Financing Insurance Workplace Legal issues Name choice Name change Driver’s license Birth certificate Passport Marriage Will Other documents
Hair removal Voice Facial feminization Hair loss Reproduction Hormones Orchiectomy Vaginoplasty Breast implants Injected silicone Breast forms Movement Tall clothing Skin care Makeup Handwriting
Transitioning early in life: psychology
Below is a published paper showing that those who transition as teens are well-adjusted
and stable after transition in proportion with the non-transgender population.
Postoperative psychological functioning of adolescent transsexuals
AB The Rorschach Comprehensive System was used to assess postoperative psychological
functioning in transsexuals who applied for sex reassignment in adolescence.
We investigated a group of 22 consecutive adolescent transsexuals, who were
otherwise psychologically well adapted. Nineteen subjects provided valid Rorschach
protocols before and after sex reassignment. The most notable change found was
an increase in X+%, reflecting a decrease in both distorted perception and idiosyncratic
perception. Little support was found for the idea of major psychological deterioration
for the patients as a group. Rather, the results suggest stability in psychological
functioning over time. The Rorschach findings are consistent with questionnaire
data from earlier studies, with the exception that the Rorschach data may point
to some improvement in reality testing.
The Rorschach Comprehensive System was used to assess postoperative psychological functioning in transsexuals who applied for sex reassignment in adolescence.
We investigated a group of 22 consecutive adolescent transsexuals, who were
otherwise psychologically well adapted. Nineteen subjects provided valid Rorschach
protocols before and after sex reassignment.
The most notable change found was an increase in X+%, reflecting a decrease
in both distorted perception and idiosyncratic perception. Little support was
found for the idea of major psychological deterioration for the patients as
a group. Rather, the results suggest stability in psychological functioning
over time. The Rorschach findings are consistent with questionnaire data from
earlier studies, with the exception that the Rorschach data may point to some
improvement in reality testing.
A more practical reason for unwillingness to start SR before 18 is that in
many countries adolescents are still legally dependent on the consent of their
parents when deciding upon medical treatment. Resistance from nonconsenting
parents thus forms an additional complicating factor in the treatment process,
while at the same time the clinician runs an increased risk of litigation.
Naturally, if treatment modes other than SR could alter extreme and life long
cross gender identities, clinicians should refrain from SR. But, as we have
pointed out elsewhere, the literature does not permit us to draw such conclusions
(Cohen-Kettenis & Kuiper, 1984). Understandable as the dilemma of conscientious
professionals who want to prevent postoperative regret may be, refusing or delaying
medical treatment is not always in the best interest of the patient. Adolescents
who were extremely cross-gendered from their earliest years may, especially
around puberty, develop other problems such as depression or social anxiety
as a consequence of their gender identity disorder. In these cases, late treatment
could have a negative impact in a variety of areas that are particularly important
during adolescence (e.g., peer relationships, romantic involvements, or academic
achievement, or all of these). This developmental arrest may in itself lead
to additional, yet avoidable problems. Second, the physical changes of puberty
(e.g., a low voice and beard growth in male-to-female transsexuals [MFs]) create
life-long traces of the biological sex when treatment is unduly postponed. Indeed,
Ross and Need (1989) found that postoperative psychopathology was primarily
associated with factors that made it difficult for postoperative transsexuals
to pass in their new gender or that continued to remind them of their transsexualism.
Thus, early treatment may be able to prevent unnecessary negative emotional
and psychological consequences. Third, on the basis of numerous follow-up studies,
it can be concluded that, in adults, unfavorable postoperative outcome is related
to a late start of the SR procedure rather than an early one (for a review,
see Cohen-Kettenis & Gooren, 1999). Age at assessment also emerged as a
factor differentiating two small groups of adult MFs with and without postoperative
regrets (Lindemalm, Korlin, & Uddenberg, 1987).
Since 1987, adolescents with gender identity disorder have been diagnosed and
treated at the gender clinic of the Department of Child and Adolescent Psychiatry,
University Medical Center Utrecht (UMCU), in collaboration with the Gender Team
of the Free University Medical Center (FUMC) in Amsterdam. The clinical procedure
is based on the Standards of Care of the Harry Benjamin International Gender
Dysphoria Association (Levine et al., 1998), a professional organization in
the field of gender identity disorders, and is described in detail in Cohen-Kettenis
and van Goozen (1997).
