Psychoneuroendocrinology 82 (2017) 59–66

Contents lists available at ScienceDirect

Psychoneuroendocrinology
journal homepage: www.elsevier.com/locate/psyneuen

Stigma and diurnal cortisol among transitioning transgender men MARK

a,⁎ b c d
L. Zachary DuBois , Sally Powers , Bethany G. Everett , Robert-Paul Juster
a
Department of Anthropology, California State University, F03-305, 1250 Bellflower Blvd., Long Beach, CA, 90804, United States
b
University of Massachusetts Amherst, Amherst, MA, United States
c
University of Utah, Salt Lake City, UT, United States
d
Columbia University, New York, NY, United States

A R T I C L E I N F O A B S T R A C T

Keywords: This study assessed diurnal cortisol functioning in relation to stigma-based transition-specific stressors
Transgender men experienced by transgender men during their transition from female to male. Sixty-five healthy transgender
Minority stress men undergoing testosterone therapy participated in in-person interviews through which transition-specific
Psychosocial stress stressors were identified. Interviews were coded according to participant reported (1) Transitioning-identity
Cortisol
stress; (2) Coming Out stress; (3) Gender-specific Public Bathroom stress; and (4) levels of general Perceived
HPA axis
Stress. Participants provided fifteen salivary samples assessing cortisol diurnal rhythm over three days.
Hierarchical linear models, adjusted for duration of time on testosterone therapy, body mass index, steroid-
related medication use, mean awakening time, and CAR, confirmed that elevated diurnal cortisol levels at
awakening were associated with transition-specific social stressors including experiencing Transitioning-identity
stress, frequent Coming Out stress, and Gender-specific Public Bathroom stress. Transitioning-identity stress and
Gender-specific Public Bathroom stress also predicted a steeper negative slope at awakening. General Perceived
Stress was not associated with elevated cortisol or slope. These results clarify the relation of increased cortisol at
awakening with a negative linear slope to perceived stigma and transition-related stress experience among
transgender men.

1. Introduction
“Transgender” and “trans” are umbrella terms to describe people
whose gender identity or presentation may not conform to conventional
social norms associated with their assigned birth sex (IOM, 2011).
Transgender people experience significant stress related to stigma-
based violence and discrimination and have disproportionally high
rates of distress, depression, and anxiety (Bockting et al., 2013;
Bradford et al., 2013). This is consistent with minority stress theory,
which states that excess stress experienced by individuals from
stigmatized social categories leads to adverse mental and physical
health outcomes (Meyer, 2003).
Existing research has focused on stigma among lesbian, gay and
bisexual (LGB) individuals (Hatzenbuehler et al., 2013; Hatzenbuehler
and McLaughlin, 2014), with few studies of transgender people
(Bockting et al., 2013; Levitt and Ippolito 2014). Moreover, the
majority of studies among transgender people are conducted among
those assigned male at birth (i.e., trans women). By contrast, trans men
who were assigned female at birth are underrepresented (Reisner et al.,
2016). Critically, few studies integrate an explicit aim to consider
socially derived stressors associated with gender identity and expres-
sion when examining psychobiological pathways linked to health and
well-being (for review, see Reisner et al., 2016).
1.1. Transition-specific and gender-related stress
Above and beyond the minority-stressors experienced by LGB
people related to sexual-orientation, transgender people experience
unique sources of stress and stigma as a result of having a gender-
identity or gender-presentation which falls outside rigid male-female
binaries (Grant et al., 2011). Additionally, unique challenges may arise
during the process of gender transition that involve: 1) social transition
related to gender presentation/expression (e.g., clothing and hair) and
identity (e.g., name and pronouns) and 2) physical transition, when
some transgender people seek hormonal therapies and surgeries to
better align their bodies with their gender identity (Bockting et al.,
2013).
For trans men who physically transition, accessing gender-affirming
treatments such as testosterone therapy and surgeries improve overall
quality of life and health (Keo-Meier et al., 2014; Newfield et al., 2006).

⁎
Corresponding author.
E-mail address: Zachary.DuBois@csulb.edu (L.Z. DuBois).

