Considerations and complexities of accurate PTSD assessment among transgender and gender diverse adults

Abstract

PTSD assessment among transgender and gender diverse (TGD) adults is complex because the literature offers little guidance on affirming assessment that accurately captures both trauma- and discrimination-related distress. This study aimed to characterize threats to precise PTSD assessment that arose during the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5). Our sample (N = 44) included trans women (38%), trans men (25%), nonbinary people (23%), and other TGD identities (14%). Participants were mostly white (75%), non-Latinx (82%), educated (91% at least some college), with a mean age of 37 years (SD = 15.5). Demographic and CAPS-5 scoring data as well as content analysis of audio-recorded CAPS-5 interviews are reported. All participants reported trauma exposure, and nearly half met PTSD diagnostic criteria (49%). Interpersonal assault was a common trauma type linked to posttraumatic symptoms (77%); 41% were sexual assaults; and 41% were discrimination-based (e.g., linked to gender identity) physical or sexual assaults. Qualitative findings suggest how and when discrimination-related experiences may threaten PTSD assessment accuracy, leading to over-pathologizing or under-detection of symptoms, for example (a) initial selection of a non-criterion A discrimination event as “worst event,” (b) linking symptoms to internalized transphobia (rather than trauma), and (c) linking victimization to gender identity/expression. Threats to PTSD assessment were more common when symptoms were linked to discrimination-based traumatic events, suggesting the importance of understanding contextual factors of index events. We offer a framework for understanding unique challenges to the assessment of PTSD among TGD people and provide recommendations for improving assessment.

Transgender and gender diverse (TGD) populations include people who do not identify with the gender assigned to them at birth, such as people who identify as transfeminine (e.g., trans women / women), transmasculine (e.g., trans men / men), nonbinary, gender fluid, and other non-cisgender identities. Community and national samples of TGD individuals report high rates of exposure to traumatic events including childhood abuse (47–55%; Reisner et al., 2014; Reisner, White Hughto, et al., 2016), intimate partner violence (26–54%; Human Rights Campaign Foundation, 2018; Reisner et al., 2014; Reisner, White Hughto, et al., 2016), physical assault (23–51%; Nuttbrock et al., 2015; Reisner et al., 2014; Shipherd et al., 2011; Testa et al., 2012), and sexual victimization (23–47%; Human Rights Campaign Foundation, 2018; Nuttbrock et al., 2015; Reisner et al., 2014; Shipherd et al., 2011; Testa et al., 2012). TGD people also experience pervasive discrimination and other minority stressors that may not be conceptualized as traumatic events (Bazargan & Galvan, 2012; James et al., 2016; Reisner, White Hughto, et al., 2016), as well as discrimination-based traumatic events, defined as discrete traumatic events that are linked to anti-TGD bias (e.g., targeted assault based on gender identity or transphobia). The majority (63%) of TGD individuals in a national sample reported experiencing a serious discrimination event such as job loss, eviction or homelessness, a lost relationship with partner or child, or denial of medical care (Grant et al., 2011).

Consistently high rates of exposure to trauma (Human Rights Campaign Foundation, 2018; Reisner et al., 2014) and discrimination (Bazargan & Galvan, 2012; Grant et al., 2011; James et al., 2016; Reisner, White Hughto, et al., 2016), as well as conceptualizations of health disparities positing that chronic marginalization increases vulnerability to physical and mental health conditions (Hendricks & Testa, 2012; Kaplan et al., 2015; Link & Phelan, 1995; Meyer, 2003; Taylor et al., 1997) would suggest higher prevalence of posttraumatic stress disorder (PTSD) in TGD populations relative to cisgender populations. However, there is wide variability in population estimates of PTSD among TGD adults based on differences in sample, measurement, and procedures. Studies using a PTSD screener developed for use in primary care (Primary Care PTSD screener; Prins et al., 2016) used with community-based samples found that 42–61% of TGD adults screened positive for probable PTSD (McDowell et al., 2019; Reisner, White Hughto, et al., 2016; Rowe et al., 2015). In another community-based sample, using the more comprehensive and stringent PTSD Checklist (PCL; Weathers et al., 2013), Shipherd and colleagues (2011) found that 18% of trauma-exposed trans women reported clinically significant PTSD symptoms. Only one study has used a structured clinical interview to examine the presence of psychiatric diagnoses in a TGD sample. Reisner, Biello and colleagues (2016) used the clinician-administered Mini-International Neuropsychiatric Interview (MINI; Sheehan et al., 1998) with a community sample of trans feminine young adults found much lower lifetime prevalence rates of PTSD (9.8%), with estimates similar to the general population (10.4% among women; Kessler et al., 1995).

The wide range of PTSD prevalence found in TGD samples suggest that some assessment practices may either under-detect PTSD symptoms or over-pathologize TGD individuals’ responses to trauma and discrimination. The studies that anchored symptoms to a specific traumatic event found more conservative rates of PTSD symptoms (Shipherd et al., 2011; Reisner, Biello et al., 2016) compared to studies that assessed symptoms generally. Anchoring symptoms to a specific traumatic event offers precision relative to measurement of symptoms without an event anchor (Priebe et al., 2018; Simpson et al., 2011), but may exclude valid PTSD symptoms that an individual associates more saliently with the discriminatory context in which the traumatic event occurred, rather than with the event itself even despite the relevance. Attempts to assess PTSD symptoms divorced from the context of discrimination not only risks under-detecting PTSD symptoms, but also may pathologize adaptive responses (Sloan & Shipherd, 2019). Particularly for discrimination-based traumatic events, an individuals’ response may meet the criteria for a PTSD symptom and be related to the anchoring event, but represent a functional adaptation that protects the individual from harm.

