinician, the in-session self-management task may be to slow down, certainly in the first parent meeting. Likely, they are not looking for productivity or new ideas at this initial session. Likely, they are more wanting to be heard, understood, and appreciated for the position in which they find themselves, that being sitting in your office for their child’s problems that appear in both private and public dimensions. 

The receptions they’ve experienced elsewhere in the aftermath of concerning, annoying, and even angering behaviors on the part of their child have often been often pressured or unpleasant. All you may need to do is to neutrally and empathetically understand as you gather the intake information and get a feel for your clients. 

If problems with empathy and remorse begin to surface during the intake, given its high correlation with more intractable relatedness issues, the clinician buckles up. This could be a long and bumpy ride. But you want that journey. Relatedness problems have universality. Their occurrences cross lines within areas such as socio-cultural, racial, ethnic, economic, sexual or gender, and/or spiritual issues. Genetic factors are minimally half the influence, and they are challenging to any clinician. Fomenting change takes skill, time, and good fortune, all three.

The format findings, weighted toward split and consecutive therapy processes, reflects the complexities and lengths of therapies involving children and adolescents. A family therapist would like to do conjoint work as much as possible for a variety of reasons, but having optional formats available that can be tailored to the specific needs and demands of the client does lessen premature terminations. One primary responsibility is to protect the process.


The gain-per-session may seem trivial, but the results using the statistic for the overall study group do make an interesting point that will be covered in the study summary that follows the sub-group section. Here, a social policy point can be made.

Being in an era of efficiency, economy, and bottom lines both private and public, mental health care is as impacted as any other industrial institution. Outpatient practitioners are part of this industry, and our work itself is affected. One of the manifestations has been a broad managed care pursuit that has developed several methods designed to produce less expensive results. This includes treatments designed to produce quicker results, and others that limit services per problem, per individual, or per family per annum. Ethical problems abound. Line therapists know this, and work thru and around the constrictions in a variety of ways, learning to live with them and learning from them.

In putting this Gain Per Session chart together, the initial intent was to simply evaluate the statistic’s utility. What emerged, though, was a concrete example of the ethical dilemmas that the designers of managed care systems face, cognizant of these ethical dilemmas or not.

The list above is organized in terms of gain per session, from highest to lowest. The additional information is the levels of difficulty the clients had at the outset, the gain made, and the number of services utilized. In particular, note where the resolved cases stood in this hierarchy.

As a diligent managed care program designer, where, how, and on what basis would you make the cutoff or shape the therapy?

Please also note, I contacted the last, .06 case after organizing this chart. The client wanted to stop counseling. Understandably, the time had come and I concurred.  After five years of clinical work with a minimal gain for the efforts, the parents were disappointed, determined, and gracefully appreciative. Now five years after leaving counseling, the young man gradually did get better. He made his way through high school, then went to a Colorado community college to prepare for a career in the skiing industry, a sport at which he had always been adept. He is employed doing what he enjoys most, traveling internationally, now getting along well with family members, great job reviews, and no other particular problems. I did not go into a detailed update on relatedness issues, except that the mother said she and father still have their worries. But things right now are fine, very good, thank you. They think about the counseling experience “often”. Their dedication had been everything.


On Validity and Reliability

To quickly review, the development and collection of this relatedness data occurred within a few weeks of the original CGAS data collection, now four years ago. The cases demonstrating pre-Axis II interactional styles were chosen first, numbering seventeen. The trait list was then developed by reviewing all eighty symptoms of the ten personality disorders in the DSM V. Thirty-one traits were selected. Then each of the identified cases were scored using the list. One result is indicated above.

The validity and reliability problems with this methodology are readily apparent. Nothing can be seen as conclusive in these forthcoming comments. Conclusions are suggestive, as said before on occasion in prior posts.

The DSM’s III – V used the same estimate that 15% of the American adult population has a personality disorder, single or mixed type. The percentage of children and adolescents was not mentioned.  Whatever that national figure may be, 30% (17/56) of a general outpatient treatment population does seem realistic. 

The 15% figure is striking. That one in seven adults has a personality disorder seems high. One in ten, maybe; one in twenty too little; but somewhere in between one and two standard deviations. Just thinking in terms of what is and what is not one standard deviation (that 15% comprising the first level of abnormality), the psychological depth of a personality disorder seems deeper than problems of anxiety, mood, and behavior (the old Axis I groups). Hypothetically, the personality disorder would then be at a rarer level of abnormality, e.g. a 1.5 standard deviation off the norm.

The reasoning is that the breakdown of self-management and coping mechanisms seems different from the basic synergy of physical, mental, emotional, and social drives that governs how one creates congruence between a sense of self and sense of world about. The latter seems biologically more fundamental. A parallel could be the statistical difference between the incidence of sexual identity and that of gender identity. The first is generally seen as a 2% phenomena while the second is .2%, or thereabouts for both. One would think gender identity to be a more deeply embedded genetic factor. In the same sense, an Axis II disorder seems deeper and a statistically more rare phenomena.

Given this perspective taken into the scoring of these seventeen cases, the point here is that my own intuition may underestimate the frequency of personality disorder. The selection of these cases be an underestimate. The assignment of observed traits could be underestimated as well, FYI.

Comments on Results

The Age Distribution of the relatedness sub-group replicates that of the study group as a whole, with a non-significant tilt toward the younger client. The 12 – 14 modal area is also the case with the study group. The relative lack of 16 – 18 year old clients may represent the difficulty that these potential clients pose to parents who would like to see them in some kind of therapy but encounter steadfast refusals. Resistance can be stiff. Note the list of defense mechanisms in the next post, and think of adolescent vehemence. 

The results of Cases Per Number of Symptoms form a normal curve. The one case with three symptoms would be at the left end of the graph, and the one with eight symptoms at the right. The resolution rate within the relatedness sub-group was 4/17 (24%). The suggestion is that the typically more difficult case, one that would warrant an equivalent adult the personality diagnosis diagnosis, begins at the point of four or five traits for a youth. On the right side of the curve, uncommonly occurring cases may present with nine or ten traits; rare situations yet higher.

