Abstract
Objective:
Transgender, non-binary, and gender expansive individuals experience disproportionately high rates of mental health symptomatology and face unique barriers to accessing care (eg, concerns about experiencing non-affirming care). This project aimed to address these demonstrated barriers by integrating gender-affirming psychiatric care within primary care.
Methods:
We established an embedded gender-affirming psychiatric care clinic within a large, urban, academic family medicine practice. Transgender, non-binary, and gender expansive adults were referred by their primary care provider for psychiatry evaluation and, if appropriate, psychotropic medication management and support connecting to outpatient psychotherapy. Patients (n = 17) completed validated measures of depression, anxiety, and working alliance at pre-treatment, each session, and post-treatment. Primary care providers (n = 6) completed measures of feasibility, acceptability, and appropriateness. We computed descriptive statistics and used Wilcoxon Signed Ranks Tests to explore changes in symptoms from pre- to post-treatment.
Results:
Patient depression and anxiety symptoms significantly improved from pre- to post-treatment. Patients utilized telehealth more often than in-person appointments. Primary care providers approved of the pilot clinic model and desired expanded offerings, particularly in-house psychotherapy.
Conclusions:
This pilot study supports the acceptability, feasibility, and preliminary effectiveness of integrating gender-affirming psychiatric care within primary care. Further research that is sufficiently powered to detect effects is needed.
Keywords
Introduction
Transgender, non-binary, and gender expansive (TNG) individuals experience disproportionately high rates of psychiatric and neurodevelopmental differences including depression, anxiety, drug and alcohol misuse, post-traumatic stress disorder, eating disorders, autism, and attention-deficit hyperactivity disorder (ADHD).1,2 Of particular concern, rates of suicidal ideation and behaviors are markedly higher in this population compared to cisgender, heterosexual peers. 1 In the U.S. Transgender Survey, 40% of over 27 000 adult TNG participants endorsed prior suicidal behavior(s)—10 times the rate in the general population. 3 Such glaring mental health disparities can be conceptualized using Hendricks and Testa’s Gender Minority Stress Model, 4 which expands upon prior work by Brooks 5 and Meyer 6 pertaining to sexual minority individuals. The Gender Minority Stress Model names unique stressors that can negatively impact TNG mental health, categorized as “distal stressors” (external; eg, discrimination, victimization, rejection, structural stigma, nonaffirmation, invalidation) and “proximal stressors” (internal; eg, gender dysphoria, vicarious stress, mental and emotional labor related to gender identity).
Despite high need, many TNG individuals do not receive mental health services. 7 The Gender Minority Stress Model helps explain the observed low prevalence of mental health service engagement among TNG individuals. Population-specific access barriers that prevent traditional mental health services from meeting patient needs include cost, fearing non-affirming clinicians (eg, discrimination, poor cultural responsiveness, low population-specific health knowledge), and non-inclusive health system infrastructures (eg, electronic health record systems, clinic facilities).3,7,8 Such harmful clinician interactions are concerningly common—33% of respondents to the 2015 U.S. Transgender Survey who had seen a medical professional in the past year reported discrimination in those interactions 3 —and far from benign, as those who experience clinician discrimination are more likely to engage in suicidal behaviors. 9 Conversely, positive experiences accessing primary care were significantly associated with increased future healthcare engagement, increased confidence in navigating healthcare generally, and decreased psychological distress.10,11 Additionally, perceived clinician affirmation has been shown to mediate ongoing engagement in mental health treatment. 12 Taken together, these findings suggest that TNG individuals need clinical services tailored to their unique needs, but that such services are infrequently available and difficult to access.3,7,13
Integrating TNG-specific mental health care with other gender-affirming health services in accessible contexts like primary care could alleviate population-specific access barriers (eg, fearing mistreatment, finding TNG-affirming clinicians) and increase the likelihood of initiation and ongoing engagement in mental health care.14-16 Integrated care models have been shown to reduce access barriers because they are often conveniently located for patients and less costly to access relative to specialty mental healthcare. They have also been robustly demonstrated to improve mental health outcomes in both general 17 and marginalized 18 populations. However, little published data to date evaluates such models with TNG adults. 19
This study aimed to address barriers to engagement in mental health care for TNG individuals by integrating gender-affirming psychiatric care with gender-affirming primary care, informed by our previous work eliciting patient and provider preferences and needs. 16 By conducting this pilot exploration of an integrated, gender-affirming care model, we specifically sought to explore preliminary implementation and effectiveness data to understand feasibility and inform potential modifications to the integrated care model in preparation for a future fully-powered trial.