Conflation of trauma and discrimination (i.e., gathering information on the impact of discrimination when attempting to assess the impact of trauma) may also result in over-pathologizing responses to trauma and discrimination. The studies that did not link symptoms to a specific traumatic event found higher rates of PTSD symptoms (McDowell et al., 2019; Reisner, White Hughto, et al., 2016; Rowe et al., 2015) compared to studies that linked symptoms to a specific traumatic event, suggesting that higher rates reported by may represent general distress rather than PTSD symptoms (Priebe et al., 2018; Simpson et al., 2011). Both chronic and discrete experiences of discrimination raise the overall level of mental health burden to the individual (Díaz et al., 2006; Ryan et al., 2009) and may result in functional adaptation or responses (e.g., vigilance, avoidance) to protect the individual from or reduce the impact of discrimination. Further, controlling for trauma exposure, discrimination has been independently linked to PTSD symptoms severity and has garnered more attention in the extant research on minority stress and PTSD in recent years (Kirkinis, et al., 2018; McDowell et al, 2019; Reisner, et al 2016; Shipherd et al., 2019b; Williams et al, 2021). Mental health co-morbidity (Simpson et al., 2011; Valentine & Shipherd, 2018), and non-pathological responses to relative risk of violence and discrimination (Berke, et al., 2022; Livingston et al., 2019; Livingston et al., 2020) have the potential to inflate detection of PTSD symptoms, especially if symptoms are not properly linked to an index traumatic event (Priebe et al., 2018; Simpson et al., 2011). This may be particularly complicated when a traumatic event is discrimination-based. Conversely, dismissing discrimination-related distress during PTSD assessment could be detrimental to the patient and result in undervaluing the resulting symptoms (e.g., Shipherd et al., 2019a; Shipherd et al., 2019b).

Discrimination is salient context in which trauma occurs for TGD people, and understanding this context is key to understanding if a symptom is adaptive or pathological (i.e., accurate assessment) as well as informing treatment. Culturally responsive PTSD assessment ideally should offer nuanced evaluation that avoids over-pathologizing recovery from trauma in the context of discrimination, while simultaneously validating the distress. High quality assessment that attends to important contextual factors and appreciates the overlap and distinction between traumatic events and discrimination is needed to inform clinical treatment recommendations (Shipherd et al., 2019a; Shipherd et al., 2019b). Moreover, a better understanding of PTSD in the context of discrimination would be useful in conceptualizing and addressing the burden of distress experienced by marginalized communities, including chronic exposure to identity-based rejection, discrimination, and violence as well as threat of such exposure, efforts to conceal TGD identity, and internalized transphobia (Hendricks & Testa, 2012; Shipherd et al., 2019b).

To date, the gold-standard of PTSD assessment (National Center for PTSD, 2020), the Clinician Administered PTSD Scale (CAPS-5; Weathers et al., 2018), has not been examined in TGD populations. The CAPS-5 has been extensively examined in cisgender populations (Blake et al, 1995; Weathers et al., 2001), and is widely viewed as the benchmark for PTSD assessment in clinical, research and forensic settings (Weathers et al., 2018). Yet, due to complex intersections of chronic trauma and discrimination often experienced by TGD communities, even clinicians trained to administer the CAPS-5 without specialized training in TGD populations may be at risk of over-pathologizing or under-detecting PTSD. Thus, clinician evaluation of PTSD should be informed by the social context of both discrimination and trauma experiences (i.e., chronic discrimination, discrimination events, traumatic events, and discrimination-based traumatic events). Attention to the social context of TGD adults reduces the risk of pathologizing adaptive responses to threat (Carter, 2007), and allows for diagnostic precision.

This is the first study to report CAPS-5 scoring data among TGD adults, and the first to provide detailed qualitative assessment of the administration process to evaluate threats to over-pathologizing and under-detection of PTSD. Conventional content analysis of audio-recorded CAPS-5 assessments (N=44) aimed to (a) characterize how (and when) gender identity and discrimination enter the CAPS-5 assessment process, and (b) identify issues that may lead to over-pathologizing versus under-detection of PTSD. Based on these findings, we offer a framework for understanding unique challenges to diagnostic assessment of PTSD among TGD adults and provide recommendations on (a) how to address these challenges with assessment, and (b) how to assess for PTSD in a way that affirms the experiences of TGD people.

Method

Participants

Participants were TGD adults who attended an in-person study visit at an academic medical center in Boston, Massachusetts between December 2018 and March 2020. Inclusion criteria were self-identifying as any gender identity other than cisgender between the ages of 18 and 65¹. We aimed to recruit 50 participants to capture large enough sub-groups of participants with and without PTSD (estimated 40–50%) to conduct qualitative analyses (i.e., estimated that 20–24 participants needed to achieve saturation in developing a coding scheme). Trauma exposure was not required for participation. Of the 48 participants enrolled, 4 were excluded from analysis based current psychotic or cognitive symptoms interfering with the assessment (n = 3) and cisgender identity (n = 1). See Table 1 for more detail on sample characteristics.

Table 1.

Sample Characteristics (N=44)

	n	%
Sex assigned at birth
Male	17	38.6
Female	27	61.4
Self-identified gender identity
Trans man or man	11	25.0
Trans woman or woman	17	38.6
Genderqueer or nonbinary	10	22.7
Othera	6	13.6
Sexual orientation
Lesbian or gay	8	18.2
Straight or heterosexual	7	15.9
Bisexual or pansexual	16	36.4
Queer	11	25.0
Otherb	2	4.5
Ethnicity
Hispanic/Latinx	8	18.2
Not Hispanic/Latinx	36	81.8
Racec
Black	6	13.6
Asian	3	6.8
White	33	75.0
American Indian or Alaska Native	1	2.3
Otherd	3	6.8
Education
Completed high school or GED program	4	9.1
Attended some college but did not complete degree	15	34.1
Completed undergraduate degree	14	31.8
Attended some post-graduate training, but did not receive degree	5	11.4
Completed post-graduate training	6	13.6
Employment Status
Employed, full-time for wages	15	34.1
Employed, part-time for wages	5	11.4
Out of work	8	18.2
Retired or disabled	8	18.2
Missing	8	18.2
Income Level
Less than $25,000	13	29.5
$25,000 to $34,999	5	11.4
$35,000 to $49,999	10	22.7
$50,000 to $74,999	2	4.5
$75,000 to $99,999	1	2.3
$100,000 to $149,999	3	6.8
$150,000 to $199,999	2	4.5
$200,000 or more	0	0
Missing	8	18.2
Index Trauma on CAPS-5 (criterion A event)
Interpersonal violence (physical or sexual)	34	77.3
Sexual violence	18	40.9
Non-interpersonal violence	10	22.7
Motor vehicle accident	4	9.1
Life threatening illness or injury	4	9.1
Sudden death of loved one (illness)	2	4.5

Note. CAPS-5, Clinician Administered PTSD Scale for the DSM-5; GED, General Education Diploma.