http://dx.doi.org/10.1016/j.psyneuen.2017.05.008
Received 19 August 2016; Received in revised form 25 April 2017; Accepted 5 May 2017
0306-4530/ © 2017 Elsevier Ltd. All rights reserved.
L.Z. DuBois et al.
Nonetheless, shifts in physical gender “cues”, including changes in male
secondary-sex characteristics (e.g., facial hair) during the first few years
of testosterone therapy may put trans men at risk for being “read” as
gender ambiguous or contradictory. This renders them more vulnerable
to gender-related stigma. Transition-related stress can be particularly
pernicious in gender segregated social spaces, including bathrooms or
locker rooms, and during social interactions in which gender is salient.
For example, a survey of urban transgender and gender non-conforming
people found 70% of respondents had experienced verbal harassment,
physical assault, and denial of access to gender-specific public bath-
rooms (Herman, 2013).
Changes in physical appearance during testosterone therapy typi-
cally require a “social transition.” This includes a “coming out” process
whereby individuals disclose their gender status to their social net-
works and request others address them using a new name/male
pronoun. In addition to “coming out” processes, social transition can
involve stress associated with the complexity of socially managing a
transitioning identity, particularly within a social context where a rigid
gender binary is expected (i.e., changes in gender recognition during
physical transition). For instance, this can occur in “mixed-company”
when social circles overlap, bringing together some people who may
know of one’s transition/gender status and others who may not. We
propose that these complex social processes and transition-related
stressors can “get under the skin and skull” via modulation of stress
physiology.
1.2. Minority stress and the hypothalamic-pituitary-adrenal axis
An understudied area of research for the transgender population is
the psychobiological link between minority stress and physical health.
A key player is the hypothalamic–pituitary–adrenal (HPA) axis respon-
sible for the production of cortisol (Sapolsky et al., 2000), which
follows a natural circadian rhythm. A normal diurnal cortisol rhythm
includes a gradual rise throughout the night followed by a 50–160%
increase in concentrations 30–45 min following awakening [cortisol
awakening response (CAR)] (Clow et al., 2010; Stalder et al., 2016),
and a decline over the day with lowest levels around bedtime (Adam,
2006).
Due to its regulatory role in multiple interconnected systems,
cortisol has been implicated as a mechanism through which stress can
lead to disease (Miller et al., 2007). Chronic stress-induced dysregula-
tion of the HPA-axis can lead to chronically high cortisol levels or to the
opposite effect where cortisol levels are decreased and a “flattening” of
the diurnal slope occurs (Carpenter et al., 2007; Ice et al., 2004). Both
chronically elevated cortisol levels and/or flattening of the diurnal
pattern are associated with negative health outcomes (Fries et al., 2005;
Marketon and Glaser, 2008). Studies examining stress and HPA-axis
activation have linked psychosocial stress to both high and low cortisol
awakening response (Chida and Steptoe, 2009; Adam et al., 2006);
however, other studies have found no significant association (Michaud
et al., 2009). Recently, Chi et al. (2015) found that stigmatization was
associated with lower cortisol levels at awakening and a flatter slope.
While little research has investigated stress and HPA-axis function-
ing among transgender populations, previous research has linked
social-evaluative threats – stressors that represent threats to the social
self – to hypercortisolemic profiles (Dickerson et al., 2004; Miller et al.,
2007). For example, African Americans who report higher levels of
perceived discrimination have higher waking cortisol levels and a
steeper diurnal decline in cortisol compared to Whites who report
experiencing discrimination (Fuller-Rowell et al., 2012). Stress linked
to devaluations of closely held identities, such as an individuals’ gender
identity and gender expression, may be a unique and important
mechanism that helps us understand established health disparities
among gender minorities. Unfortunately, there is a lack of field-based
research on stress and social evaluative threats during gender transi-
tion.
60
Psychoneuroendocrinology 82 (2017) 59–66
To the best of our knowledge, only one study has examined the
HPA-axis among trans men and trans women. Colizzi et al. (2013)
reported that elevated general perceived stress and cortisol prior to
hormonal therapy was related to significantly lower levels of general
stress and cortisol following 12 months of hormonal therapy. The
Colizzi study examined psychobiological distress only before and after
hormonal treatment; however, no psychosocial data was gathered over
the course of the first year of transition or following a year, as in the
current study. While transitioning itself is beneficial and helps to align
the body with one’s internal gender identity, stress experienced during
transition may reflect unique experiences of stigma and minority-stress.
In addition to general perceived stress, transgender-specific stressors
may be related to elevated levels of diurnal cortisol. Previous research
has shown that during the early phases of transition, psychosocial stress
related to transition-specific stressors can cause physiological changes
in blood pressure and levels of inflammation (DuBois, 2012). In
particular, stress related to “coming out” as transgender was associated
with a diminished decline in nocturnal blood pressure (i.e., blood
pressure while sleeping) (DuBois, 2012). Understanding the links
between transition-specific stressors and HPA-axis functioning will
provide complementary insights to further understand gender-identity
health disparities and develop public health strategies to improve trans
men’s health.
In the current analysis, we hypothesize that (1) trans men reporting
high transition-specific stress (i.e., Transitioning-identity stress; Coming
Out stress; Gender-specific Public Bathroom stress) will exhibit elevated
diurnal cortisol levels at awakening and (2) that trans men reporting
high transition-specific stress will exhibit a steeper linear slope or rate
of change in cortisol.
2. Methods
2.1. Participants
As part of the Transition Experience Study, sixty-five healthy trans
men were recruited from rural western Massachusetts, Boston, and
southern Vermont. Each participant was 18 years of age or older,
assigned a female sex designation at birth but expressed a male/trans
masculine gender identity, and was using testosterone therapy as part
of his transition. Participants were ineligible for participation if they
were currently taking medication for any cardiovascular or immune-
related conditions.
As trans men represent a relatively hard-to-reach population,
recruitment efforts focused on area support groups, social networks,
and snowball techniques. Interview and biomarker data were con-
ducted and collected during the period of 2009–2012 by the principal
investigator of the study (LZD) who identifies as an out, transgender
man. All interviews took place in the participants’ home, in private
meeting rooms at area LGBT health clinics and the University of
Massachusetts Amherst. Responses were transcribed onto the interview
instrument and digitally recorded. Consent forms were discussed and
participants documented their consent by writing their initials into a
notebook rather than providing their full names on the consent forms.
Each participant received $50 as remuneration. Further description of
the methodology of this study has been published previously (DuBois,
2012). The project was approved by the Institutional Review Board for
research involving human participants at the University of Massachu-
setts Amherst.
In addition to the in-depth interview, each participant provided
anthropometric and bio-impedance measures during in-person sessions.
