Volume 1, Number 1, July - September 1997
Self and Gender: Narcissistic Pathology and Personality Factors in Gender Dysphoric Patients. Preliminary Results of a Prospective Study.
Citation: Hartmann U, Becker H, Rueffer-Hesse C (1997) Self and Gender: Narcissistic Pathology and Personality Factors in Gender Dysphoric Patients. Preliminary Results of a Prospective Study. IJT 1,1, http://www.symposion.com/ijt/ijtc0103.htm
This paper presents preliminary results concerning the relationship of self and gender
in patients requesting or seriously considering sex change. Specific attention is paid on
pathological features in regulatory processes of the self-system as well as on personality
factors associated with different types of gender disorders. Based on the results of
retrospective analyses a prospective study was designed to identify subtypes of gender
dysphoric patients based on a scrupulous psychiatric and psychpathological evaluation. The
evaluation procedure consists of (i) clinical interviews, (ii) a structural interview
according to concepts of Kernberg, and (iii) a set of self-developed and standardized
questionnaires. The results indicate significant psychopathological aspects and
narcissistic dysregulation in a substantial proportion of patients. Different subtypes of
self-(dys)regulation seem to emerge which are discussed with special reference to
differential diagnosis and prognostic factors.
The two intriguing concepts of self and gender are of major importance for the field of gender dysphoria, but at the same time both are complex and controversial. In recent years, the clinical, etiological, and psychopathological diversity of gender dysphoric patients had to be increasingly recognized by professionals. Looking at the remarkably different gender and developmental backgrounds of individuals with gender problems, many - including DSM-IV - have done away with the term 'transsexualism' as a distinct diagnostic category. This nosological shift, however, should be accompanied by improvements in the precision of differential diagnoses and clinical subtyping allowing a better fine-tuning of clinical managment. Existing approaches have mainly been restricted to gender and sexual orientation variables whereas personality and psychopathological factors associated with gender disorders have rather been neglected.
|Based on the results of a retrospective analysis of all patients that
have consulted our gender dysphoria team at the psychiatric outpatient clinic of Hannover
Medical School during a one-year period, a prospective study was designed to identify
subtypes of gender dysphoric patients by means of a scrupulous psychiatric and
psychological evaluation (Becker & Hartmann 1994). This contribution will concentrate
on pathological features in the regulation of the self-system and on some associated
personality factors. A number of preliminary empirical results of the first 25 consecutive
patients of our prospective study will be presented with a special focus on the results of
the psychometric instruments we have employed. Since the number of 5 biological females is
too small for statistical comparisons, the data presented here only refer to biological
Our preliminary results indicate significant psychopathological aspects and
narcissistic dysregulation in most of our gender dysphoric patients. Among biological
males different subtypes of self-(dys-) regulation and corresponding MMPI-profiles seem to
emerge. Results of the narcissism inventory indicate that of the 4 main dimensions (the
threatened self, the traditional narcissistic self, the ideal self, the hypochondriac
self) scales covering aspects of the ,threatened self' show the most significant
deviations while a number of patients do not have a negative body-self. The implications
of these results should be considered when thinking about differential diagnoses and
Table 1: 'Components of evaluation procedure'
The main components of our evaluation procedure are summarized in table 1. All patients were interviewed, usually independently by different team members. After that, all patients went through a structured interview - based on concepts of Kernberg (1984) for severe personality disorders - addressing relevant aspects of self-pathology, narcissistic regulation and object-relations. In addition, all patients were asked to fill out a set of both self developed and standardized questionnaires including the MMPI in its short version, the 16PF, the Rosenzweig Picture-Frustration-Study, the Narcissism Inventory (Deneke & Hilgenstock 1989), the Androphilia-Gynephilia-Index and the Cross-Gender-Fetishism scale, both designed by Blanchard (1985, 1989).
Table 2: 'Sociodemographic data'
The sociodemographic data show that the mean age of 30 years (range 17 - 45) does not
differ in the androphilic and gynephilic groups, which will be compared in most of the
following analyses. In the same way educational level and marital status are equal in both
groups whereas the vocational situation of gynephilic patients is significantly worse.
The results of the other three standardized psychometric instruments will only be touched upon before concentrating on the 'narcissism inventory'.
Figure 1: The MMPI-Scales
A quick look at the clinical scales of the MMPI shows that overall most scores are above the normal T-value-range of 40 to 60 indicating pronounced psychopathological features for the whole group of our patients. On the other hand, only the MF-scale has values above 70 which of course is no surprise in a sample like this, which also applies to the PD-scale. Looking at the two subgroups, one can see that the scores of the gynephilic patients are clearly higher for the so-called 'neurotic trias' of Hypochondria, Depression and Hysteria, the most valid scales of the German version of the MMPI. This suggests that the gynephilic patients of our sample have more neurotic symptoms, especially of the somatization and psychosomatic type and it also shows that emotional problems are expressed in a body language.
