This study was developed as an extension of the key informant interview process for the multilevel TransHealthGUIDE intervention project, a National Institutes of Health–funded U01 multilevel intervention designed to improve access to care and support for gender-diverse youths. The primary youth-facing intervention will be a mobile app to provide resources and support, and a key informant interview process was designed to improve the utility of the final product by making certain it would address participant needs. Initial key informant interview guides were developed through an iterative process of discussion with researchers and staff members who either identified as part of the transgender umbrella or who had substantial clinical or research experience with this population. After the initial round of key informant interviews were completed, there was a recognized need for feedback from larger numbers of diverse young TGD individuals than were accessible for the interviews, which this protocol was designed to elicit. The goal of this survey was to identify critical content areas and styles for the proposed app-based intervention.

After review, this protocol was declared exempt by the Boston Children’s Hospital Institutional Review Board (IRB-P00043127).

Participants were recruited using the Prolific survey platform, as the site is known to have a large, diverse sample available including many transgender young adults. Recruitment on Prolific allows for the selection of an anonymous, targeted demographic sample within their overall participant panel consisting of an international sample of individuals ages 18 years and older. In this case, our recruitment strategy was to request a sample of young adults between the ages of 18 and 25 years who lived in the United States and identified as transgender according to platform screening questions, which may be updated by users at any time. Two separate samples of 50 individuals who identified as White and 50 individuals who identified as a race other than White were recruited for this protocol to mirror our intended demographics for the overarching project. The platform stops issuing invitations, and deletes unanswered invitations, when the prespecified sample size is reached. Although the study automatically closed after 100 responses were collected, 2 individuals experienced technical difficulties when trying to take the survey and did not complete the protocol before the study slots were full. They anonymously reached out to the study team expressing that they wished to participate and were given the opportunity to do so. As such, a total of 102 individuals completed the survey. Of note, 4 individuals who were classified as non-White by the platform screener self-identified as White on the survey, and they were classified as White in the analysis.

Individuals who were eligible to participate in the study received an invitation using the Prolific platform [23] between August 29, 2022, and August 31, 2022. If after reading the study description they chose to participate, they were directed to the survey (Multimedia Appendix 1), which was housed on the Boston Children’s Hospital Research Electronic Data Capture (REDCap) site [24,25]. After completing the survey, which took on average slightly longer than the estimated 10 minutes (13 minutes, 3 seconds), they were returned to Prolific platform using a completion code. Participants were approved for a US $3 payment within 24 hours of completing the survey. This corresponded to greater than US $13 per hour, which is substantially higher than the required minimum of US $8 per hour required by Prolific. Participants remained fully anonymous.

The survey, which included both structured and open-ended response elements, was designed to include the core domains that had been established during the key informant process but achieve greater granularity in areas central to the technical and behavioral design aspects of the project. Items were also included in the survey that specifically addressed areas where key informants had indicated a need for additional data. Prior to launch, the survey was reviewed by a diverse range of staff to assure it was both comprehensive and affirming of identity.

Qualitative data from open-ended questions were analyzed by question or topic to understand specific issues around content development for TGD youths. A conventional content analysis of open-ended survey responses was performed by 2 reviewers (ERB and MQ) in Microsoft Word [26,27]. Reviewers independently analyzed and identified both themes and the number of people who endorsed them. Discrepancies, primarily in phrasing of themes, were resolved through a process of discussion. The number of nonnull respondents for each question is included in the caption of the analysis. Null responses were inclusive of only “no,” “no response,” or “n/a” typed into the field, and these were not counted or analyzed. Quantitative analyses were descriptive as no a priori hypotheses were being tested, given the exploratory nature of this study. Individuals were categorized as BIPOC if they endorsed at least 1 racial or ethnic identity other than White. Data are presented as frequencies and percentages for categorical variables or means and SDs for continuous variables. Descriptive statistics are stratified and presented for BIPOC, White, and all TGD young people in the sample. All analyses were implemented in Stata (version 16.0; StataCorp).

