Training in LGBTQ-affirmative Cognitive-Behavioral Therapy: A Randomized Controlled Trial Across LGBTQ Community Centers

Abstract

Objectives:

This randomized controlled trial examined whether an 11-week synchronous (i.e., real-time) online training in LGBTQ (lesbian, gay, bisexual, transgender, queer)-affirmative CBT could lead to increased uptake of this practice at LGBTQ community centers across 20 US states and internationally.

Methods:

A total of 121 mental health providers (M_age = 37.74; 74.4% cisgender; 76.9% LGBTQ; 60.3% non-Hispanic/Latinx White) were randomized to receive the 11-week training either immediately (n = 61) or after a 4-month wait (n = 60). At baseline and 4- and 8-months after baseline, participants self-reported their LGBTQ-affirmative competency, cultural humility, and knowledge of the minority stress theory and practice skills underlying LGBTQ-affirmative CBT. To objectively assess uptake of LGBTQ-affirmative CBT, participants demonstrated, through simulated practice, how they would respond to two video-based clinical vignettes.

Results:

Compared to waitlist, participants in the immediate training condition reported greater improvements in self-reported cultural competence (d = 1.24), minority stress knowledge (d = 0.78), LGBTQ-affirmative CBT knowledge (d = 0.78), and LGBTQ-affirmative CBT skills familiarity (d = 0.91) and use (d = 0.96); effects persisted 8 months post-baseline. Cultural humility showed no significant difference by condition (d = 0.07). In objectively coded assessments of simulated practice, participants in the training condition demonstrated greater uptake of LGBTQ-affirmative practice skills (d = 0.82).

Conclusions:

Findings preliminarily suggest that mental health providers can be trained to deliver LGBTQ-affirmative CBT using the low-cost, efficient reach of online training. This training can help disseminate evidence-based mental healthcare to LGBTQ individuals and its implementation across practice settings.

In every context in which such disparities have been examined, lesbian, gay, bisexual, transgender, queer, and other sexual or gender minority (LGBTQ) individuals represent one of the highest-risk populations for stress-sensitive mental health problems, especially depression, anxiety, substance use, and suicidality (Bränström, 2017; Day et al., 2017; Lipson et al., 2019; Plöderl & Fartacek, 2005; Semlyen et al., 2016). The most plausible source of this disparity lies in the multifaceted manifestation of stigma-related stress that starts early in LGBTQ development and persists across much of the lifespan (Bränström & Pachankis, in press; Bränström et al., in press; Rice et al., 2019). As a pervasive influence on development and daily life, stigma increases the need for mental health services by increasing stressful demands (e.g., with discrimination) and undermining coping resources (e.g., social support; Meyer et al., 2008). As a fundamental cause of poor mental health, stigma also poses structural barriers to mental healthcare, potentially including seeking and receiving evidence-based care (Hatzenbuehler et al., 2013).

Given their greater exposure to stigma-related stress and associated mental health challenges, LGBTQ people seek mental healthcare at a higher frequency than the general population (Cochran et al., 2017). At the same time, evidence suggests that LGBTQ individuals face several barriers to receiving mental healthcare. Like the general population, LGBTQ people report experiencing shame and discomfort when discussing personally distressing experiences, which can serve as a barrier to seeking mental healthcare (Ferlatte et al., 2019; Moore et al., 2020). In addition, LGBTQ individuals seeking therapy report LGBTQ-specific barriers to care including difficulty finding an LGBTQ-affirmative provider (Ferlatte et al., 2019), anticipated discrimination from providers (Cronin et al., 2021; Kcomt et al., 2020; McNamara & Wilson, 2020, and mistrust of providers (McNamara & Wilson, 2020; Owens et al., 2007; Shipherd et al., 2010; Zullo et al., 2021). Such expectations of poor treatment are often justified because mental health providers report relatively little training in LGBTQ-specific issues (Alessi et al., 2016) and can display biases in therapeutic practice (Bishop et al., 2021; Morris et al., 2020; Nedela et al., 2022; O'Shaughnessy & Speir, 2018; Shelton & Delgado-Romero, 2011; Spengler et al., 2016) and diagnosis (Eubanks-Carter & Goldfried, 2006; Rodriguez-Seijas et al., 2021), although not all mental health providers demonstrate such bias (Thompson et al., 2019). A general lack of LGBTQ-affirmative training has been documented starting with graduate training for mental health providers, including few clinical opportunities to work with LGBTQ clients (O’Hara et al., 2013; Rodriguez-Menendez et al., 2017; Sherry et al., 2005; Soulliard et al., 2021). Cost, time, and travel barriers can interfere with receiving training after graduate school. Gender minority individuals, bisexual individuals, and racial/ethnic minority individuals face compounded barriers to care, including disproportionate exposure to socioeconomic strain and engaging with therapists who lack the training to acknowledge unique and intersectional identity-related experiences (Ferlatte et al., 2019; Mizock & Lundquist, 2016; Moore et al., 2020; Steele et al., 2017).

An additional structural barrier to mental health for LGBTQ individuals is the relative lack of evidence-based mental health treatments created specifically for LGBTQ clients (O'Shaughnessy & Speir, 2018). Without such treatments, mental health clinicians lack concrete evidence-based guidance for implementing LGBTQ-affirmative professional directives, such as those promulgated by professional organizations (e.g., American Psychological Association, 2015, 2021). Recently, however, such treatments have been developed and tested in small open trials (Budge et al., 2021; Craig & Austin, 2016), non-randomized trials with a waitlist comparison (Craig, Eaton, et al., 2021), and randomized controlled trials (Pachankis, Harkness, Maciejewski, et al., in press; Pachankis et al., 2015; Pachankis et al., 2020); accumulating evidence suggests that these treatments can improve LGBTQ individuals' mental health. The most rigorously examined of these treatments to date, referred to as LGBTQ-affirmative CBT (Pachankis, Harkness, Jackson, et al., in press; Pachankis, Jackson, et al., in press), is based in the cognitive-behavioral principles and techniques, as well as adjunctive interventions (e.g., motivational interviewing), contained in the Unified Protocol for the Transdiagnostic Treatment of Emotional Disorders (Unified Protocol; Barlow et al., 2017). LGBTQ-affirmative CBT adapts the techniques and general principles of the Unified Protocol to specifically address LGBTQ individuals’ cognitive, affective, and behavioral reactions to minority stress in ways that raise consciousness about minority stress, are self-empowering, and draw upon LGBTQ community strengths and personal resilience (Pachankis, 2014; Pachankis, Harkness, Maciejewski, et al., in press; Rodriguez-Seijas et al., 2019). Across randomized controlled trials, LGBTQ-affirmative CBT shows promise for reducing depression, anxiety, and substance use among sexual minority men (Pachankis, Harkness, Maciejewski, et al., in press; Pachankis et al., 2015) and gender-diverse sexual minority women (Pachankis et al., 2020). LGBTQ-affirmative CBT contains nine modules that present CBT principles and techniques in an LGBTQ-affirmative manner. These LGBTQ-affirmative adaptations were generated through deep consultation with LGBTQ community members and clinical experts (Pachankis, 2014; Pachankis et al., 2022; Scheer et al., in press). The treatment has been effectively delivered in individual and group formats and implemented in community-based clinics (e.g., Jackson et al., 2022; Pan et al., 2021), supporting its promise for broader implementation beyond the controlled trials in which it was originally developed and tested.