A few months before retiring, a mother and father came in with household safety and welfare concerns over their eldest,18 year old son. They were seeking either therapeutic help or other recommendations about what to do. At stake was having the young man leave their home, an ultimately sad resort. The problems were serious, multiple, of long duration, and rather than tempering as he neared graduation the issues were accumulating, including graduating itself. The parents themselves had reached a stage of resignation, or so they appeared, as they went through their list with an application of togetherness. Their other two kids, at least, were OK.

Taking down the presenting problems in that friendly, nodding, indecipherable professionalism, my pen stopped dead during their last example of challenging behaviors. 

A few weeks earlier, Grady was leaving school through the gym. The girl’s volleyball team was doing dig drills in the middle of the court as the coach stood and observed from the bench. The young man was heading past her toward the rear exit of the gym. That was the first sign. One veers away from a coaching coach while traveling though a gym or anywhere, not toward her, unless a good reason exists. His was not a good reason.

Stopping right next to the rapidly alarmed young woman, he casually propositioned her with a raunchy, vilest innuendo. She paled, and then reddened, and then yelled at him like only a coach can yell a “get out” that echoes off the walls, all happening within a few of short seconds. The team immediately knew what occurred without having heard anything distinctly. Her girls all started moving quickly toward her. The client hasted  out the door. All women were mad.

Instead of continuing my note-taking, I found myself just locking gazes with the two mortified folks, likely agape.. Something about this was almost predatory and not merely stupid, but I couldn’t know.

Grady and I actually did OK together during the individual evaluation session. He took the questions seriously, worked to provide answers, if not the insights the quality of thinking was there. Other directedness or self-awareness one would anticipate from a high school senior was amiss. Malice, though, was not a forethought.The outlook, still, was dubious. 

I met with the three of them next, skipping the summary review with the parents. The work was laying out the problems, eliciting what each of the three wanted, and clarified the minimal   Expectations of living at home. Grady was alternately argumentative and accepting, affectively almost disconnected from the situation’s severity.

Later that day, I took out the list of relatedness traits, completed only a month or so earlier, to use for the first time at work. He had nine, possibly ten. I was nevertheless willing to do the work. He didn’t leave, though, and was respectful even in his complaints and disregards. He never displayed that reptilian look of someone gauging prey.

Before the fifth session he left home at his own initiative. 

I met with the parents two more times, working with grief reactions and planning, and referring them to was a local Tough Love group that had a very good reputation. I was  willing to do the work, offered to see anyone, at any time, for anything. I wouldn’t have been surprised if he returned, wouldn’t if he didn’t, and he didn’t. So be it.

Beyond-eight traits are out there, uncommon if not rare, but they are present. And they’re tough to treat, some probably thoroughly untreatable given contemporary clinical mental health technology, but one does not know. Over the years, I saw two boys who wouldn’t surprise me if either or both made the A or B Section front pages of the local Times. That wasn’t Grady. Whether his tangle of traits is in reality a Gordian Knot or not remains to be seen. The pain of the parents suggested substance within the family, and that’s where healing begins.

To continue with the commentary, when the relatively high total of empathy-remorse problems became evident, the non-relatedness portion of the study group, N = 39, was also scored for comparison’s sake. The result was 9/39  (22%) for the non-relatedness group,  compared to 13/17  (76%). The overall frequency of 22/56 (40%), though, would demand a specific response.

The findings also note that those thirteen unresolved empathy and remorse problems of the  entire study group were all assigned to the thirteen unresolved relatedness cases.

 To the degree this finding can be generalized, empathy and remorse difficulty is a key marker for potentially more difficult clinical problems. As an aside, the benefits of doing some kind of on-going data collection than can aid and abet intervention developments are evident. But even having a specific attention in mind does not necessarily mean treatment success within the realm of pre-Axis II issues. Client defense mechanisms have enormous impact. More on that in the Relatedness Summary, planned for Post 29, including one list of those defense mechanisms.

Last comment:  52/56 of the study group were males. All relatedness cases were males. Take a second look at the top five scoring traits above. Not much in the way of occurrence or pattern differentiations between the genders is made in the Personality Disorder Sections of the DSM’s III – V, but those top five certainly seem more male than female. An interesting comparison would be for some similar work done with 50 or 60 females. Personality is part genetic and part experience (50 – 50 according to one study, get into that later) and comparing this kind of data results between similar groups of females and males would be interesting in that light.

Note: Refer to Post 7: Gender Distribution for explanations of this disparity between males and females in my practice if not read already.

Discussions about clinical approaches to these cases will occur in the summary post of the relatedness sub-group section. 




Origin of ‘Relatedness’. 

For anyone not familiar with the term, please note that “Axis II” is a holdover from earlier DSMs. The 5-Axis categorization system of the DSM III and IV was deleted by the DSM V. The term had become professional shorthand for personality disorder. 

Interactional traits that have the appearance and effect of adult personality disorders begin to appear as early as mid-childhood, but predominantly in adolescence. As a group, these youth are among the most difficult with whom to do effective therapy. The traits can be debilitating to the patient and disruptive to the environments in which these young people live and function.  Little research or other work has been done to systematically identify them, e.g. the depth and breadth of the problem, how clinical work is impacted, and just how to go about doing the work. This section introduces data concerning pre-Axis II clients, including identification, frequency, severity, and outcomes.

The DSM’s III – V have had the same formal guidelines for diagnosing a child or adolescent personality disorder. If the cluster of traits in question has a childhood version of an adult personality disorder diagnosis, the clinician uses that version. For example, would be a conduct disorder diagnosis for a youth who, as an adult, would be diagnosed with an anti-social personality disorder. For those disorders that do not have a childhood antecedent, the diagnosis can be made for both adults and children. Currently, using the same diagnosis for both populations  would apply to seven of the ten recognized personality disorders. An example would be a schizotypal personality disorder.