Methods
Context
All research activities were approved by the University of Pennsylvania IRB. The pilot clinic was instantiated within a single large, urban, outpatient family medicine practice affiliated with the University of Pennsylvania Health System. The practice is staffed by 60 clinicians, including resident and attending physicians and nurse practitioners. Study activities occurred between July 19, 2023 and April 17, 2024. The integrated gender-affirming psychiatric pilot (IGAP) clinic was staffed by a resident psychiatrist (TGG) and a psychiatry/family medicine double-boarded attending psychiatrist (RK). Appointments were offered in person and via telehealth.
Participants
Patient participant eligibility criteria were: 1) 18+ years old, 2) TNG, 3) enrolled in primary care services at the practice, and 4) open to psychiatric evaluation and medication management. Exclusion criteria were: 1) only psychiatric diagnosis is mild depression or mild anxiety (eligible for existing integrated mental health services in the practice), and 2) exclusively seeking an ADHD or autism evaluation or Social Security disability paperwork.
Primary care providers (PCPs) who had engaged with the IGAP clinic were invited to complete brief surveys. PCP respondent eligibility was confirmed by 1) verifying clinical role, and 2) eliciting the number of interactions with the IGAP clinic (eg, direct patient care, formal referrals, discussions about potential referrals, and consultation about specific patients, culturally responsive approaches to care, and related clinical topics) and excluding respondents reporting “0” interactions.
Clinical Model
Initial IGAP clinic intake appointments were scheduled for 90 min, and follow-up appointments were scheduled for 30 min. During intake appointments, the IGAP clinician explored the presenting concern, conducted a review of psychiatric symptoms, administered the Patient Health Questionnaire-9 (PHQ-9) 20 and the Generalized Anxiety Disorder-7 (GAD-7) 21 survey, verified medical history, and elicited a psychiatric history (diagnoses, care, treatments), substance use history, gender-affirming care history, family psychiatric history, and social history. Then, the clinician shared their working understanding of the patient’s experience in terms of diagnostic labels, elicited patient feedback on that formulation, presented the patient with recommended workup and treatment options (including referral to psychotherapy, as needed), collaboratively refined an initial treatment plan, and addressed any remaining patient questions.
At subsequent appointments, the clinician reviewed patient goals, verified medication adherence, reviewed test results (if applicable), and elicited updates on symptoms, medication side effects, psychosocial stressors, substance use, and gender-affirming care since the prior appointment. The PHQ-9 20 and GAD-7 21 were also administered at each encounter. Appointments were scheduled at clinically indicated intervals (ranging from 2 weeks to 2 months), and the transition back to PCP-only management was offered and discussed with patients after multiple appointments with no medication changes and stable symptomatology. While patients received care in the IGAP clinic, messages were exchanged with PCPs for coordinating care as indicated. Transition to PCP-only follow-up included a message to the PCP summarizing the care received in the IGAP clinic and the medications that will become the PCP’s responsibility for further management. The PCP was invited to contact the IGAP clinicians with questions at any point and to return the patient to IGAP clinic care if indicated in the future.
Study Procedures
PCPs were notified about the IGAP clinic and study eligibility criteria in a recurring administrative meeting and by email. To make a referral, PCPs sent a potential patient participant’s electronic health record chart with their consultation question(s) to the IGAP clinician study team members to review; follow-up messages were exchanged as needed for clarification. Then, the study team confirmed eligibility and interest in participation with the patient by phone prior to scheduling an intake appointment.