Broader implementation of evidence-based practice can be achieved by training providers in such practice (Lehane et al., 2019) and training providers to deliver LGBTQ-affirmative evidence-based care, specifically, represents a promising way to reduce the barriers to LGBTQ-affirmative mental healthcare described above. In fact, accumulating research suggests that formal training in LGBTQ-affirmative healthcare can potentially reduce provider bias and increase provider competence, knowledge, and self-efficacy for delivering such care. Most such evidence comes from observational studies (e.g., Phelan et al., 2017), open trials (e.g., Lelutiu-Weinberger & Pachankis, 2017), and studies of healthcare training outside of mental healthcare specifically (Jaffer et al., 2016; Lelutiu-Weinberger et al., 2016; White Hughto et al., 2017). For instance, studies have examined transgender competence training in medical clinics (Lelutiu-Weinberger et al., 2016) and correctional settings (Jaffer et al., 2016; White Hughto et al., 2017), finding that the trainings were associated with reduced patient complaints and provider bias and increased motivation to care for transgender patients. Three studies have focused on training mental health providers to become more LGBTQ affirmative and found in open trials that a 1-2-day training can be associated with increases in self-reported LGBTQ-affirmative skills, attitudes, and self-efficacy for care and reductions in self-reported LGBTQ bias (Craig, Iacono, et al., 2021; Lelutiu-Weinberger & Pachankis, 2017; Pepping et al., 2018). Notably, no LGBTQ-affirmative training intervention has been tested in a randomized controlled trial capable of establishing its efficacy (Bettergarcia et al., 2021).

The present study tested the preliminary efficacy of training mental health providers to deliver LGBTQ-affirmative CBT. The setting for the present study was LGBTQ community centers across the US and internationally. Since the beginning of the LGBTQ rights movement, LGBTQ community centers have served as frontline settings for LGBTQ-affirmative mental healthcare (Martos et al., 2017; Silverstein, 1997). LGBTQ community centers include any organization whose goal is to support the LGBTQ community by offering formal programming, ranging from support groups to comprehensive health services. Today, these centers continue to play a central role in supporting the mental healthcare needs of diverse LGBTQ communities. Because they offer care for free or reduced fee, they are particularly likely to reach those who might not be able to otherwise afford LGBTQ-affirmative care (Martos et al., 2019). Of the more than 270 LGBTQ community centers across the US and internationally, about 40% provide mental health services according to a recent unpublished report (CenterLink & Movement Advancement Project, 2020). A recent survey of executive directors and chief operating officers of 60 US LGBTQ community centers with any paid staff found that 48% had more than five mental health providers on staff, with 70% having social workers, 62% having counselors, 22% having psychologists, and 8% having psychiatrists on staff (Pachankis, Clark, et al., 2021), suggesting that mental health service provision occupies an important part of these centers’ mission. Although one small, unpublished study found that the most common mental health problems addressed at these centers include LGBTQ-specific concerns (Rogers, 2012), the treatment approaches used at these centers to address these concerns remains unknown, especially since LGBTQ-affirmative evidence-based practice has not existed until recently. In the above-mentioned survey of executive directors and chief operating officers of LGBTQ community centers, 62% reported offering CBT and 62% also reported offering supportive therapy; however, when and how these distinct types of approaches are delivered and for what specific presenting concerns remains unknown (Pachankis, Clark, et al., 2021). At the same time, 83% of respondents indicated that their mental health staff would benefit from training in LGBTQ-affirmative CBT and 100% of respondents indicated that they would provide administrative support for their staff to receive such training, suggesting high need and support for training in this approach (Pachankis et al., 2021).

Motivated by this background, the present study employed a waitlist-controlled trial of mental health providers working across LGBTQ community centers to test the efficacy of an 11-week training in LGBTQ-affirmative CBT. Based on previous evidence that training in LGBTQ-affirmative approaches is equally efficacious when delivered in-person and remotely (Lelutiu-Weinberger et al., 2021) and to ensure broad reach to diverse geographic settings while overcoming access barriers (i.e., provider cost, time, and travel), we delivered the training synchronously (i.e., in real time) online. Outcomes included self-reported LGBTQ-affirmative competency, cultural humility (i.e., a clinician’s openness to accepting how their own identities and experiences may bias their awareness of their clients’ intersecting identities and experiences), and knowledge of the minority stress theory and practice skills underlying LGBTQ-affirmative CBT. An objectively coded video-based simulated practice assessment provided a further opportunity to assess uptake of LGBTQ-affirmative CBT skills.

Method

Participants

A total of 121 mental health providers across 50 LGBTQ community centers enrolled in the study. Table 1 provides demographics and sample characteristics of participants by group condition. On average, participants were 37.74 (SD=10.55) years old. Slightly over half were cisgender women (54.5%), over three-quarters reported a sexual minority identity (76.9%), and one-quarter (24.8%) reported a gender minority identity. Overall, 78.5% of the sample identified as sexual and/or gender minority individuals, and approximately a quarter (24.0%) reported both sexual and gender minority identities. Over one-third (39.7%) reported a racial/ethnic minority identity. The majority reported having earned a master’s degree (78.5%) and working full-time (54.5%) at their respective LGBTQ community center with predominantly LGBTQ adults between the ages of 18-50 (82.6%). Participants represented LGBTQ community centers in 20 different US states, Canada, and Uganda. The number of participants from a single LGBTQ community center ranged from 1 to 12 (M = 4.01, SD = 3.11). Over one-third reported an eclectic or integrative theoretical orientation (36.4%), and a majority reported individual therapy as their primary mode of intervention (81.0%). Participants, on average, reported working in the mental health field for 6.13 (SD = 7.11) years and working exclusively or predominantly with LGBTQ clients for 3.58 (SD = 4.39) years.

Table 1.