For practical purposes in fieldwork, though, a different set of guidelines and conventions has been used. The personality disorder diagnosis itself was rare. For most of us, the diagnosis was never used. Just the term ‘personality disorder’ in the context of outpatient child and adolescent therapeutic work was rarely heard, even in settings like staffings, consults, community trainings, or continuing education workshops. Other explanatory diagnoses and concepts can be used in its stead. This included elite treatment settings. While working on the inpatient medical units at Seattle Children’s Hospital from ’78 – ’84, I frequently spotted for the inpatient psychiatry unit’s social worker in her absence. In those highly comprehensive, multi-disciplinary  staffings, I cannot recall ‘personality disorder’ being used, even informally. 

Clinicians are cautious about this label for a couple of reasons. One, a personality diagnosis could well be premature. A clause in the basic DSM definition of personality disorders states that the traits have to be enduring and appear in multiple settings. Since a youth’s development is incomplete, young clients could theoretically “grow out of it”, and often do. 

Two, a caution that borders on prohibition concerns the lack of a fully developed sense of identity. A young person would likely have difficulty incorporating the negative connotation of something so fundamental as their personality and continue to see themselves as they had.  Rather than understanding the problem as some aspect of themselves needing help, the child or adolescent could see themselves as defective, damaged, aberrant, etc. That self-perception can become part of their identity set. The colloquial dictum has been that a clinician can’t give a child or adolescent a personality disorder diagnosis.

A concern about the term ‘personality disorder’ being shared with adults raises the same concern about doing so with an adolescent or child, even though the clinician had been thinking so. The question then arises, why use a mental health term that is largely not shared with the client when other terms more viable are available to the clinician for sharing their formulations with the client? Is the use of ‘personality’ as the global reference point for this class of disorder really necessary? 

Hence the term ‘relatedness’. However, “relatedness” is not being offered as an alternative, but only as an example of the possibilities. Do keep in mind that the thoughts in the last paragraph are personal opinions.

Lack of remorse and lack of empathy were initially listed separately. The first half dozen lack-of-remorse cases were also scored as having lack of empathy. The two were listed in different personality disorders. Stopping to reassess, the qualities of each seemed sufficiently similar. Continuing to see them as separate seemed to be a scoring misrepresentation that made these particular clients a step more troubled or dysfunctional than was really the case, in my opinion. So, the two traits were combined into one. 

“Listen to the nurses. They know what’s going on’

Thanks to Cathy Baker, RN

Hospital Mentor and Guide



The DA/PA posts conclude the global study data, posts 6 – 23. Comprehensive review will be in Study Summary Section, following Sub-Groups.


Sub-Group Section  – 

Relatedness, or pre-Axis II symptoms

Suicidal Ideation

Parent Configuration #4 – Single mothers

Sexual Identity



Unresolved depression, suicidal ideation

Clients with CGAS gains under 5

Clients with CGAS gains over 24

Clients Length of Treatment over over 2 years

Clients Length of Treatment under 7 weeks

Termination Type 3 – 5 – unilateral and administrative terminations

Format 3 – Conjoint Family followed by Individual Therapy

Clinical errors

Axis II summary

Study Summary Section

Review of Findings


Therapy Process Section

Setting Up A Practice

Initial Contact

Assessment: parent interview, child evaluation, summary session

Initial family sessions

Middle Work


Context Section

Publishing Estimates

Sub-Group Section – 15 – 20 posts, planned completed in Jan/Feb

Study Summary – 3 – 4 posts, planned completed in March

Therapy Process – 30 – 40 posts, planned completed late summer 2020

Context Section – Planned completed – Spring, 2021

I believe that changes made in this compassionate treatment, as described in earlier posts, are more effectively sustained because of the family’s intimate awareness of how the growth occurred, and how to keep everyone involved and on track. This contention would be an interesting part of research that examines what works and how among therapy types.

For the reader interested in doing child, adolescent, and family therapy, one thought to entertain as this work proceeds is where this discipline will fit in the future. Prescriptive therapies and other programs designed for children and adolescents will proliferate as the patient population grows larger. Competition can get noisy, advertising seductive. Compassionate-type therapies will continue to be quiet, patient, few quick fixes, no bells, no whistles. Private practice will remain viable and less impacted by efficiency strictures within mental health industry policies. The solo office may become the only viable setting in which long-term therapy can occur.

A market for family therapy will always be there, but may be limited. Income may be a challenge. Creativity, determination, and health help. If you’re still tempted to go in this noble direction, take heart and keep the faith. To paraphrase a Joni Mitchel lyric, ‘sacrifice and satisfaction, you know they’re the same release’ (from People’s Parties).

“Listen to the nurses. They know what’s going on.”

Thanks to Kittee Berfelz, RN,

Guide and Mentor at Seattle Children’s Hospital


In addition to being an alternative measurement of effectiveness, diagnostic and problem area data can be used to assess specific areas of both process strength and need for particular attention within specific clinical groups. The following data charts compare the outcomes of three age groupings: 5 – 11; 12 – 13; and 14 – 18.  While this division created an arithmetic balance, the three could also be seen as representing three distinct stages of growth: childhood; transitional; and adolescence. 

Six Diagnostic and Problem Areas are used, including: anxiety, depression, behavior;  and family, school, and social. For the purposes of this particular analysis, relatedness and the problem area of community are not included. Relatedness does not constitute an Axis I diagnostic category for children and adolescents. As indicated in an earlier post, the community problem area was insufficiently designed, and only involves relatedness cases.

Generalizability of these results to other practices is limited due to the validity issues already described. The section is meant as a demonstration of how the data system can be used by the individual practitioner, and how the results could have been translated into clinical work.