Psychometric scales and background questionnaires with demographic
and biobehavioral data were distributed prior to meeting and were
collected in person during the interview. Participants were asked to
indicate if they were taking any steroid-based medications (e.g., for
colds, allergy, or asthma) both on the background questionnaire and
during the in-person interview. Salivary sample collection materials
L.Z. DuBois et al.
and both written and verbal instructions detailing procedures were
provided during the interview and participants then collected their
saliva after completing the interview and receiving instruction.
2.2. Biological measures
2.2.1. Anthropometry
Height was measured with a freestanding portable anthropometer.
Body weight was measured using a portable digital scale (Tanita, model
BC-550T). Body mass index (BMI) was then calculated from weight
(kg)/height (m2).
2.2.2. Saliva sampling
Participants collected salivary samples over three consecutive days
after the interview was completed: (1) upon awakening (while still in
bed), (2) thirty minutes post-waking (Clow et al., 2010) (prior to
coffee/tea and brushing teeth), (3) mid-morning (approximately 4 h
post waking, avoiding meal time), (4) mid-afternoon (approximately
8 h post-waking, avoiding meal time), and (5) at bedtime (prior to
brushing teeth). Participants used the “passive-drool” technique, using
a straw and polypropylene vial to collect their saliva (Ice, 2007). Each
participant was instructed to refrain from brushing their teeth, smok-
ing, or consuming dairy, alcohol, or caffeine 30 min prior to sample
collection. In addition, a compliance questionnaire was filled out with
each sample collection (Powers et al., 2016).
Sleep patterns, including information on average wake times and
average hours of sleep, may play a role in the functioning of the HPA-
axis and were therefore also collected (Clow et al., 2010; Zeiders et al.,
2011). For the waking sample, the questionnaire asked participants to
note the following: time to bed the night prior, approximate time that
they fell asleep, number of times they awoke, final wake time, and
method used to wake (i.e., alarm). Text-messages or phone call
reminders prompted participants to collect their mid-morning and
mid-afternoon samples. Study participants labeled their samples with
the date and time of collection and then refrigerated them until
retrieved. Samples were then frozen in a lab-grade freezer (−20°) until
being shipped on dry ice to the University of Dresden, Germany for
analysis in the laboratory of Dr. Clemens Kirschbaum. Analysis was
conducted using a chemiluminescence immunoassay (CLIA) with high
sensitivity of 0.16 ng/ml (IBL-International; Hamburg, Germany). Intra-
and inter-assay coefficients of variation were below 8%.
An average cortisol level for each of the diurnal assessment time
points was calculated by averaging the cortisol levels at each time point
across all days of data collection (Juster et al., 2011, 2013, 2016a,b).
This resulted in 5 cortisol scores representing the average diurnal level
(nmol/L) at each time point (waking, 30-min post waking, mid-
morning, mid-afternoon, and bedtime) for each participant.
2.2.3. Time on testosterone therapy
Testosterone therapy initiates and maintains male secondary sex-
characteristics (i.e., facial hair, lower voice) for trans men. The duration
of time an individual has been on testosterone determines the degree to
which male secondary sex characteristics are apparent and can
influence the transition-specific stress experience assessed. Time on
testosterone therapy was coded as a continuous variable that ranges
from .04 to 13.83 years.
2.3. Transition-related psychosocial stress measures
Participants were asked to discuss their transition experience during
the in-person interview and were prompted to answer specific questions
about the three transition-specific stressors that were previously
identified ad-hoc by the lead author based on findings from his
unpublished pilot-study (Transitioning-identity stress, Coming Out
stress, and Gender-specific Public Bathroom stress). As described above,
findings related to Coming Out stress (referred to as “Out stress”) have
61
Psychoneuroendocrinology 82 (2017) 59–66
been published previously in relation to ambulatory blood-pressure
measures (DuBois, 2012).
2.3.1. Transitioning-identity stress
Transitioning-identity stress refers to a sense of stress related to
having a public gender identity and body that is in flux. During in-
person interviews, each participant was asked: Do you feel that you are
known by more than one gender in your public life (e.g., some people
know you as a man, others as a trans man, and others know you as a
woman)? If participants answered yes to this question, they were asked
to address the following: (1) Would you prefer to be known by others
consistently as a man, trans man or in some other way (e.g., as gender
non-binary); and (2) Do you feel stress related to how others know you
in terms of your social or public gender identity?
A dichotomous variable was then created during analysis of the
interview data according to the following criteria: two independent
coders coded participants positively as experiencing Transitioning-
identity stress if they reported to be both a) socially identified as
sometimes male, sometimes female, sometimes transgender instead of
having a singular and consistent public male or trans male gender
identity and also b) feeling that the experience was stressful/distressing
(i.e., preferring a consistent singular social identity) (yes = 1).
Participants were coded as negative (0 = referent) for experiencing
Transitioning-identity stress if they reported no stress/distress related
to their transitioning social identity regardless of status. All differences
in coding were resolved by consensus. Inter-rater reliability (Cohen’s
Kappa) was .75, indicating excellent agreement (Fleiss, 1971).
2.3.2. Coming Out stress
Coming Out stress refers specifically to stress associated with the
process and/or experience of being out as transgender. During in-
person interviews, each participant was asked: Is being “out” or the
process of “coming out” a source of stress or anxiety for you?
Participants were instructed to select an answer from the following
choices: never, sometimes/infrequently, most of the time, and always and
responses were coded 0–3 (DuBois, 2012).
2.3.3. Gender-specific Public Bathroom stress
During the in-person interview, participants were asked to rate their
stress experience when using gender-specific public bathrooms on a
Likert Scale ranging from 0 (low stress) and 10 (very high stress).
2.4. General perceived stress
Each participant completed Cohen and colleagues’ Perceived Stress
Scale (1983), which is the most commonly used measure of general
perceived stress shown to predict numerous health outcomes (Ice,
2007). Ten questions ask participants to rate how often they felt or
thought in specified ways during the last month and are summed for
each participant, yielding scores ranging from 0 (low perceived stress)
to 40 (high perceived stress). The coefficient alpha for the scale was .91,
indicating a high degree of internal consistency among the items on the
scale. The means of the individual items ranged from 1.07 to 2.59, with
a mean on the total scale of 16.85 (SD = 7.22). Overall, the partici-
pants’ responses on the scale indicated that they perceived themselves
as experiencing a moderate amount of stress.
2.5. Statistical analysis
Bivariate correlations and one-way ANOVAs were used to identify
potential confounders of the relation of stress variables to cortisol
assessments. Growth curve modeling was used to plot temporal
trajectories of participants’ diurnal rhythms and to predict variance in
these trajectories from participants’ transition-related stress
(Transitioning-identity stress, “Coming Out” stress, and Gender-specific
Public Bathroom stress) or generalized non-transition-specific stress
L.Z. DuBois et al. Psychoneuroendocrinology 82 (2017) 59–66