Demographics and Sample Characteristics by Condition

Characteristic	Immediate Intervention (n = 61)		Waitlist Control (n = 60)		Condition comparisons
	n	%	n	%
Age					t = 2.13, p = .04
M	39.74		35.72
SD	11.36		9.33
Gender identitya					χ2 = 0.98, p = .32b
Cisgender man	14	23	11	18.3
Cisgender woman	29	47.5	37	61.7
Transgender man	6	9.8	0	0.0
Transgender woman	1	1.6	1	1.7
Gender non-binary	4	6.6	8	13.3
Gender non-conforming	1	1.6	2	3.3
Genderqueer	6	9.8	1	1.7
Other	2	3.3	3	5.0
Sexual orientation					χ2 = 0.56, p = .76c
Asexual	1	1.6	1	1.7
Bisexual	6	9.8	7	11.7
Gay	13	21.3	9	15.0
Heterosexual/straight	14	23.0	14	23.3
Lesbian	9	14.8	9	15.0
Pansexual	7	11.5	6	10.0
Queer	11	18.0	13	21.7
Other	0	0.0	1	1.7
Race/ethnicity					χ2 = 0.50, p = .48d
Asian	1	1.6	2	3.3
Black/African American	5	8.2	8	13.3
Native Hawaiian/Pacific Islander	1	1.6	0	0.0
White, Hispanic/Latinx	4	6.6	4	6.7
White, Non-Hispanic/Latinx	39	63.9	34	56.7
Multiracial	11	18.0	11	18.3
Missing	0	0.0	1	1.7
Hispanic/Latinx					χ2 = 1.75, p = .19
Yes	15	24.6	9	15.0
No	46	75.4	51	85.0
Education degree					χ2 = 6.46, p = .17
High school/GED or associate’s degree	2	3.2	0	0.0
Bachelor’s degree	6	9.8	14	23.3
Master’s degree	50	82.0	45	75.0
Doctoral degree	3	4.9	1	1.7
Employment status					χ2 = 10.79, p = .013
Full-time (40 hours per week)	37	60.7	29	48.3
Part-time (less than 40 hours per week)	19	31.1	12	20.0
Practicum student/intern	3	4.9	13	21.7
Other	2	3.3	6	10.0
Theoretical orientation					χ2 = 1.71, p = .43e
Cognitive Behavioral (CBT)	10	16.4	17	28.3
Existential	0	0.0	1	1.7
Humanistic/Person-Centered	15	24.6	14	23.3
Psychodynamic/Psychoanalytic	2	3.3	3	5.0
Solution-Focused	9	14.8	4	6.7
Third-wave CBT	1	1.6	0	0.0
Eclectic or Integrative	23	37.7	21	35.0
Missing	1	1.6	0	0.0
Primary mode of psychotherapy					χ2 = 4.28, p = .23
Individual therapy	52	85.2	46	76.7
Family therapy	1	1.6	0	0.0
Group therapy	7	11.5	9	15.0
Other	1	1.6	5	8.3
Age of clients					χ2 = 0.91, p = .92
<5 to 12	2	3.3	2	3.3
13-17	5	8.2	7	11.7
18-30	29	47.5	29	48.3
31-50	23	37.7	19	31.7
51-65+	2	3.3	3	5.0
Number of years working in mental health					t = 1.12, p = .26
M	6.85		5.40
SD	7.42		6.77
Number of years working with LGBTQ clients					t = 1.92, p = .06
M	4.33		2.82
SD	5.37		2.95

^aParticipants were able to select multiple genders for their gender identity. Percentages are based on total sample size for each identity.

^bGender was dichotomized as cisgender vs. transgender/nonbinary when conducting condition comparisons.

^cSexual orientation was grouped as gay or lesbian vs. other sexual minority identity vs. heterosexual when conducting condition comparisons.

^dRace was dichotomized as racial/ethnic minority vs. White, Non-Hispanic when conducting condition comparisons.

^eTheoretical orientation was grouped as CBT vs. other orientation vs. eclectic or integrative when conducting condition comparisons.

Procedure

Recruitment and Screening

All study procedures were approved by the Yale University Human Subjects Committee and preregistered on clinicaltrials.gov (record NCT04559698). From September to October 2020, we recruited participants through an online flyer emailed to listservs of mental health providers working at CenterLink-affiliated LGBTQ community centers. CenterLink is an international nonprofit organization consisting of over 270 member LGBTQ community centers in the US, Canada, Australia, China, and Uganda. The study trainers also held an informational webinar in October 2020 to offer mental health providers at CenterLink-affiliated LGBTQ community centers the opportunity to learn more about the LGBTQ-affirmative CBT training and this study.

Participants completed a brief online screener on Qualtrics to confirm the following eligibility criteria: (1) age 18 or older; (2) fluent in English; (3) licensed or unlicensed mental health provider (e.g., clinical/counseling psychologist, pre-doctoral psychology intern, counselor, martial and family therapist, social worker, crisis counselor, third-year or greater graduate practicum student/extern in a mental health training program); and, (4) currently practicing in or formally affiliated with a CenterLink-affiliated LGBTQ community center, confirmed by a CenterLink director. All mental health providers received emails confirming their eligibility to participate in the LGBTQ-affirmative CBT training.

Study Design

Figure 1 provides an overview of the study flow and summary of participants who completed each study stage, including those who were ineligible, were lost to follow-up, or withdrew. Participants completed online study measures at three time-points (baseline, 4-months post-baseline, 8-months post-baseline) to assess efficacy of the training intervention. Prior to completing the baseline surveys, participants were asked if they would like to participate in the research component of the training (i.e., complete the study measures). Participants who opted out of the research component (n = 13) were assigned to attend one of the two trainings and then ended the online survey. Participants who chose to complete the research component provided informed consent prior to completing the baseline assessment. A total of 121 participants who agreed to participate in the research component were randomized, using the Qualtrics randomizer, into one of two groups: (1) immediate intervention condition (n = 61) or (2) waitlist control condition (n = 60).

Figure 1.

Study structure and participant flow through study phases.

Participants randomized to the immediate intervention condition attended the 11-week online training from October 2020 to January 2021. During this period, waitlist participants received one email per month reminding them of the 4-months post-baseline assessment and participation in their upcoming training. At the end of the training for the immediate intervention condition, in February 2021, all participants completed the 4-months post-baseline assessment. Participants randomized to the waitlist control condition attended the same 11-week online training from March 2021 to May 2021. At the end of the training for the waitlist condition, in June 2021, all participants completed the 8-months post-baseline assessments. During this second training, participants in the immediate intervention condition received one email per month reminding them of the upcoming 8-months post-baseline assessment. Participants received $50 for completion of each assessment (possible total of $150 for the entire study).

Participants were also able to earn one continuing education (CE) credit per session attended (total of 11 CEs for the entire training) from the American Psychological Association’s (APA) Division 44 (Society for the Psychology of Sexual Orientation and Gender Diversity) CE Program. The Division 44 CE Committee approved the training for CE credits for licensed psychologists and other mental health professionals (e.g., licensed clinical social workers, licensed mental health counselors). Participants pursuing CE credits completed an APA-required feedback form after each attended session. These feedback forms were submitted to the Division 44 CE Committee at the end of both trainings and kept separate from the study data collected at baseline, 4-months post-baseline, and 8-months post-baseline. All participants received the same training and study procedures regardless of whether they chose to pursue CE credits.

The study employed two safeguards against contamination between the two arms, in which participants in the immediate training condition might share training content with those in the waitlist condition. First, we asked participants in the immediate training group to not share any training materials with their colleagues in the waitlist control group. Second, during the 4-months post-baseline assessment, we asked participants in the waitlist control group (i.e., those who were about to receive the training) whether they had received any training materials (e.g., slides, handouts) from their colleagues who attended the first training. No participants in the waitlist control group reported receiving materials from those in the first training.

Training Intervention

The training was directly based on the content of LGBTQ-affirmative CBT treatment materials described in detail elsewhere (Pachankis, 2014; Pachankis, Harkness, Jackson, et al., in press; Pachankis, Jackson, et al., in press; Pachankis et al., 2022), which have been shown to be preliminarily efficacious for reducing sexual minority men’s and gender-diverse sexual minority women’s depression, anxiety, and substance use (Pachankis, Harkness, Maciejewski, et al., in press; Pachankis et al., 2015; Pachankis et al., 2020). Four psychologists served as the trainers across all sessions, including three licensed clinical psychologists (first, second, and third authors) and a counseling psychologist (seventh author). The trainers identified as sexual minority individuals and included a White cisgender gay man, an Asian American cisgender queer/gay man, a White cisgender bisexual woman, and a Black cisgender gay/queer man. Each of the 11 sessions took place on Zoom for a duration of 1-hour. During the first weekly session, the four trainers provided a background of the LGBTQ-affirmative CBT intervention, including its reliance on CBT-based case conceptualization informed by minority stress theory (Pachankis et al., 2022), background on the treatment development process (Pachankis, 2014), and evidence of the treatment’s efficacy (Pachankis, Harkness, Maciejewski, et al., in press). For weekly sessions 2-10, the trainers worked in dyads to present material from each of the nine modules of LGBTQ-affirmative CBT. For the final weekly session, all four trainers provided a final review of the modules, as well as facilitated a discussion on additional considerations for implementing the treatment as raised by participants. With minimal exception (e.g., two trainers once switching training sessions due to a necessary absence), the four trainers occupied the same role in both trainings.