PCPs received $5 (Visa ClinCard) as a token of gratitude for completing brief surveys about their experience with the clinic 6 months after the pilot clinic was launched. All collaborating PCPs (attending and resident physicians) had continuity panels and continued seeing patient participants for primary care during and after study participation.
Measures
Psychiatric Symptoms
Patients rated depression and anxiety symptomatology at each IGAP appointment via the PHQ-9 20 and the GAD-7, 21 respectively. These reliable and valid measures20,21 assess frequency of symptoms included in the DSM-5 diagnostic criteria for either major depressive disorder (PHQ-9; 9 items) or generalized anxiety disorder (GAD-7; 7 items) and interference with functioning over the prior 2 weeks (Not at all: 0; Several days: 1; More than half of days: 2; or Nearly every day: 3). Items are summed to a total symptom severity score: 0 to 4 no or minimal, 5 to 9 mild, 10 to 14 moderate, >15 moderately severe to severe.
Working Alliance
We assessed patient perspectives of working alliance with the clinician prescribing their psychiatric medications—their PCP at baseline, then for the IGAP at all subsequent appointments—using the Hatcher Working Alliance Inventory-Short Revised (WAI-SR). 22 This 12-item measure queries about individual experiences with mental health care on a 5-point Likert scale (Seldom = 1, Sometimes = 2, Fairly often = 3, Very often = 4, Always = 5). Items are summed to a total score (12-60); higher scores indicate better working alliance. The WAI-SR was administered via REDCap, a secure HIPAA-compliant platform.
Acceptability, Feasibility, and Appropriateness
PCPs completed the Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM), and Feasibility of Intervention Measure (FIM) 23 in via REDCap24,25 about their perceptions of the IGAP clinic. The AIM, IAM, and FIM are each validated 4-item measures of perceived acceptability, appropriateness, and feasibility, respectively, of an intervention. 23 Items are rated on a 5-point Likert scales (Completely disagree = 1, Disagree = 2, Neither agree or disagree = 3, Agree = 4, Completely agree = 5), then averaged (1.0-5.0). Higher average scores indicate greater acceptability, appropriateness, and feasibility, respectively.
Additional metrics of feasibility and acceptability were determined via chart review at the conclusion of the study period and included number of appointments, duration of care, number of telehealth versus in-person appointments, and internal psychotherapy referral status). Optional open-ended response survey questions were also included (Appendix 1).
Analysis Plan
Consistent with recommendations for pilot studies, 26 our primary goal was to examine the feasibility of the IGAP clinical model. We computed descriptive statistics (ie, means, standard deviations) for all measures. While not sufficiently powered, given that this was a feasibility pilot, for patients with a clinically significant baseline PHQ-9 or GAD-7 score we also conducted Wilcoxon Signed-Rank Tests to explore change in symptoms from pre- to post-treatment. We also compared WAI-SR ratings of prior PCP medication management to ratings of their IGAP clinic psychiatric care using Wilcoxon Signed Ranks Tests. To account for variation in the number of IGAP clinic WAI-SR ratings per patient due to variation in number of appointments attended, we computed an average psychiatric WAI-SR score for each patient. Analyses were conducted using SPSS (Chicago, IL).
Results
We enrolled 19 patient participants in the IGAP clinic. One was lost to follow-up, and 1 moved away after the intake appointment. These results reflect the 17 patients who engaged in at least 2 IGAP appointments. Collaborating PCPs (n = 10) were invited to complete the AIM, FIM, and IAM; 6 PCPs (3 attending physicians and 3 resident physicians) completed surveys. Demographics were collected by open-ended self-report (Table 1).
Patient Participant Demographics (N = 17).
Participants may have more than 1 diagnosis.