Figure 2: The scales of the '16PF'
In the 16PF significant deviations from the normal range - which is marked by the two horizontal lines in figure 2 - can be found in the primary factors C, H, I, O and Q3. This suggests that our gender disordered patients have a significantly lower ego strength, are more emotionally disturbed and have problems in coping with disappointments. The markedly deviant scores in primary factor 'I' describe our patients as highly sensitive, with rich inner lives, but also as impatient, demanding, with high expectations and a tendency to avoid responsibilities. Primary factor 'H' indicates that the patients of our sample are low in their self-confidence and rather inihibited, cautious and socially introverted. Looking at the differences between our subgroups, one can see that they are of minor importance in this test. The only statistically significant differences or trends are in primary factors 'L' (F1,17 = 10.89; p < 0.01) and 'Q3' (F1,17 = 3.29; p = 0.08), suggesting that the androphilics have a more sceptical attitude towards other people, try to rely more on their own opinion and have a tendency to be disputatious and resentful. The gynephilics are more spontaneous and guided by momentary impulses and ideas without clearcut future conceptions.
Figure 3: The scales of the 'PFT'
The Rosenzweig Picture-Frustration-Test is a well-known semiprojective instrument designed to measure the ways of coping with frustration and aggression. The six main categories of the PFT are depicted on figure 3. Again, the two lines mark the normal range between stanines 4 and 6. Figure 3 shows that for both the whole sample and the gynephilic subgroup all scores are well within the normal range while the androphilics do show some deviations. They are relatively low on category 'Extrapunitivity' and high on 'Impunitivity' suggesting a strong inclination to evade conflicts and to delude themselves about obstacles or the frustrational character of a given situation. Looking at the three reaction types one can see that androphilics are low on obstacle-dominance and very high on need-persistence which confirms the impression that their need-persistence i.e. their emotional pressure to reach a specific goal is so predominant that the obstacles encountered tend to be denied. The low scores on category 'Extrapunitivity' inidicate that their assertiveness, their ability to get their way in a constructive manner is below average. The gynephilic group is significantly lower in need-persistence (F1,16 = 3.87; p = 0.06) and higher in obstacle-dominance (F1,16 = 4.02; p = 0.06). Thus, compared to the androphilics they are well aware of the obstacles in their way and even tend to be blocked by them without feeling the intense urge for a quick solution.
The results of the Narcissism Inventory
The 'Narcissism Inventory' (NI) is a questionnaire developed at the Hamburg University Medical School in the 1980ies. It was designed to assess a number of theoretically and clinically relevant aspects of the organization and regulation of the narcissistic personality system. It consists of 163 items belonging to 18 scales which cover a wide range of different modes of narcissistic autoregulation. These 18 scales are grouped into 4 main dimensions according to the results of a factor analysis.
Figure 4: The 4 dimensions of the 'Narcissism inventory'
These 4 dimenions are called the 'threatened self', the 'classic narcissistic self', the 'idealistic self' and the 'hypochondriac self'. On figure 4 the results of these 4 dimensions are depicted for our sample. In interpreting the t-values it must be taken into account that we do not yet possess norms for a normal, non-clinical sample but only for a clinical sample consisting of individuals with diagnoses ranging from psychosomatic disorders and neurotic depression to narcissistic personality. Naturally, this circumstance erects narrow limits to an interpretation referring to the normal population. For the inspection of the scores in this diagram it implies that a t-score of 50 is average compared to a patient sample and scores above 50 can be viewed in our preliminary analysis as a clinically substantial finding.
Figure 4 shows that overall the highest scores can be found in the dimensions 'the threatened self' and 'the idealistic self', the first indicating a marked instability of the self-system with fluent transitions between an arduously maintained and a progressive decompensation. An analysis of the single scales of the 'threatended self' shows high scores in 'derealization/depersonalisation', 'archaic retreat' and - expectedly - 'negative body image'. This dimension has significant correlations up to .7 to a number of MMPI-scales such as depression, psychopathic deviate, paranoia and psychasthenia and also to the 16PF-scales emotional disturbance and sensitivity. The value of the dimension 'the idealistic self' goes back to high scores in the scales 'object-devaluation' and 'symbiotic self-protection'.
Looking once more at differences between the subgroups the diagram shows that androphilic patients are higher in 'the threatened self' and especially 'the idealistic self'. Among the single scales statistically significant differences can be found in 'derealisation/depersonalisation' (F1,18 = 7.23; p < 0.05), 'archaic retreat' (F1,18 = 3.39; p = 0.08) and 'symbiotic self-protection' (F1,18 = 6.85; p < 0.05), all with higher scores for the androphilics. Gynephilic patients are higher (but not statistically significant) in the dimensions 'the hypochondriac self' and the 'classic narcissistic self' which is largely due to high scores in the scale 'narcissistic rage'.
Subgroups of patients according to self-regulatory mechanisms
Using the 4 dimensions of the Narcissism Inventory we have performed a cluster analysis of our cases to see how this statistical procedure groups our patients and to compare this solution to our clinical impression. We have used the Ward algorithm and after a careful analysis of the cluster agglomeration schedule have decided for the 4-cluster-solution. The main features of these clusters were then determined by univariate and multivariate statistical procedures. The main cluster characteristics relating to narcissism are summarized in table 4. The number of cases is small, especially in clusters 3 and 4, allowing only a tentative interpretation but on the other hand all clusters have a good correspondence to our clinical opinion.