^aSpecified responses included: agender/nonbinary (n = 1), gender fluid (n = 2), gender nonconforming (n = 1), nonbinary trans woman (n = 1), nonbinary man (n = 1).

^bSpecified responses included: biromantic asexual (n = 1) and not applicable currently (n = 1).

^cParticipants could select multiple racial identity categories.

^dSpecific responses included multiracial (n = 2) and West-Indian American (n = 1).

^eSpecific responses included: some college (n=2) and trade school (n=1).

Procedures

Participants were recruited using paper and electronic flyers distributed to local providers and clinics that specialize in TGD health, a tabling event at a local conference for TGD people and allies, and via a snowball sampling approach. All participants were provided with copies of recruitment materials to distribute to their networks. Recruitment materials used the following study name: “Transgender and Gender Nonconforming Trauma and Resilience Study.”² Recruitment materials specified that we would ask participants about experiences of trauma and discrimination. Interested individuals were pre-screened for inclusion criteria prior to scheduling a 3-hour in-person study visit. This manuscript focuses on demographic and CAPS-5 scoring and process data collected during a larger study visit that included completion of questionnaires and semi-structured interviews. All assessments were audio recorded, allowing for qualitative analysis of the interview process. Participants were remunerated $50 or $60 for their time, depending on time of enrollment (procedures were slightly modified mid-study resulting in higher remuneration). All procedures were approved by the Boston University Medical Campus Institutional Review Board.

Clinicians who administered the CAPS-5 were three doctoral-level psychologists with dual expertise in PTSD and sexual and gender minority health (2 were sexual or gender minorities). Clinicians completed a standardized web training on the CAPS-5 developed by the National Center for PTSD (NC-PTSD). The study PI completed a supplemental in-person training with NC-PTSD researchers to support training and supervision of the CAPS-5 for this project. The three study clinicians met weekly in the first 3 months of the study to establish reliability in scoring and consulted with NC-PTSD researchers as needed for clarification on scoring rules. After analyses, the research team assembled an expert panel to support interpretation of findings. The panel consisted of 3 clinical psychologists with expertise in treating PTSD in TGD adults.

Measures

Demographic Information

Age, gender identity, sex assigned at birth, sexual orientation, race, ethnicity, education level, and military service data were gathered using a questionnaire.

PTSD Diagnosis

The Clinician-Administered PTSD Scale for DSM-5 (CAPS-5; Weathers et al., 2018; American Psychiatric Association, 2013) was used to assess for current PTSD diagnosis based on symptoms present in the past month. The CAPS-5 is a structured clinical interview that includes assessment of PTSD criteria through identification of an index event involving direct or indirect exposure to actual or threaten death, serious injury, or sexual violence (criterion A); severity and frequency scores across four symptom clusters (intrusion, avoidance [criterion B], negative alterations in cognitions and mood [criterion C], and marked alterations in arousal and reactivity [criterion E]); determination of the duration of symptoms (criterion F), presence of significant distress or impairment (criterion G), and involvement of dissociative features. Trained CAPS-5 providers follow standard prompts and rating scales to assist in delineating PTSD symptoms from other types of psychological distress (e.g., sleep disturbance due to nightmares of trauma versus worry about work deadline) and promote standardization across interviewers and settings. Assessors were trained to validate a patient’s distress associated with all symptoms and capture the distress where appropriate diagnostically (e.g., PTSD, depression, etc.). Accurate assessment allows for determination of diagnosis and symptoms severity and is used to inform treatment conceptualization and planning.

The interview anchors assessment of the presence/absence and severity of each DSM-5 criterion to the index trauma. While items 1–8 are inherently trauma-related, the interviewer must determine the trauma-relatedness of items 9 and 11–20 as “definite,” “probable,” or “unlikely”. In a modified procedure for our study, clinicians also gathered information on whether the selected index trauma was discrimination-based (perceived by participant to be due to TGD identity). We used the probe, “In your opinion, was this event related or not related to your gender identity or expression? [If yes] How so?”

The CAPS-5 demonstrates strong interrater reliability (κ = .78–1.00) and test–retest reliability (κ = .83) for diagnosis, as well as high internal consistency (α = .88) and interrater reliability (ICC = .91) and good test–retest reliability (ICC = .78) for symptom severity scores (Weathers et al., 2018). The CAPS-5 also demonstrates good discriminant validity with assessments of anxiety, depression, and other behavioral health conditions (r’s = .02 to .54; Weathers et al., 2018).

Data Analysis

We used conventional content analysis (Hsieh & Shannon, 2005) to generate preliminary codes reflecting how and when TGD discrimination and other minority stress experiences entered the standard CAPS-5 assessment. We applied a team-based approach to developing a codebook (Patton, 2015), which started with individual review of audio recording to generate an exhaustive list of potential codes. Codes reflect how and when gender identity entered the clinical assessment, meaning that codebook development was directed by naturalistic observation of themes related to gender identity that participants brought up or that clinicians administering the CAPS-5 described in field notes, and organized by where in the assessment particular themes arose in terms of diagnostic criteria or symptom cluster. This approach to analysis of the CAPS-5 assessment is inductive (based on themes arising naturally during assessment) and pragmatic (allowing clinicians and those reviewing assessment field notes to identify where in the assessment and for what symptom clusters themes are relevant to). Moreover, our codebook development allowed for flexibility in capturing (1) themes on the social context of TGD people, and (2) themes on assessment issues that may lead to over-pathologizing or under-detection of PTSD in TGD people.

The coding team consisted of two ciswomen master’s level students with training in mental health counseling who served as primary coders as well as a nonbinary bachelor’s level research assistant and a ciswoman doctoral level psychologist as secondary coders (to provide clarification or resolve discrepancies). The team met weekly to discuss codes, and group codes into themes. We continued to review audio recordings until no novel observations (codes, themes) were presented. The final codebook listed 59 codes grouped into 15 higher level themes, and included ‘location’ codes so that we could determine which symptom clusters may be more important to providing affirming assessment. Codes were later grouped into two major threats to CAPS assessment (“validity” and “context”). Validity codes were those that may be problematic in terms of the accuracy of assessment of PTSD in this population. Context codes reflected understanding of social context of TGD lived experiences, and were relevant to rapport building and affirming assessment, yet were unlikely to impact scoring decisions. In terms of inter-coder agreement, we set a benchmark of mean of 80% agreement across >20% of interviews. Primary coders attained excellent agreement (91.6%) across 10 (22.2%) interviews prior to shifting to individual coding of the remainder of interviews (n = 44). The team met regularly throughout the single coding process to review progress and resolve discrepancies. Each time a code was assigned, a timestamp, as well as linked symptoms or symptom clusters, were documented. Timestamps representing strong examples of each code were selected for transcription.