The relatively low number of children with depressive symptoms was surprising. The total of DA/PA issues seen in Group 1 were a third lower than the other two groups, but the depression score of 3 was well below any other line. The assessment for depression symptoms was uniform and applied to all new children and adolescents, and if questions arose about the child’s capacity to answer, the parent(s) would be consulted during the assessment. The scoring seems valid. 

All three of the clients included in the depression category, whose ages were 7, 8, and 10, reported suicidal thoughts. The eight year old had random, repeated thoughts over a period of months. The others were more blurts during emotional interchanges.The children’s schools had not experienced suicidal thoughts or behaviors with these children. A couple of teachers did report other depressive concerns like irritability and teary reactions to normal happenstance. These symptoms, including the irritability and excessive sensitivity resolved by the end of treatment. Many dissipated shortly after the assessment and therapy processes began.

Given that the numbers of DA/PA areas of concerns for the youngest group were roughly two/thirds of the older two groups, even this low number of depression issues may have been within the range of norm. Another possibility is that younger children may be more prone to manifest depression in anxious and/or behavioral terms. 

As a very general observation, the general receptivity of youth to family therapy increases as the child is younger. 

The 46% resolution rate of anxiety problems within the 14 – 18 year-old group was likely due to two prevailing issues. A few of the group may have had sexual identity worries which, if so, can be profound. Another few had processes that terminated due to administrative issues. 

Sexual identity was seen as a possible explanation of anxiety problems that either arose or worsened as adolescence progressed. Most within the study group were in this age bracket. They just a portion of  the only clients with anxiety. Probability ranged from degrees of possibility to probable, so they were an inferred and unconfirmed group.

These particular clients tended to do well in other areas of concern. In the exquisitely difficult work with a portion of these boys who seemed more probable and were suffering, this one possible underlying cause of the anxiety problem was not a focus of exploration due to presumed reluctance and potential negative consequences. In other ways, though, the therapy was commonly experienced as a validation of their abilities, self-perceptions, and self-worth, resulting at least in a diminishment of worry. Much of the anxiety, though, remained at termination. This particular group will also be covered in the following Sub-Group Section.

In the early 20-teen years, the Evidence Based Treatment movement was just beginning to gain traction in the mental health business, and was not utilized in my practice. If the initial clinical approach failed to generate improvement, an appropriate EBT would be sought and used now.

The low resolution rates for both anxiety and school were impacted by a high number of administrative terminations in the 14 – 18 year-old group. Not much can be done about processes driven out by circumstance except to learn what one can. This would be particularly true if some clinical issue became a prompt of one sort or another. Axis II parent or parent-figures do form a portion of the administratively terminated client and family. The problem is that the client is commonly still anxious. The Termination Type 5 group will also be reviewed in the Sub-Group Section. 

The high resolution rate for depression is also notable. A common aphorism used in the past is ‘activity alleviates anxiety’. By the same token, one can conceive that  ‘connectedness alleviates depression. Although possible, connectedness may not necessarily resolve depression but can lessen depression’s drag on one’s being. Simply experience of effective family therapy enhances connectedness. 

The DA/PA data conveys the point that a multiplicity of issues accompanies most any new case. The chart above indicates that 80% of clients come in with more than kind of emotional or behavioral problem. Similarly, 80% come in with more than one problems area of concern. Therapies have a field of responsibility that tends to everything within its power for that given case, in that given situation, in some way or another.

 Focused therapies can carve out specific symptoms and problems areas to treat, but a broader approach is quite likely necessary for tending to the range of issues that are outpatient material in nature. Tending can be simply watching, or making a referral, or added to the treatment problem list. Cover as much as possible, feasible, and ethical. ‘Some way or another’ is the art of what we do.



The Diagnostic and Problem Area (DA/PA) data can be used to evaluate the particular impacts that a treatment process has on specific clinical populations. Several of the analyses in the upcoming Sub-Group Section will use this tool. DA/PA results can also be aggregated to evaluate overall changes in the study group. DA/PA results closely resemble those of the CGAS, and may be a feasible alternate to the CGAS for measuring clinical change. The advantage would be that a DA/PA system specifically outlines what symptomatic areas have been resolved and those that still remain. DA/PA also provides a more concrete compliment to periodic process reviews and termination discussions with the client(s). Under those circumstance of reviews and termination meetings, the system may well be a more effective reminder and motivator.

Aggregating the Data

Any calculation involving data based on combined diagnostic area and problem area data (DA/PA) is presuming equal impacts between the two areas upon the clients, and equal weights between the four variables of each group. Supports for the validity of these presumptions are, at best, theoretical. Suggestive and interesting substantiations do occur, however, within the forthcoming results. 

One inferential theory is that the development of personality is equally impacted by genetics and environment, loosely represented here by the Diagnostic Area and Problem Area tables, respectively. This notion was a finding in a University of Minnesota study of identical twins raised in separate environments. The research was reported on an NPR Weekend Edition some 20 years ago. No finding states that the four different clusters of diagnostic symptoms have the same relative impacts on the human being. Neither does the notion that the four different problem areas have the same relative impacts have any empirical support, to the best of my knowledge. We would all know about these things if they were tried and true. 

Still, the results are interesting. Approaching them as having the potential to be indicative of something substantive is but one more example of abductive reasoning. See what you think. Does a validity exist when the use of the measurements finds useful congruences?

DA/PA Pre and Post Distributions

The X axis represents the Combined DA/PA values of the study group consisting of nine integer values. The range is from eight to zero, or worst to best in terms of functioning.  As in the pre and post CGAS histograms, the pre values form a normal curve while the post scores are in the form of a classic poisson curve, where the mode is on the right wall of the graph with a long left tail.

Also similar to the CGAS results, the combined DA/PA histogram for clinical gain is a normal curve. The X-axis takes the number of DA/PA resolutions from zero to six, the maximum number of resolutions occurring within the study group. 

Another Comparative View

Two notable factors in this chart include: the statistical similarity comparison of the Pre and Post scores  with other findings; and a possible issue with validity.