(general Perceived Stress Scale). We used Hierarchical Linear Modeling, Table 1

Version 7 (Raudenbush and Bryk, 2002) to specify a 2-level model that Sample characteristics (N = 65).
accounted for the dependency of cortisol scores over the day (Adam and
Characteristic Statistic
Gunnar, 2001). The Level 1 (within-participant) model defined three
parameters – intercept, linear slope, and quadratic term – that Demographics
characterized participants’ curvilinear cortisol trajectories. The inter- Age, M (SD) 31.8 (9.1)
Race/ethnicity, n (%) White 48 (74)
cept and linear parameter coefficients were allowed to take on different
values for each participant while the quadratic term was fixed. The Socioeconomic
Level 2 (between participant) model included participant stress pre- Years of education, M (SD) 17.3 (2.9)
College degree earned, n (%) 44 (67.7)
dictors to explain variance in the Level 1 coefficients. The relationship Current job status
between each stressor and cortisol trajectories was examined indepen- Works full-time, n (%) 34 (52.3)
dently adjusting for steroid-related medicines, BMI, years on testoster- Works part-time, n (%) 16 (24.6)
one, and time of waking. In addition to controlling for these potentially Unemployed/laid off, n (%) 15 (23.1)
Monthly income, M, (SD) $1986 ($1639)
confounding variables, CAR was dichotomized and included as a
Financial status
control variable at Level 2, due to evidence that the CAR is subject to Makes less than needs/seeks work, n (%) 30 (46.9)
regulatory influence independent of the rest of the diurnal cycle (Clow Makes exactly what needs/no wiggle room, n (%) 13 (20.3)
et al., 2004). Makes just enough to be comfortable, n (%) 16 (25)
Makes plenty/can be generous, n (%) 5 (7.8)
Accrued debt from transition costs 20 (31)
3. Results
Transitional/Physical
Body Mass Index kg/m2 30.2 (6.7)
3.1. Descriptive and inferential statistics
Years on testosterone, M (SD) 3.3 (3.4)
Top surgery, n (%) 61.5
Sample characteristics are shown in Table 1. Bivariate correlations Hysterectomy, n (%) 23.1
and one-way ANOVAs were used to identify potential confounders of Sexuality
diurnal cortisol at trend level (p < .10). We identified the following Sexual orientation
covariates that were included in our main analyses: years on testoster- Heterosexual, n (%) 15 (23.1)
one treatment (r = .233, p = .066), steroid-based medications Gay (prefers men and/or transmen), n (%) 9 (14.5)
Queer (prefers women), n (%) 16 (24.6)
(r = −.244, p = .062), BMI (r = −.307, p = .017), and mean awa- Queer (flexible/bisexual), n (%) 25 (38.5)
kening time (r = −.577, p < .001). Variables that were not related to Sexual identity
diurnal cortisol (p > .10) included age, sexual orientation, education, Man or guy, n (%) 32 (49.2)
relationship status, number of children, caffeine consumption, cigarette Trans man, n (%) 20 (30.8)
Genderqueer/queer, n (%) 7 (10.8)
smoking, alcohol use, and hours of sleep per night and were thus
Context dependent, n (%) 4 (6.2)
excluded from analyses. Associations among the three transition-related Not female, n (%) 2 (3.1)
stress variables ranged from low to moderate [Gender-specific Public
Psychosocial
Bathroom stress and Coming Out stress, r = .16; Gender-specific Public Perceived Stress Scale scores M (SD) 16.4 (6.7)
Bathroom stress and Transitioning-identity stress, r = .41; Coming Out Victimized due to gender presentation
stress and Transitioning-identity stress, r = .32] (Correlation table not Yes, before transition, n (%) 12(18.5)
shown but available upon request). Yes, before and after transition, n (%) 7 (10.8)
Yes, after transition, n (%) 17 (26.2)
For the purposes of illustrating the relation of our primary stress
variables to diurnal cortisol trajectories (Fig. 1), we employed pre- Transition-specific Stressors
Reports Transitioning-identity stress, n (%) 29 (44.6)
liminary repeated measures analysis of covariance (RM-ANCOVA) for
Reports Coming Out stress, n (%) 29 (44.6)
diurnal cortisol with the aforementioned covariates entered as a Levels of Gender-specific Public Bathroom Stress, M (SD) 4.2 (3.1)
function of (A) Transitioning-identity stress (no: n = 36, yes:
Relational
n = 29), (B) Coming Out stress (infrequent: n = 36, frequent: Relationship status
n = 29) (C) Gender-specific Public Bathroom stress (no/low < 4: Single, n (%) 15 (23.1)
n = 31, medium/high 4+: n = 33), and (D) general Perceived Stress Dating/In a relationship, n (%) 15 (23.1)
(no/low < 16: n = 31, medium/high 16+: n = 32). Partnered/married, n (%) 35 (53.8)
Lives with partner, n (%) 32 (62.7)
Current relationship preceded transition 23 (46.9)
3.2. Hierarchical linear models
Health Behaviors
Medications for allergies or colds, n (%) 14 (7.2)
HLM models were centered on the sample collected at awakening so Mean wake-time M (SD) 8:01 AM (2:11)
that the intercept represented HPA-axis activation at the beginning of Smoker, n (%) 13 (20)
each participant’s day (Adams et al., 2006). In each of the first 3 Consumes alcohol, n (%) 53 (82.8)
models, one of three specific transition-related stress variables Exercises regularly, n (%) 44 (68.8)

(Transitioning-identity stress, Coming Out stress, or Gender-specific

Public Bathroom stress) was added as a predictor to explain variability
in the cortisol intercept and slope. Table 2 presents the results for Model
1: Transitioning-identity stress, Table 3 presents the results for Model 2:
Coming Out stress, Table 4 presents the results for Model 3: Gender-
specific Public Bathroom stress. A final model was run with non-
transition-specific general stress (Perceived Stress Scale) entered as the
predictor. Table 5 presents the results for general Perceived Stress.
These models provided statistical tests of the association between
transition related and non-transition related stress and participants’
(a) cortisol level at awakening (intercept at centered sample), and (b)
linear slope (rate of change at awakening). Steroid-related medications,
BMI, years on testosterone, mean awakening time, and CAR were
included in all models as controls.
3.2.1. Diurnal cortisol predicted by Transition-related stress variables
Table 2 shows that experiencing Transitioning-identity stress pre-
dicted higher cortisol levels at the awakening sample. Participants who
experienced Transitioning-identity stress had an average cortisol read-
ing of the awakening level of 1.43 higher points than participants
without Transitioning-identity stress. Transitioning-identity stress also
predicted a steeper falling slope at the awakening sample.

62
L.Z. DuBois et al. Psychoneuroendocrinology 82 (2017) 59–66

Fig. 1. Estimated mean ( ± SE) diurnal cortisol as a function of transitional factors. Specifically, cortisol diurnal trajectories are shown for illustrative purposes as a function of (A)
Coming Out stress, (B) Transitioning-identity stress, (C) Gender-specific Public Bathroom stress, and (D) general Perceived stress. Values are adjusted for years of testosterone treatment,
medication, body mass index, and mean awakening time.

Table 2 Table 3
Transitioning-identity stress predict diurnal cortisol trajectory. Coming Out stress predict diurnal cortisol trajectory.