Characteristics of the study sample are displayed in Table 1. Overall, 102 gender-diverse individuals between the ages of 18 and 25 were recruited into the study. While the goal was to recruit a balanced sample, the platform screening misclassified some individuals who self-identified as White as BIPOC, which led to a slight oversampling of White participants.

Of the BIPOC study population, approximately half identified as Hispanic or Latino/a/x, a third identified as Asian, and a quarter identified as Black. Individuals were allowed to select multiple racial identity options, and more than half (24/46) of individuals in this group selected more than 1 racial identity for their self-description. Geographically, there was a slightly higher proportion of individuals who grew up in the South of the United States compared to the Northeast, Midwest, or West, particularly within the BIPOC study population. There was also substantial religious diversity among the population’s caregivers. The largest proportion of individuals identified as having been raised Catholic (40/102), followed by Orthodox (16/102), atheist (18/102), and agnostic (15/102). The mean Patient Health Questionnaire-2 score of 2.6 suggested a tendency toward mild depression in this population (Table 1).

When asked about what information sources they had used to access information about gender and health as a teenager, social media was by far the most commonly used option (84/102). However, more than half of individuals also identified using websites (75/102), YouTube (64/102), and friends (62/102) as sources of information (Table 1). The most common topics participants had searched for were “talking about gender identity” (69/102), “nonmedical ways to express gender” (64/102), “finding support” (63/102), “finding web-based community” (60/102), “finding mental health support” (57/102), “telling people about their gender or coming out” (51/102), “how to transition in various contexts” (49/102), “sexuality and/or sexual health” (49/102), “medical gender care” (49/102), “social transition” (48/102), and “legal transition” (45/102). A substantial fraction (70-94/102) of individuals stated that the information they found for most of those topics was at least moderately useful. However, less than 80% of people stated that information was at least moderately useful for several of the most commonly searched topics, including nonmedical ways to express gender, medical transition, and sexual health (Table 2).

When individuals were asked to identify the specific websites they preferred for accessing information about gender and health, the most commonly identified sites were YouTube (64/102), Reddit (62/102), Twitter, (53/102), TikTok (48/102), and Tumblr (44/102). When asked what types of content they liked for learning about gender and health, the most popular styles and formats of content were personal stories (70/102), followed by text (63/102), short videos (1-3 min, 60/102), and infographics (59/102). By far, individuals’ preferred sources for gender-focused content were trans people in general (84/102) and trans friends (83/102). Individuals could select multiple response options (Table 3).

Qualitative data were analyzed by question or topic to understand specific issues around content development for TGD youths. Approximately half the sample (49/102) responded to at least 1 open-ended question, and 35 of 102 (34%) responded to 2 or more. In the first question young adults listed several topics on which they would have liked information when they were growing up. These ranged from general topics, such as understanding gender identity development and evolution, to more specific topics reflecting their individual circumstances and needs. Young adults commented on the need for support in finding TGD-affirming medical care, navigating familial, relational, and cultural contexts, and finding community. There were also requests for information about narrower topics, such as neopronouns and voice training (Textbox 1).

When asked about additional topics they thought should be included in the final app, there were multiple requests for information in broad areas, such as “LGBT topics” and comprehensive lists of identities. A number of young people mentioned the need for information that addressed complexities of intersectionality and cross-cultural identity development around gender. Requests were made for specific inclusion of topics relevant to transmasculine and nonbinary youths (Textbox 2).

BIPOC young adults endorsed several specific challenges related to race and ethnicity when trying to access information or care. Some of these concerns reflected difficulty communicating with parents about concepts, such as nonbinary gender, that were difficult to discuss in highly gendered languages (eg, Spanish) and cultural contexts. Others involved navigating racial stereotypes or feelings of isolation around their intersecting racial and gender minority identities (Textbox 3).