The Supplemental Materials outline the objectives of each of the 11 training sessions. The trainers prepared presentation slides for each session and engaged participants using various strategies (e.g., role plays, deidentified video clips of therapists and clients) and Zoom functions (e.g., anonymous polls, chat box discussions, and breakout rooms to review and discuss content). Participants were provided copies of the presentation slides and client handouts from each respective module prior to each training session. At the beginning of the training and at points throughout, the trainers were explicit about the evidence base of the treatment materials, including the fact that they have only been tested for efficacy with sexual minority men and gender-diverse sexual minority women, most of whom identified as transgender, non-binary, gender fluid, or genderqueer. At the same time, the trainers discussed how LGBTQ-affirmative CBT principles and techniques could potentially benefit gender minority individuals across the gender spectrum, because these principles and techniques are rooted in minority stress theory and because transgender and gender-diverse individuals have found LGBTQ-affirmative treatment approaches to be acceptable (Austin et al., 2018; Pachankis et al., 2020). The trainers stressed that future research is needed into approaches that address the distinct concerns of the full gender spectrum. This approach was taken to address the fact that many mental health providers working at LGBTQ community centers provide care to the full spectrum of the sexual and gender minority population and that minority stress, as addressed in LGBTQ-affirmative CBT, underlies numerous presenting concerns of this population, while also transparently acknowledging the more limited state of evidence for this treatment approach as applied to transgender individuals.

Outcome Measures

All measures were administered at baseline, 4-months post-baseline, and 8-months post-baseline, except the acceptability questionnaire, which was only administered immediately following the intervention (i.e., at 4-months post-baseline for the immediate intervention condition and 8-months post-baseline for the waitlist control).

Sexual Orientation Counselor Competency Scale (SOCCS) – Skills Subscale

The SOCCS (Bidell, 2005) is a self-report measure of one’s knowledge, attitudes, and clinical skills related to working with lesbian, gay, and bisexual clients. For the present study, participants only completed the 11-item Skills subscale, modified to be inclusive of transgender clients and to specifically assess clinical skills in LGBTQ-affirmative CBT (e.g., changing items from “I feel competent to assess the mental health needs of a person who is LGB in a therapeutic setting” to “I feel competent to assess the mental health needs of a person who is LGBTQ using an LGBTQ-affirmative CBT framework”). Items are rated on a scale from 1 (not at all true) to 7 (totally true). The SOCCS Skills subscale has demonstrated good internal consistency (α = .91) and one-week test-retest reliability (r = .83) among a sample of predominantly White, heterosexual psychology undergraduate and graduate psychology students and doctoral-level counselors (Bidell, 2005). Internal consistency coefficients (Cronbach’s α) across the present study for the full sample were .90 at baseline, .91 at 4-months post-baseline, and .83 at 8-months post-baseline. We computed an average total score in which higher scores represented higher levels of self-reported clinical skills in LGBTQ-affirmative CBT.

Multidimensional Cultural Humility Scale (MCHS)

The MCHS (Gonzalez et al., 2021) is a 15-item self-report measure of cultural humility, defined as “the ability to maintain an interpersonal stance that is other-oriented (or open to the other) in relation to aspects of cultural identity that are most important to the client” (Hook et al., 2013, p. 354). Therapist’s cultural humility has been associated with positive outcomes in psychotherapy, particularly among culturally diverse populations (Hook et al., 2016; Owen et al., 2016). We modified the MCHS to assess cultural humility specifically when working with LGBTQ clients (e.g., changing items from “I am comfortable asking my clients about their cultural experience” to “I am comfortable asking my LGBTQ clients about their cultural experience”). Items are rated from 1 (strongly disagree) to 6 (strongly agree). The MCHS has demonstrated acceptable internal consistency (α = .79) and construct validity among a sample of predominantly White, female counselors working with diverse clients in a variety of mental health settings (Gonzalez et al., 2021). Internal consistency coefficients (Cronbach’s α) across the present study for the full sample were .68 at baseline, .72 at 4-months post-baseline, and .75 at 8-months post-baseline. We computed a sum score with higher scores representing higher levels of cultural humility in LGBTQ-affirmative CBT.

Minority Stress Knowledge

To assess knowledge of minority stress theory, we developed a 10-item multiple choice assessment with questions including four or five response options, of which one was the correct response. We developed each question based on tenets of minority stress theory (Brooks, 1981; Meyer, 2003) and principles of LGBTQ-affirmative CBT (Pachankis, Harkness, Maciejewski, et al., in press), as reviewed in the 11-week training. To develop this measure, the second author created an initial set of questions based on minority stress theory as applied to LGBTQ-affirmative CBT. The first author reviewed and provided feedback on the initial draft of items. The first and second author then solicited feedback from the five remaining authors, all of whom are experts in minority stress and LGBTQ mental health. The first author incorporated this feedback into the final version of the assessment. A sample question included, “Out of the following examples, which one does NOT capture an example of minority stress experienced by a LGBTQ person?” Other items provided brief case scenarios, such as, “Haruto identifies as a 22-year-old Japanese-American, pansexual, genderqueer (they/them) person. They come to therapy with several concerns, including anxiety around whether to note their affirmed pronouns on their law school application, coming out to their family, and worries of losing connections within the Japanese community if they disclose their sexual and gender identities. What minority stress factors appear to be most salient for Haruto?” We computed a total score ranging from 0 to 10, with participants earning one point for each correct response. Higher scores represented greater knowledge of minority stress. The 10 questions are provided in the Supplemental Materials.

CBT/LGBTQ-Affirmative CBT Knowledge

To assess content knowledge of general CBT and LGBTQ-affirmative CBT skills, we developed a 10-item multiple choice assessment with items including four or five response options. Items were based on the LGBTQ-affirmative CBT treatment (Pachankis, 2014; Pachankis, Harkness, Maciejewski, et al., in press; Rodriguez-Seijas et al., 2019) presented over the course of the 11-week training and on questions from a previously developed CBT knowledge test (Myles & Milne, 2004). To create this scale, we followed the same iterative process that was used to create the Minority Stress Knowledge assessment. Sample questions include, “Because many LGBTQ people experience isolation, guilt, and shame in the context of a society that often invalidates their experiences and denies some of the rights and respect afforded to others, LGBTQ people may feel that they do not have the right to express their personal needs, wants, or desires. CBT most directly addresses this via what intervention?” We computed a total score ranging from 0 to 10, with participants earning one point for each correct response. Higher scores represented greater knowledge of minority stress. The 10 questions are provided in the Supplemental Materials.