Awakening Sample Awakening Sample

Fixed Effect Estimate SE p Fixed Effects Estimate SE p

Cortisol Level Cortisol Level

Intercept 11.630 .613 < .001 Intercept 11.616 .649 < .001
Years on Testosterone .054 .072 .45 Years on Testosterone .024 .080 .76
Body mass index − .137 .054 .01 Body mass index − .135 .053 .01
Steroid-based Medications − .112 1.137 .92 Steroid-based Medications .181 .979 .85
Time Awake −.130 .157 .41 Time Awake −.128 .171 .46
Cortisol Awakening Response .46 .667 .49 Cortisol Awakening Response .475 .729 .52
Transitioning Identity Stress 1.427 .580 .02 Coming Out Stress .960 .427 .03

Linear Slope Linear Slope

Intercept −1.38 .116 < .001 Intercept −1.384 .118 < .001
Years on Testosterone −.009 .006 .09 Years on Testosterone −.005 .006 .38
Body Mass Index .010 .004 .02
Body Mass Index
Steroid-based Medications −.028 .106 .79
Steroid-based Medications −.038 .093 .68
Time Awake .024 .014 .08
Time Awake .024 .015 .12
Cortisol Awakening Response −.023 .045 .61
Cortisol Awakening Response −.031 .051 .55
Transitioning Identity Stress −.136 .041 .002
Coming Out Stress −.058 .030 .06
Curvature Across Diurnal Trajectory
Curvature Across Diurnal Trajectory
Intercept .050 .006 < .001
Intercept .050 .006 < .001

Bold values indicate p = .05 or less.

Bold values indicate p = .05 or less.

Table 3 shows that more frequent Coming Out stress predicted

Coming Out stress, participants had an average increase in cortisol level
significantly higher cortisol levels at the awakening sample, and
at awakening of .96 higher points.
showed a strong trend toward a steeper falling slope at awakening.
Table 4 shows that Gender-specific Public Bathroom stress also
The model indicated that, for every one-unit increase in frequency of
predicted higher awakening cortisol: For every one-unit increase in

63
L.Z. DuBois et al. Psychoneuroendocrinology 82 (2017) 59–66

Table 4 vide evidence that trans men experiencing greater transition-related

Gender-specific Public Bathroom stress scores predict diurnal cortisol trajectory. stress exhibit elevated cortisol concentrations at awakening. Following
these elevated levels, trans men with Transitioning-identity stress or
Awaking Sample
higher levels of Gender-Specific Public Bathroom stress exhibited
Fixed Effects Estimate SE p steeper, faster linear slope. Our findings thus confirm others who have
found the awakening point (Adam et al., 2006) and diurnal slope
Cortisol Level
important to examine (Adam and Gunnar, 2001). These findings
Intercept 11.663 .595 < .001
Years on Testosterone .041 .061 .50 provide insights into the various ways minority stressors – specific to
Body Mass Index − .152 .056 .009 the challenges faced by trans men during their transition – can “get
Steroid-based Medications .011 1.195 .99 under the skin and skull” and impact measures of biological stress.
Time Awake −.334 .167 .05 Overall, these findings expand those of Colizzi et al. (2013) who
Cortisol Awakening Response .409 .682 .55
identified higher waking cortisol levels among transgender men and
Public Bathroom Stress .335 .095 < .001
transgender women prior to their beginning physical transition. Our
Linear Slope
findings clarify that higher waking cortisol levels continue during
Intercept −1.388 .115 < .001
Years on Testosterone −.007 .004 .10 transition for trans men who experience stress related to the use of
Body Mass Index .011 .004 .01 gender-specific public bathrooms, managing a transitioning social
Steroid-based Medications −.032 .096 .74 identity, or their coming out process. More broadly, our findings can
Time Awake .042 .015 .006 be interpreted in light of literature demonstrating hypercortisolemic
Cortisol Awakening Response −.022 .046 .63
Public Bathroom Stress −.027 .008 < .001
profiles in relation to perceived risk of humiliation or disrespect
(Dickerson et al., 2004; Miller et al., 2007). These findings have been
Curvature Across Diurnal Trajectory
shown both in lab-based studies of social-evaluative threat (Dickerson
Intercept .050 .006 < .001
and Kemeny, 2004) and naturalistic studies where feelings of anger and
Bold values indicate p = .05 or less. of being “overwhelmed” were associated with higher than average
diurnal cortisol levels (Adam et al., 2006).
Table 5 The trans men in this study who experienced either Transitioning-
General Perceived Stress scores do not predict diurnal cortisol trajectory. identity stress or higher levels of Gender-specific Public Bathroom
Stress exhibited steeper slopes rather than the flattened profile generally
Awakening Sample
associated with chronic stress (Barker et al., 2012). This steeper decline
Fixed Effects Estimate SE p may signal healthy recovery from the increased cortisol levels experi-
enced earlier in the day. This finding is in line with previous research
Cortisol Level that found a relationship between perceived discrimination and steeper
Intercept 11.550 .617 < .001
slopes among a different minority population – African Americans – but
Years on Testosterone .029 .075 .70
Body Mass Index − .160 .055 .005 not Whites (Fuller-Rowell et al., 2012). Fuller-Rowell et al. (2012)
Steroid-based Medications −.400 1.084 .71 argued that awareness of racism in daily life might serve an important
Time Awake −.089 .162 .58 protective mechanism against HPA-axis hyperactivity.
Cortisol Awakening Response .696 .688 .32 In line with this interpretation, steep diurnal decline may indicate
Perceived Stress .072 .041 .09
resilience in this population because trans men are frequently aware
Linear Slope that there is often stigma associated with their transgender identity and
Intercept −1.381 .118 < .001
they anticipate stress around gender-specific public bathroom use.
Years on Testosterone −.005 .006 .36
Body Mass Index .011 .004 .01 Though it is important to recognize the potentially negative effects of
Steroid-based Medications −.003 .104 .98 increased morning and total daily cortisol, the finding of a steeper
Time Awake .022 .015 .15 diurnal slope among men who report experiencing transition-related
Cortisol Awakening Response −.043 .047 .36 stressors may reflect a healthy response to psychosocial stress asso-
Perceived Stress −.002 .003 .47
ciated with stigma. Just as a flattened cortisol profile can reflect
Curvature Across Diurnal Trajectory elevated bedtime levels rather than an overall blunted response
Intercept .050 .006 < .001
(Lovell et al., 2011), this steep decline from the morning peak to
Bold values indicate p = .05 or less. bedtime seems to reflect an increased awakening response followed by
a healthy evening decline.
Gender-specific Public Bathroom stress, participants had an average Our finding linking transitioning-related stressors to elevated diur-
cortisol reading at the awakening sample of .34 higher points. Gender- nal cortisol production suggests a potential mechanism whereby
specific Public Bathroom stress also predicted a steeper falling slope at minority stress can lead to health disorders. Our findings differ from
awakening. those of Chi et al. (2015) who showed that stigma was associated with
lower cortisol levels at awakening among HIV positive children. Their
study, however, also included measures of resilience and demonstrated
3.2.2. Diurnal cortisol predicted by general Perceived stress that children who exhibited high levels of resilience perceived lower
Table 5 shows that general Perceived stress was not significantly levels of stigma, which in turn shaped their diurnal cortisol rhythm.
related to cortisol level at the awakening sample, or the slope at the Regrettably, we did not measure resilience and focused on adults,
awakening sample. which may explain the difference between study results and should be
explored further.
4. Discussion Our finding that more frequent Coming Out stress predicted
elevated cortisol levels at waking also differs slightly from that of
This study examined transgender specific minority stress and Juster et al. (2013) who found that disclosure in terms of sexual
diurnal cortisol profiles among transitioning trans men undergoing orientation was associated with lower diurnal HPA-axis activity (Juster
testosterone therapy. Several key findings emerged, each supporting et al., 2013) and lower perceived stress when they are socially
our overall hypothesis that specific transition-related stressors are supported (Juster et al., 2016a,b). Our study specifically measures
associated with amplified HPA-axis production. Specifically, we pro- stress related to the process of coming out as transgender, not sexual