In post-hoc analyses, we used t-tests to examine differences in validity and context codes between assessments anchored to discrimination-based versus non-discrimination-based traumatic events.

Observational and qualitative data and are not publicly available, however, the corresponding author welcomes inquiries. Syntax and qualitative codebooks utilized to conduct reported analyses are available upon request. This study is not a clinical trial, therefore the study hypotheses and analytic plan were not preregistered.

Results

CAPS-5 scoring data were used to characterize the sample (see Table 1 for demographic information and trauma event type; see Table 2 for scoring summaries). The mean age of participants (N=44) was 36.8 years (SD = 15.5). All participants (100%) reported trauma exposure on the CAPS-5. Twenty-one (48.8%) participants met criteria for PTSD diagnosis (see Table 2). Eighteen (41.9%) participants selected a discrimination-based traumatic event as their index event for the CAPS-5 assessment. Most participants selected assault (physical or sexual) as their index trauma (n = 34, 77.3%), with 40.9% (n =18) experiencing sexual violence (Table 1). Of note, although there was a significant difference (t(41) = 4.7, p <.001) in functional impairment between those who met criteria for PTSD (M = 2.9, SD = .56) and those who did not (M = 2.0, SD = .62). However, those without PTSD still reported moderate level of impairment.

Table 2.

CAPS-5 Summary

	Total (n = 43)		PTSD (n = 21)a		Possible Range of Scores
	M	SD	M	SD
Criterion A (n, %)
Yes	42	95.5	20	95.2
Probable	2	4.5	1	4.8
Criterion B (Items 1–5)
Severity	4.44	2.71	5.90	2.36	0–20
Symptom	1.81	1.33	2.52	1.03	0–5
Criterion C (Items 6–7)
Severity	2.67	2.1	3.76	1.58	0–8
Symptom	1.07	0.77	1.57	0.51	0–2
Criterion D (Items 8–14)
Severity	5.72	3.83	8.1	2.47	0–28
Symptom	2.21	1.67	3.33	1.11	0–7
Criterion E (Items 15–20)
Severity	4.37	2.57	5.95	1.94	0–24
Symptom	1.7	1.06	2.43	0.68	0–6
Criterion B-E
Total Severity	17.21	8.79	23.71	4.95	0–80
Total Symptom	6.79	3.82	9.86	1.98	0–20
Criterion G
Subjective Distress	2.7	0.71	3	0.55	0–4
Social Functioning	2.49	1.06	3	0.95	0–4
Occupational or other area	2.14	1.01	2.62	0.87	0–4
Global Validity	1.67	0.87	1.67	0.80	0–4
Global Severity	2.65	0.75	3.1	0.54	0–4
Dissociative Symptoms
Severity	0.51	1.03	0.57	1.12	0–10
Symptom	0.16	0.43	0.24	0.54	0–2
PTSD diagnosis specifiers (n, %)
With dissociative symptoms			4	19.0
With delayed onset (≥6 months)			3	14.3

Note. CAPS-5, Clinician Administered PTSD Scale for the DSM-5; PTSD, posttraumatic stress disorder.

^aCAPS-5 scoring data from one participant was excluded.

Table 3 reports our validity and context codes, and the proportion of the sample where a code was observed and when during the assessment the code was assigned (i.e., during which criterion A – E). For example, the code “describes how they were targeted because of their gender identity or expression [self-blame]” most commonly came up during the assessment of criterion A and criterion D. Table 3 also lists which criterion of the CAPS-5 assessment each code was associated with, including criterion A traumatic event, criterion B-E symptom clusters, and dissociation. The most common validity codes were (a) first selecting a non-criterion A discrimination event as “worst event,” (b) reporting symptoms, but linking to internalized transphobia rather than trauma; and (c) noting how gender expression may have increased risk for violence at time of the traumatic event. The most common context codes described a range of non-affirming experiences, including (a) social and occupational exclusion, and how these produced a (b) broad vulnerability to interpersonal violence and (c) poor adaptation following trauma. In Table 4, we present detailed clinical vignettes and participant quotes that exemplify validity considerations (i.e., those that may threaten assessment accuracy) and our clinical recommendations (developed in consultation with the expert panel) to address these challenges.

Table 3.

Proportion of CAPS-5 Assessments with Validity and Context Codes for TGD Persons, and PTSD Criterion Where Considerations Presented (N=44)

	n	%	PTSD Criterion
			A	B	C	D	E
Validity Codes
Initially names a non-criterion A discrimination event as index trauma (“worst event”)	26	57	x
Describes the presence of symptom, but did not link to index trauma (scored as “0-absent”)	16	35			x	x	x
Links gender expression to violence risk (feminine expression or ‘not passing’ increases risk)	11	24	x	x	x	x	x
Unable to disentangle discrimination- v. trauma-related symptom expression, even after probing	10	22	x	x	x	x	x
Social or occupational impairment is higher than expected based on PTSD symptom severity	9	20				x
Describes how they were targeted because of their gender identity or expression (self-blame)	8	17	x			x
Reports the presence of negative internalized beliefs about self (gender-related)	8	17	x			x
Observation that ongoing minority and general stressors impair participant’s ability to link symptoms to discrete traumatic event	6	13	x	x	x	x	x
Describes symptom onset as predating traumatic event, in relation to discrimination	5	11				x	x
Comments that chronic/pervasive discrimination is more impactful than a discrete traumatic event	5	11	x
Describes inappropriate guilt related to “burden” of their gender identity on their social supports	5	11	x			x
Links media portrayal of discriminatory policies (e.g., military ban, bathroom bill) to symptoms	5	11		x		x
Links media portrayal of violence against trans women to symptoms	4	9	x			x	x
Describes gender dysphoria “body detachment” when asked about traumatic dissociation	4	9
Links global fears (not based on personal history) about transphobic violence to symptoms	3	7				x	x
Describes verbal assaults (e.g., slurs) as trauma reminders	3	7		x	x
Context Codes
Features of the trauma
Traumatic event was a discrimination-based event based on gender identity or expressiona	18	41	x
Perpetrator of traumatic event was family member (family of origin)a	13	28	x			x
Perpetrator of traumatic event was peer or romantic partner	12	26	x
TGD community was the context for interpersonal violence (TGD-identified peer or partner)	6	13		x		x
Structural discrimination and social determinants
Describes difficulties with employment	22	48	x	x		x	x
Describes difficulties with housing loss, homelessness, and household conflict	14	30	x	x		x	x
Ongoing crises impair ability to report on trauma-related distress (hierarchy of needs)	6	13	x		x	x
Exclusion and rejection
Describes experiences of familial rejection as ‘traumatic’ or linked to symptomsb	16	35	x		x	x
Describes experiences of peer rejection as ‘traumatic’ or linked to symptoms	9	20	x	x		x
Describes experiences of transphobia within one’s racial/ethnic group or religious community	7	15	x		x	x
Describes isolation, exclusion, or avoidance of TGD community post-trauma (when this was context for violence)	3	7	x		x
Describes complicated bereavement when loss of someone who was not gender affirming	2	4	x
Decision to disclose trauma to social supports was based on how affirming these supports were of gender identity	2	4		x	x
Gender dysphoria and gender affirming services
References gender affirming talk therapy when discussing symptoms	6	13			x	x	x
References gender affirming surgical care when discussing symptomsb	5	11	x			x	x
References gender affirming medical care (e.g., hormone therapy) when discussing symptomsb	3	7		x			x
Describes distress associated with reminders that body does not match gender identitya	3	7	x	x		x	x

Note. CAPS-5, Clinician Administered PTSD Scale for the DSM-5; PTSD, posttraumatic stress disorder; TGD, transgender and gender diverse.

^aLinkage to discrimination event was determined by additional probing questions; standard assessment of criterion A on the CAPS-5 only revealed discrimination for 8 (17.8%) of participants.

^bLinked to dissociation sub-type items.

Table 4.

Detailed Description of Validity Challenges and Recommendations for Affirming Assessment.

Challenge	Example	Recommendations
Initially names a non-criterion A discrimination event as “worst event”	Participant refers to “worst event” of harassment and discrimination (including “being outed”) in the workplace due to their gender identity and presumptions of sexual orientation: “…my bosses pulled me into the office…they asked…’are you gay?’, ‘are you bisexual?’…[so] they get on their walkie talkies…and said [I’m] gay…and all throughout [the state]…people…didn’t want to work with me anymore…when they found out I referred to myself as “her” they said ‘well I don’t want to work with him, he might have AIDS…ever since then my life went downhill” (60-year-old woman).	Allow space to discuss extremely stressful events to establish necessary rapport for assessment.
		Validate the distress associated with non-criterion A discrimination events prior to narrowing to a discrete traumatic event.
		This information is important for case conceptualization and treatment mapping. For example, treatment may focus on distress associated with a pervasive hostile work environment.
Traumatic event was discrimination-based	Participant has perception that they were targeted for sexual abuse in childhood (age of 6) due to their gender expression. The participant described how family members encouraged interaction between this participant and the perpetrator, so the participant could learn how to behave more traditionally masculine (“macho”, violent): “In the South, you’re taught to be this little boy. If you’re anything outside of that, it’s not the normal. A lot of times because I was not ‘normal’—[I] didn’t even know what I was--I was exposed ..to a lot of different people that [my family] may have thought could be [traditional male] role models. One [of these men] molested me” (33-year-old trans woman).	Conduct assessment as usual with discrimination-based traumatic event.
		Acknowledge universal risk factors for violence (predators focus in on the socially isolated or excluded, align with parents to gain access, engage in grooming behaviors).
		Acknowledge unique risk factors for violence, being targeted based on gender expression or some other aspect of identity.
Participant described the presence of symptom, but did not link to index trauma (scored as “0-absent” in standard administration)	Participant described engaging in avoidance of people (and confined spaces) due to gender-related harassment and perceived physical threat. However, participant links avoidance to proximal threats, not to distal trauma: “I’m trying to use the stairs in my building…I know it sounds crazy but I…live on the tenth floor, take the elevator to the eighth floor, then I walk around, try to get to another set of elevators…but it helps me to feel more relaxed…[with the landlord and building residents], bullying is happening, discrimination, …all those crazy things that I’m doing work” (48-year-old trans woman).	Attempt to collaboratively parse out which event memory is being re-experienced, and leading to symptoms (discrimination event v. discrimination-based traumatic event v. other traumatic event).
		Err on the side of linking to trauma, when there is reasonable impression that current discrimination is triggering distal trauma. Note that clients may not perceive these linkages.
	Another participant describes being unable to disentangle whether mistrust of cis women is linked to distal trauma (cis woman perpetrator) or discrimination by ciswomen during her gender transition: “I have a general mistrust of women. I had an attacker that was a [cis]woman, … [when] I was developing [friendships with ciswomen] it was very hard to trust…I felt like I had this “male privilege” attached to me that I didn’t really ask for…[I experienced] cattiness [from ciswomen] that I didn’t expect when I was transitioning” (36-year-old woman).
Participant links gender expression to overall risk for violence (i.e., feminine presentation or ‘not passing’ increases risk) Social and occupational impairment may be higher than expected, and not driven by trauma alone.	Participant refers to making choices in gender expression, i.e., appearing more masculine in order to avoid risk of violence: “I need to be as dangerous-seeming as possible; I need to be as loud as possible… There’s also an element of I need to be, uh, as masculine as possible. I need to respond to this in a low, loud, threatening voice. … I need to follow this person [in a threatening manner]. I can’t let them attack people” (24-year-old nonbinary person).	Validate choices around gender expression that may improve actual safety (not avoidance) while also not reinforcing overgeneralized fear or sense of control over others’ behaviors.
		Consider the relative costs and benefits of concealment across contexts. Consider developmental context, and that gender expression development can influence social or occupational exclusion or engagement.
	Participant experiences functional impairment in terms of completing tasks, managing finances, and gaining employment due to concentration problems, as well as limited relationships with family and past friends: “I always [have problems with concentration]…in terms of…trying to apply for jobs, and dealing with responsibility type-stuff, like trying to get food stamps and getting them all the documentation they need.” “It’s nothing new, but I am cut off from my family, so it’s something that’s always there and I do think about pretty regularly…I’ve definitely cut out a lot of friends, even before I came out as trans…I definitely had friends that were transphobic or queer-phobic” (29-year-old nonbinary person).
		Consider the contribution of discrimination to social and occupational impairment.
		Trauma, gender identity development, and discrimination exposure may happen on different timescales, and temporal sequencing can be relevant in contextualizing assessment of functioning.
	Another participant refers to coming out as transgender later in life and experiencing rejection by their wife, who physically assaulted them when they chose to seek gender affirming medical and surgical care: “My late wife and I were separated, we were in constant battles over my transition… I was in denial, I wanted to still be married to her and could not understand why she could not love the soul of the person just because the outside had changed” (55-year-old woman).
Participant reports the presence of negative internalized beliefs about self (gender- related)	Participant describes negative beliefs about self that are linked to internalized transphobia as well as discrimination, harassment, and interpersonal violence (discrimination events): “I know I’m not doing anybody any harm, I know [being transgender] is intrinsic to who I am, but to be called bad things still has [me] seriously think…maybe I am a bad person. [These thoughts] don’t go away…when you’re called a sexual degenerate” (65-year-old trans woman).	Affirm that societal discrimination and stereotypes are inaccurate and harmful.
		Treatment may additionally focus on reducing internalized transphobic beliefs.
Observation that ongoing minority stressors impair participant’s ability to explicitly link symptoms to discrete traumatic event	Participant describes how stress related to housing discrimination (based on gender identity) and threat of homelessness make it difficult to link trauma symptoms to distal (trauma) stressors: “Right now, I always feel that I’m in danger, but that’s just because I don’t have a home anymore… it’s not about the assault…you just have to be careful about anybody” (44-year-old trans woman).	Pause assessment and address client’s experience of crisis and provide resources and advocacy steps.