In a carefully selected group of applicants, the often-assumed association
between transsexualism and psychopathology has not been found (Cohen, de Ruiter,
Ringelberg, & Cohen-Kettenis, 1997; Cohen-Kettenis & van Goozen, 1997;
Smith et al., 2001). This significantly contributed to the decision to start
hormone therapy between the ages of 16 and 18 in these adolescents. This happens
in two phases: first, hormones with reversible effects (for MFs, antiandrogens
to block further masculinization of the body; for female-to-male transsexuals
[FMs], progestins to suppress menstruation); second, estrogens to feminize the
MFs and androgens to masculinize the FMs. If applicants do not fulfill the rather
strict treatment eligibility criteria, treatment is denied or postponed until
Two studies were carried out to examine the effectiveness of sex reassignment
for adolescent transsexuals. Cohen-Kettenis and van Goozen (1997) conducted
an ex post facto study on postoperative functioning of the first 22 consecutive
adolescent transsexual patients who had attended the gender clinic at the UMCU
and who had undergone SR. They concluded that starting the SR procedure before
adulthood resulted in favorable postoperative functioning, provided that careful
diagnosis had taken place in a specialized gender team and that the criteria
for starting the procedure early had been strict. To check the reliability of
the findings, a prospective study was performed involving the next 20 consecutive
adolescents who had undergone SR and 27 adolescents whose application for SR
had been rejected (Smith et al., 2001). Again, theresults did not confirm the
concern that psychological functioning would deteriorate after SR.
In both studies, psychological functioning was measured by means of well-known
reliable and valid self-report personality questionnaires. As a group, the treated
adolescent transsexuals did not show signs of severe psychopathology, neither
before nor after treatment. Compared to Dutch normative groups, the mean follow-up
scores of the patients in both studies were all within the average range. However,
some clinicians, such as Lothstein (1984), criticize the use of self-report
questionnaires with transsexuals. Lothstein argued that transsexuals suffer
from borderline personality pathology and stated that the intact reality testing
of individuals with such a pathology is only expected to become impaired in
unstructured situations. So self-report questionnaires may be too structured
to uncover this phenomenon. In addition, the possibility was mentioned that
transsexuals intentionally try to "fake good" on self-report measures
in order to be referred for SR (before treatment) or downplay negative outcomes
as a psychological defense (after treatment).
Therefore, information was gathered on the psychological functioning of adolescent
transsexual applicants for SR, making use of the Rorschach test. This instrument
is thought to be less subject to influences of conscious steering in responding.
Part of the collected data was used in a study to determine the extent to which
psychopathology is necessarily associated with adolescent transsexualism (Cohen
et al., 1997). Areas of psychological functioning associated with fundamental
psychological disturbances were assessed by means of the Rorschach Comprehensive
System (Exner, 1995). As a group, the adolescent transsexuals did not show the
marked degree of psychopathology encountered in psychiatric groups on the variables
investigated. The results supported the findings among adult transsexuals that
major psychopathology is not associated with the development of transsexualism
(Cole, O'Boyle, Emory, & Meyer, 1997; Fleming, Jones, & Simons, 1982;
Mate-Kole, Freschie, & Robin, 1988; Pauly, 1981).
The aim of the present follow-up study is to examine the level of psychological
functioning of adolescent transsexuals before and after SR treatment. Specifically,
we aim to detect potential differences in psychological functioning, which might
not become apparent with structured questionnaires. Therefore, a less structured
instrument is employed: the Rorschach, following the procedure of the Comprehensive
System (Exner, 1995).
A design involving random assignment to early versus late treatment or SR treatment
versus non-SR treatment would be methodologically preferable over a design without
a control group. However, as is repeatedly pointed out in follow-up studies
among transsexuals, it is for ethical reasons not possible to create a late
or nontreated control group. Moreover, chances are low that patients, after
being informed about its purpose, would agree to participate in such a study.
Twenty-two patients who had participated in the Cohen et al. (1997) study and
had received sex reassignment surgery were requested to participate in a follow-up
study making use of the Rorschach test (about half of these patients had participated
in the first questionnaire follow-up study and the remaining patients had participated
in the second). Two patients refused participation. As one patient did not provide
a protocol with 14 or more responses, valid re-test protocols were available
for 19 patients.
The mean age of the 19 patients was 22.5 years (SD = 2.09, range, 18-27) at
follow-up. Six patients were MFs and 13 were FMs. MFs and FMs did not differ
significantly in terms of age.
The Rorschach was administered to patients in accordance with the procedures
for the Comprehensive System (Exner, 1995). Psychologists trained in the Comprehensive
System administered the Rorschach protocols. These psychologis were not the
clinicians who referred patients for hormone treatment.
The mean period of length between the pre-SR and post-SR testing sessions was
58.5 months (SD = 14.5, range, 40.5-87.2). Length of time between the two testing
sessions was not significantly associated with gender, with age at time of first
testing, nor with age at time of second testing.
The ethics' committees of both the UMCU and the FUMC approved the study.
Univariate two-tailed paired t tests were used to determine if there were changes
between pre- and post-treatment psychological functioning of the adolescent
transsexuals. An independent second psychologist coded the post-SR protocols.
Summary statistics for 69 Rorschach Comprehensive System variables were studied
for differences between pre-SR and post-SR measurements. This was done for the
full group of patients and for MFs and FMs separately (see Table I).
Differences between pre- and post-SR measurements with an associated univariate
two-tailed at p [less than or equal to] .05 were found for 13 variables. The
univariate p of two of these variables was less than .01: form quality (X+%)
was greater after SR than before and the frequency of form dominated color responses
(FC) was less after SR than before.
For a description and interpretation of the significant variables, and of X-%
(one of the main variables indicating distorted perception), following Exner's
guidelines (Exner, 1995) for interpretation, see Appendix. For a more comprehensive
description of all the variables and of the blots, see Exner (1995).
We observed that there were significant differences for a little less than
one fifth of the Rorschach variables between the pre-SR and post-SR measurements.
The differences were not systematic in the sense that they were all in one direction
or even within one domain of functioning. We found, for example, a decrease
in frequency for a number of determinants (m, C', FC, and FD), but a significant
increase in another (V). The changes were within normal ranges for a number
of variables. Thus, it is difficult to find a coherent interpretation for these
There was a sharp increase in X+%. This can be seen as reflecting a greater
amount of conventionality in processing. The increased X+% is attributable to
a decrease in both distorted perception (X-%) and idiosyncratic perception (Xu%).
In our earlier study of pre-SR adolescent transsexuals (Cohen et al., 1997),
we found the adolescent transsexuals to have an X-% intermediate between a psychiatric
adolescent and a student control group. In that study, it was the only finding
consistent with the idea that adolescent transsexuals exhibit an underlying
psychopathological disturbance. Therefore, it is interesting to note that as
a group the patients in this study showe improvements and obtained X-% values
nearly equal to those of the student controls in the earlier study. It may be
that after SR there is a diminution in psychological conflict, resulting in
less impingement on conventionality. Prior to sex reassignment, transsexuals
experience a serious psychological discrepancy between their physical self and
th eir feeling of a real self. SR is sought as a means of resolving this discrepancy.
If SR is successful, the discrepancy can be eliminated or reduced, lowering
the possible strain and debilitating effect of the discrepancy on reality testing.
We found little or no support for the idea that there was major psychological
deterioration for the patients as a group. By and large, the results suggest
stability in psychological functioning over time and that there were some areas
in which improvement was evident.
A number of limitations of the study need to be addressed. One limitation is
formed by the size of the sample. For statistical purposes, the total number
of subjects was small. However, the sample is based on a sizeable proportion
of the total population of adolescent transsexuals requesting SR in the Netherlands
and, until larger samples are available, we must make best use of the modest
data available to us.
A second limitation concerns the possibility of selection bias in our sample.
Firstly, adolescents who apply for SR have parents who are generally supportive
of treatment (though they may be unhappy about their child's transsexualism).
The adolescents in our sample may enjoy more favourable circumstances than adolescent
transsexuals whose parents are not supportive of treatment. Secondly, adolescents
with less extreme or more fluctuating cross-gender identities are more likely
not to pursue SR this early in life. In any event, we do not have data on the
number of adolescent transsexuals not applying for SR. We can therefore not
claim that our patients are representative of all Dutch adolescent transsexuals.
Our conclusions pertain to the population of psychologically well-functioning
adolescent transsexuals that applies for SR before adulthood and completes SR
in adulthood. (One of the selection criteria for early SR is psychological stability.)
A third limitation concerns experimenter bias due to the possibility that the
psychologists administering and scoring the Rorschach were not blind to the
potential problems of the individuals tested or perhaps even the study hypotheses.
Care was taken to avoid additional bias in the coding process by having independent
psychologists, who were not familiar with the patients or their condition, carry
out a second coding procedure of the post-SR protocols.