Familiarity with and Use of LGBTQ-Affirmative CBT Skills

This 14-item questionnaire was developed for the present study to assess self-reported familiarity with and use of the LGBTQ-affirmative CBT skills covered in the training. To create this scale, we followed the same iterative process that was used to create the Minority Stress Knowledge assessment and CBT/LGBTQ-Affirmative CBT Knowledge assessment. The assessed skills were based on modules from the LGBTQ-affirmative CBT treatment, which itself was adapted from modules in the Unified Protocol. The instructional prompt for all items was: “Based on your current familiarity with LGBTQ-affirmative CBT skills, please answer the following items.” For the seven familiarity items, participants were asked how familiar they were with each skill (e.g., “How would you rate your familiarity with cognitive appraisal and reappraisal as an intervention?”). Items were rated on a scale from 1 (not at all familiar) to 5 (extremely familiar) and summed. Seven parallel items asked about participants’ use of LGBTQ-affirmative CBT skills in their clinical practice with LGBTQ clients (e.g., “How often do you use cognitive appraisal and reappraisal as an intervention with your LGBTQ clients?”). Items were rated on a scale from 1 (not at all) to 5 (very often) and summed to derive a single score. Internal consistency coefficients (Cronbach’s α) in the present study for the skills familiarity items were .85 at baseline, .82 at 4-months post-baseline, and .84 at 8-months post-baseline. For the skills use items, Cronbach’s α was .83 at baseline, .77 at 4-months post-baseline, and .78 at 8-months post-baseline. Items covered the following seven skills addressed in the training: motivational interviewing, antecedent-behavior-consequence model, cognitive appraisal, emotion driven behaviors and alternative behaviors, assertiveness, behavioral experiments and emotional exposures, and relapse prevention.

LGBTQ-Affirmative CBT Skills Measured Through Simulated Practice

A simulated practice assessment was developed based on the so-called “deliberate practice” approach. Deliberate practice refers to a method of applying ongoing effort into improving one’s performance in psychotherapy and has been shown to be an effective method of improving clinical skills (Rousmaniere et al., 2016). One main goal of deliberate practice in psychotherapy focuses on transferring content knowledge to experiential performance. Based on theory and research on deliberate practice, we developed two videos of fictional LGBTQ clients, portrayed by two paid LGBTQ actors. Each video lasted approximately 90 seconds. Participants were provided the following instructions prior to viewing the videos, “You will be presented with two brief video clips of fictional clients. Imagine that the individuals in both videos are your own LGBTQ clients who you are seeing for an initial appointment.” (Video scripts presented in Supplemental Materials.) The first video depicted a client experiencing general life stress not related to their LGBTQ identity (i.e., stress with managing their time at work). The second video depicted a client experiencing a stressful concern related to their sexual identity (i.e., stress due to lack of identity support from one’s family and others in the LGBTQ community). After watching each video, participants responded to the following prompt: “Based on the video clip, please write a paragraph (i.e., approximately 5-8 sentences) describing what clinical approaches you would use in your clinical treatment in order to address the challenges discussed by this client. Your response should include as much detail as possible. For example, you may choose to begin your prompt as such: ‘Over the next few sessions, I would…’”

To objectively code participant responses to this prompt to compare uptake of LGBTQ-affirmative CBT skills from before to after training across conditions, the first three authors developed a 22-item coding form based on LGBTQ-affirmative CBT skills covered in the training (e.g., “Discussed mindfulness from a minority stress framework [i.e., present-focused, nonjudgmental awareness, anchoring in the present]”). Each item was coded using a rating of 0 (did not mention at all), 1 (briefly discussed [e.g., mentioned/named skill]), and 2 (discussed in detail [e.g., mentioned/named skills AND provided example]). We computed an average score for each participant across ratings for each of the two videos.

Two bachelors-level research assistants independently coded the 587 responses (baseline = 235; 4-months post-baseline = 189; 8-months post-baseline = 163). Both coders carefully reviewed the LGBTQ-affirmative CBT treatment manual and client workbook upon which the training content was based. The second and third authors, both clinical psychologists, also trained the two coders in identifying the LGBTQ-affirmative CBT principles and intervention strategies that were covered in those materials and the training. Prior to coding, the second and third authors provided several example responses for the coders to code independently, followed by reviewing their codes together. Bachelors-level coders have been shown to yield reliable and valid coding of clinical content in previous studies (Beidas et al., 2012; Dorsey et al., 2018).

All responses were given unique identification codes so that coders were masked to the assessment time-point and condition of each response. Inter-rater reliability was assessed via intra-class correlation (ICC) estimates based on a two-way mixed effects model. The second author calculated ICCs for the average total of LGBTQ-affirmative CBT skills items and the average total of general CBT skills items. The coders initially met with the second and third authors after every 20 coded responses and in cases where the coders could not reach consensus in order to resolve the discrepancy. ICC estimates were deemed indicative of poor reliability for values less than .5, moderate reliability between .5 and .75, good reliability between .75 and .9, and excellent reliability greater than .9 (Koo & Li, 2016; Portney & Watkins, 2000). Once excellent inter-rater agreement was reached, one of the research assistants coded the remaining responses in intervals of 100 responses, while the other research assistant coded 20 of each set of 100 responses to continue assessing inter-rater reliability. The second and third author continued to review the 20 responses coded by both coders and facilitated consensus meetings for coders to discuss and resolve discrepancies. Overall, the average ICC for the LGBTQ-affirmative CBT skills items was .87.

Acceptability of LGBTQ-Affirmative CBT Training

We created a measure of training acceptability for the present study based on previous research regarding training in LGBTQ-affirmative care (Lelutiu-Weinberger & Pachankis, 2017; White Hughto et al., 2017). Seven questions assessed participants’ perceptions of the trainings’ interestingness, informativeness, and helpfulness for implementing LGBTQ-affirmative care (e.g., “How informative was the training”) on a scale ranging from 1 (not at all informative) to 5 (extremely informative). Internal consistency coefficients (Cronbach’s α) for the immediate intervention and waitlist control conditions were .87 and .91, respectively.

Analytic Plan

Intervention efficacy was assessed using an intent-to-treat analysis and included all participants (n = 121). First, to test the effectiveness of randomization, we examined differences in baseline participant demographic and professional characteristics between the immediate intervention condition (n = 61) and waitlist control (n = 60) using t-tests for continuous measures and chi-square tests for categorical measures. Age, fulltime employment status, and number of years working in LGBTQ mental healthcare differed between conditions (p < .10), with the waitlist control group containing participants with a younger mean age, who were less likely to be working fulltime, and who reported fewer years working in LGBTQ mental healthcare. Thus, we included fulltime employment status and number of years working in LGBTQ mental healthcare as covariates in subsequent analyses. We did not include age as a covariate because of its high correlation (r = .50) with number of years working in LGBTQ mental healthcare. Dependent variables were assessed for normality using skewness thresholds of ±2 and kurtosis thresholds of ±7 (Byrne, 2010; Hair et al., 2010), and all dependent variables were considered normal.