		Validate that general stress and immediate crises take priority [hierarchy of needs], and may inhibit the ability to get an accurate assessment of PTSD at this time. Plan to reassess PTSD symptoms.
Participant describes how self-blame cognitions reflect internalized transphobia	Participant describes how family of origin espoused homophobic and transphobic beliefs, which led them to think they were “bad” or there was “something wrong with them.” So trauma was contextualized from the frame that they must have done something to deserve it (self-blame): “I was thinking that [the sexual assault] is my fault…[because] in my culture I remember my mom saying that [she] would prefer to see [me] dead than to see [me] gay” (48-year-old trans woman).	Due to discrimination and rejection histories, TGD individuals may be more likely to express internal attributions of blame.
Participant describes symptom onset as predating traumatic event, in relation to discrimination	Participant described extreme negative beliefs related to trust that originate from gender-based discrimination and predated the traumatic event. Participant also notes how trauma confirmed and strengthened this belief: “ [Negative beliefs about self and others] were reinforced by [the sexual assault]; they didn’t change, there was just more evidence” (30-year-old trans man).	Affirm that societal discrimination and stereotypes are inaccurate and harmful, and may naturally make it difficult to trusting others.
		Further questioning needed to parse out linkage to prior discrimination vs linkage to trauma.
		Focus of treatment would be on the utility of these cognitions rather than accuracy.
Participant comments that chronic/pervasive discrimination is more impactful than a discrete traumatic event	Participant describes the burden of ongoing persistent discrimination, rejection, and exclusion on mental health as a larger contributor to trauma-like symptoms (over discrete traumas): “The worst thing that has actually happened to me is just the more general existing as being trans, it’s not like one particular event…it’s kind of a slow trauma that builds up that’s just recognizing the fact that that most people in the world kind of despise me, and there’s a lot of people who would kill me if they could” (27-year-old trans woman).	Validate.
		Relevant for case conceptualization and treatment planning.
Participant describes inappropriate guilt related to “burden” of their gender identity on their social supports (e.g., family)	Participants describe beliefs that they are burdensome to their families (due to TGD identity), and describe intense guilt and shame during assessment: “It’s blaming myself for things I shouldn’t blame myself for…it’s guilt. … She loves me, she supports me, but she also can’t handle it [my gender identity/expression] at the same time. And so she’ll say ‘No no no, be yourself at Christmas’ and all that, but I know that’s going to destroy her if I do. And then I feel guilty for even asking about it” (32 year old nonbinary trans woman).	Consider relevance of gender identity to guilt and shame, and believes that one is burdensome to others.
		Acknowledge that concealment has pros and cons and is context dependent (and based on the likelihood of affirming responses from others).
	Another participant describes: “[I] struggle with shame around being trans and things with my family and not being out to my family and feeling like I am hard to love or I am burdensome…” (25-year-old nonbinary person).	Differentiate concealment due to guilt or shame from adaptive concealment for safety, convenience, or emotional self-preservation.
Participant links media portrayal of discriminatory policies (e.g., military ban, bathroom bill) to symptoms	Participant links recent news reports of discriminatory government policies against TGD people to negative cognitions about others and the world (trust, safety): “I served my country honorably, I didn’t change my gender until after. But now [the president] wants to kick all the ‘tranny’s’ out of the military, and we’re good soldiers…[Hearing about] continual…abuse [coming] from the highest office in the land is very triggering…[I feel] anger at [the] discrimination I am experiencing from my own government…and it is impossible to get away from” (60-year-old trans woman).	Understand as systemic invalidation from credible institutions.
		Consider relevance of invalidation in their traumatic event to determine linkage to trauma. May be a trauma reminder.
		Acknowledge that minority group related threat is both an additional stress exposure and a potential trauma reminder.
Participant links media portrayal of violence against trans women to symptoms	Participant describes strong negative beliefs about safety precipitated by personal experiences with police violence as a trans woman and hearing about acts of violence committed against other trans women in the context of help-seeking: “The world is completely dangerous, because more and more and more, I’m starting to see when girls like me, not just [women] of color, [but] trans women… [go for] help, when they want to talk, I would see a picture or I would hear like a death or a murder… [then] I’m in this new ocean of shit [referring to own re-experiencing symptoms of police mistreatment] …It could be just seeing a police car, any type of thing [on social media] …it’ll give me a [bad] feeling…I’ve [moved up] appointments [because I was worried about being out at night] …you know, I’m constantly worried” (33-year-old trans woman).	Understand as systemic invalidation from credible institutions.
		May be a trauma reminder and exacerbate symptoms, but is likely not a stand-alone traumatic event.
		Acknowledge that minority group related threat is both an additional stress exposure and a potential trauma reminder.
	Participant describes media portrayals of murders of TGD people as traumatic when assessing trauma exposure: “Seeing human suffering on the news and the TV, understanding the human suffering that the gender [diverse] community is going through still to this day is traumatic…an average of one [TGD person] a month is being killed…” (65-year-old woman).
Participant described phenomenon of gender dysphoria “body detachment” when asked about traumatic dissociation	Participants may describe body detachment experiences that are part of the clinical presentation to gender dysphoria (with gender-based triggers). These responses may or may not be linked to trauma reminders: “I feel…pretty good about my body generally. I recently got top surgery and am on hormones. Like, it’s different now than before I medically transitioned. It was a lot more difficult with the experience of hovering over myself and like, watching myself…But that happened more when I was misgendered, and I don’t get misgendered really that much anymore” (25-year-old nonbinary person).	Clinically differentiate “body detachment” related to gender dysphoria from dissociative responses to trauma reminders.
		Body detachment in the context of gender dysphoria can be adaptive when used with other coping strategies.
Participant links global fears (not based on personal history) about transphobic violence to symptoms	Participants without personal histories of violence may endorse global fears (vigilance) of transphobic violence: “I feel like it’s on guard against, like, I don’t know, harassment, and stuff like that, even though I don’t receive that much harassment…I’m very, very vigilant of, like, how strangers are looking at me. When I’m standing waiting for the train, like, I’m afraid someone will kick me in front of the train or something, um, even though I have no evidence someone would do that…I’m kind of just always watching people’s eyes, and like, if someone is like, frowning at me…it makes me alert and uncomfortable in a way that’s probably unwarranted” (27-year-old trans woman).	Avoid over-pathologizing responses. For example, threat perceptions may be accurate and reflect adaptive vigilance, rather than pathological “hypervigilance.”
		Acknowledge realistic risks to safety for TGD people.