64
L.Z. DuBois et al.
orientation, and is not exclusively referring to the outcome of dis-
closure. Our findings represent an important addition to the literature
given that despite recent shifts in visibility and support for transgender
rights, stigma and discrimination remains pervasive, posing a substan-
tial threat to the health and well-being of the transgender population
(IOM, 2011). Particularly during transition, gender identity and
expression may not conform uniformly to culturally prescribed binary
male-female categories, thus increasing the risk for both stigma and
discrimination.
In addition to providing support for our hypothesis, our findings
provide further evidence that stigma affects health and therefore speak
to contemporary socio-political issues in the United States and abroad.
Importantly, the significant associations between levels of Gender-
specific Public Bathroom stress and cortisol levels directly address
contemporary arguments regarding access to gender-specific public
spaces. For example, in 2016, a series of “bathroom bills” were
proposed in at least 18 states (Kralik, 2016). Many of these “bathroom
bills” seek to criminalize the use of gender-specific bathrooms by
transgender people and require they use the bathroom that corresponds
with their assigned sex as listed on their birth certificate rather than
according to their gender identity and physical presentation (Schilt and
Westbrook, 2015; Wang et al., 2016). Particularly in the wake of the
recent rescinding of protections for transgender students in public
schools, these findings point to the potential serious negative effect that
denial of this protection can have on transgender people.
The policy debate surrounding these bills has an enormous impact
for the lives of transgender people, and transgender youth in particular
who are highly vulnerable to stigma (IOM, 2011). In addition to mental
health and quality of life effects, further research could examine
whether these stigmatizing and discriminatory experiences related to
using bathrooms directly affect physical health by forcing trans people
to suppress their basic bodily needs, which can increase individual’s
risk for urinary tract or kidney infections among a host of other
negative health outcomes (Schuster et al., 2016).
The absence of a relationship between general perceived stress in
our study and cortisol suggests that for people in gender transition,
general measures of perceived stress inadequately assess stress experi-
ences and its effects at the biological level. This finding is consistent
with others that have shown that stress responses are not indiscriminant
and that the specific characteristics of the stressor and of the individual
both contribute to variation in stress physiology (Kudielka et al., 2009).
Future research should focus on developing psychometrics that capture
the unique experiences of trans populations that may be sensitive to
biological processes.
To the best of our knowledge, this study is the first to examine
psychosocial stress experiences and diurnal cortisol during gender
transition among trans men. Results from this study linking transi-
tion-specific stress to HPA-axis functioning are particularly informative
given the extremely high rates of discrimination and victimization
experienced by the trans population (Grant et al., 2011; Lombardi et al.,
2002; Reisner et al., 2014) and the potential for stigma-experience to
negatively affect both mental and physical health (White Hughto et al.,
2015).
4.1. Limitations
Despite these contributions, it is important to note limitations of this
study. First, this study was designed to examine the stress experiences
of trans men during transition, thus these findings are not generalizable
to trans women undergoing gender transition, trans men who are not
utilizing testosterone as part of their transition or other people who
identify as transgender but who are not transitioning. Second, the study
sample size is small. Notwithstanding and considering that trans men
represent a hard-to-reach population, the number of participants and
high level of compliance with study protocols is promising for future
research.
65
Psychoneuroendocrinology 82 (2017) 59–66
4.2. Conclusions
This study expands minority stress models and applies a social
evaluative threat framework within real-life settings by highlighting the
negative effects of stigma, prejudice, and discrimination on transgender
individuals (Bockting et al., 2013; Dickersen et al., 2004; Meyer, 2003).
In this study that provides a first step forward in biomarker assessment
among trans men, we characterize specific identity and transition-
related stressors in relation to psychobiological pathways linked to
health. The types of stressors that trans men described are indicative of
managing social relationships (Transitioning-identity stress and Coming
Out stress) and reflect cultural norms pertaining to traditional binary
distinctions imposed on both gender presentation and gender identity
(levels of Gender-specific Public Bathroom stress). These gender-
minority and social-evaluative threat experiences illuminate the im-
portance of facilitating the transition process itself while simulta-
neously providing sufficient support and coping resources for trans
men during transition, particularly within sociocultural contexts where