To examine intervention efficacy, we used linear mixed models with maximum likelihood estimation and an unstructured covariance matrix to test the Condition × Time interaction for all intervention outcomes including self-reported LGBTQ cultural competence; LGBTQ cultural humility; minority stress knowledge; and LGBTQ-affirmative CBT knowledge, skills familiarity, and skills use; and objectively coded demonstration of LGBTQ-affirmative simulated practice skills. Primary analyses between conditions were limited to baseline (time = 0; pre-intervention for immediate condition, 4-months pre-intervention for waitlist condition) and 4-months post-baseline assessment (time =1; post-intervention for immediate condition, pre-intervention for waitlist condition) and examined the Condition × Time effect of receiving immediate intervention (condition = 1) versus receiving the 4-months waitlist (condition = 0). Thus, these models compared pre-intervention and post-intervention outcomes in the immediate intervention group with 4-months pre-intervention and pre-intervention outcomes in the waitlist control group. Effect sizes (Cohen’s d) for linear mixed models were calculated as mean pre-post change in the immediate intervention group minus the pre-post change in the waitlist control group, divided by the pooled baseline standard deviation (Morris, 2008).

To examine longer-term persistence of intervention efficacy, we secondarily limited analyses to the immediate intervention group and compared 4-months follow-up (time = 2) to both pre-intervention (time = 0) and post-intervention (time = 1). If differences from pre-intervention were detected at 4-month follow-up but not post-intervention this would indicate delayed efficacy. Differences between post-intervention and 4-months follow-up would reveal whether efficacy was maintained, increased, or decreased from post-intervention. Effect sizes (Cohen’s d) were calculated as mean change from pre-intervention and post-intervention to 4-month follow-up divided by the pre-intervention immediate intervention condition standard deviation accounting for correlation between time points (Morris & DeShon, 2002).

All results were evaluated at p < .05. We report means, standard errors, 95% confidence intervals, and effect sizes.

Results

Condition comparisons.

We found significant Condition × Time interaction effects in models comparing the immediate intervention (pre- to post-intervention) with the waitlist condition (4-months pre-intervention to pre-intervention) (see Figure 2). Significant interactions suggested relative improvements in self-reported LGBTQ cultural competence (d = 1.24), minority stress knowledge (d = 0.78), LGBTQ-affirmative CBT knowledge (d = 0.78), LGBTQ-affirmative CBT skills familiarity (d = 0.91), and LGBTQ-affirmative CBT skills use (d = 0.96) for participants in the immediate intervention condition compared to those in the waitlist condition. Similarly, analyses based on our simulated practice coding demonstrated relative improvements in use of LGBTQ-affirmative skills favoring participants in the immediate intervention condition (d = 0.82; see Table 2). No significant Condition × Time interaction was found for LGBTQ cultural humility (d = 0.07).

Figure 2.

Condition × Time comparison between immediate intervention and waitlist condition.

Table 2.

Condition by Time Comparisons and Long-Term Follow-Up for Cultural, Minority Stress, and CBT Outcomes

Variable (Possible Range)	Immediate Intervention n=61		Waitlist Control n=60		Condition X Timea n=121			Longer-term Efficacyb n=61
	M	SE	M	SE	Est.	95% CI	d	Est.	95% CI	d
LGBTQ Cultural Competence (1-7)					1.77***	[1.29, 2.26]	1.24
4-month pre-intervention	-	-	3.30	0.19
Immediate pre-intervention	3.31	0.17	3.14	0.18				−1.95***	[−2.26, −1.64]	1.53
Post-Intervention	5.05	0.15	5.03	0.19				−0.33**	[−0.56, −0.09]	0.23
4-month follow-up	5.33	0.13	-	-				referent time
LGBTQ Cultural Humility (15-90)					0.48	[−2.33, 3.30]	0.07
4-month pre-intervention	-	-	75.28	0.83
Immediate pre-intervention	76.48	0.90	75.76	1.03				−2.16*	[−4.15, −0.16]	0.25
Post-Intervention	77.34	0.88	77.27	1.27				−1.27	[−2.89, 0.35]	0.18
4-month follow-up	78.55	0.74	-	-				referent time
Minority Stress Knowledge (0-10)					1.15***	[0.46, 1.83]	0.78
4-month pre-intervention	-	-	7.15	0.19
Immediate pre-intervention	7.02	0.19	7.04	0.26				−0.77***	[−1.19, −0.34]	0.57
Post-Intervention	8.05	0.21	7.63	0.34				−0.21	[−0.23, 0.64]	−0.16
4-month follow-up	7.82	0.24	-	-				referent time
LGBTQ-affirmative CBT Knowledge (0-10)					1.39***	[0.70, 2.08]	0.78
4-month pre-intervention	-	-	5.48	0.21
Immediate pre-intervention	5.36	0.24	5.46	0.26				−1.02***	[−1.60, −0.45]	0.54
Post-Intervention	6.77	0.30	6.80	0.32				−0.21	[−0.31, 0.73]	−0.17
4-month follow-up	6.48	0.28	-	-				referent time
LGBTQ-affirmative Skills Familiarity (7-35)					5.06***	[3.19, 6.92]	0.91
4-month pre-intervention	-	-	19.00	0.66
Immediate pre-intervention	18.30	0.76	19.70	0.76				−6.77***	[−8.14, −5.40]	1.26
Post-Intervention	24.34	0.66	25.46	0.79				−1.04	[−2.09. 0.02]	0.26
4-month follow-up	25.57	0.66	-	-				referent time
LGBTQ-affirmative Skills Use (7-35)					5.02***	[3.31, 6.74]	0.96
4-month pre-intervention	-	-	17.53	0.66
Immediate pre-intervention	16.77	0.67	18.46	0.77				−6.82***	[−8.20, −5.45]	1.35
Post-Intervention	22.93	0.72	23.61	0.77				−0.93	[−2.14, 0.28]	0.24
4-month follow-up	24.02	0.69	-	-				referent time
Simulated Practice (LGBTQ-affirmative CBT Skills)c (0-2)					0.12***	[0.05, 0.18]	0.82
4-month pre-intervention	-	-	0.12	0.02
Immediate pre-intervention	0.11	0.02	0.14	0.02				−0.33***	[−0.16, −0.05]	0.87
Post-Intervention	0.24	0.03	0.31	0.03				0.02	[−0.05, 0.10]	−0.14
4-month follow-up	0.22	0.03	-	-				referent time

^*p<.05.

^**p<.01.

^***p<.001.

^aCondition X Time effects compare immediate pre-intervention and post-intervention measures for the immediate intervention group against 4-month pre-intervention and immediate pre-intervention measures for the waitlist control group, adjusting for fulltime employment status and number of years working in mental healthcare with LGBTQ clients.

^bLonger-term efficacy models compare the pre- and post-intervention measures against the 4-month follow-up measures in the immediate intervention group only.

^cn=118 in full sample and n=58 in the immediate control group for deliberate practice outcomes due to non-response missingness from three participants in the immediate intervention group.

^dFor CBT Skills Deliberate Practice means and standard deviation are reported for raw scores and models are conducted with square root transformed scores to account for violations of normal distribution.

Longer-term follow-up.