Young adults raised several potential privacy and safety concerns about the proposed project. Some young adults expressed fears about hacking and doxing (publishing someone’s private or identifiable information on the internet with malicious intent), while others were more concerned about family members or parents finding the app on their phone. They were very clear that strong data protection—as well as password protection and a way to rapidly exit the app or switch to an app not related to lesbian, gay, bisexual, transgender, queer (LGBTQ+) topics—would be necessary to minimize risk to participants. Several people also stated that the app should not be explicitly gender-related in appearance to protect users. There were fewer concerns about accessibility, although the need for alt text (descriptive text about an image’s contents) and closed captioning were mentioned (Textbox 4).

TGD young adults endorsed searching for a wide variety of information about gender affirmation, health, and social needs on the internet. However, importantly, their answers also suggested that, as young adults, they have become aware of a number of topics that would have been useful for them to learn about when they were younger, topics that at the time they did not know to look for. This suggests a need for broader education and discussion of gender identity development across society, as the lack of language for a concept can make it difficult to locate information about that concept. While there is no high-quality evidence suggesting that learning about gender diversity leads to increased adoption of gender-diverse identities (ie, the debunked social contagion hypothesis), our clinical experience suggests that many TGD young people struggle to understand how they fit into the world when they are raised in environments that don’t discuss their existence as a possibility [12,28,29].

When looking at the specific topics of interest, the following were all of interest to the TGD young people in this study: hormones, and surgical care, other medical gender-affirming care, mental health support or therapy, and sexuality or sexual health. However, the most searched topics reflected more general questions and concerns about gender identity and nonmedical ways to express it, something that has also been reported in previous studies of web-based behavior [30]. Given the data collected, we cannot ascertain the reason why fewer participants reported searching for information on medical care for gender affirmation. Reluctance to search for information on medical care could result from concerns about privacy or lack of desire for medical affirmation, or both. This finding may also reflect the gender identities of the individuals surveyed, which skewed heavily toward nonbinary identities, a group that may be less likely to be interested in medical gender affirmation as compared to binary transgender youths. The heavy weighting of our survey questions toward medical affirmation reflected the planned intervention being designed by medical providers, but it may also reflect some unconscious assumptions around trans normativity that posit a gender journey inclusive of some degree of medical and surgical affirmation being of interest to the majority of transgender youths, when that is not necessarily the case [31-33]. Indeed, that concern was raised by several respondents, and a de-emphasis on binary and medicalized genders has been incorporated into the app design.

More than 20% of study participants stated they had trouble finding information about nonmedical gender affirmation, which may include changing one’s name and pronouns; or using hairstyling, clothing, makeup, padding, or body-modifying garments to create an appearance that more accurately reflects a person’s internal sense of self. Nonmedical gender affirmation may also involve binding or tucking to conceal highly gendered areas of the body that are associated with discomfort or dysphoria. Both improper binding and tucking have the potential to lead to injury, ranging from skin breakdown [34] and infection to testicular torsion [35]. Accurate information about the benefits of social gender affirmation may be useful to help young people enact them in as safe a way as possible and to advocate for their needs with family, school, and other environments.

TGD youths expressed a significant desire for resources that facilitate connection to other members of their community. This likely reflects both the importance of social engagement for well-being and the challenges in identifying support experienced by members of marginalized or multiply marginalized groups, particularly during the COVID-19 pandemic [36]. An interest in community connection may also underlie participants’ strong preference for learning information about gender and health from other TGD people. Given the preponderance of politically motivated misinformation campaigns targeted at TGD youths, a preference for trans-produced content may also reflect conscious or subconscious attempts to screen for content that is less likely to be transphobic or otherwise hostile—TGD young people scanning for safety. While not yet studied in TGD communities, there is also growing literature about the advantages of being treated by others with similar identities for people from racially minoritized groups [37-39].

Participants’ interest in content from TGD people aligns with a strong preference among respondents for learning specifically from the personal stories of other TGD people, a theme that was also articulated in key informant interviews performed with youths and caregivers as part of the overarching project (data not yet published) and also reported by other investigators doing similar work [40]. Historically, digital storytelling has served as an effective learning tool for patients, caregivers, and providers and a mechanism for community outreach in other areas [41]. This suggests tremendous potential for its use in a TGD health intervention. While youths using internet sources for health information used Google to initiate searches and visited medical websites more often than social media platforms, TGD youths primarily used social networking websites for gender-related and health information [28]. This difference may be due to lack of accessible, age-appropriate, and reliable gender-affirming information on health-specific websites [42].