As in every previous statistical depiction of before-and-after clinical status, the data of the Pre scoring forms a normal curve while the Post data becomes a poisson curve. The termination data here comes closest to the pure version with the mode at the right wall of the graph and then a long left tail. The question becomes the relationship, if any, between all these similar findings.

In particular, the DA/PA Termination data suggests two relatively separate groups in their numbers per line. numbers.  40 cases have 0 – 2 remaining DA/PA issues, while16 cases have 3 – 5 remaining. Similarly, the clinical gains of the 16 unilaterally terminating, Termination Type 3 – 5 cases. They had about half half of the average gain of the 41 mutually terminating cases, Term- Type 1 and 2 (ref. Post 14). 15 cases ended with CGAS scores of 60 or less, while 41 terminated with sores of 61 or more (Post 12). The same roughly 70% – 30% type of split occurs in the informal taxonomy of prognosis at termination as well (Post 20 – Results Section).

These graphic “tail” groups are not all the same clients. The specific common factor of the 70 – 30 split  may suggest that these measurements produce defining clinical objectives in and of their own right. For example,  0 – 2 DA/PA at termination may be a reasonable clinical goal in that dimension. Mutually decided terminations may be a reasonable desired outcome. A termination CGAS of 61 could be seen as a minimally satisfactory gain, particularly for those clients beginning in the serious and severe ranges of disturbance. Aa positive functional prognosis at termination may be yet another way to evaluate clinical casework.

The finding in this data that legitimately raises questions about validity involve the 17 cases that were scored as having 6 and 7 DA/PA at beginning of treatment, and all moved to 5 or fewer at termination. Translated, this means that none of those cases, virtually all of which were in the severe and serious areas at the beginning of treatment, terminated with little or no change. A review of the ratings for this group verified the scoring. If one were to accept the equivalence of 2 DA/PA to 10 CGAS points, all those cases minimally made a movement into the next higher CGAS level.

Most any clinician who has experienced the difficulties of working with severely disturbed kids and their families would be reasonably dubious about the finding. A group of 4 or 5, maybe. 17 such cases and no quick, precipitous drop-out? That would seem at best Improbable. 

Regardless of theoretical belief and therapeutic approach, a generalist child and adolescent clinician knows the experience of a clinical flame-out involving the seriously or severely troubled client, and knows that unfortunate event gets sprinkled throughout a career. If a full-time practitioner sees 30 – 40 child and adolescent cases averaging 20 – 30 sessions per case per year (and that’s a full load given the attentions and complications the child and adolescent cases need and present) one is likely to see upwards of three to four cases in which the client is in the severely disturbed, 31 – 40 CGAS range range, and a higher number of seriously impaired. This is unless these situations are successfully screened out during the inquiry contact, which a few clinicians will do just to avoid the inevitable complications.

During my last year of practice, not included in this study group, two severe cases were seen in which any kind of substantive progress did not occur. One was an older adolescent with pervasive conduct problems and associated relatedness traits. Out of need for security, the parents were considering eviction. By the fourth session, the boy had left home on his own. The functional and competent parents came for another couple of sessions to review options. They were referred to a regarded local Tough Love group. The second client was similar in behaviors and traits, altho several years younger. After a year-long process that saw problems get better and the worsen as he hit adolescence, the boy was becoming a community threat. Then came an episode of moderate vandalism in the waiting room, the only such occurrence during the entire practice. I was inadvisedly meeting with the mother alone for the last half of the last session before my vacation. I returned to a post-vacation cancellation message terminating the process, presumably in part due to the vandalism. but also due to lack of financial support for the boy’s  care.

The year before the two year study group also had two such cases. One was an older adolescent of an immigrant East Asian family. Over the phone, the description seemed one of depression with withdrawal. The four session assessment found him to have seriously impaired communication, relational, and probably insight skills to the point that a schizotypal diagnosis needed to be ruled out. was possible. The family was referred for a psychiatric evaluation. The second was a junior high schooler who was refusing to go to school, tantrums, infantalized behaviors at home, and social withdrawal. Within four sessions of an on-going family process, he had returned to school. According to both the teacher and school counselor, he was acting within normal bounds at school. What I needed to do was see the parents separately at that point to begin sorting out these differing presentations, but I persisted with the conjoint process that was in part focused on communication balance within the family. Shortly thereafter, the parents terminated, opting for a neurologist.

These four drop-outs lacking even partial resolutions are normal. They become part of the work. We always work to get better, and casework shortfalls do offer lessons to be learned. The experience becomes heuristic. We think our way through and make adjustments as indicated. When the result is in some part due to clinical decisions and operations, we also have to feel our way to a sense of resolution, knowing we’ll see something of the sorts again, and again… that’s part of the learning as well.

To have zero of these kinds of cases over two years within this study group I view as a something of a statistical anomaly. 

Statistical Relationship Between CGAS and DA/PA Change?

Presuming a correlation between CGAS and DA/PA scores, the average overall CGAS gain of 13.4 would equal the 2.7 average DA/PA reduction. A two-point reduction in DA/PA would be basically equivalent to a 10 point CGAS gain. For example, the resolution of a depressive condition with 3 or 4 symptoms, combined with defiance at home or non-compliance with schoolwork could be considered tantamount to a 10 point gain. Validity and reliability of a DA/PA scoring system and of CGAS valuations would have to be established for that suggestion to be statistically evaluated.