Table 3: 'Main characteristics of clusters'
The 9 cases combined in cluster 1 have high scores only in 'the idealistic self' which means they have a marked narcissistic pathology primarily in their object-relations. Their prevailing self-regulation patterns indicate that they have a profound fear of being disappointed and hurt by others. To protect themselves against this they tend to emphasize their autonomy and their moral superiority. There is a strong ambivalence between a longing for another person and impulses to avoid and escape any close relationship. The cases of cluster 1 also have a strong tendency to identify themselves with specific highly valued personal ideas, a self-regulation mode serving the purpose to stabilize and protect the self. As you can see from the second slide, they do not show significant psychopathology in the MMPI or 16PF, but in the PFT have a strong need-persistence and tend to deny any obstacles in their way.
The 5 cases grouped into the second cluster have by far the most significant psychopathology, emotional disorders and severe narcissistic dysregulation in all 4 dimensions. In these individuals their auto-regulation-modes are always on the edge of decompensation and the gender dysphoria appears as only one facet in a profoundly disturbed personality.
The two small clusters 3 and 4 differ from the larger clusters in a respectively particular manner. The cases in cluster 3 have no narcissistic and general psychopathology but are socially isolated and introverted, they feel socially unattractive and live more or less in disguise. They have a strong need-persistence, tend to play obstacles or frustrations down and hope that the desired sex change will turn their lives to the better.
The patients in cluster 4 do not seem to have a true gender dysphoria but rather a
disturbed body-relation which is more of the hypochondriac, dysmorphophobic or
somatization type. Accordingly, they do not reject their body and do not have a negative
body-self in the NI. They feel easily blocked by conflicts or frustrations, which seem to
be expressed in a body-language.
At this stage, the provisional status of our data only permits some few conclusions. The central findings of this questionnaire analysis support the view of a great heterogeneity of gender disordered males which not only extends to the already well known gender and sexuality variables but also to general personality pathology and especially the different modes of self-regulation. We could identify a significant narcissistic pathology in most of our patients, but the regulation-modes afflicted by this pathology differ widely. The cluster analysis has yielded an interesting and clinically reasonable subtyping of our patients with two larger subgroups of which one is marked by severe narcissistic and personality pathology where the gender dysphoria appears as only one facet in this profound pathology, as a rather desperate attempt at stabilizing a fragmented self. In the other larger cluster there is no substantial personality pathology, but one might speculate that the gender dysphoria is part of a deeper problem in object-relations, for which the transsexual wish probably serves as an imagined solution. The data analysis has indicated that the sexual orientation does account for some variance in our sample, but in a multivariate view it does not seem to be a significant predictor. Thus, by our preliminary analysis the notion that gynephilics have more substantial personality and gender pathology could not be confirmed. However, the complete analysis of our data including the developmental and biographic variables as well as the results of the structured interview appears to be promising and may change this impression.
In closing, the cumulative evidence of our study so far is consistent with the view
that gender dysphoria is a disorder of the sense of self as has been proposed by Beitel
(1985) or Pfäfflin (1993). The central problem in our patients is about identity and the
self in general and the transsexual wish seems to be an attempt at reassuring and
stabilizing the self-coherence which in turn can lead to a further destabilization if the
self is already too fragile. In this view the body is instrumentalized to create a sense
of identity and the splitting symbolized in the hiatus between the rejected body-self and
other parts of the self is more between good and bad objects than between masculine and
feminine. The results obtained so far confirm the conviction that we have to maintain a
clinical perspective in the field of gender dysphoria and must continue to improve our
understanding of this enigmatic and fascinating problems.
Becker H, Hartmann U. (1994): Geschlechtsidentitätsstörungen und die Notwendigkeit der klinischen Perspektive. Ein Beitrag aus der psychiatrischen Praxis. Fortschritte der Neurologie Psychiatrie 62: 290 - 305.
Kernberg O. (1984): Severe Personality Disorders. Yale University Press: New York.
Deneke FW, Hilgenstock, B. (1989): Das Narzißmus-Inventar. Huber:Bern.
Blanchard R. (1985); Research methodes for the typological study of gender disorders in males. In: Gender Dysphoria. Development, Research, Managment. Steiner, BW ed. Plenum: New York.
Blanchard R. (1989): The concept of autogynephilia and the typology of male gender dysphoria. The Journal of Nervous and Mental Disease 177: 616 - 623.
Beitel A. (1985): The spectrum of gender identity disturbances. An intrapsychic model. In: Gender Dysphoria. Development, Research, Managment. Steiner, BW ed. Plenum: New York.
Pfäfflin F. (1993): Transsexualität. Beiträge zur Psychopathologie, Psychodynamik und zum Verlauf. Enke: Stuttgart 1993.
Correspondence and requests for materials to: