Anxiety and Depression in Transgender Individuals: The Roles of

Transition Status, Loss, Social Support, and Coping

Stephanie L. Budge and Jill L. Adelson
University of Louisville
Kimberly A. S. Howard
Boston University
Objective: The purpose of the current study was to examine facilitative and avoidant coping as mediators
between distress and transition status, social support, and loss. Method: A total of 351 transgender
individuals (n 226 transgender women and n 125 transgender men) participated in this study.
Participants completed measures on transgender identity, family history of mental health concerns,
perceptions of loss, coping, depression, and anxiety. Results: The rates of depressive symptoms (51.4%
for transgender women; 48.3% for transgender men) and anxiety (40.4% for transgender women; 47.5%
for transgender men) within the current study far surpass the rates of those for the general population.
Structural equation modeling (SEM) was used to analyze the data2 separate models were hypothesized,
based on reports of anxiety or depression. The SEM results suggest that the processes for transgender
women and transgender men are primarily similar for depression and anxiety; avoidant coping served as
a mediator between transition status and both distress variables. Social support was directly related to
distress variables, as well as indirectly related through avoidant coping. Conclusion: Results suggest the
need for practitioners to focus on interventions that reduce avoidant coping strategies, while simultane-
ously increasing social support, in order to improve mental health for transgender individuals. Individuals
who are in the beginning stages of their transition will use different coping strategies than those who are
in later stages; interventions should be adjusted on the basis of the transition status of transgender clients.
Keywords: transgender, anxiety, depression, coping, social support
Transgender individuals are particularly vulnerable to mental
health concerns and psychological distress (Nuttbrock, Rosen-
blum, & Blumenstein, 2002). Regarding specific psychological
distress, research has shown that transgender individuals report
higher levels of both anxiety and depression than the population as
a whole. Depression has been reported to affect 16.6% of the total
population, and combined anxiety disorders affect 28.8% of the
United States population (Kessler, Berglund, Demler, Jin, & Wal-
ters, 2005). Within the literature, rates of depression for transgen-
der individuals range from 48% to 62% (Clements-Nolle, Marx,
Guzman, & Katz, 2001; Nemoto, Bodeker, & Iwamoto, 2011;
Nuttbrock et al., 2010). Anxiety and overall distress rates for
transgender individuals range from 26% to 38% (Hepp, Kraemer,
Schnyder, Miller, & Delsignore, 2005; Mustanski, Garofalo, &
Emerson, 2010).
Although the rates of depression, anxiety, and overall distress
indicate above-average rates for the transgender population, there
are only several studies that examine potential explanations for
these findings. Nuttbrock et al. (2010) report that there is a sig-
nificant positive relationship between gender-related abuse and
depression in transgender women. The authors note that social
stressors and/or ostracism from peers can explain a large amount
of the depression that is reported by the individuals in their study.
Budge et al.s (2012) qualitative study indicates that distress varies
on the basis of several factors, including the individuals transition
process, coping mechanisms used, and level of social support.
Although both of these studies provide insight into possible ex-
planations for contributing factors to transgender individuals dis-
tress, there continues to be a lack of generalizable information
regarding the actual process through which individuals cope with
and experience depression and anxiety.
Coping mechanisms have been theorized to buffer the effects of
psychological distress due to stigma, internalized homo(or trans)pho-
bia, and experiences of discrimination and violence (Meyer, 2003).
Internalized transphobia is a result of internalizing negative mes-
sages based on societal expectations of what is considered to be
normal (i.e., not deviating from gender norms) (Bockting &
Coleman, 2007). As transgender individuals negotiate in a world
that is rarely accepting of them, it is critical that the types of coping
mechanisms (both adaptive and maladaptive) are better under-
stood. It has been theorized that there are two general categories of
This article was published Online First February 11, 2013.
Stephanie L. Budge and Jill L. Adelson, Department of Educational and
Counseling Psychology, Counseling, and College Student Personnel, Uni-
versity of Louisville; Kimberly A. S. Howard, School of Education, Boston
University.
We thank Bruce Wampold, (Department of Counseling Psychology,
University of WisconsinMadison), David Kaplan (Department of Edu-
cational Psychology, University of WisconsinMadison), Lee Nelson
(Department of Counseling and Family Therapy, University of Missouri
St. Louis), and Teresa Bear (Department of Psychology, Alverno College)
for their statistical assistance and helpful comments on a draft of this
article.
Correspondence concerning this article should be addressed to Stephanie
L. Budge, Department of Educational and Counseling Psychology, Coun-
seling, and College Student Personnel, University of Louisville, Louisville,
KY 40292. E-mail: [email protected]
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
Journal of Consulting and Clinical Psychology 2013 American Psychological Association
2013, Vol. 81, No. 3, 545557 0022-006X/13/$12.00 DOI: 10.1037/a0031774
545
coping styles: (a) emotion-focused coping and (b) problem-
focused coping (Lazarus & Folkman, 1984); for the purposes of
this study, these are described as facilitative and avoidant coping,
respectively. Avoidant coping occurs when individuals try to pre-
vent an emotional response to the stressor, for example, using
avoiding behaviors or cognitions, minimizing the problem, trying
to detach oneself from the outcomes of a problem, or overeating or
drinking. Facilitative coping occurs when an individual seeks
social support, learns new skills, changes behavior to positively
adapt, and finds alternative means to seek happiness.
The coping literature related to transgender individuals is sparse
and primarily focuses on transgender youth. For transgender
youth, the role of coping within the family unit was related to
making meaning of gender within the family and accepting the
youths gender identity (Wren, 2002). Coping, via collective self-
esteem, has been found to temper psychological distress related to
the adoption of an adult transgender male identity (Sanchez &
Vilain, 2009). Contrary to other findings, higher levels of aspects
of facilitative coping were found to be related to higher levels of
depression, trauma symptoms, mental health problems, and inter-
nalizing/externalizing problems for transgender youth (Grossman,
Daugelli, & Frank, 2011). Although several of these studies link
coping to distress, none indicate different types of coping mech-
anisms and their relationship to mental health concerns for trans-
gender individuals.
Just as coping appears to play a central role during the process
of gender transitioning, social support also appears to influence
transgender individuals well-being and smoothness of gender
transition (Budge et al., 2012). The importance of social support is
not unique to the transgender population; humans are social beings
who form attachments from the moment they are born (Bowlby,
1969). Additionally, humans have a fundamental need to belong
that is adaptive (Leary, Tambor, Terdal, & Downs, 1995). The
buffering hypothesis provided by Cohen and Wills (1985) theo-
rizes that others can provide direct support, and thus help individ-
uals cope with stressful events by being able to reveal stress to
others and have them take on part of the burden of the stress (by
listening or providing material help).
Typically, social support is indicated to be important for cis-
gender (e.g., individuals whose sex assigned at birth is congruent
with their gender identity) men and women; however, women tend
to use social support more than men (Dwyer & Cummings, 2001).
Social support for transgender individuals has been shown to be
particularly important in order to learn about medical resources,
speak out about political concerns, and use social networks (Pinto,
Melendez, & Spector, 2008), yet there are major deficits in social
support for many transgender individuals (Budge et al., 2012).
Furthermore, transgender individuals perceive that they receive
less social support from family members when compared with
their nontransgender siblings (Factor & Rothblum, 2007). Lower
levels of social support in transgender individuals are related to
more unprotected anal sex in transgender women (Golub, Walker,
Longmire-Avital, Bimbi, & Parsons, 2010) and also predict more
depressive symptoms (Nemoto et al., 2011). Although the litera-
ture indicates that social support is important in decreasing distress
for transgender individuals, it is unclear what types of social
support (e.g., family, friends, romantic) contribute to lower levels
of distress.
Research findings are clear in regards to social supportthe
more social support an individual experiences, the less distress will
be evident. Although these findings have been replicated manifold
in research, less is known about the experiences of loss related to
not having social support. Budge et al.s (2012) results indicate
that many transgender participants in the study reported loss on a
multitude of levelsfrom social support loss (family, friends, and
romantic partners) to financial loss. It has been established that
loss is greatly related to psychological distress (Park, 2008). When
not in reference to the death of a loved one, loss is most often
related to the shattering of assumptions, hopes, and worldview
(Gluhoski & Wortman, 1996; Schwartzberg & Janoff-Bulman
1991). Harvey and Miller (1998) indicate that a pile-up of losses
often leads to increased distress that will contribute to a need to
adapt to and cope with significant stressors. Though preliminary
investigations shed light on the concept that transgender individ-
uals do indeed experience loss, the extent of loss and the outcomes
from loss are unknown.
Loss and social support are interwoven in their contribution to
psychological distress and well-being. In addition to these factors,
it is also possible that transgender identity may contribute to the
level of distress experienced. Social identity theory (Tajfel &
Turner, 1986) postulates that there is a strong link between self-
concept and group identification. When individuals feel that their
identity is not positively perceived, they may seek to leave their
group or find other ways to achieve a positive identity (R. Brown,
2000). Devors (2004) theoretical model of transgender identity
development indicates that, as transgender individuals move from
anxiety regarding their nonconforming identity, they begin to
accept their identity and eventually experience prideful feelings.
Bockting and Colemans (2007) theoretical model describes sev-
eral processes of transgender identity development, such as pre-
coming out, coming out, exploration, intimacy, and identity inte-
grationthis model indicates that transitioning is a developmental
process whereby individuals will achieve more of a sense of self as
they move through the stages. Research indicates that the further
along transgender individuals are in their identity and transition
process (hereafter referred to as transition status), the more well-
being they report (Budge et al., 2012). Although these findings are
informative, there is a dearth of information regarding specific
mental health outcomes (e.g., depression and anxiety) and an
understanding of the specific relationships that occur during the
identity process, coping, social support, and loss.
In the current study, our objective was to determine the rela-
tionship of transition status, perceptions of loss, social support, and
coping on the mental health outcomes of depression and anxiety
for transgender men and women by using structural equation
modeling (SEM). Although several studies have focused on social
support and specific types of coping as they relate to depression
and distress, none of these studies has included the amount of
perceived loss and the transition status of transgender men and
women. In addition, social support is typically included as the
coping mechanism; in this study, we hypothesized that facilitative
and avoidant coping would mediate the relationship between spe-
cific types of support and mental health outcomes. As well, we
hypothesized that facilitative and avoidant coping would mediate
the relationship between perceived loss, transition status, and
mental health outcomes. Previous studies have determined gender
differences for men and women related to social support and
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
546
BUDGE, ADELSON, AND HOWARD
coping (e.g., Dwyer & Cummings, 2001), thus we were interested
in testing differences between transgender men and women in the
current study.
Method
Participants
For the purposes of the current study, only transgender men
(n 125) and transgender women (n 226) were included in
the analyses. Participants ages ranged from 18 to 78 (M
40.28, SD 12.85). For race/ethnicity, participants indicated
the following: 86.3% White (n 303), 2.6% Hispanic/Latino
(n 9), 1.1% Black/African American (n 4), 1.1% Asian/
Asian American (n 4), 0.6% Native American (n 2), 7.4%
biracial (n 26), and 0.6% multiracial (n 2). Transgender
women (n 226) reported the following sexual orientations:
29.2% bisexual (n 66), 32.7% lesbian (n 74), 18.1%
straight/heterosexual (n 41), 4.4% queer (n 10), 9.7%
questioning (n 22), 1.3% gay (n 3), and 3.3% (n 10)
identified identities in an open-ended category (e.g., pansexual,
asexual, etc.). Transgender men (n 125) reported the follow-
ing sexual orientations: 19.2% bisexual (n 24), 4.0% lesbian
(n 5), 20% straight/heterosexual (n 25), 37.6% queer (n
47), 3.2% questioning (n 4), 9.6% gay (n 12), and 4.2%
(n 8) identified identities in the open-ended category.
Measures
Demographic questions. Participants identified their gender
identity by filling out an open-ended question where they could
provide the best fitting identity label; thus, individuals who exclu-
sively identified with a binary gender identity (e.g., transgender
man, male-to-female transsexual) were included in the current
study. Participants transition status was assessed via an item that
asked respondents to rate on a 5-point scale where they felt they
were in their transition process (e.g., from 1 I have been
thinking about transitioning, but have not taken any steps to make
any changes to 5 I have made most of the changes I have
wanted to make and consider myself living full time with my
transgender identity); this question was developed on the basis of
transgender identity theory (e.g., Devor, 2004) and qualitatively
tested (Budge et al., 2012). Family history of anxiety and depres-
sion were assessed by asking participants whether they had bio-
logical family members (limited to parents, siblings, and grand-
parents) who had been diagnosed with anxiety or depression and to
indicate which family members were provided with the diagnosis.
This question was asked due to the high heritability of depression
and anxiety (Hettema, Neale, & Kendler, 2001; Sullivan, Neale, &
Kendler, 2000), thus it was important to control for family history
of these mental health concerns.
Multidimensional Scale of Social Support (MSPSS; Zimet,
Dahlem, Zimet, & Farley, 1988). The MSPSS is a 12-item,
7-point Likert scale with three 4-item subscales: Significant Other
(SO), Family (F), and Friends (FR). Previous research (Yeh et al.,
2008) has yielded coefficient alphas for each of the three sub-
scales: .92 for SO, .87 for F, and .85 for FR. Sample items for this
scale include: My family tries to help me, I can talk about
problems with my friends, and There is a special person who is
around when I am in need. For the current study, the following
coefficient alphas emerged: .95 for SO, .93 for F, and .93 for FR.
Cronbachs alpha for the overall scale in the current study was .91.
Item-level responses were included in the analyses.
Ways of Coping (Revised) (WC-R; Folkman, Lazarus,
Dunkel-Schetter, DeLongis, & Gruen, 1986). The WC-R scale
is a 66-item questionnaire that asks respondents about specific
stressful encounters and the thoughts and acts that people use to
cope with these stressful demands. Out of the eight scales, the
following four are negative coping mechanisms: Confrontive Cop-
ing, Distancing, Self-Controlling, and Escape Avoidance. The last
four scales are positive coping mechanisms: Planful Problem Solv-
ing, Accepting Responsibility, Positive Reappraisal, and Seeking
Social Support. Examples of items include the following: I try not
to act too hastily or follow my first hunch, Wish that I could
change what is happening or how I feel, Accept it, since nothing
can be done, and Im changing or growing as a person in a good
way. Participants respond to a 4-point Likert scale (0 does not
apply and/not used; 3 used a great deal). The instrument is
scored by summing the items that comprise each subscale. The
scale was normed on a community sample of middle-aged hetero-
sexual married couples (Folkman et al., 1986). Coefficient alphas
from the validation of the original study ranged from .70 to .88 for
the subscales.
As the validation study for the WC-R was based on heterosexual
couples, it was anticipated that the transgender population may
differ in meaningful ways from the validation sample; thus, an
exploratory factor analysis (EFA) was conducted on all 66 items in
the inventory. The fit of the model was adequate based on the
root-mean-square error of approximation (RMSEA):
2
(1645)
2763.40, p .001; RMSEA .039, CI
90
[.036, .042]; standard-
ized root-mean-square residual (SRMR) .035. The item loadings
were not similar to those from the original sample for which the
measure was tested. It was our interpretation that two clusters were
indicative of the true latent variables underlying the construct of
coping. Thus, the EFA for a two-factor model was examined. The
fit of the model was adequate for a two-factor model based on the
RMSEA:
2
(2014) 5016.59, p .001; RMSEA .058, CI
90
[.056, .060]; SRMR .061. For the first factor, 19 items loaded
higher than .4 (.41.73). For the second factor, 16 items loaded
higher than .4 (.44.73). All other items were not included, either
because they did not load more than .4 on any item or if there were
two or more loadings higher than .4 on any single item. Inspection
of the items that loaded onto Factor 1 (Facilitative Coping) re-
vealed that they clustered around the concept of facilitative coping
as a coping mechanism; that is, the participants were able to use
the coping mechanism to help them with their problem. An exam-
ple of a facilitative coping item from the WC-R is as follows:
Looked for the silver lining, so to speak; tried to look on the
bright side of things. The items that loaded onto Factor 2
(Avoidant Coping) clustered around the concept of avoidance; that
is, the participants avoided their problems in order to ignore the
problem or chose not to deal with the issue. An example of an item
from this factor is as follows: Tried to make myself feel better by
eating, drinking, smoking, using drugs or medication, etc. Sub-
scale sum scores for Facilitative Coping and for Avoidant Coping
were used in this study, and the coefficient alpha for the current
study was .87 for each of those subscales.
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
547
TRANSGENDER ANXIETY, DEPRESSION, AND COPING
Transgender Perception of Loss Scale. The Transgender
Perception of Loss Scale is a nine-item scale that asks participants
to indicate the extent to which you have experienced loss that has
occurred since you began to identify as transgender. This scale
was created for the purposes of this study. Loss is measured in the
following areas: employment, housing, finances, healthcare, pa-
rental, sibling, romantic partner, children, and friends. These con-
structs were created from qualitative themes derived from a data
set related to transgender emotional and coping processes (Budge
et al., 2012). Participants respond to a 5-point Likert scale (1 no
loss at all, 2 a little loss, 3 some loss, 4 much loss, and 5
extreme loss). Sum scores on the measure were used in the anal-
ysis. Cronbachs alpha for the current study was .80.
Center for Epidemiologic Studies Depression Scale (CES-D;
Radloff, 1977). The CES-D measures depressive symptoms in
the general population. This scale is a 20-item measure that asks
questions regarding the major components of depressive symptom-
atology, such as depressed mood, loss of appetite, sleep distur-
bance, and feelings of worthlessness. Sample items include the
following: I felt hopeful about the future and I was bothered by
things that usually dont bother me. Respondents answer items on
a 4-point Likert scale, where a 0 indicates rarely or none of the
time, and a 3 indicates most or all of the time during the past week.
This measure has been shown to have high internal consistency and
concurrent validity with clinical diagnosis (Boisvert, McCreary,
Wright, & Asmundson, 2003). The scale is scored by summing all of
the items per respondent and using the total score to indicate the
cutoff for depression. Previous studies have found internal consis-
tency reliability at .90 for the general population (Katz, Petracca,
& Rabinowitz, 2009; Schleichter, Harris, Catley, & Nazir, 2009).
This scale has been used with transgender men and women, where
the typical cutoff point of 16 was used to indicate depression
(Clements-Nolle et al., 2001). Cronbachs alpha for the current
study, which included the sum scores, was .94.
Burns Anxiety Inventory (BAI; Burns, 1998). The BAI is a
33-item inventory that assesses symptoms of anxiety. This self-
report measure asks respondents to indicate feelings of stress or
anxiety (thoughts, feelings, and physical symptoms) over the past
7 days. Examples of items on this scale include Feeling tense,
stressed, uptight or on edge and Fears that something terrible
is going to happen. Respondents answer items on a 4-point Likert
scale, where 0 not at all and 3 a lot. The measure is scored
by adding all items together for a total score. The inventory allows
for valid inferences and reliable scores (Burns, 1998); it is a part
of standard clinical practice, assesses a wide range of anxiety
symptoms, and is sensitive to change during treatment (Persons,
Roberts, & Zalecki, 2003). This measure has been found to have
internal consistency of .92 (Kring, Persons, & Thomas, 2007;
Stice, Presnell, & Spangler, 2002). Cronbachs alpha for the cur-
rent study, which included the sum scores, was .95.
Procedure
The instruments were posted online. Participants were recruited
through social networking sites, and e-mails were sent to univer-
sity and community lesbian, gay, bisexual, transgender centers
around the United States. After the survey was posted, participants
were given 2 months to participate. All procedures were reviewed
and approved by the Institutional Review Board at a large mid-
western university. A total of 603 participants logged in to partic-
ipate in the current study, with a completion rate at 74%. Because
there can be risk of participants completing the survey more than
one time, we checked participants data with IP addresses to
determine duplicates; this is recommended practice with data
collected online (Kraut et al., 2004). There were four instances of
repeated IP addressesin two instances, the participant had
stopped the survey after the demographic information, but had
finished at the second try, thus the first responses were deleted. In
the other two instances, demographic information differed among
all responses and thus were retained; many computers are shared
(e.g., roommates, libraries, families), and it is common practice to
retain responses where the demographic information differs (Gos-
ling, Vazire, Srivastava, & John, 2004). Only participants who
completed the entirety of the survey were included in the final data
set (N 446), which included the following identities: transgender
women, transgender men, genderqueer individuals (e.g., individu-
als who typically do not identify with the gender binary of male/
female), and crossdressers. Because hypotheses were specific to
comparing transgender women and men, genderqueer and cross-
dressing individuals were not included in the current study.
Analytic Strategy
All analyses were conducted using Mplus 7 (Muthn & Muthn,
19982011). First, we examined the hypothesized hybrid model
for all participants. In this model (see Figure 1), the F, FR, and SO
subscales of the MSPSS are latent constructs, each measured by
the four items on the subscale. These three subscales are indicators
of a higher order factor, Social Support. Transition Status, Loss,
and Social Support were hypothesized to predict two different
outcomes in two separate models: (a) Depression, controlling for
family history of depression and (b) Anxiety, controlling for fam-
ily history of anxiety. Family history was included as a control
variable because previous research has indicated a possible genetic
link between depression and anxiety (Hettema et al., 2001; Sulli-
van et al., 2000); we wanted to control for the possibility of genetic
factors contributing to mental health concerns. We also included
age and income as control variables (not included in the diagram
for visual clarity). Additionally, the relationships between transi-
tion status, loss, and social support and the outcome were predicted
to be mediated through both Facilitative Coping and Avoidant
Coping. Finally, based on prior theory (e.g., affect phobia;
McCullough et al., 2003), we specified that Avoidant Coping is a
predictor of Facilitative Coping. Adequate fit of the model was
indicated by a nonsignificant chi-square. In addition, the compar-
ative fit index (CFI), the RMSEA, and the SRMR were used. We
used Hu and Bentlers (1999) recommended guidelines for accept-
able fit of CFI .95, RMSEA .06, and SRMR .08. Addi-
tionally, we examined whether RMSEA was statistically signifi-
cantly greater than .05 and whether the upper bound of the
confidence value exceeded .10, which may indicate a serious
problem (Browne & Cudeck, 1993). To identify potential misfit in
the hypothesized measurement model, we examined modification
indices combined with theoretical rationale. To examine whether
being a transgender man or woman moderated the hypothesized
processes in the model, we estimated a multiple-group hybrid
model and assessed whether the estimated path coefficients varied
across the two groups using the chi-square difference test for more
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
548
BUDGE, ADELSON, AND HOWARD
restrictive models and using the Wald test to examine specific
parameters in the model.
Results
Preliminary Results
Prior to conducting SEM analyses, frequencies and independent
t tests were performed on the psychological distress measures to
determine the extent of distress, as well as differences between
groups. A total of 51.4% (n 116) of transgender women reported
symptoms of depression higher than or at the clinical cutoff of 16
points (M 20.08, SD 13.81); similarly, 48.3% (n 65) of
transgender men reported depression symptoms higher than the
cutoff (M 19.03, SD 13.44). For anxiety, 40.4% (n 91) of
transgender women reported significant anxiety symptomatology
(e.g., higher than or at the clinical cutoff of 21) results (M21.94,
SD 17.76), whereas 47.5% (n 64) of transgender men re-
ported significant anxiety symptomatology (M 24.91, SD
19.35). The t-test results between transgender men and women
were not statistically significant for depression, t(322) 0.66, p
.51, or anxiety, t(309) 1.38, p .17. Means, standard devi-
ations, and correlations among measures are presented in Table 1.
Primary Results: Depression
Measurement model. Using the two-step approach recom-
mended by Kline (2011), we first examined a measurement model
with no structural paths (i.e., the structural portion was just-
identified with all correlations) to examine model fit prior to
estimating the full structural model. The original hypothesized
model had the following model fit:
2
(139) 406.98, p .001;
RMSEA .074, CI
90
[.064, .083], p .001; CFI .940;
SRMR .049. With the exception of SRMR, all fit indices
indicated that the model fit was not in the a priori-specified
acceptable range. Thus, we examined the modification indices and
the wording of the items on the MSPSS to determine potential
respecifications. We added three residual correlations between
MSPSS items (one for each subscale): (a) MSPSS3 (My family
really tries to help me) with MSPSS11 (My family is willing to
help me make decisions), unlike the other two items on the F
subscale, these items were about the respondents family actively
trying or being willing to help; (b) MSPSS9 (I have friends with
whom I can share my joys and sorrows) and MSPSS12 (I can
talk about my problems with my friends), unlike the other two
items on the FR subscale, these items were about communicating
with friends; and (c) MSPSS5 (I have a special person who is a
real source of comfort to me) and MSPSS10 (There is a special
person in my life who cares about my feelings). Additionally, we
added a residual correlation between the FR subscale and the SO
subscale given that individuals have more choice over who their
friends and significant others are than they do over their family;
thus, individuals are likely to choose friends and significant others
who provide a certain level of support, whereas they cannot choose
their family based on the level of support. The measurement model
Figure 1. Hypothesized hybrid mediation model predicting depression and anxiety (separately). (Exogenous
variable correlations omitted for visual clarity.) MSPSS Multidimensional Scale of Social Support.
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
549
TRANSGENDER ANXIETY, DEPRESSION, AND COPING
with these four additional correlations had the following model fit:

2
(135) 275.89, p .001; RMSEA .055, CI
90
[.045, .064],
p .20; CFI .969; SRMR .044, indicating good model fit.
Multiple-group analysis. We estimated the full hybrid model
and conducted the multiple-group analyses, beginning with estab-
lishing measurement invariance or partial measurement invariance
between the two groups (transgender women and men) and then
examining whether there were differences in the structural paths.
The full hypothesized structural model (with the respecifications to
the measurement model) provided good model-data fit,
2
(142)
299.34, p .001; RMSEA .057, CI
90
[.048, .066], p .11;
CFI .964; SRMR .050. Next, we estimated a model in which
all parameters were unconstrained between the two groups. The
model fit indices were still within the acceptable ranges,
2
(284)
460.18, p .001; RMSEA .060, CI
90
[.050, .070], p .05;
CFI .960; SRMR .059.
We established measurement invariance with the exception of
one residual correlation. When we constrained the factor loadings
and all residual covariances to be equal between the two groups,
the model fit was slightly less than adequate,
2
(297) 479.89,
p .001; RMSEA .060, CI
90
[.050, .070], p .05; CFI .958;
SRMR .061. Unconstraining MSPSS3 and MSPSS11 from the
F subscale but constraining the other three residual correlations to
be equal resulted in good model fit indices,
2
(296) 464.34, p
.001; RMSEA .058, CI
90
[.047, .067], p .11; CFI .962;
SRMR .060, and also was preferred over the unconstrained
model,
2
(12) 4.16, p .98, and the model with all correlations
constrained,
2
(1) 15.55, p .001, thus indicating that the
measurement models were equivalent for transgender men and
women with the exception of the residual correlation between
MSPSS3 and MPSS11.
Before examining whether the specified structural paths were
moderated by being a transgender woman or man, we examined
whether the correlations among the exogenous variables, which
were in the model as an assumption, could be constrained between
the two groups. When we constrained all of those correlations
(among Transition Status, Loss, Social Support, Age, and Income),
the CFI was .958, and the chi-square difference test indicated that
the model with those correlations unconstrained was preferred,

2
(10) 28.03, p .002; however, with the correlations between
Transition Status and Social Support and among Social Support,
Loss, Age, and Income constrained to be equal and with the
correlations between Transition Status and Loss, Age, and Income
freely estimated in the two groups, the model was preferred over
the model with all correlations constrained,
2
(3) 22.05, p
.001, and over the model with none of these correlations con-
strained,
2
(7) 5.98, p .54. Although most of the exogenous
variable correlations were equivalent for transgender men and
women, there was a positive relationship between Transition Sta-
tus and Loss and between Age and Transition Status and a negative
relationship between Income and Transition Status for transgender
women, whereas there was no relationship between Transition
Status and Loss, Age, or Income for transgender men.
Our primary interest was in the structural paths in the model and
whether those were moderated by group (transgender men or
women). To do this, we first examined the structural paths in the
final model, with all loadings and most correlations constrained to
be equal. The structural paths from this model are provided in
Table 2. Next, we conducted Wald tests to examine whether each
structural path was equal between the two groups. As shown in
Table 2, the only structural paths that were statistically signifi-
cantly different between the two groups were Depression regressed
on Facilitative Coping and Depression regressed on Income. Thus,
we reran the model with all other structural paths constrained to be
equal. This final model had adequate model fit,
2
(320) 488.39,
p .001; RMSEA .055, CI
90
[.045, .065], p .19; CFI .962;
SRMR .064, and was preferred over the more parameterized
model without the structural paths constrained to be equal,

2
(17) 18.07, p .38, and over the more parsimonious
model with all structural paths constrained to be equal,
2
(2)
6.04, p .049. Therefore, we used this as our final model to
examine the hypothesized relationships.
Final model. The final structural model with unstandardized
path estimates is shown in Figure 2. As expected, family history of
depression has a positive relationship with ones current level of
depression, hence the need to control for this variable. The model
explains 55% and 59% of the variability in depression in trans-
gender women and men, respectively, and it explains 23% and
22% of the variability in Facilitative Coping and 32% and 31% of
the variability in Avoidant Coping in transgender men and women,
respectively (with all these R
2
values being statistically signifi-
cant). As shown in Table 3, Loss does not have a statistically
Table 1
Means, Standard Deviations, and Correlations Among Major Variables
Scale M SD 1 2 3 4 5 6 7 8 9 10 11
1. Family History of Depression 1.56 0.50
2. Family History of Anxiety 1.75 0.43 .43

3. Age 40.28 13.85 .14

.14

4. Income 3.72 2.11 .13

.11 .41

5. Transition Status 39.68 13.66 .94 .07 .03 .07 .15

6. CES-D 56.06 18.42 .95 .21

.15

.11

.20

.22

7. BAI 16.69 7.88 .80 .29

.27

.20

.24

.14

.81

8. Loss 44.84 9.43 .87 .11 .11 .14

.10 .26

.27

.23

9. Facilitative Coping 34.12 9.59 .87 .02 .07 .02 .02 .13

.18

.08 .11

10. Avoidant Coping 58.17 16.46 .91 .18

.16

.21

.25

.26

.67

.65

.19

.01
11. MSPSS 39.68 13.66 .94 .04 .02 .12

.11

.15

.41

.28

.28

.25

.30

Note. Income was assessed on an 8-point scale, with 3 $20,000$29,999 and 4 $30,000$39,999. CES-D Center for Epidemiologic Studies
Depression Scale; BAI Burns Anxiety Inventory; MSPSS Multidimensional Scale of Social Support.

p .05.

p .01.
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
550
BUDGE, ADELSON, AND HOWARD
significant direct, indirect, or total effect on Depression. Although
Transition Status does not have a statistically significant direct
effect on Depression, it does have a statistically significant nega-
tive indirect effect on it through Avoidant Coping. Finally, the
level of Social Support that one receives has statistically signifi-
cant direct, indirect, and total effects on Depression. The greater
the Social Support one receives, the lower their Depression. The
indirect effect between Social Support and Depression goes
through Avoidant Coping, with the indirect effect explaining 36%
of the total relationship between Social Support and Depression for
transgender women and 32% of that relationship for transgender
men (i.e., partial mediation). Thus, Avoidant Coping serves as a
mediator for the relationship between Transition Status and De-
pression and between Social Support and Depression. However,
Facilitative Coping does not serve as a mediator for the relation-
ship between any of the predictors and Depression.
In the examination of the moderating effect of being a trans-
gender man or woman, six parameters were found to be different.
In this final model, the correlation between MSPSS3 and MPSS11
on the F subscale was only .108 (p .26) for transgender women,
but it was .539 (p .001) for transgender men, suggesting a
stronger residual correlation between these two indicators of Fam-
ily Support. The correlation between Transition Status and Loss
was statistically significant for transgender women (r .334, p
.001), suggesting that the further along they are in their transition
status, the greater their feelings of loss; however, that relationship
was not statistically significant for transgender men (r .091, p
.28). Additionally, although age and income were not related to
Transition Status for transgender men (r .003, p .97; r .05,
p .61; respectively), they were related to Transition Status for
transgender women (r .19, p .002; r .19, p .002;
respectively), meaning that for transgender women, there is a small
relationship indicating that older individuals and individuals with
less income are further in their transition status. However, Income
predicted Depression in transgender men ( 0.15, p .049),
indicating that transgender men with greater income are slightly
less depressed, but not in transgender women ( 0.03, p .55).
Finally, although the effect of Facilitative Coping on Depression
was found to be different using the Wald test, the structural path
for both groups was not statistically significant and did not result
in different conclusions concerning direct, indirect, and total ef-
fects in the model.
Primary Results: Anxiety
Measurement model. Similar to the Depression analyses, we
used the two-step approach recommended by Kline (2011) and
first examined a measurement model with no structural paths (i.e.,
the structural portion was just-identified with all correlations) to
examine model fit prior to estimating the full structural model. The
original hypothesized model had the following model fit:

2
(139) 417.91, p .001; RMSEA .076, CI
90
[.067, .084],
p .001; CFI .937; SRMR .050. With the exception of
SRMR, all fit indices indicated that the model fit was not in the a
priori-specified acceptable range. Thus, we made the same respeci-
fications for this model as we did with the Depression measure-
ment model. The measurement model with the four additional
correlations (MSPSS3 with MSPSS11, MSPSS9 with MSPSS12,
MSPSS5 with MSPSS10, and Friends with Significant Other) had
the following model fit:
2
(135) 284.24, p .001; RMSEA
.056, CI
90
[.047, .065], p .13; CFI .966; SRMR .044,
indicating good model fit.
Multiple-group analysis. Having established the measure-
ment model for the Anxiety model, we then estimated the full
Table 2
Structural Paths (Unstandardized) From Depression and Anxiety Models With Measurement Constraints and Test of Equality of
Those Paths Between Groups
Depression Anxiety
DV IV
Trans women
estimate (SE)
Trans men
estimate (SE) Wald test p
Trans women
estimate (SE)
Trans men
estimate (SE) Wald test p
Facilitative Coping Transition Status 0.33 (1.12) 1.10 (1.54) 0.60 .44 0.18 (1.22) 0.95 (1.52) 0.35 .55
Loss 0.39 (0.13)

0.53 (0.23)

0.32 .57 0.41 (0.14)

0.49 (0.22)

0.09 .77
Social Support 5.49 (1.70)

4.31 (2.19)

0.19 .66 5.86 (1.96)

4.14 (2.17) 0.36 .55

Avoidant Coping 0.27 (0.10)

0.06 (0.14) 3.79 .05 0.31 (0.11)

0.06 (0.14) 4.45 .04

Age 0.06 (0.06) 0.06 (0.09) 1.22 .27 0.09 (0.07) 0.06 (0.10) 1.71 .19
Income 0.21 (0.49) 0.67 (0.96) 0.19 .66 0.38 (0.52) 0.92 (1.02) 0.24 .62
Avoidant Coping Transition Status 2.99 (0.98)

1.82 (1.42) 0.40 .53 2.54 (1.06)

1.75 (1.41) 0.21 .65

Loss 0.11 (0.11) 0.16 (0.21) 0.04 .85 0.06 (0.12) 0.18 (0.21) 0.24 .62
Social Support 3.19 (1.15)

2.69 (1.64) 0.06 .80 3.62 (1.27)

2.78 (1.64) 0.17 .68

Age 0.10 (0.05)

0.23 (0.08)

1.90 .17 0.14 (0.05)

0.24 (0.08)

1.18 .28
Income 0.89 (0.42)

0.78 (0.89) 0.013 .91 0.46 (0.45) 0.45 (0.94) 0.00 .99
Distress Facilitative Coping 0.01 (0.11) 0.39 (0.14)

5.01 .03 0.11 (0.16) 0.23 (0.19) 1.79 .18

Avoidant Coping 0.72 (0.11)

0.57 (0.13)

0.78 .38 0.90 (0.17)

1.20 (0.17)

1.48 .22
Transition Status 0.25 (1.10) 0.82 (1.61) 0.09 .77 0.57 (1.57) 1.61 (2.13) 0.15 .69
Loss 0.03 (0.15) 0.35 (0.25) 1.82 .18 0.18 (0.21) 0.19 (0.33) 0.00 .98
Social Support 5.19 (2.07)

0.55 (2.01) 2.63 .11 3.29 (2.77) 0.52 (2.73) 0.51 .48
Family History 2.59 (1.33) 4.02 (1.86)

0.39 .53 9.22 (2.38)

3.75 (2.70) 2.32 .13

Age 0.07 (0.06) 0.04 (0.10) 0.05 .83 0.06 (0.09) 0.10 (0.14) 0.06 .81
Income 0.51 (0.49) 2.91 (1.11)

8.01 .005 0.30 (0.75) 3.57 (1.45)

5.63 .02
Note. DV dependent variable; IV independent variable; trans transgender.

p .05.

p .01.

p .001.
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
551
TRANSGENDER ANXIETY, DEPRESSION, AND COPING
hybrid model and conducted the multiple-group analyses. First, we
established measurement invariance or partial measurement invari-
ance between the two groups (transgender men and women), and
then we examined whether there were differences in the structural
paths. The model fit statistics for the full hypothesized structural
model (with the respecifications to the measurement model)
indicated good model-data fit,
2
(142) 296.24, p .001;
RMSEA .056, CI
90
[.047, .065], p .12; CFI .964;
SRMR .048. Additionally, a model in which all parameters
were unconstrained between the two groups exhibited good model
fit,
2
(284) 449.43, p .001; RMSEA .058, CI
90
[.048,
.068], p .09; CFI .962; SRMR .056.
Similar to the Depression model, we established measurement
invariance but allowed one residual correlation to vary between the
two groups. When we constrained the factor loadings and all
residual covariances to be equal between the two groups, the
model fit was adequate,
2
(297) 469.69, p .001; RMSEA
.058, CI
90
[.048, .068], p .09; CFI .960; SRMR .057.
However, given the finding with the Depression outcome regard-
ing the residual correlation between MSPSS3 and MSPSS11 from
the F subscale, we tested whether allowing it to be freely estimated
but constraining the other three correlations to be equal resulted in
better model. Indeed, the partial measurement invariance model
had strong model fit indices,
2
(296) 453.56, p .001;
RMSEA .056, CI
90
[.045, .066], p .18; CFI .964;
SRMR .057; was preferred over the model with that correlation
constrained,
2
(1) 16.13, p .001; and was preferred over the
unconstrained model as well,
2
(12) 4.16, p .98, thus indi-
cating that the models were equivalent for transgender women and
men with the exception of the residual correlation between
MSPSS3 and MPSS11.
Similar to the Depression model, we could not constrain all
three correlations among the exogenous variables to be equal
across the two groups, with the CFI being .959, and the chi-square
difference test indicating that the model with those correlations
unconstrained was preferred,
2
(10) 29.95, p .001; however,
a model that constrained correlations between Transition Status
and Social Support and among Social Support, Loss, Age, and
Income and that allowed correlations between Transition Status
and Loss, Age, and Income to be freely estimated in the two
groups was preferred over the model with none of these correla-
tions constrained,
2
(7) 5.19, p .64, and over the model with
all of these correlations constrained,
2
(3) 24.76, p .001. This
indicates that the exogenous variable correlations were equivalent
for transgender men and women with the exception of the corre-
lation between Transition Status and Loss, Age, and Income, in the
same direction as was found in the Depression analyses.
We were primarily interested in the structural paths in the model
and whether those paths were moderated by group (transgender
women and men). We first examined the structural paths in the
Figure 2. Unstandardized path coefficients (SE) for structural causal paths in final constrained Depression and
Anxiety models. Several parameters were not constrained to be equal between groups. For those parameters, the
first estimate (SE) is for transgender women, and the second is for transgender men. (Exogenous variable
correlations omitted for visual clarity.)

p .05.

p .01.

p .001.
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
552
BUDGE, ADELSON, AND HOWARD
final model, with the constraints described above applied to the
model (see Table 2). Similar to in the Depression model, there was
a statistically significant difference between the two groups for
Anxiety when regressed on Income. However, unlike in the De-
pression model, there was not a statistically significant difference
between the two groups for Anxiety when regressed on Facilitative
Coping. Although in the Depression model the difference between
the two groups for Facilitative Coping on Avoidant Coping was
just above our alpha level, in the Anxiety model, there was a
statistically significant difference in the effect of Avoidant Coping
on Facilitative Coping. As a result, we reran the model with all
structural paths constrained to be equal besides Anxiety regressed
on Income and Facilitative Coping regressed on Avoidant Coping.
This final model had adequate model fit,
2
(320) 472.41, p
.001; RMSEA .053, CI
90
[.042, .063], p .32; CFI .965;
SRMR .062, and was preferred over the more parameterized
model without the structural paths constrained to be equal,

2
(17) 13.66, p .69, and over the more parsimonious model
with all structural paths constrained to be equal,
2
(2) 8.20, p
.02. Therefore, we use this as our final model to examine the
hypothesized relationships.
Final model. The final structural model with unstandardized
path estimates is shown in Figure 2. As expected, family history of
anxiety has a positive relationship with ones current level of
anxiety, hence the need to control for this variable. The model
explains 47% of the variability in anxiety in transgender women
and 53% of the variability in anxiety in transgender men. Addi-
tionally, it explains 23% and 26% of the variability in Facilitative
Coping and 33% and 34% of the variability in Avoidant Coping in
transgender women and men, respectively (with all these R
2
values
being statistically significant). As shown in Table 3, Loss does not
have a statistically significant direct, indirect, or total effect on
Anxiety, just as it did not have any effects on Depression. Al-
though Transition Status does not have a statistically significant
direct effect on Anxiety, it does have a statistically significant
negative indirect effect through Avoidant Coping. Finally, the
level of Social Support that one receives has statistically signifi-
cant indirect and total effects on Anxiety, but once the indirect
effect is accounted for, there is not a statistically significant direct
effect of Social Support on Anxiety. The greater the Social Support
one receives, the lower ones Anxiety. The indirect effect between
Social Support and Anxiety goes through Avoidant Coping. Al-
though the direct effect between Social Support and Anxiety is not
statistically significant, the indirect effect only explains 67% of the
total effect between the two constructs. Thus, Avoidant Coping
serves as a mediator for the relationship between Transition Status
and Anxiety and between Social Support and Anxiety, but Facil-
itative Coping does not serve as a mediator for the relationship
between any of the predictors and Anxiety.
In the examination of the moderating effect of being a trans-
gender man or woman, six parameters were different in the Anx-
iety model. The correlations that were different between the two
groups were the same as those found in the Depression model:
correlation between MSPSS3 and MPSS11 .108 (p .26) for
transgender women, .544 (p .001) for transgender men; corre-
lation between Transition Status and Loss .334 (p .001) for
transgender women, .076 (p .37) for transgender men; correla-
tion between Transition Status and Age .208 (p .001) for
transgender women, .007 (p .94) for transgender men; and
correlation between Transition Status and Income .201 (p
.001) for transgender women, .041 (p .66) for transgender men.
Again, this suggests a stronger residual correlation between these
Table 3
Total, Indirect, and Direct Unstandardized Effects for Final Depression and Anxiety Models (With Measurement and
Structural Constraints)
Effect
Depression trans
women estimate (SE)
Depression trans
men estimate (SE)
Anxiety trans
women estimate (SE)
Anxiety trans
men estimate (SE)
Between Loss and Distress
Total 0.09 (0.15) 0.01 (0.16) 0.27 (0.19) 0.27 (0.20)
Indirect 0.04 (0.08) 0.04 (0.09) 0.05 (0.12) 0.04 (0.12)
Through Facilitative Coping 0.02 (0.04) 0.09 (0.06) 0.02 (0.05) 0.02 (0.05)
Through Avoidant Coping 0.06 (0.07) 0.06 (0.07) 0.07 (0.11) 0.07 (0.11)
Through Avoidant Coping & Facilitative Coping 0.001 (0.001) 0.003 (0.004) 0.001 (0.002) 0.000 (0.001)
Direct 0.05 (0.13) 0.05 (0.13) 0.22 (0.17) 0.22 (0.17)
Between Transition Status and Distress
Total 2.25 (1.19) 2.15 (1.24) 2.95 (1.63) 2.97 (1.65)
Indirect 1.62 (0.63)

1.52 (0.69)

2.25 (0.97)

2.27 (0.99)

Through Facilitative Coping 0.01 (0.04) 0.04 (0.20) 0.001 (0.05) 0.001 (0.05)
Through Avoidant Coping 1.64 (0.63)

1.64 (0.63)

2.28 (0.98)

2.28 (0.98)

Through Avoidant Coping & Facilitative Coping 0.02 (0.04) 0.08 (0.06) 0.03 (0.07) 0.008 (0.02)
Direct 0.64 (0.91) 0.64 (0.91) 0.70 (1.24) .70 (0.57)
Between Social Support and Distress
Total 6.23 (1.56)

6.99 (1.71)

5.65 (1.96)

5.68 (1.93)

Indirect 2.43 (0.73)

3.19 (0.83)

4.01(1.26)

4.04 (1.30)

Through Facilitative Coping 0.22 (0.45) 1.08 (0.60) 0.25 (0.60) 0.25 (0.60)
Through Avoidant Coping 2.23 (0.60)

2.23 (0.60)

3.81 (1.11)

3.81 (1.11)

Through Avoidant Coping & Facilitative Coping 0.02 (0.05) 0.11 (0.10) 0.04 (0.11) 0.01 (0.04)
Direct 3.80 (1.55)

3.80 (1.55)

1.64 (1.98) 1.64 (1.98)

Note. trans transgender.

p .05.

p .01.

p .001.
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
553
TRANSGENDER ANXIETY, DEPRESSION, AND COPING
two indicators of Family support for transgender men that the further
along transgender women are in their transition status, the greater
their feelings of loss, the older they are, and the less income they
have (without the same relationships holding for transgender men).
Additionally, Income predicted Anxiety in transgender men
( 0.18, p .008), indicating that transgender men with
greater income are slightly less anxious, but not in transgender
women ( 0.02, p .75). Finally, although Avoidant Coping
did not predict Facilitative Coping for transgender men ( 0.07,
p .57), it did for transgender women ( 0.24, p .01).
However, that did not result in different conclusions concerning
the direct, indirect, and total effects in the model.
Discussion
Our primary goal in this study was to examine depression and
anxiety in transgender male and female populationswith empha-
sis given to understanding the relationships among loss, social
support, transition status, and coping that contribute to, or allevi-
ate, distress. Prior to testing our main hypotheses, our results
indicate that 51% of transgender women and 48% of transgender
men report significant depressive symptomatology. These rates are
in line with other studies that have examined depression rates for
transgender men and women (Clements-Nolle et al., 2001; Nemoto
et al., 2011; Nuttbrock et al., 2010). However, the rates of anxiety
for this study appear to be greater than those that have been
previously reported in the literature; 47% of transgender men and
40% of transgender women report significant anxiety symptom-
atology, both of which are higher than any of the upper limits of
26%38% reported within the literature (Hepp et al., 2005; Mus-
tanski et al., 2010).
Both the anxiety and depression models indicate similar results.
This finding is not surprising on the surface, as there is a large
amount of comorbidity between depression and anxiety (Mathew,
Pettit, Lewinsohn, Seeley, & Roberts, 2011). However, according
to minority stress theory, individuals who have invisible minority
identities may experience more anxiety due to hypervigilance and
experiences of discrimination (Meyer, 2003). Instead, it appears
that both anxiety and depression are indicative of distress for
transgender men and women and that avoidant coping served as a
mediator between transition status and depression/anxiety. Tran-
sition status was negatively related to avoidant coping, indicating
that the further along individuals were in their identity process, the
less avoidant coping they used, but that the more avoidant coping
individuals used, the more depression and anxiety they reported.
Thus, it appears that individuals who are in the beginning stages of
their transition process use more avoidant coping, and thus expe-
rience more distress. Previous studies have shown that more dis-
tress may be exhibited in the beginning stages of the gender
transition process (Budge et al., 2012). Studies have also shown
that using more cognitive avoidant coping strategies increases
symptoms of depression and anxiety for nontransgender popula-
tions (Blalock & Joiner, 2000; Holahan, Moos, Holahan, Brennan,
& Schutte, 2005). However, previous studies have not looked at
any of these variables in the context of how identity, coping, and
distress influence and relate to one another.
The multigroup model results also indicate that social support is
directly and indirectly related to depression and anxiety, with
avoidant coping as a mediating variable. There is a large body of
research that indicates that more social support is related to lower
levels of depression and anxiety (Wareham, Fowler, & Pike, 2007;
Wester, Christianson, Vogel, & Wei, 2007; Wright & Perry, 2006);
this finding was also present in the current study. Social support is
less commonly researched as being related to distress measures
with coping as a mediating variable. However, Valentiner, Hola-
han, and Moos (1994) indicate that coping should be conceptual-
ized as a mediating variable, especially when personal and social
resources are directly predicting mental health outcomes. The
findings in this study indicate that the less social support a trans-
gender individual has available, the more avoidant coping he or
she will use, and in turn, more depressive and anxiety symptoms
will be reported.
It should be noted that facilitative coping was not seen as a
mediator in either of the models in which social support and
transition status indirectly predicted depression and anxiety. This
null finding is of particular importance, given that several of the
main interventions for depression and anxiety incorporate facili-
tative coping mechanisms into treatment planning (e.g., Coping
With Depression course; Lewinsohn, Clarke, Rohde, Hops, &
Seeley, 1996). Instead, it appears that it is the absence of avoidant
coping that may result in less psychological distress. This finding
is logical when considering how avoidance causes distress. Re-
search has shown that the avoidance of emotional experiences, via
suppression, may not have an impact on subjective experiences but
leads to increases in sympathetic arousal (Gross, 1998; Gross &
Levenson, 1997). Conceptualized as ironic process (Wegner &
Zanakos, 1994), the more an individual attempts to avoid a situ-
ation, the more they will think about what they have been attempt-
ing to repress. This process appears to be used often with trans-
gender individualsBudge et al. (2012) found that denial and
suppression of transgender identity were central to the beginning
stages of transgender identity and emotional avoidance. It is pos-
sible that interventions that have been focusing on increasing
facilitative coping may be able to reconceptualize treatment plan-
ning to intervene on avoidance strategies used for coping.
An additional finding from this study was that gender identity
moderated several variables within the models. In both models,
transgender men endorsed using more family help with making
decisions and providing support than transgender women. There is
limited research regarding transgender men and family members,
thus there is not current scholarly evidence that is consistent with
this particular finding. It is possible that transgender men were
socialized to be more involved with their families when they were
being raised female, as womens roles in families are more pre-
scribed than mens (Knudsen & Wrness, 2001), thus, transgender
men may gain some advantages of support from this process.
Transitioning to being male has been perceived as more accept-
able, due to the power status of men in society (Schilt, 2010),
which may be easier for family members to provide support to
transgender men. An additional variable that was moderated by
gender identity was loss as it relates to transition status. As
transgender women proceeded through their transition process,
they reported experiencing more loss than transgender men. Trans-
gender women may experience more distressing emotions, dis-
crimination, and relationship concerns related to not being able to
pass (e.g., be perceived as women; Gagn & Tewksbury, 1998).
Not being able to pass has been linked to specific losses, such as
job loss (Budge, Tebbe, & Howard, 2010); at the same time,
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
554
BUDGE, ADELSON, AND HOWARD
transgender individuals have reported that the further along they
are in their identity process, the more hopes for the future they
exhibitsuch as being read as female (Budge et al., 2012). There-
fore, it is possible that the further along transgender women are
within their transition process, the more loss they will also report
due to their hopes about their appearance not coming to fruition.
Additional findings that were gender specific were related to age
and income. Results from this study indicate that when transgender
women are further along in their transition, they tend to be older
and have less income. Grant et al. (2011) report that 15% of
transgender women report yearly household income of $10,000 or
less and that 16% of transgender men report the same level of
income; it is possible that the findings from this study indicate that
age and transition status are important factors to consider when
providing treatment to transgender women when compared with
transgender men. Also related to income, transgender men who
reported having higher incomes also reported less depressive
symptomology. It appears that income is a protective factor for
transgender men, but not transgender women; higher incomes and
income disparities are highly related to traditional masculinity
(Coughlin & Wade, 2012); thus, it may be that transgender men
experience less depression when they feel they are fulfilling gen-
der roles. The final finding from this study was that avoidant
coping predicted facilitative coping as they relate to anxiety for
transgender women, but not transgender men. As noted in the
literature review, women seek social support at higher rates than
men (Dwyer & Cummings, 2001); thus, it could be that transgen-
der women are also able to access social support after they have
first used avoidance mechanisms, because gender socialization
deems this an acceptable trajectory of coping for women (Rosario,
Shinn, Mrch, & Huckabee, 1988).
This study is the first of its kind to examine transgender distress
and transitioning processes through the types of coping mecha-
nisms used. Nonetheless, as with all studies, several limitations can
be identified. First, the sample was fairly homogenous regarding
racial identification. The experiences and results represented in
this study are more likely representative of White transgender
populations and may not encompass the transitioning processes
and experiences of distress for transgender individuals of color. It
would be beneficial to conduct a study regarding distress and
well-being for individuals of color to understand the complex
dynamics of minority stress for both race and transgender identity.
Furthermore, recruitment and access issues can be considered
limitations of this study. It may be considered a limitation to
conduct research online, as there are many individuals who do not
have access to the Internet or have the knowledge to navigate
online survey systems. Also, because this population is marginal-
ized and transgender identities can remain invisible, the sample
likely represents individuals comfortable enough with their iden-
tities to respond to a survey about identity, instead of individuals
at the beginning of their transition process who were not yet ready
to answer questions regarding their transition. As with all online
research, it is also difficult to verify the identity of participants; we
checked all IP addresses to ensure that participants only took the
survey once. However, as with most self-report data, it is impos-
sible to verify whether participants are being truthful about their
transgender identity. In order to address issues related to comfort
with identity, future research should focus on conducting studies of
individuals who are in the beginning stages in order to understand
more about the process of what it is like for them at the formative
stages of their transgender identity.
In addition, genderqueer and crossdressing individuals were
excluded from the analyses for this study. Crossdressers internal
gender identity typically matches their sex assigned at birth, but
their gender expression is typically indicative of a different gender;
genderqueer individuals most often do not identify as male or
female, but somewhere in-between, or they reject the notion of
gender completely. Because of these identity factors, comparing
these participants with transgender men and women does not fit
the scientific rationale for the current study. However, genderqueer
and crossdressing individuals are often ignored within scientific
literature. Research regarding coping mechanisms and mental
health outcomes for all transgender populations are essential in
order to understand the spectrum of gender identities and their
multifaceted psychological processes.
There were limitations related to the measurement of social
support. The MSPSS scale assesses for the social support of
transgender individuals families of origin, as well as friends and
significant others. We were primarily interested in the presence or
lack of social support from different types of relationships; how-
ever, this does not capture whether or not transgender individuals
are gaining social support from families of choice. Future research
should assess both family-of-origin support and family-of-choice
support.
The results from this study can meaningfully impact the inter-
ventions that are used with transgender individuals. First, mental
health professionals should be aware that the identity process of
the client, regardless of transgender male or female identity, is
related to the amount of avoidant coping used and the level of
distress they experience. This is particularly important for inter-
ventions, given that identity processes are often not at the forefront
of many types of interventions. In addition, preventative interven-
tions may be particularly useful in order to help reduce the amount
of loss that is experienced by transgender women. Determining
areas where clients are worried about loss (e.g., potential loss of
employment in the future) may lead to interventions such as
writing a letter to a human resources department (see M. L. Brown
& Rounsley, 1996). Because social support appears to be an
essential component of mental health, it is important that interven-
tions match the clients entire system, and does not exclusively
focus on the individual. Depending on the age and family makeup
of the client, family/couple therapy can be particularly useful. For
example, family therapy with clients who are younger and are
possibly dealing with individuation from their family will look
different from family therapy with clients who are older and have
a partner and children. Raj (2002) notes that, from a theoretical
standpoint, a client-centered approach to working with transgender
clients is the most applicable. He states that psychotherapy must
provide an affirmative stance related to the clients identity and
that this is essential to promoting a sense of support and, ulti-
mately, better mental health. Affirmative stances include using
correct language related to the clients identity, normalizing trans-
gender identity, and promoting self-empowerment (Raj, 2002).
Beyond interventions, it is important to raise awareness that the
rates of distress within the sample of the current study are stag-
geringly highwhich indicates that there are factors that are
contributing above and beyond what would be experienced within
the general population. This does not indicate that transgender
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
555
TRANSGENDER ANXIETY, DEPRESSION, AND COPING
individuals should be more pathologizedinstead, it supports the
hypothesis that marginalized populations will experience distress
at higher rates. This is indicative that societal changes must occur
in order to ease distress experienced by transgender populations.
References
Blalock, J. A., & Joiner, T. E. (2000). Interaction of cognitive avoidance
coping and stress in predicting depression/anxiety. Cognitive Therapy
and Research, 24, 4765.
Bockting, W., & Coleman, E. (2007). Developmental stages of the trans-
gender coming out process: Toward an integrated identity. In R. Ettner,
S. Monstrey, & E. Eyler (Eds.), Principles of transgender medicine and
surgery (pp. 185208). New York, NY: The Haworth Press.
Boisvert, J. A., McCreary, D. R., Wright, K. D., & Asmundson, G. J. G.
(2003). Factorial validity of the Center for Epidemiologic Studies-
Depression (CES-D) scale in military peacekeepers. Depression and
Anxiety, 17, 1925.
Bowlby, J. (1969). Attachment and loss, Vol. 1: Attachment. New York,
NY: Basic Books.
Brown, M. L., & Rounsley, C. A. (1996). True selves: Understanding
transsexualism for families, friends, coworkers, and helping profession-
als. San Francisco, CA: Jossey-Bass.
Brown, R. (2000). Social Identity Theory: Past achievements, current
problems, and future challenges. European Journal of Social Psychol-
ogy, 30, 745778.
Browne, M. Y. C., & Cudeck, R. R. (1993). Alternative ways of assessing
model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural
equation models (pp. 136162). Newberry Park, CA: Sage Publicatinos.
Budge, S. L., Katz-Wise, S. L., Tebbe, E. N., Howard, K. A. S., Schneider,
C. L., & Rodriguez, A. (2012). Transgender emotional and coping
processes: Facilitative and avoidant coping throughout gender transition-
ing. The Counseling Psychologist. Advance online publication. doi:
10.1177/0011000011432753
Budge, S. L., Tebbe, E. N., & Howard, K. A. S. (2010). The work
experiences of transgender individuals: Negotiating the transition and
career decision-making processes. Journal of Counseling Psychology,
57, 377393. doi:10.1037/a0020472
Burns, D. D. (1998). Therapist toolkit. Unpublished manuscript, Los Altos,
CA.
Clements-Nolle, C., Marx, R., Guzman, R., & Katz, M. (2001). HIV
prevalence, risk behaviors, health care use, and mental health status of
transgender persons: Implications for public health intervention. Amer-
ican Journal of Public Health, 91, 915921.
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering
hypothesis. Psychological Bulletin, 98, 310357.
Coughlin, P., & Wade, J. C. (2012). Masculinity ideology, income dispar-
ity, and romantic relationship quality among men with higher earning
partners. Sex Roles, 67, 311322.
Devor, A. H. (2004). Witnessing and mirroring: A fourteen stage model of
transsexual identity formation. Journal of Gay & Lesbian Psychother-
apy, 8, 4167.
Dwyer, A. L., & Cummings, A. L. (2001). Stress, self-efficacy, social
support and coping strategies in university students. Canadian Journal
of Counseling, 35, 208220.
Factor, R. J., & Rothblum, E. D. (2007). A study of transgender adults and
their non-transgender siblings on demographic characteristics, social
support, and experiences of violence. Journal of LGBT Health Research,
3, 1130.
Folkman, S., Lazarus, R. S., Dunkel-Schetter, C., DeLongis, A., & Gruen,
R. J. (1986). Dynamics of a stressful encounter: Cognitive appraisal,
coping, and encounter outcomes. Journal of Personality and Social
Psychology, 50, 9921003.
Gagn, P., & Tewksbury, R. (1998). Conformity pressures and gender
resistance among transgendered individuals. Social Problems, 45, 81
101.
Gluhoski, V. L., & Wortman, C. B. (1996). The impact of trauma on world
views. Journal of Social and Clinical Psychology, 15, 417429. doi:
10.1521/jscp.1996.15.4.417
Golub, S. A., Walker, J. J., Longmire-Avital, B., Bimbi, D. S., & Parsons,
J. T. (2010). The role of religiosity, social support, and stress-related
growth in protecting against HIV risk among transgender women. Jour-
nal of Health Psychology, 15, 11351144.
Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we
trust Web-based studies? A comparative analysis of six preconceptions
about Internet questionnaires. American Psychologist, 59, 93104.
Grant, J. M., Mottet, L. A., Tanis, J., Harrison, J., Herman, J. L., &
Keisling, M. (2011). Injustice at every turn: A report of the National
Transgender Discrimination Survey. Washington, DC: National Center
for Transgender Equality and National Gay and Lesbian Task Force.
Gross, J. J. (1998). Sharpening the focus: Emotion regulation, arousal, and
social competence. Psychological Inquiry, 9, 287290.
Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: The acute effects
of inhibiting negative and positive emotion. Journal of Abnormal Psy-
chology, 106, 95103.
Grossman, A. H., Daugelli, A. R., & Frank, J. A. (2011). Aspects of
psychological resilience among transgender youth. Journal of LGBT
Youth, 8, 103115.
Harvey, J. H., & Miller, E. D. (1998). Toward a psychology of loss.
Psychological Science, 9, 429434.
Hepp, U., Kraemer, B., Schnyder, U., Miller, N., & Delsignore, A. (2005).
Psychiatric comorbidity in gender identity disorder. Journal of Psycho-
somatic Research, 58, 259261.
Hettema, J. M., Neale, M. C., & Kendler, K. S. (2001). A review and
meta-analysis of the genetic epidemiology of anxiety disorders. Ameri-
can Journal of Psychiatry, 158, 15681578.
Holahan, C. J., Moos, R. S., Holahan, C. K., Brennan, P. L., & Schutte,
K. K. (2005). Stress generation, avoidance coping, and depressive symp-
toms: A 10-year model. Journal of Consulting and Clinical Psychology,
73, 658666.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance
structure analysis: Conventional criteria versus new alternatives. Struc-
tural Equation Modeling, 6, 155.
Katz, J., Petracca, M., & Rabinowitz, J. (2009). A retrospective study of
daughters emotional role reversal with parents, attachment anxiety,
excessive reassurance-seeking, and depressive symptoms. The American
Journal of Family Therapy, 37, 185195.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., & Walters, E. E. (2005).
Lifetime prevalence and age of-onset distributions of DSM-IV disorders
in the national comorbidity survey replication. Archives of General
Psychiatry, 62, 593602.
Kline, R. B. (2011). Principles and practice of structural equation mod-
eling (3rd ed.). New York, NY: Guilford Press.
Knudsen, K., & Wrness, K. (2001). National context, individual charac-
teristics and attitudes on mothers employment: A comparative analysis
of Great Britain, Sweden and Norway. Acta Sociologica, 44, 6779.
Kraut, R., Olson, J., Banaji, M. R., Bruckman, A., Cohen, J., & Couper, M.
(2004). Psychological research online: Opportunities and challenges.
American Psychologist, 59, 105117.
Kring, A. M., Persons, J. B., & Thomas, C. (2007). Changes in affect
during treatment for depression and anxiety. Behaviour Research and
Therapy, 45, 17531764.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New
York, NY: Springer.
Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995).
Self-esteem as an interpersonal monitor: The sociometer hypothesis.
Journal of Personality and Social Psychology, 68, 518530.
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
556
BUDGE, ADELSON, AND HOWARD
Lewinsohn, P. M., Clarke, G. N., Rohde, P., Hops, H., & Seeley, J. R.
(1996). A course in coping: A cognitive-behavioral approach to the
treatment of adolescent depression. In E. D. Hibbs & P. S. Jensen (Eds.),
Psychological treatments for child and adolescent disorders: Empiri-
cally based strategies for clinical practice (pp. 109135). Washington,
DC: American Psychological Association.
Mathew, A. R., Pettit, J. W., Lewinsohn, P. M., Seeley, J. R., & Roberts,
R. E. (2011). Co-morbidity between major depressive disorder and
anxiety disorders: Shared etiology or direct causation? Psychological
Medicine, 41, 20232034.
McCullough, L., Kuhn, N., Andrews, S., Kaplan, A., Wolf, J., & Hurley,
C. L. (2003). Treating affect phobia: A manual for short-term dynamic
psychotherapy. New York, NY: Guilford Press.
Meyer, I. H. (2003). Prejudice, social stress, and mental health in lesbian,
gay, and bisexual populations: Conceptual issues and research evidence.
Psychological Bulletin, 129, 674697.
Mustanski, B. S., Garofalo, R., & Emerson, E. M. (2010). Mental health
disorders, psychological distress, and suicidality in a diverse sample of
lesbian, gay, bisexual, and transgender youths. Research and Practice,
100, 24262432.
Muthn, L. K., & Muthn, B. O. (19982011). Mplus users guide (6th
ed.). Los Angeles, CA: Author.
Nemoto, T., Bodeker, B., & Iwamoto, M. (2011). Social support, exposure
to violence, and transphobia: Correlates of depression among male-to-
female transgender women with a history of sex work. American Journal
of Public Health, 101, 19801988.
Nuttbrock, L., Hwahng, S., Bockting, W., Rosenblum, A., Mason, M.,
Macri, M., & Becker, J. (2010). Psychiatric impact of gender-related
abuse across the life course of male-to-female transgender persons.
Journal of Sex Research, 47, 1223.
Nuttbrock, L., Rosenblum, A., & Blumenstein, R. (2002). Transgender
identity affirmation and mental health. International Journal of Trans-
genderism, 4. Retrieved from http://www.wpath.org/journal/www.iiav
.nl/ezines/web/IJT/97-03/numbers/symposion/ijtvo06no04_03.htm
Park, C. L. (2008). Testing the meaning making model of coping with loss.
Journal of Social and Clinical Psychology, 27, 970994.
Persons, J. B., Roberts, N. A., & Zalecki, C. (2003). Anxiety and depres-
sion change together during treatment. Behavior Therapy, 34, 149163.
Pinto, R. M., Melendez, R., & Spector, A. (2008). Male-to-female trans-
gender individuals: Building social support and capital from within a
gender-focused network. Journal of Gay and Lesbian Social Services,
20, 203220.
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for
research in the general population. Applied Psychological Measurement,
1, 385401.
Raj, R. (2002). Toward a transpositive therapeutic model: Developing
clinical sensitivity and cultural competence in the effective support of
transsexual and transgendered clients. International Journal of Trans-
genderism, 6. Retrieved from http://www.wpath.org/journal/www.iiav
.nl/ezines/web/IJT/97-03/numbers/symposion/ijtvo06no02_04.htm
Rosario, M., Shinn, M., Morch, H., & Huckabee, C. B. (1988). Gender
differences in coping and social supports: Testing socialization and role
constraint theories. Journal of Community Psychology, 16, 5569. doi:
10.1002/1520-6629
Sanchez, F. J., & Vilain, E. (2009). Collective self-esteem as a coping
resource for male-to-female transsexual individuals. Journal of Coun-
seling Psychology, 56, 202209.
Schilt, K. (2010). Just one of the guys: Transgender men and the persis-
tence of gender inequality. Chicago, IL: University of Chicago Press.
Schleicher, H. E., Harris, K. J., Catley, D., & Nazir, N. (2009). The role of
depression and negative affect regulation expectancies in tobacco smok-
ing among college students. Journal of American College Health, 57,
507512.
Schwartzberg, S. S., & Janoff-Bulman, R. (1991). Grief and the search for
meaning: Exploring the assumptive worlds of bereaved college students.
Journal of Social and Clinical Psychology, 10, 270288.
Stice, E., Presnell, K., & Spangler, D. (2002). Risk factors for binge eating
onset in adolescent girls: A 2-year prospective investigation. Health
Psychology, 21, 131138.
Sullivan, P. F., Neale, M. C., & Kendler, K. S. (2000). Genetic epidemi-
ology of major depression: Review and meta-analysis. American Journal
of Psychiatry, 157, 15521562.
Tajfel, H., & Turner, J. (1986). The social identity theory of intergroup
behavior. In S. Worchel & W. Austin (Eds.), Psychology of intergroup
relations (2nd ed., pp. 724). Chicago, IL: Nelson-Hall.
Valentiner, D. P., Holahan, C. J., & Moos, R. H. (1994). Social support,
appraisals of event controllability, and coping: An integrative model.
Journal of Personality and Social Psychology, 66, 10941102.
Wareham, S., Fowler, K., & Pike, A. (2007). Determinants of depression
severity and duration in Canadian adults: The moderating effects of
gender and social support. Journal of Applied Social Psychology, 37,
29512979.
Wegner, D. M., & Zanakos, S. (1994). Chronic thought suppression.
Journal of Personality, 62, 615640.
Wester, S. R., Christianson, H. F., Vogel, D. L., & Wei, M. (2007). Male
gender role conflict and psychological distress: The role of social sup-
port. Psychology of Men and Masculinity, 8, 215224.
Wren, B. (2002). I can accept my child is transsexual but if I ever see him
in a dress Ill hit him: Dilemmas in parenting a transgendered adoles-
cent. Clinical Child Psychology and Psychiatry, 7, 377397.
Wright, E. R., & Perry, B. L. (2006). Sexual identity distress, social
support, and the health of gay, lesbian, and bisexual youth. Journal of
Homosexuality, 51, 81109.
Yeh, C. J., Okubo, Y., Ma, P.-W. W., Shea, M., Ou, D., & Pituc, S. T.
(2008). Immigrant high school students cultural interactions, accultur-
ation, family obligations, language, and social support. Adolescence, 43,
775790.
Zimet, G. D., Dahlem, N. W., Zimet, S. G., & Farley, G. K. (1988). The
Multidimensional Scale of Perceived Social Support. Journal of Per-
sonality Assessment, 52, 3041.
Received April 10, 2012
Revision received December 17, 2012
Accepted December 26, 2012
T
h
i
s
d
o
c
u
m
e
n
t
i
s
c
o
p
y
r
i
g
h
t
e
d
b
y
t
h
e
A
m
e
r
i
c
a
n
P
s
y
c
h
o
l
o
g
i
c
a
l
A
s
s
o
c
i
a
t
i
o
n
o
r
o
n
e
o
f
i
t
s
a
l
l
i
e
d
p
u
b
l
i
s
h
e
r
s
.
T
h
i
s
a
r
t
i
c
l
e
i
s
i
n
t
e
n
d
e
d
s
o
l
e
l
y
f
o
r
t
h
e
p
e
r
s
o
n
a
l
u
s
e
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
u
s
e
r
a
n
d
i
s
n
o
t
t
o
b
e
d
i
s
s
e
m
i
n
a
t
e
d
b
r
o
a
d
l
y
.
557
TRANSGENDER ANXIETY, DEPRESSION, AND COPING