Feasibility
Sixteen patients (94%) elected to have at least 1 IGAP appointment by telehealth and 67% of IGAP appointments were conducted by telehealth. Patients reported selecting telehealth to overcome transportation barriers and/or scheduling constraints, and in open-response feedback underscored the importance of telehealth services in their ability to access care. All enrolled patients were offered psychotherapy referrals (all 8 without current therapy accepted, 9 declined because they had pre-existing care, and 1 with pre-existing care requested a new referral). At the end of the trial period, 2 patients had initiated psychotherapy based on our referrals (1 within the same health system; 1 in community private practice); 7 remained on the health system’s waitlist. Patients attended a mean (SD) 3.4 (1.9) IGAP appointments over a mean (SD) 126.8 (98.0) days. At the end of the 9-month study period, 6 patients had returned to PCP-only management and 11 patients remained in ongoing psychiatric care through the IGAP clinic; no patients who were transitioned to PCP-only management returned to the IGAP clinic for additional specialist management during the study duration.
PCP participants reported a mean (SD) 2.7 (1.5) interactions with the IGAP clinic. On the FIM, all PCP participants deemed the intervention feasible (M = 4.6, SD = 0.5). To support ongoing feasibility, PCPs suggested further formalizing referral and communication processes, especially around prescribing controlled substances.
Acceptability
Per the AIM, all PCP participants deemed the intervention acceptable (M = 4.8, SD = 0.4). In open ended responses, PCPs reported that the intervention made a positive impact on their patients. For example, 1 PCP shared, “Overall good experience for transgender patients who also have psychiatric needs beyond the scope of primary-mental-health-care.”
Appropriateness
All PCP participants deemed the intervention appropriate on the IAM (M = 4.8, SD = 0.4). Further, they requested integrated psychiatry be expanded, especially to specifically meet the needs of patients with a history of trauma.
Effectiveness
Baseline scores were clinically elevated (≥5) among 7 (41%) patients on the PHQ-9 and among 16 (94%) patients on the GAD-7 (see Table 2). Clinically-elevated symptoms improved significantly on both measures of depression from pre- (M = 14.8, SD = 5.8) to post-treatment (M = 3.8, SD = 3.5; Z = −2.0, P = .043) and anxiety from pre- (M = 12.6, SD = 4.5) to post-treatment (M = 5.4, SD = 3.5; Z = −3.3, P < .0013).
Medication Prescriptions for Patient Participants at Baseline and Post-Treatment.
In the 12 months preceding their IGAP intake appointments, 2 patients (12%) had 1 or more emergency department (ED) visit for psychiatric symptoms. During study participation, no participant presented to an ED for psychiatric symptoms.
Diagnosis and Treatment
Many patients’ psychiatric diagnoses were updated based on IGAP evaluations. Corrected/refined diagnoses include: major depression → bipolar disorder (1), bipolar disorder → PTSD + substance use disorder ± borderline personality disorder (2), OCD → autism (1). Specified diagnoses included: anxiety disorders (2), depressive disorders (1). Additional new diagnoses included: anxiety disorders (4), depressive disorders (2), eating disorders (2), neurodevelopmental disorders (2), obsessive compulsive disorder (OCD; 1), PTSD (1), substance use disorders (predominantly nicotine and cannabis; and 9), complex bereavement (1).
All 12 patients (71%) on baseline psychotropic medications had their medications adjusted during study participation; of the 5 without baseline psychotropic medications, 3 (18%) started medications and 2 (12%) remained on no medications (see Table 2). The 5 patients who endorsed nicotine use were offered medication to support cessation; 3 accepted and initiated nicotine replacement therapy and/or bupropion.
Satisfaction
All patient participants who completed the WAI-SR (n = 14) reported a strong working alliance with the IGAP team (Table 2). Those who completed both baseline and intervention surveys (n = 11) reported higher working alliance with IGAP clinicians (M = 55.4, SD = 5.1) than with the PCP previously managing psychiatric medications (M = 42.4, SD = 10.4; Z = −3.0, P = .003).
Discussion
This 9-month pilot study demonstrated feasibility of an integrated gender-affirming psychiatric care program within a family medicine clinic. Additionally, while not sufficiently powered due to the pilot nature of this study, all patients attending multiple visits reported improvement in depression and anxiety symptoms over the course of treatment. Given prior literature evidencing higher acuity mental health symptoms among TNG individuals who receive primary care in a setting that does not offer integrated psychiatric services, 27 this study highlights the potential value of integrated services for this population.