male-female gender categories are rigidly enforced. Finally, these
findings speak to the importance of anti-discriminatory transgender
policies, as well as increased education to reduce stigma, prejudice, and
discrimination against transgender people in order to improve trans-
gender health.
Acknowledgements
We are grateful to all of the trans men who participated in this
study. We also thank The Fenway Institute and Tapestry Health Clinics
for providing safe and private meeting rooms for conducting interviews.
We thank Lisa Fiorenzo, Aline Sayer, and the UMass Center for Research
on Families for statistical consultation.
References
Adam, E.K., Gunnar, M.R., 2001. Relationship functioning and home and work demands
predict individual differences in diurnal cortisol patterns in women.
Psychoneuroendocrinology 26 (2), 189–208.
Adam, E.K., Hawkley, L.C., Kudielka, B.M., Cacioppo, J.T., 2006. Day-to-day dynamics of
experience: cortisol associations in a population-based sample of older adults. Proc.
Natl. Acad. Sci. U. S. A. 103 (45), 17058–17063.
Adam, E.K., 2006. Transactions among adolescent trait and state emotion and diurnal and
momentary cortisol activity in naturalistic settings. Psychoneuroendocrinology 31
(5), 664–679.
Barker, E.T., Greenberg, J.S., Seltzer, M.M., Almeida, D.M., 2012. Daily stress and cortisol
patterns in parents of adults and children with a serious mental illness. Health
Psychol. 31, 130–134.
Bockting, W.O., Miner, M.H., Swineburne, R., Hamilton, A., Coleman, E., 2013. Stigma,
mental health, and resilience in an online sample of the US transgender population.
Am. J. Public Health 103 (5), 943–951.
Bradford, J., Reisner, S.L., Honnold, J.A., Xavier, J., 2013. Experiences of transgender-
related discrimination and implications for health: results from the Virginia
Transgender Health Initiative Study. Am. J. Public Health 103 (10), 1820–1829.
Carpenter, L.L., Carvalho, J.P., Tyrka, A.R., Wier, L.M., Mello, A.F., Mello, M.F.,
Anderson, G.M., Wilkinson, C.W., Price, L.H., 2007. Decreased adrenocorticotropic
hormone and cortisol responses to stress in healthy adults reporting significant
childhood maltreatment. Biol. Psychiatry 62 (10), 1080–1087.
Chi, P., Slatcher, R.B., Li, X., Zhao, J., Zhao, G., Ren, X., Zhu, J., Stanton, B., 2015.
Perceived stigmatization, resilience, and diurnal cortisol rhythm among children with
parents living with HIV. Psychol. Sci. 26 (6), 843–852.
Chida, Y., Steptoe, A., 2009. Cortisol awakening response and psychosocial factors: a
systematic review and meta-analysis. Biol. Psychol. 80 (3), 265–278.
Clow, A., Thorn, L., Evans, P., Hucklebridge, F., 2004. The awakening cortisol response:
methodological issues and significance. Stress 7, 29–37. http://dx.doi.org/10.1080/
10253890410001667205.
Clow, A., Hucklebridge, F., Thorn, L., 2010. The cortisol awakening response in context.
Int. Rev. Neurobiol. 93, 153–175.
Cohen, S., Kamarck, T., Mermelstein, R., 1983. A global measure of perceived stress. J.
Health Soc. Behav. 24 (4), 385–396.
Colizzi, M., Costa, R., Pace, V., Todarello, O., 2013. Hormonal treatment reduces
psychobiological distress in gender identity disorder, independently of the
attachment style. J. Sex. Med. 10 (12), 3049–3058.
Dickerson, S.S., Kemeny, M.E., 2004. Acute stressors and cortisol responses: a theoretical
integration and synthesis of laboratory research. Psychol. Bull. 130 (3), 355–391.
Dickerson, S.S., Gruenewald, T.L., Kemeny, M.E., 2004. When the social self is threatened:
shame, physiology, and health. J. Pers. 72 (6), 1191–1216.
L.Z. DuBois et al.
DuBois, L.Z., 2012. Associations between transition-specific stress experience, nighttime
dip in blood pressure, and C-reactive protein levels among transgendered men. Am. J.
Hum. Biol. 24 (1), 52–61.
Fleiss, J.L., 1971. Measuring nominal scale agreement among many raters. Psychol. Bull.
76 (5), 378–382.
Fries, E., Hesse, J., Hellhammer, J., Hellhammer, D.H., 2005. A new view on
hypocortisolism. Psychoneuroendocrinology 30 (10), 1010–1016.
Fuller-Rowell, T.E., Doan, S.N., Eccles, J.S., 2012. Differential effects of perceived
discrimination on the diurnal cortisol rhythm of African Americans and Whites.
Psychoneuroendocrinology 37 (1), 107–118.
Grant, J.M., Mottet, L., Tanis, J.E., Harrison, J., Herman, J., Keisling, M., 2011. Injustice
at Every Turn: A Report of the National Transgender Discrimination Survey. Natl.
Gay Lesbian Task Force U.S.
Hatzenbuehler, M.L., McLaughlin, K.A., 2014. Structural stigma and hypothalamic-
pituitary-adrenocortical axis reactivity in lesbian, gay, and bisexual young adults.
Ann. Behav. Med. 47 (1), 39–47.
Hatzenbuehler, M.L., McLaughlin, K.A., Slopen, N., 2013. Sexual orientation disparities in
cardiovascular biomarkers among young adults. Am. J. Prev. Med. 44 (6), 612–621.
Herman, J., 2013. Gendered restrooms and minority stress: the public regulation of
gender and its impact on transgender people’s lives. J. Publ. Manag. Soc. Policy 19
(1), 65–80.
Institute of Medicine (IOM), 2011. The Health of Lesbian, Gay, Bisexual, and Transgender
People: Building a Foundation for Better Understanding. The National Academies
Press, Washington DC.
Ice, G., Katz-Stein, A., Himes, J., Kane, R., 2004. Diurnal cycles of salivary cortisol in
older adults. Psychoneuroendocrinology 29 (3), 355–370.