A fourth limitation is the absence of a treatment control group. Ideally, a
different research design in which eligible SR applicants would be randomly
assigned to either a treatment group or a nontreatment control group would have
been methodologically more desirable. As we pointed out earlier, such a study
is for ethical and practical reasons impossible to conduct.
In their questionnaire study, Cohen-Kettenis and van Goozen (1997) found a
few differences between pre- and post-SR adolescent transsexuals. On the short
version of the MMPI (NVM; Luteyn, Kok, & van der Ploeg, 1980) and the Dutch
Personality Inventory (NPV; Luteyn, Starren, & van Dijk, 1985), an increase
was observed in the extroversion, dominance, and self-esteem subscales and a
decrease in the inadequacy subscale. Pre- and post-SR mean scores were all within
average ranges of Dutch norms. This result was confirmed in the second questionnaire
study (Smith et al., 2001). Our Rorschach findings are therefore consistent
with the questionnaire investigations with regard to the stability of psychological
functioning throughout the SR treatment period, with the exception that the
Rorsohach data may point to some improvement in perceptual accuracy, indicating
enhanced reality testing. The fear that the adolescents' psychological functioning
will deteriorate as a consequence of an early start of the SR procedure is not
substantiated by the Rorschach findings. If anything, their functioning changes
in a more healthy direction.
Many applicants for SR are not good candidates for SR and probably will never
be. In some, SR is sought as a solution for non-gender problems. Careful diagnostic
procedures are used to keep "questionable" SR applicants from receiving
SR. When adverse factors are present in the psychological profile of an adolescent
applicant, it is probably prudent to maintain a conservative policy of delaying
the start of hormone treatment until adulthood.
APPENDIX: DESCRIPTION AND INTERPRETATION OF THE SIGNIFICANT VARIABLES
WsumC: the weighted sum of all color responses (each FC [Form Color] = 0.5;
each CF [Color Form] = 1.0; each C [Color] = 1.5). This index variable is seen
as an indication of the amount of emotional energy that is used when the subject
responds to the environment.
EA: experienced actual (the sum of M and WsumC) is regarded as an indication
of the resources available to the individual in guiding behavior. (M: human
movement responses are coded for responses involving the kinaesthetic activity
of a human, or of an animal or fictitional character in human-like activity.)
C': achromatic color responses are responses in which the percept is referred
to as being either black, white, or gray. The amount of C' in the protocol is
seen as an indication of the amount of suppression of emotion.
V: vista responses are coded for responses in which there is a perception of
three dimensionality based on shading characteristics of the blot. The presence
of vista responses is considered to suggest the presence of painful experience
in the perception of self.
X+%: the percentage of responses that are conventional (and adequate). A response
is considered conventional when it is offered to a particular area of the blot
by at least 2% of the population.
X-%: the percentage of responses showing poor correspondence between the verbalized
percept and the contours of the blot. The presence of more than a few of such
responses is considered as distorted perception.
Xu%: the percentage of unusual responses (form adequate but not conventional).
These are responses, which are immediately recognized as adequate, but are relatively
rarely offered in the population.
Zf: the frequency of responses exhibiting organizational activity. This frequency
provides important information concerning the extent to which the subject has
organized the stimulus field and whether that effort has been efficient. This
frequency is regarded as an indication of the amount of cognitive effort exercised
to organize the environment.
W: whole responses are a specific form of organizational activity and are regarded
as the number of responses involving the entire blot.
FC: form color responses are form-dominated responses in which color I integrated.
Blends responses are those having more than one determinant (e.g.: color and
movement). The number of Blends responses in a protocol is seen as an indication
of emotional complexity.
Pure H is coded for responses involving full human percepts (seeing a total
person as opposed to for example only a head or a hand). Presence of Pure H
responses is regarded as, among other things, interpersonal maturity.
FD: form dimension is coded for responses involving three-dimensionality not
based on shading in the blot. The presence of FD responses is considered as
the use of introspection.
Note. Only two-tailed p values [less than or equal to] .10 have been listed.
Mean values have been rounded off. Intercoder agreement Kappas: 0.63 for Colour
Codes, 0.74 for FM, 0.48 form, 0.81 for V.0.78 for Conventional Form Quality
(X+%). and 0.85 for Organizational Activity (Zf).
Received April 2, 2001; revision received July 13, 2001; accepted July 30,