The observed prevalence of baseline symptomatology in this sample was comparable to broader samples of TNG young adults for depression, 28 but higher for anxiety. 29 Anecdotally, some patients attributed elevated scores on these measures to acute stressors related to their LGBTQIA+ identities, despite subjective overall improvements in their mental health. This underscores the importance of considering population-specific psychosocial stressors in clinical care. Future research utilizing population-specific metrics of quality of life and mental well-being, such as the Gender Minority Stress and Resilience Scale, 30 the Gender Congruence and Life Satisfaction Scale, 31 and the Multidimensional TNB Resilience Scale 32 are needed to better understand the patient experience.
Despite relatively infrequent interactions with the pilot clinic, collaborating PCPs approved of this model and requested expanded mental health care service offerings, namely in-house psychotherapy, for their LGBTQIA+ patients. Importantly, all patients were interested in psychotherapy yet most remained on a waitlist during the entire 9 month study period. Future work is needed to explore the potential added value of integrated, culturally-responsive, psychotherapy to medication management within a clinic like IGAP to increase timely access to psychotherapy. Such efforts should utilize existing evidence-based TNG-affirming psychotherapy models33,34 when possible. Telehealth was also extremely popular with patients and identified as an essential component in the intervention’s accessibility.
Of note, pilot clinic participation led to updated psychiatric diagnoses for 15 (88%) patient participants. While likely partially attributable to receiving specialist care, this may also indicate that TNG patients were able to receive care that approached their mental health with more nuance and understanding of how trauma and social determinants of health intersect with gender identity and TNG experiences more broadly. Further research is warranted in this area.
Limitations of this work include our small sample size within a single setting, which limits generalizability; future studies with adequately powered samples and comparison groups are needed, as well as studies in other clinics/health systems to examine generalizability. Subgroup analyses were not possible due to the small sample size; future work should also examine if patient characteristics or receipt of various psychotherapy models influence outcomes.
Our results are specific to the patient population included in the study and the model would not be expected to address access barriers for individuals who do not already have an established relationship with primary care. Because this was integrated within an existing family medicine clinic’s operations, appointment frequency depended on clinical need. We elected not to create a waitlist comparison group due to ethical concerns around withholding indicated mental health services, which limits our ability to compare the reported outcomes to those of peers who did not participate in this pilot. Future larger studies expanding upon this work would benefit from inclusion of comparison groups, as ethically feasible. Of note, an openly trans/non-binary clinician (author TGG) provided care in the IGAP clinic, which may have influenced clinical and/or implementation outcomes such as perceived patient-clinician working alliance. Future studies are needed that examine the impact of patient preference regarding clinician and patient concordance on specific identities. We specifically chose to situate this preliminary study within a practice that frequently provides gender-affirming primary care services, and accordingly most frequently collaborated with the PCPs who have a clinical focus in this area; future work, especially in more general settings, could also benefit from specifically evaluating clinicians on cultural and clinical competence for treating TNG patients (eg, using the Transgender Knowledge, Attitudes, and Beliefs Scale and Transgender Clinical Competency Scale). 35
In sum, this pilot study supports the acceptability, feasibility, and preliminary effectiveness of integrating gender-affirming psychiatric care with gender-affirming primary care. Integrated care models that are tailored to the needs of patients with historically marginalized and minoritized identities offer promise for reducing health disparities and increasing access to care that is clinically effective and non-stigmatizing.
Footnotes
Appendix 1
Ethical Considerations
This protocol #853727 was approved by the University of Pennsylvania Institutional Review Board.
Consent to Participate
Participants provided written informed consent.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the EPSP Pilot Grant (University of Pennsylvania Psychiatry); TGG’s research time was funded by National Institute of Health R25MH119043 (University of Pennsylvania Psychiatry).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Per IRB approved protocol and for participant anonymity, non-aggregate data will not be available by request.