Ice, G., 2007. Measuring emotional and behavioral response. Measuring Stress in
Humans: A Practical Guide for the Field. Cambridge University Press, New York
(pp. 60).
Juster, R.P., Sindi, S., Marin, M.-F., Perna, A., Hashemi, A., Pruessner, J.C., Lupien, S.J.,
2011. A clinical allostatic load index is associated with burnout symptoms and
hypocortisolemic profiles in healthy workers. Psychoneuroendocrinology 36 (6),
797–805.
Juster, R.P., Smith, N.G., Ouellet, E., Sindi, S., Lupien, S.J., 2013. Sexual orientation and
disclosure in relation to psychiatric symptoms, diurnal cortisol, and allostatic load.
Psychosom. Med. 75 (2), 103–116.
Juster, R.P., Raymond, C., Desrochers, A.B., Bourdon, O., Durand, N., Wan, N., Pruessner,
J.C., Lupien, S.J., 2016a. Sex hormones adjust “sex-specific” reactive and diurnal
cortisol profiles. Psychoneuroendocrinology 63, 282–290.
Juster, R.P., Ouellet, É., Lefebvre-Louis, J.P., Sindi, S., Johnson, P.J., Smith, N.G., Lupien,
S.J., 2016b. Retrospective coping strategies during sexual identity formation and
current biopsychosocial stress. Anxiety Stress Coping 29 (2), 119–138.
Keo-Meier, C., Herman, L., Reisner, S.L., Pardo, S., Sharp, C., Babcock, J., 2014.
Testosterone treatment and MMPI-2 improvement in transgender men: a prospective
controlled study. J. Consult. Clin Psychol. 83 (1), 143–156.
Kralik, Joellen, 2016. Bathroom Bill Legislative Tracking. National Conference of State
Legislatures. (Posted 30 Aug. 2016/Accessed 19 December 2016). http://www.ncsl.
org/research/education/-bathroom-bill-legislative-tracking635951130.aspx.
Kudielka, B.M., Hellhammer, D.H., Wuest, S., 2009. Why do we respond so differently?
Reviewing determinants of human salivary cortisol responses to challenge.
Psychoneuroendocrinology 34 (1), 2–18.
Levitt, H.M., Ippolito, M.R., 2014. Being transgender: navigating minority stressors and
66
Psychoneuroendocrinology 82 (2017) 59–66
developing authentic self-preservation. Psychol. Women Q. 38 (1), 46–64.
Lombardi, E.L., Wilchins, R.A., Preising, D., Malouf, D., 2002. Gender violence:
transgender experiences with violence and discrimination. J. Homosex. 42 (1),
89–101.
Lovell, B., Moss, M., Wetherell, M.A., 2011. With a little help from my friends:
psychological, endocrine and health corollaries of social support in parental
caregivers of children with autism or ADHD. Res. Dev. Disabil. 33, 682–687.
Marketon, J.I.W., Glaser, R., 2008. Stress hormones and immune function. Cell Immunol.
252 (1–2), 16–26.
Meyer, I.H., 2003. Prejudice, social stress, and mental health in lesbian, gay, and bisexual
populations: conceptual issues and research evidence. Psychol. Bull. 129, 674–697.
Michaud, K., Matheson, K., Kelly, O., Anisman, H., 2009. Impact of stressors in a natural
context on release of cortisol in healthy adult humans: a meta-analysis. Stress 12 (5),
177–197.
Miller, G.E., Chen, E., Zhou, E.S., 2007. If it goes up, must it come down? Chronic stress
and the hypothalamic-pituitary-adrenocortical axis in humans. Psychol. Bull. 133 (1),
25–45.
Newfield, E., Hart, S., Dibble, S., Kohler, L., 2006. Female-to-male transgender quality of
life. Qual. Life Res. 15 (9), 1447–1457.
Powers, S.I., Laurent, H.K., Gunlicks-Stoessel, M., Balaban, S., Bent, E., 2016. Depression
and anxiety predict sex-specific cortisol responses to interpersonal stress.
Psychoneuroendocrinology 69, 172–179.
Raudenbush, S.W., Bryk, A.S., 2002. Hierarchical Linear Models: Application and Data
Analysis Methods, 2nd ed. Sage, Newbury Park, CA.
Reisner, S.L., White, J.M., Dunham, E.E., Heflin, K., Begenyi, J., Cahill, S., The Project
Voice Team, 2014. Discrimination and Health in Massachusetts: A Statewide Survey
of Transgender and Gender Nonconforming Adults. The Fenway Institute, Fenway
Health, Boston.
Reisner, S.L., Poteat, T., Keatley, J., Cabral, M., Mothopeng, T., Dunham, E., Holland,
C.E., Max, R., Baral, S.D., 2016. Global health burden and needs of transgender
populations: a review. Lancet 388 (10042), 412–436.
Sapolsky, R., Romero, L., Munck, A., 2000. How do glucocorticoids influence stress
responses? Integrating permissive, suppressive, stimulatory, and preparative actions.
Endocr. Rev. 21 (1), 55–89.
Schilt, K., Westbrook, L., 2015. Bathroom battle grounds and penis panics. Contexts 14
(3), 26–31.
Schuster, M.A., Reisner, S.L., Onorato, B.A., 2016. Beyond bathrooms—meeting the
health needs of transgender people. Perspect. N. Engl. J. Med. 375, 101–103.
Stalder, T., Kirschbaum, C., Kudielka, B.M., Adam, E.K., Pruessner, J.C., Wüst, S.,
Dockray, S., Symth, N., Evans, P., Hellhammer, D.H., Miller, R., Wetherell, M.A.,
Lupien, S.J., Clow, A., 2016. Assessment of the cortisol awakening response: expert
consensus guidelines. Psychoneuroendocrinology 63, 414–432.
Wang, T., Solomon, D., Durso, L.E., McBride, S., Cahill, S., 2016. Policy Brief: State Anti-
Transgender Bathroom Bills Threaten Transgender People’s Health and Participation
in Public Life. Fenway Inst. & Cent. Am. Prog.
White Hughto, J.M., Reisner, S.L., Pachankis, J.E., 2015. Transgender stigma and health:
a critical review of stigma determinants, mechanisms, and interventions. Soc. Sci.
Med. 147, 222–231.
Zeiders, K.H., Doane, L.D., Adam, E.K., 2011. Reciprocal relations between objectively
measured sleep patterns and diurnal cortisol rhythms in late adolescence. J. Adolesc.
Health 48 (6), 566–571.