Note. PTSD, posttraumatic stress disorder; TGD, transgender and gender diverse.

We found that CAPS-5 assessments of discrimination-based traumatic events (versus other traumatic events) had more validity codes (M = 3.83, SD = 2.07 v. M = 2.12, SD = 1.96, t(41) = 2.76, p = .01). There were no differences in frequency of context codes for discrimination-based versus other traumatic events (M = 3.56, SD = 2.31 v. M = 2.68, SD = 1.63, t(41) = 21.44, p =.16). This finding suggests that gender identity is most relevant to accurate PTSD assessment when the event is discrimination-based (i.e., identity is related to the trauma context), and that a style of assessment that affirms TGD experiences is equally necessary regardless of trauma type.

Discussion

This is the first study to publish on threats to quality PTSD assessment using the CAPS-5 for PTSD with a TGD sample (N = 44). All participants reported at least one lifetime traumatic event. We found that nearly half (48%) of participants met criteria for PTSD. Our PTSD estimate is in the high end of the range of previous estimates that relied on the PC-PTSD (42 – 61%; McDowell et al., 2019; Reisner, White Hughto, et al., 2016; Rowe et al., 2015), and much higher than prevalence based on PCL (18%; Shipherd et al., 2011) and clinician-administered MINI interview (10%; Reisner. Biello et al., 2016). In part, the variability in PTSD diagnosis estimates are likely multiply determined, including the sample studied (clinical versus non-clinical) and whether the PTSD symptom assessment strategy clearly anchored the symptoms to a traumatic event. Some self-report measures of PTSD (e.g., PCL) perform well in the general population when appropriately anchored to a traumatic event; whereas there is limited data on some interview approaches (e.g., MINI) to the assessment of PTSD (National Center for PTSD, 2020). The wide variability in PTSD symptom reports in the literature underscore the need for additional study of high quality assessment of PTSD symptoms in the study of trauma recovery in TGD samples.

CAPS-5 assessments were anchored to discrimination-based traumatic events 41% of the time, which is consistent with prior work examining exposure to discrimination-based traumatic events (e.g., 42%; Shipherd et al., 2011). Of note, this rate emerged only when directly queried; without direct probing only 18% of events were described by participants as discrimination-based. Thus, for a large minority of trauma-exposed TGD people, discrimination was a characteristic of the index trauma (i.e., part of the trauma context) however this linkage was only made upon culturally competent inquiry. Our qualitative analysis further elucidated how integral discrimination is to the trauma recovery process, and how similar clinical features may arise from both discrimination and trauma exposure. This overlap in clinical features can complicate assessment and treatment planning (Livingston et al., 2019; Livingston et al., 2020; Shipherd et al., 2019a). Given that validity concerns were more common when assessments were anchored to a discrimination-based traumatic event, we argue that understanding this aspect of the trauma context is essential to accurate and affirming assessment of PTSD.

Our content analysis of CAPS-5 assessments highlighted key conceptual, clinical, and assessment challenges relevant to TGD populations. We provided our recommendations to address some of these challenges to assessment (see Table 4 for the complete list). For example, we believe that it is important to: (a) Validate all extremely stressful experiences of TGD people (even when not a traumatic event) and remain collaborative across the assessment, conceptualization, and treatment planning process. (b) Ask additional probing questions to try to disentangle responses to discrimination versus trauma. (c) Given high rates of multiple traumas in a lifetime (Shipherd et al., 2011), err on the side of rating a symptom as related to traumatic events when discrimination is exacerbating prior trauma response by using trauma-relatedness probes: “Did [symptom endorsed] get worse after [index traumatic event]?” For populations with high trauma exposure (multiple events, multiple perpetrators) and experiences of discrimination that can lead to PTSD or PTSD-like symptoms, the link between proximal distress and the exact index trauma can get muddled (Salomaa et al., 2022). This is not to say that symptoms are not related to index trauma events, but that the overlay of repeated, chronic, and pervasive interpersonal harms can make it difficult (or unreasonable) for interviewees to make the link between the index trauma the symptom (see Table 4). (d) Understand that vigilance in the presence of actual risk for violence is not hypervigilance. (e) Acknowledge that discrimination and trauma may have shared risk and recovery pathways, as well as unique features. Treatment planning may involve extending existing treatments for PTSD to address discrimination-based stress (e.g., reappraisal of esteem and blame cognitions).