These results raise a second question, that being the pattern of decreasing gains as the initial CGAS rises with the exception of the 31 – 40 range being lower than the 41 – 50. This could be a function of the small N of five. The result could also indicate the greater difficulty of engaging the severely disturbed youth and generating a process of positive gain. Compare the behavioral descriptors of the two categories again:

41 – 50  Serious – Moderate  degree of interference in most social areas or severe impairment of functioning in one area, such as might result from, for example,suicidal preoccupations and ruminations, school refusal and other forms of anxiety, obsessive rituals, major conversion symptoms, frequent anxiety attacks, poor to inappropriate social skills, frequent episodes of aggressive or other anti-social behavior with some preservation of meaningful social relationships

31 – 40  Severe – Major impairment of functioning in several areas and unable to function in one of these areas i.e. disturbed at home, in school, with peers, or in society at large, e.g. persistent aggression without clear instigation, markedly withdrawn or isolated behavior due to either mood or thought disturbance; suicidal attempts with clear lethal intent; such children are likely to require special schooling and/or hospitalization or withdrawal from school (but this is not sufficient criterion for inclusion in this criterion). 

When a case in the severe range begins with a capacity for change, the results can be impressive. Of the five cases in this study, one gained 35 points, albeit over several years and a high number of sessions. The other four had more or less intransigent relatedness issues, including little in the way of substantive relations that could provide help and support of positive change. Three of those situations did make gains, moving from the severe to the serious range. The improvements were mostly in family relationships and behavior, but on the whole the changes were not to a point where a substantial change in overall functionality occurred.  The cases plateaued and discontinued.

On the other hand, those cases that began in the 40’s, or serious range, did have meaningful relationships that persevered through their other attitudes and behaviors. This factor led them to be more receptive to the positive reinforcement that enhanced motivations for further change. Most of the cases that made gains of 20 points and greater came from this group.

The pattern of decreased gain from the 41 – 50 decile down through the 71 – 80 decile for both CGAS results and the equivalent DA/PA could be seen another version of the maxim, ‘the longer in treatment, the better one gets’.





The last major data source to be developed in this project was the specific areas of difficulty that the child or adolescent client was experiencing pre and post. The data was divided into two tables, one identifying diagnostic areas, and the other life problem areas.The four diagnostic areas are anxiety, depression, behavior, and relatedness. The four problem areas are family relationships, school performance and behavior, social relationships, and community involvement.

Existing anxiety and depression problems were invariably disclosed over the course of the first two intake sessions, the first being with the parents and the second with the child. The evaluations were supplemented during the child interview by using the eighteen-point symptom list under Generalized Anxiety Disorder in the DSM III R, and the ten-point symptom list under Major Depression in the DSM IV. Evaluating depression was part and parcel of every initial assessment, whether depressive symptoms had been disclosed or not. Use of this depression inventory was done in some part as a way to ease into the exploration of possible suicidal ideation, the question of which is the last of the ten. 

When present, behavior problems were always indicated by the parents, and occasionally confirmed by the child. The depth and extent of the behavior issues were subsequently explored during the assessments and into the therapy process itself.

“Relatedness” is my term for youth demonstrating pre-Axis 2 patterns of interpersonal relational and behavioral dysfunction, seventeen clients in this study. Aspects of this diagnostic category were often inferred by the parents at the beginning, but really became evident over a period of observation, information gathering, and clinical interactions.

Development of the four problem areas essentially used the concept of concentric life circles, the center being the  internal state of the person as indicated by the four diagnostic categories above. The first surrounding circle represents family relationships. The second is designated for work or school; in this study, school. The next concerns social relationships and activity. Last is community activity, defined as interactions and behavior that range from the socially contributing to anti-social. Part-time jobs, for example, would be considered within this category, as would, say, shoplifting. 

Specific diagnoses were generally not used in the identification of diagnostic or problem areas. The assessment did include: inquiries about the problem history; family history of similar diagnostic problems and other mental health issues; previous treatments and treatment outcomes; family efforts to deal with the problem(s); impacts of the problems on family members including the marriage; and any other concurrent treatments.

Determining the severity of any client diagnostic or problem area for the purposes of this study was not part of the data collection. The small overall N made further levels of analysis unfeasible, and lack of documentation in this specific regard made any attempt to assign severity scores inherently unreliable. The data is essentially a binary ‘yea’ or ‘nay’, for any and all of the eight categories.

The “community” problem area was narrowly applied, focusing more on the anti-social problems that may have arisen, such as theft, aggression, and vandalism. Issues with adults other than family, such as coaches, extra-curricular teachers and tutors, extracurricular activity adult leaders and supervisors were included within the “social” circle. Patterns of problems with part-time employers were included in the community category, although this rarely occurred and did not occur within this cohort.



The number of anxiety problems at intake would have been significantly lower thirty years ago.

Behavior would have been the most frequent problem area years ago, but also remember that the clientele is overwhelmingly boys

Depression rates are higher, witness the well-publicized increases in suicidal ideation, attempts, and commissions, but not to the extent of anxiety increases. One could reasonably speculate that a portion of the increase in depression is a function of increases in chronic worrying by children and adolescents. 

Relatedness issues afflicted about 30% of the study population at intake, which is similar to data I’ve seen concerning the percentage of adult Axis II disorders receiving outpatient mental health care. 

Nothing about resolution rates in the treatment of youth “relatedness” or pre-Axis II problems could be found, but one in five may well be within a broad range of norm. 

As a matter of note, client diagnoses (as distinguished from diagnostic areas) become the formal clinical baseline of a case. As used here, diagnostic and problem areas become reference points for clinical progress. The general focus in therapy sessions is the client’s presenting problem of the day and symptoms being experienced in these diagnostic and problem areas. The work is to alleviate or eliminate symptoms, enhance relationships (particularly family), and improve problem solving skills. The baseline diagnoses themselves did not arise as a topic very often in therapy processes. When they did, the reason usually involved administrative insurance matters. The focus of both the overall process and the session work itself were client determined as much as possible and feasible.

Thanks to Russ Roepcke, Psychologist

Director, Karma Clinic, Counterpoint Clinic

1970 – 1985





Four patterns stood out for the study, including: the single mother group; the 16 – 18 age group; an amorphous cluster of cases out of the various small statistical groupings having noticeably less progress than the rest, thus raising the question of commonality among them; and the 12 – 13 year old group. 