In the immediate intervention condition, treatment effects were generally maintained at 4-month follow-up, shown by non-significant effects between post-intervention and 4-month follow-up (see Table 2) and similar levels of outcomes from post-intervention to 4-month follow-up (see Figure 3) for self-reported minority stress knowledge, LGBTQ-affirmative CBT knowledge, LGBTQ-affirmative skills familiarity, and LGBTQ-affirmative CBT skills use, and objectively coded demonstration of simulated practice of LGBTQ-affirmative skills. Self-reported LGBTQ cultural competence continued to improve post-intervention indicated by the significantly lower cultural competence score at post-intervention to 4-months follow-up (d = 0.23). Change in self-reported LGBTQ cultural humility in the immediate intervention condition can be considered delayed (see Figure 3), evidenced by the significant change from pre-intervention to 4-month follow-up (d = 0.25) but lack of difference between pre-intervention and post-intervention.

Figure 3. Longer-term efficacy for immediate intervention condition comparing pre-intervention and post-intervention to 4-month follow-up.

Note. *Indicates significantly lower score pre-intervention or post-intervention compared to 4-month follow-up.

Given that one participant resided and worked outside of the North American context, namely in Uganda, we performed a set of sensitivity analyses removing that participant. These analyses showed no differences in magnitude, direction, or significance compared to the models run with the full sample.

Acceptability.

In terms of acceptability, 74.4% of participants attended more than half of the training and 33.9% attended all 11 sessions. All participants (100%) reported moderate-to-high interest in the training. On average, participants identified Module 2 (Nature and Emotional Impact of Minority Stress) and Module 4 (Mindful Awareness & Minority Stress) as the most helpful. Participants identified the first and last modules, Module 1 (Motivation Enhancement for Treatment Engagement) and Module 9 (Relapse Prevention), respectively, as the least helpful. Almost all participants described the training as moderately-to-extremely informative (97.9%), felt that the four trainers were knowledgeable about the intervention (100%), and believed the training helped them to gain more knowledge about LGBTQ-affirmative CBT (98.8%). Additionally, most participants described the training as moderately-to-extremely helpful in building motivation to change their interactions with LGBTQ clients (96.8%), developing a better understanding of sexual and gender minority identities (96.3%), and preparing to further interact with and care for their LGBTQ clients (94.6%).

Discussion

This waitlist-controlled trial establishes the preliminary efficacy of training mental health providers to deliver LGBTQ-affirmative CBT, an intervention with increasing empirical evidence for improving LGBTQ people’s mental health (Jackson et al., 2022; Pachankis, Harkness, Maciejewski, et al., in press; Pachankis et al., 2020). In this study, mental health providers working at LGBTQ community centers across the US, Canada, and Uganda were trained to deliver this treatment that applies CBT tenets to addressing the cognitive, affective, and behavioral targets through which minority stress undermines LGBTQ people’s mental health (Pachankis et al., 2022). Compared to participants on the waitlist, those who immediately received the training reported significantly greater improvements in expected outcomes, including self-reported LGBTQ cultural competence, minority stress knowledge, LGBTQ-affirmative CBT knowledge, and LGBTQ-affirmative CBT skills familiarity and use; these results persisted four months after training. To determine actual uptake of the therapeutic skills taught in the training and to overcome known limitations of self-report, participants also completed an objectively coded simulated practice exercise, in which participants who received the training demonstrated objectively greater uptake of minority-stress-informed LGBTQ-affirmative practice skills compared to those on the waitlist. Otherwise, only one outcome was not significantly affected by the training, namely self-reported LGBTQ cultural humility. The training was deemed to be highly acceptable, with the majority of participants completing more than half of all training sessions and indicating that the training was interesting, informative, and helpful in instilling new knowledge and motivation for providing care to LGBTQ clients.

This study represents the first randomized controlled trial preliminarily establishing the efficacy of training mental health providers in any type of LGBTQ-affirmative care. Results complement and lend further empirical support to the benefit of such care as derived from existing observational studies (e.g., Phelan et al., 2017), open trials (e.g., Lelutiu-Weinberger & Pachankis, 2017), and studies of healthcare training outside of a mental health context (Jaffer et al., 2016; Lelutiu-Weinberger et al., 2016; White Hughto et al., 2017). This existing research has found that exposure to formal (e.g., training) and informal (e.g., exposure to role models) supports for LGBTQ-affirmative care are associated with outcomes such as reduced provider anti-LGBTQ bias, increased provider motivation to care of LGBTQ patients, and increased provider skills and self-efficacy in delivering such care. By employing an experimental design, the present study shows that receiving formal training in LGBTQ-affirmative CBT specifically plays a causal role in increasing a comprehensive array of self-reported LGBTQ-affirmative knowledge and attitudes, and self-reported and objectively demonstrated skills in delivering LGBTQ-affirmative CBT. Because the training tested in this study was delivered synchronously online and during work hours to the centers in which participants were affiliated, these results support a cost-effective, efficient means to ultimately enhance the mental healthcare available to LGBTQ individuals. Indeed, an existing study finds no difference in online versus in-person delivery of LGBTQ-affirmative provider trainings (Lelutiu-Weinberger et al., 2021), further supporting the future broad implementation of this training.

Results of this study suggest that an 11-hour training in LGBTQ-affirmative CBT can increase mental health providers’ ability to uptake advancements in intervention science that encourage providers to combine core CBT tenets with emerging theory and research demonstrating the impact of minority stress on LGBTQ individuals’ mental health (Balsam et al., 2019; Hatzenbuehler, 2009; Meyer, 2003). Indeed, minority stressors such as anxious expectations of rejection, identity concealment, and self-stigma are postulated to at least partially explain the substantial mental health disparities affecting this population, as they represent psychosocial reactions to the stigmatizing societal structures that disproportionately affects LGBTQ people (Meyer, 2003). To the extent that providers can be trained to address these minority stressors using adapted treatments such as CBT for this purpose, they can play a key role in addressing these population disparities. The training approach tested in this study, by training providers to directly respond to psychosocial manifestations of minority stress in LGBTQ clients’ lives, also represents a means through which providers can learn to concretely assess and address minority stress in treatment as increasingly encouraged by professional organizations (e.g., American Psychological Association, 2021).

This study was conducted in the distinct context of LGBTQ community centers, which predominantly serve LGBTQ clients. Therefore, presumably all participants were operating amidst workplace norms of LGBTQ affirmation and over three-quarters of participants identified as LGBTQ themselves. This distinct context could have affected results in opposing ways. On the one hand, given participants’ professional and personal supports for delivering LGBTQ-affirmative care, some training effects could have been diminished compared to a situation in which this training had been tested in a more general context. Specifically, if participants had already been exposed to the importance of LGBTQ cultural competence and humility and general LGBTQ-affirmative therapeutic skills, they might have already attained a ceiling on these outcomes, thereby diminishing any effect that could be found resulting from this training. At the same time, even providers who possessed these indicators of a more general LGBTQ-affirmative therapeutic stance might not have been previously exposed to LGBTQ-affirmative CBT skills specifically, thereby protecting these more specific outcomes from this prior influence. On the other hand, given that providers were operating in LGBTQ-affirmative work settings and that many identified as LGBTQ themselves, outcomes might have been enhanced compared to a more general setting given that providers might have been particularly motivated to engage in the training and demonstrate uptake of the treatment approach covered in this training. Research with directors of LGBTQ community centers shows that the large majority supports providing resources, including staff time, for their mental healthcare staff to receive this type of training, perhaps suggesting another facilitator of motivation and uptake of this training (Pachankis et al., 2021). Notably, about three-quarters of participants reported a theoretical orientation other than CBT, suggesting that the CBT basis of the treatment approach covered in the training might have been relatively novel to many participants.