As part of this formative evaluation, we were interested in determining whether TGD young people had strong preferences for styles of information as well as source when seeking information related to gender and health. Previous research has suggested a preference for interactive content and visual content (videos, images, and animation) among adolescents seeking general health information on the internet, but it was unclear whether these preferences would convey to TGD young people [43]. In this study, the most popular platform for information-seeking was YouTube (video-based platform or application—not inherently social or interactive). Reddit (text- and photo-based discussion forum) and Twitter (multimedia platform—highly interactive) were also very highly endorsed; both are text-based platforms. Surprisingly, although TikTok (short videos—strong social content) was brought up frequently in individual key informant interviews and was the second most frequently used social media platform among US teenagers in 2021, it was not one of the top information sources in this survey [44]. This may reflect how survey questions were framed: in this survey, participants were asked to reflect on the sources they previously used, which may have been prior to TikTok’s creation. The preference for text over videos of any length was also reflected in the types of content for which young people expressed a preference, suggesting the importance of varied content types to support young people’s information-seeking behaviors.

Another important aspect of intervention development raised by participants was the need to protect the data and privacy of individuals interacting with any TGD-focused resources and media. For youths whose TGD identity is unknown to their caregivers, disclosure of this information may put them at risk of abuse or housing loss. TGD youths with unsupportive caregivers have become unhoused as well as experienced both physical and emotional abuse [45,46]. In addition, in the current regulatory environment, supportive caregivers may be at risk of having custodial rights revoked, or undergoing investigation by child protective organizations [12]. Given safety concerns, and in line with suggestions from participants, plans for our proposed interventions now include ensuring information is available in a form that is not clearly visibly branded as trans to a casual observer and creating a quick escape option similar to those seen on websites and apps designed as resources for individuals experiencing abuse [47].

This study has several limitations. Individuals must be aged 18 years or older to join the Prolific platform, so we were unable to survey those aged under 18 years. This is potentially important, as youths aged under 18 years may have different preferences due to a different life stage and our planned educational intervention has a target audience that includes individuals aged 13-25 years. In addition, there is a risk of selection bias; the sample of TGD young people using Prolific is almost certainly not representative of the population of TGD young people as a whole, likely being at a minimum more computer literate and having needed to hear about the platform to enroll. Participants were asked to recall informational sources, unmet needs, and topics from their early teens, which may have introduced recall bias; the accuracy of self-reporting on past information-seeking behaviors is unknown. Preferred information sources have also likely shifted in the years since the participants reached the age of majority, as young people’s social media preferences change substantially over time. Strengths of this exploratory study include enrolling a purposive sample of TGD young adults from across the United States comprised of 50% BIPOC youths, gathering exploratory quantitative and qualitative data informed by findings from key informant interviews, and centering on the experiences of young TGD people.

The findings of this study underscore the need for accessible, expert-informed information for TGD youths, particularly more information generated for the TGD community, by the TGD community. TGD-identified and TGD health and research experts have suggested a “nothing about us without us” framework [48,49], which can be adapted for use in research with TGD young people. TGD youths are looking, with mixed success, for information on a wide range of topics, from general overviews of gender identity to more specific questions about gender affirmation, building community, and sexual health. There is a need for intentional interaction between community, researchers, and medical professionals to generate and codevelop trustworthy resources on these topics. Findings suggest that involvement of and presentation by trans people is a key element of establishing trust with TGD youths. Further, there is a need for information to be presented using multiple methods (ie, video, text, and audio) to meet the varied preferences of youths. Finally, youths identify web-based safety as a critical area for content providers, and individuals working in this field should recognize and address the fears that TGD youths have in both web-based and in-person spaces.