Seven of the fifteen pivot charts had anomalous results involving the single mothers group. Single mother life experiences seem clearly different from the others. These irregular results included: lower average gain; a lower average number of sessions;  curiously, a higher average length of treatment; the clients being all early to mid-adolescent boys, arguably the most difficult parental management period with boys who are in some way or another struggling; the highest rate of conjoint family work; and a higher rate of administrative terminations  (financial problems, insurance coverage, need to move, job change, or shift change at work). In discussing this group, finding some dominant factor is difficult.

The findings certainly underscore the daily impediments impacting single mothers, simply as matters of life. These include such things as frequent problems with finances, time, sleep, dysphoria, isolation, lack of paternal financial support, lack of paternal parenting support, and paternal opposition to treatment that occasionally includes outright sabotage of a process, mostly via persistent lobbying.

A dearth of professional education concerning the single parent population added to the problem of recognizing inhibited clinical progress. Discovering this lower improvement rate for this group was initially surprising, but eventually made sense. This kind of embedded clinical problem or shortcoming is less likely to occur if practitioners maintain a simple data base and analytic structure or system.

To address the issue in the office, the first adjustment probably would have been to develop a list of possible impediments to regular treatment processes that single parents encounter. The list could be used like a depression or anxiety inventory during the assessment. Perhaps the initial assessment process itself could be extended to an additional hour with the single parent to more thoroughly evaluate.

The work could be toward corrective measures, accommodations, and protecting the process potential. The list could be then be used occasionally to augment individual check-ins with the mother and her view of the process. Also, developing a list of additional community resources available to the single parent for additional support and services could prove useful.

In reviewing the 16 – 18 year age group’s tendency to have shorter processes, the conclusion was that only half of them really belonged in the study. Those that did belong had processes initiated by the parent(s). The others sought counseling for themselves via their parents’ insurance. In the latter group, the parent(s) were seen for the first session as per usual. The work thereafter was individual. The client’s issues tended to be circumscribed, the work relatively short-term, and the outcomes generally meeting the goals. The parents came in or talked over the phone a couple of times during these processes, more as a check-in during their boy’s counseling, with his knowledge both before and after. These were more individual adult cases than child and family work.

The third group to be considered here is not as specific. Rather, in a variety of forms they represent that tail of the poisson curve (ref. Post 12). These were cases that simply didn’t fair as well, each appearing in different places among the pivot charts. When considered as a whole, they constituted a group of size, significance, and added concern. 

The intents of the pivot chart work were twofold: to identify particular clinical groups that warranted further attention, which has been done; and to understand further what factor(s) are impacting the less improved. Lacking any cogent results in this exercise that began to explain lower outcomes, other options at this point in the study had to be considered.

One notion was to create a new organization of the study group, apply data in regards to adult Axis II problems and youth “relatedness” problems to the new taxonomy, and review the results. 

The adult Axis II data had been the last basic piece to be collected, now three years ago. While not an expert in adult personality disorders, forty years of experience in the field of child and family mental health does create a certain clinical sense of Axis II involvement and their often disruptive, skewing impacts on a given case.  

The relatedness data was organized more systematically and most likely holds a higher degree of validity. Both the data gathering and subsequent analyses of the adult Axis II issues and client relatedness problems will be covered at greater length in the Sub-Group Study Section.

The 56 cases were divided into three levels based on my view of the youth’s functional prognosis at termination. The first ( Group C) were those who at termination continued to have difficulties with personal change, family dysfunctions, and impeding environmental circumstances that all together seemed problematic. 5 were so identified. The second (Group A)  were those whose prospects seemed good. Barring traumas, major losses, or other misfortunes, the clients would likely be operating from the mid-70’s and upwards on the CGAS. They totaled 41. The remainder (Group B) formed the middle of the three tiers, those whose outlooks were questionable. These clients had personal, familial, and environmental strengths to be encouraging, and sufficient unresolved or persistent negatives to create concern.  


Prognosis outlook was strongly influenced by the relatedness factors of the client youth. All clients in Group C  had relatedness difficulties, and they were among the highest in the numbers of traits that each exhibited. Group B was also impacted by relatedness problems, but at a rate that was 50% less than C. Those youth were generally in the more moderate range of “relatedness” concerns, i.e. fewer traits. Group A had two clients with these relatedness difficulties out of the 41, both being among those with the fewest traits. 

The 2 – 1 ratio between C and B for the percentage of relatedness issues among the groups was similar to the data concerning the adult Axis II issues within the broad family nexus (parents, step-parents, partners,  parents with and without visitation rights, adoptive parents, guardian parent(s) etc). Group A, on the other hand, had a percentage of Axis II-impacted cases that was close to that of Group B. Group A’s adult Axis II cases were disproportionately high compared to the low number of client relatedness cases.

The surprise in this very informal result of data dredging was the number of cases that had an adult Axis II issue within the overall group itself, close to half the total cases. Even more surprising was that within the family constellations of the good prognosis group, almost a third showed some evidence(s) of Axis II involvement, again within that broad range of characters. That could lead to questions about the Axis II scoring itself, but similar data with which to compare these numbers doesn’t appear to be available. These numbers theoretically could be within in the range of norm for an outpatient child and adolescent clinical practice.

The primary intrinsic and determinate factors of progress seem to be the client’s pathology and relational difficulties. Family and other environmental issues certainly contribute, but to lesser degrees. Given the overall softness of the data, that generality is probably as much as can be said. However, the result that identifies adult Axis II and youth relatedness issues as having the major impact inhibiting clinical gain does intuitively make clinical sense.

One inference is that the work in child and adolescent mental health is primarily about the client, and the client was, indeed, the usual focus of work in this process. Occasionally, though, the management problems posed by an Axis II – involved adult leads to an implicit change in focus, at least temporarily. In those instances, session participation may shift from conjoint to a split-session in which the adult issues can be addressed more directly. That flexibility is a key element of the “family-base” type of treatment approach as opposed to a purely conjoint one. In any and all events, the family’s clear participation in therapy is instrumental to fostering growth, guiding the changes, and later maintenance of change.