Results of this study position the field for several future implementation research directions regarding LGBTQ-affirmative CBT. First, as noted above, given that the present trial took place with providers affiliated with LGBTQ community centers, future research might test the generalizability of the training tested here in settings that might not currently possess the strong institutional support of these centers (Pachankis et al., 2021). Future research that seeks to test the efficacy of such training in other settings such as general non-LGBTQ-specific mental healthcare clinics, college counseling centers, and Veterans Administration hospitals and clinics might assess implementation barriers and facilitators beforehand to ensure institutional buy-in and success of the training. This research could also test the efficacy of delivering such training to mental health trainees in formal training programs in order to instill career-long LGBTQ-affirmative practice skills from the start (Boroughs et al., 2015; Hope & Chappell, 2015). To the extent that this research takes place in diverse geographic contexts that span a range of structural stigma climates, researchers can examine whether structural stigma moderates the efficacy of this training and the mechanisms through which it might do so (Hatzenbuehler & Pachankis, 2021). Second, future implementation research might develop and test ways in which to embed this training in practice settings to sustain its benefits. For instance, many settings, including LGBTQ community centers as documented here, employ relatively junior mental healthcare staff, including trainees, and experience high turnover (Pachankis et al., 2021). To the extent that practice support can be embedded within the setting, the training benefits can reach those who might not have been present at the formal training or those who might need ongoing support in delivering LGBTQ-affirmative CBT. Communities of practice, or networks of providers who together learn new practice skills and provide motivational and instrumental support to each other as they acquire these skills, represent one implementation strategy for ensuring sustainability after the initial training (Barwick et al., 2009). Future research could randomize sites to receive this type of adjunctive support to determine its incremental utility. Third, future research might test the efficacy of training in LGBTQ-affirmative care as applied to other types of approaches, including dialectical behavioral therapy (DBT; Cohen et al., 2021), acceptance and commitment therapies (ACT; Yadavaia & Hayes, 2012), interpersonal psychotherapies (Budge, 2013), and emotion-focused therapies (Medley, 2018), all of which now have clinical guidance and case examples of their successful integration with LGBTQ-affirmative principles, even as they await efficacy testing.

Study results must be interpreted in light of several design limitations. First, as noted above, results must be contextualized within the specific setting in which this training was delivered, namely LGBTQ community centers, a setting with particularly strong facilitators of training success in treatments such as LGBTQ-affirmative CBT (Pachankis et al., 2021). Second, several of the measures utilized in this study were either created specifically for the study or adapted from existing measures. This approach was necessary given that no measures existed to capture some of the distinct outcomes targeted by the training, namely the uptake of LGBTQ-affirmative CBT. Although based on the content of the training, which itself was based on the content of the empirically supported treatment materials, these measures were not validated ahead of time. Likewise, simulated practice videos, while developed based on typical case presentations in trials of LGBTQ-affirmative CBT by therapists from those trials, were not validated in advance. That the training impacted most outcomes in the expected direction, however, lends preliminary support for the validity of these measures. Notably, the one self-report measure that did not evince significant change (i.e., LGBTQ cultural humility) was only modestly adapted for LGBTQ individuals from its existing form and this scale’s moderate inter-item reliability could perhaps explain this pattern of results. Third, the follow-up period was limited to four months and future research could examine longer-term training effects. Fourth, although the training content strove to be inclusive of the full spectrum of sexual and gender minority individuals affected by minority stress, the specific LGBTQ-affirmative treatment approach taught in the training has only been tested in randomized controlled trials of sexual minority men and gender-diverse sexual minority women, most of whom identified as transgender, non-binary, gender fluid, or genderqueer. Although open pilot studies have found LGBTQ-affirmative CBT to be acceptable to transgender individuals (Austin et al., 2018), this population can face distinct mental health treatment needs not necessarily represented by the minority stress focus of the LGBTQ-affirmative CBT principles and techniques covered in the training (e.g., Budge, 2015) Future research is needed to test the efficacy of training approaches that focus on these distinct treatment needs. Finally, study outcomes were limited to the perspective of mental health providers and did not include change in outcomes among the LGBTQ clients who such a training is ultimately expected to benefit. Research that links therapist competence in LGBTQ-affirmative CBT to client improvements in mental health is needed to establish the real-world impact of such training (e.g., Jaffer et al., 2016); indeed, most existing assessments of LGBTQ-affirmative training has been limited to therapists (e.g., Pantalone & Abreu, 2021). Current evidence is, in fact, mixed regarding whether therapist CBT competence is reflected in client outcomes (Davidson et al., 2004; Haug et al., 2016; Kuyken & Tsivrikos, 2009; Rapley & Loades, 2019; Simons et al., 2010). Assessing the impact of LGBTQ-affirmative CBT training in real-world clinical practice would also allow for addressing a limitation of our simulated practice assessment, which only assessed the selection and delivery of specific LGBTQ-affirmative CBT techniques rather than uptake of a broader LGBTQ-affirmative approach to case conceptualization. Future research that studies the application of this training to actual practice would be well-suited to incorporating established measures of CBT case conceptualization (e.g., Kuyken et al., 2016), as adapted to LGBTQ-affirmative CBT (Bucci et al., 2016; Easden & Kazantzis, 2018). This research would also allow for studying whether specific LGBTQ-affirmative CBT techniques can be appropriately matched to the client’s presenting concern rather than just being delivered at all, an important aspect of evidence-based intervention tailoring (Zilcha-Mano et al., 2022).

In conclusion, in this first randomized controlled trial of the preliminary efficacy of training mental health providers in LGBTQ-affirmative CBT, the training was associated with positive improvements across most self-reported outcomes and objectively coded demonstrations of skills uptake. LGBTQ-affirmative CBT served as the basis of the training given its focus on addressing LGBTQ-specific minority stress (Hatzenbuehler, 2009; Pachankis, Harkness, Maciejewski, et al., in press), preliminary efficacy in improving LGBTQ individuals’ mental health in recent randomized controlled trials (Pachankis, Harkness, Maciejewski, et al., in press; Pachankis et al., 2015; Pachankis et al., 2020), and utility for translating professional guidance for LGBTQ-affirmative care into clinical practice (American Psychological Association, 2021). Accumulating evidence of the efficacy of training in this approach can continue to empower mental health providers and the settings in which they work to address the substantial mental health disparities facing LGBTQ populations through the uptake of identity-affirmative, evidence-based clinical practice.

Public Health Significance Statement.

Training mental health providers in LGBTQ-affirmative CBT can reduce the substantial mental health disparities facing sexual and gender minority populations by reducing barriers to receiving evidence-based mental healthcare. This randomized controlled trial found that, compared to waitlist, an 11-week training in LGBTQ-affirmative CBT based in minority stress theory yielded significantly greater uptake of self-reported LGBTQ cultural competence, minority stress knowledge, and LGBTQ-affirmative CBT knowledge, and skills familiarity, and objectively coded skills use for mental health providers working at LGBTQ community centers across the US and internationally. Dissemination of LGBTQ-affirmative CBT through efficacious training programs embedded within frontline settings might represent an efficient means for improving mental healthcare for the LGBTQ population.