Our findings add to conceptual work that aims to understand distinct and overlapping experiences and symptoms of trauma and minority stress among TGD people (Livingston et al., 2019; Livingston et al., 2020; Salomaa et al., 2022; Shipherd et al., 2019b). This is the first study to examine aspects of this conceptual model as they arise during the gold standard assessment of PTSD. We extend this work by providing concrete recommendations of how to apply this conceptual framework to the assessment process. Beyond PTSD assessment, our findings highlight the importance of cultural competence in working with TGD people, including understanding key contextual factors. Some issues that arose during the CAPS-5 assessment were violence-vulnerability factors such as rejection and victimization within familial and peer groups and discrimination in housing and employment (which led to increase vulnerability for violence in the context of homelessness and poverty; see Table 4). Contextual factors common in TGD adults (regardless of PTSD diagnosis) may explain why subjective distress, and social and occupational impairment were uniformly high across participants. These findings suggest the importance of cultural competency training in assessment and clinical practice with TGD adults. We refer the reader to other works on cultural competency in working with TGD populations (e.g., “Adult transgender care: An interdisciplinary approach for training mental health professionals”; Kauth & Shipherd, 2018)

Together, these findings suggest that, in a trauma-exposed sample, even participants who did not meet PTSD diagnostic criteria were experiencing high distress and impairment. This is further supported when comparing the education level of our participants to their income and employment information. In Massachusetts the average annual household income is $79,835 (Data USA; 2021), only 6 participants (13.6%) reported income in that range; whereas many participants were roughly half that income level. During the timeframe of the current study (December 2018 to March 2020) the unemployment rate for Massachusetts ranged from 2.7% to 3.3% (U.S. Bureau of Labor Statistics, 2021), while 13.6% of our sample reported being out of work. In a sample where the majority of participants had some college education, their underemployment and comparatively low levels of income are notable.

Constraints on Generality

Although this study addressed a gap in the literature by considering best practices in PTSD assessment of TGD adults, there are limitations to highlight. This study included a small number (N = 44) of heterogeneously identified TGD people (e.g., trans men, trans women, genderqueer, nonbinary, etc.). We collapsed findings across identities, which may have masked potential sub-group differences. The sample was relatively homogenous in age at 37 years (± 15). Moreover, this sample was predominately non-Hispanic white and recruited at an urban medical center with mostly some college education. Challenges to the assessment of PTSD among people with other intersecting minoritized or marginalized identities (racial minority, rural, elderly, etc.) were not considered in this evaluation, but are worthy of future study given similar structural inequities that may drive PTSD disparities and complexities of assessment. The study benefitted from snowball sampling and recruitment near a safety net hospital clinic for TGD people. However, like all studies on this topic, it was a community sample of convenience. These recruitment strategies may have inadvertently drawn a sample with worse mental health due to medical engagement and peers who were aware of their trauma experiences. Future studies could address these limitations by leveraging more nuanced sampling strategies.

Clinical Implications

Our findings suggest that TGD adults experience a range of extremely stressful experiences, including experiences that are chronic and pervasive as well as discrete severe events that are relevant to PTSD assessment, case conceptualization, and treatment planning. Discrete traumatic events lend themselves to traditional methods of assessment for PTSD, however, upon assessment, our participants often reported that chronic and pervasive experiences of invalidation and harm were significant drivers of mental health distress. Indeed, equal levels of subjective distress and impairment were expressed by this trauma-exposed sample, regardless of PTSD diagnosis. Given this, clinicians working with TGD individuals must operate from an integrated conceptual framework, and gather information pertaining to both discrimination and trauma experiences that are common in this population. Additionally, we recommend that clinicians asked directly about whether index events were discrimination-based as this information is relevant to the evaluation of clinical features that follow. In our study, we found that standard criterion A assessment only identified that events were discrimination-based 18% of the time (compared to 42% when we directly asked). This discrepancy highlights the importance of clinician probing, rather than relying on respondents to offer this connection spontaneously.

Similar to PTSD assessment, discrimination assessment elicits information about stressor exposure and associated distress. We recommend that clinicians work collaboratively with clients to determine if gender identity is related (or not related) to presenting clinical concerns. This includes asking probing questions during PTSD assessment to understand linkages between minority stress and endorsed clinical features. Lack of systematic, thorough, and gender affirming assessment can lead to both over-pathologizing as well as under-detection of mental health diagnoses. Additionally, an integrated case conceptualization is necessary for treatment selection and sequencing, as relevant interventions for this population may include singular intervention (e.g., PTSD treatment provided in affirming way), combined treatments (dually addressing PTSD and discrimination-based stress), or adjunctive therapies (addressing discrimination-based stress).

Future Research

Future research is needed to refine integrated conceptual models for discrimination and traumatic stress. Most importantly, there is a need to engage TGD people in all aspects of the research process, to ensure that this work is guided by those who stand to benefit from research products, including valid and affirming assessment of PTSD. Additionally, future studies may aim to conduct formal validation research on the CAPS-5 among diverse TGD populations (e.g., diverse in gender identity, race, age, geography). Finally, clinical research is needed to establish best practices for PTSD treatment among TGD populations. Despite the lack of empirical testing, there is no reason to believe that TGD people cannot benefit from current evidence-based treatments for PTSD, however, our research and clinical experiences suggest that these treatments may need to be adapted to dually address responses to discrimination (especially those that have shared risk and recovery pathways with trauma).

Conclusion

Our study has contributed to the field by reporting scoring and process data gathered through the administration of the CAPS-5 with a sample of TGD adults. Our findings highlight how common trauma exposure is among TGD people, given that 100% of our sample reported exposure to at least one traumatic event, despite this not being an inclusion criterion. Over 40% of CAPS-5 assessments were anchored to a discrimination-based traumatic event. Approximately, half of the sample met criteria for PTSD diagnosis. Our in-depth content analysis provided detail on exactly how and when in the assessment gender identity and minority stress experiences were more likely to arise. Based on the issues identified, we provided our own expert recommendations on how to administer the CAPS-5 in a way that is rigorous while also affirming the lived experienced of TGD people.

Public Significance Statement.

We are the first to characterize the complexities of assessing posttraumatic stress disorder (PTSD) among transgender and gender diverse (TGD) adults, using the gold standard clinician interview assessment. All participants reported trauma exposure, with 48% meeting PTSD diagnostic criteria. Our study details how and when discrimination-related experiences may threaten PTSD assessment accuracy, leading to over-pathologizing or under-detection of symptoms. We also provide recommendations for assessing PTSD in TGD adults.