In the aftermath of termination, the most difficult cases like those in Group C could generate frustration, be ultimately often sad, but certainly educative. Watching and noting those patterns unfold and operate during the objectivity of therapeutic work is important self-education.

Lastly, the 12 – 13 year old boys group did stand out positively in a couple of the findings, hardly enough to make sweeping generalizations but intriguing. When engaged in a therapy process and having a vested interest in the outcome, these clients usually approached family work with a certain enthusiasm that pleasantly surprised the parents. Still children in ways but developing their more adult-like formal reasoning skills, they can still be readily influenced in the directions that their emerging independence takes. When working with this age group, family therapy may indeed be the treatment of choice.

Thanks to Carlah Lytle, MSW, R.I.P.

Practicum and Social Work Supervisor

POST 18 – Pivot Charts


Posts 18 and 19 use pivot charts to look at the interactive relationships between variables, including age distribution, number of sessions, length of treatment, parent configuration, format, and termination type. The search is for interesting patterns that emerge as the data gets organized by these charts. 

The first question is whether the noticed pattern is a meaningful departure from the norm, or simply an anomaly.  For example, take the average CGAS values of the five different parent configurations in Post 14. That result separated both the single mother group and adoptive families from the other three configurations. The two had similarly low averages. The single mother’s group routinely encounter  difficult life circumstances that could account for their disparity in clinical results. That difference was appearing in other study findings as well. Upon further investigation, though, the adoptive family group seemed to be a random collection of particularly unusual cases for that particular clinical population which in turn presumably led to lower outcome averages. Both groups will be discussed further in the Sub-Group Study Section, but one is a departure from the norm as established by other results, and the other is an anomaly.

In reviewing these charts and discovering patterns that separate particular groups from others, consider a tier of possibilities. Does the observed pattern seems potentially meaningful? Does a broader pattern of results involving the identified group in question exist within the data set? Can similar findings in other studies be located? Is the overall data set sufficiently large to run statistical significance tests? The answer here is ‘no’. However, do the findings nevertheless warrant some kind of clinical consideration toward process adjustment(s) to see if the results of therapy can be positively effected? Perhaps with enough data over time or in combination with similar data from other sources, the sample may become of sufficient size to begin formal statistical evaluations.

This investigative research process is called “data dredging”. The term refers “to the use of data mining methods to sample parts of a larger population data set that are too small for reliable statistical inferences to be made about the validity of any patterns discovered.” (Wikipedia entry on Data Mining).

The quote continues: “These methods can, however, be used in creating new hypotheses to test against the larger data populations”. A definitive caution in the Wikipedia entry on “Data Dredging” itself states that “dredging would be a misuse of data analysis”.

As an aside, the thinking behind dredging also represent examples of abductive reasoning. If the underlying pattern is of sufficient concern but cannot be statistically substantiated, experimenting with minor clinical process changes, additions, or deletions within the bounds of ethics codes can still be warranted. Examine the results, re-hypothesize and reformulate as indicated, examine the results, re-hypothesize, etc., until consistent, positive results are occurring. 

As the lowest form of statistical truth searching, lower than data mining which at least has a theoretical target to seek, the name ‘dredging’ does give off a connotation of being impure. But, data mining and dredging can still be a part of the search for truth. 

These two posts can be seen as a small educational opportunity. If the reader is so inclined, spend a few minutes perusing through the presented charts. They begin with a series of five with the primary variable being Age Distribution, then four by the Number of Sessions variable that begins Post 19, then three via the Length of Treatment, and so on through the post. Pick out what appear to be patterns that do not seem to fit with the overall distribution of the particular chart.  I am going to be presenting and discussing my own observations at the end of each primary variable series. 

One socio-political factor to consider has to do with how mental health services will be allocated in the future. The main point of dredging herein is to enhance treatment through understanding case characteristics and management, process effectiveness, and developing strategies and tactics pinpointing certain clinical groups. In the broader picture, though, rising national mental health needs combined with increasing proportions of the population unable to afford mental health treatment may lead to broad managed care approaches that limit the numbers of services available per person per problem per period of time (commonly a year). Some of the data herein does tangentially illuminate problems inherent in capping mental health care.

Age X # Sessions

Age X Length of Treatment

Age X Parent Configuration

Age X Format

Age by Termination Type


Age  X  #Sessions  –  As indicated before, the length of treatment in terms of both session usage and time in treatment are self-determined, i.e. the primary decision rests with the client and parents. The clinician generally plays an advisory role. With those conditions in mind, the moderate upward slope of the age X   # of sessions from the oldest to the youngest suggests that session usage in inversely proportional to age. The youngest tended to have more sessions than the oldest, and the inferred line of significance appears to be straight or nearly straight. If this data were available on a national basis for general outpatient child and family therapy, a similar finding would not be surprising, but is only conjecture here. 

X  LoT  –  The significance line here would be close to level. The difference between this finding and the one above, both of which are measuring roughly equivalent data, would likely be the added time of the split process cases bring to the total time. Neither finding is particularly important to this study.

X  Parent Configuration  –  The disproportionate number of 12 – 16 year old boys in the single mother parent configuration could be seen as a poignant reminder of sole parenting’s inherent stresses, given the difficulties many families experience as individuals of this age group emotionally and dramatically burgeon toward adulthood.

X  Format – The concentration of 12 – 16 year olds in conjoint and split session processes is also related to the single mother group, where eight of ten were in conjoint and one other was in the split-session format.  Also, the 12 – 13 year age group in particular stands out here and in previously noted findings as one which may be particularly well-suited to the conjoint format, having started with lower average CGAS scores and a relatively high gain average that used a low average number of sessions.

Thanks to Alan Leider, MD,

1920 – 2010

Consultant, teacher, mentor