Drivers of Mental Health Disparities for Diverse, Rural, LGBTQ+ Communities

Transcript

SHAUN DAVIS:  Hi, everyone.  My name is Dr. Beshaun Davis.  I am here from NIMH.  I'd like to welcome you to our Office of Disparities Research and Workforce Diversity webinar series.  Today we're kicking off our webinar series with a presentation on rural LGBTQ mental health by Dr. Sarah Murray and a team of her collaborators from across Johns Hopkins University and Emory University.

So, first off, I want to welcome you all to this webinar series and thank you all for coming.  Just for, by way of introduction, I'll sort of share a little bit about sort of the goals of the webinar series.  In general, we have this webinar series every year, and we tend to focus on promoting research from investigators focusing on minoritized populations, women's mental health, rural and American, Alaska Indian and Alaska Native mental health, workforce diversity, social determinants of health and a wide variety of topics related to disparities in mental health.

So, we're kicking off with this one focusing on rural LGBTQ mental health for a couple of reasons.  To start, oftentimes in the research on the LGBTQ+ community, we tend to focus on people in urban or sort of suburban areas, and oftentimes the LGBTQ community in rural settings is kind of left behind.  There's often people talk a lot about the sort of "it gets better" messaging in that campaign from a few years ago, where oftentimes the message was, it gets better, you move to a big city and things get better for you.

But there are tons of queer folks who live in rural communities and their mental health concerns are just as important, right? 

So, the work presented today will highlight some of the concerns of folks in rural communities as well as looking at unique risk and protective factors linked to suicidality, depression and a variety of mental health outcomes for folks in these communities.

So, I'm going to hand it over to Dr. Murray in a moment, but before I do, I just want to briefly introduce her, and she can introduce her wonderful team who are all here and will be co‑presenting today.

So, Dr. Murray is an associate professor at Johns Hopkins University, where she does a variety of work on sort of minoritized populations and particular work focused on minority stress and the impact of that on mental health concerns.  Some of those key themes will be present today in this presentation.

And so, by way of just making sure we have enough time for them to present and to get all of the wonderful insights from this study that's underway, I'm going to go ahead and stop there and hand it over to Dr. Murray to continue with introductions of her team and to get started with the presentation. 

Towards the end we'll do a moderated Q&A.  So, if you have questions, please put them in the chat throughout, and we will answer them in the last 30 minutes of the webinar.  So, thank you all for attending, and I'll hand it over to Dr. Murray. 

SARAH MURRAY:  Fantastic, thank you so much.  I'll get my screen shared for everyone.  Okay.  Good afternoon, folks.  My name is Sarah Murray.  Thanks so much to Dr. Davis for the introduction.  I am thrilled to be here today to share with you some of our ongoing efforts at better understanding stigma and discrimination and how they may drive mental health disparities for diverse, rural, sexual and gender minority communities.

In particular to tell you about our ongoing NIMH‑funded study, on note as we begin, we're going to try to be consistent with terminologies as we go through, but I recognize that there are multiple terms that people use when talking about gender and sexual diverse communities and that they may not always perfectly coincide with the way people describe their communities or their own identities.  So, we're using the term "sexual and gender minorities, consistent with the NIH research office, but we do appreciate that this is an imperfect term.

The work we're presenting today is supported by a big, wonderful team.  I feel so fortunate to get to work with all of the smart and thoughtful researchers listed here on this screen.  Our team is mainly divided across two institutions.  So, Prism Health at Emory Rollins School of Public Health under the leadership of Dr. Travis Sanchez, who is a co‑leader but also someone I consider a mentor, and at the Johns Hopkins School of Public Health.  And not to be forgotten we also have Dr. Jennifer Glick, who is now at Louisiana State University.

Given our large team and collaborative structure, you won't only be hearing from me today, but as Dr. Davis introduced from some of the other folks who are really essential to making this work happen.  So, Savannah Winter is program coordinator at REALM and focuses on LBGTQ community building resiliency and empowerment and is coming from a background of community advisory work. 

In addition, Mariah Valentine‑Graves is program director of REALM as well as two other large national survey studies hosted by Prism Health.  She has a passion for health equity, harm reduction and technology‑informed and innovative recruitment and retention of research participants and has stayed at Emory after completing her MBH. 

And at Hopkins, my colleague/assistant/research professor Dr. Shoshanna Fine, whose research encompasses two frequently overlapping areas ‑‑ assessing risk and protective factors for the health and psychomental health and psychosocial development among disadvantaged adolescents around the world and evaluating culturally appropriate prevention and intervention strategies.

Kirsten Siebach is at MSW and is currently a PhD student with interests in the impact of the structural environment on the well‑being and mental health of the LGBTQ+ community. 

And I'm a psychiatric epidemiologist whose research focuses on using mixed methods to understand and intervene upon pathways between violence and stigma and mental disorders globally.  
The zen of much my research focuses on better understanding experiences of stigma and discrimination among sexual and gender minority adults.  And I really have a passion for intervening to build environments in which sexual and gender diverse folks can fully be themselves and thrive.

I also just want to note that this presentation does contain discussion of suicidality.  And on the screen, we have listed some of the resources that we remind our participants of.  And we want to remind you of them as well, in case someone you know or yourself should ever need them. 

We're not necessarily endorsing any one of these in particular, but as someone for whom mental health treatment has been really instrumental, I just want to take a moment to encourage us to take care of ourselves and each other.

Moving into our rationale for the study, I think many of you will be familiar with some of those disparities in mental health that are documented as being experienced by SGM folks.  But briefly and globally, overall, gay, lesbian, bisexual individuals have been found to have over twice the odds of reporting a past‑12‑month mental disorder relative to heterosexual individuals.  In the U.S. specifically, we see elevated prevalence of mental illness and suicide among sexual minority adults relative to heterosexual adults and among individuals who are transgender or gender diverse relative to cisgender folks. 

To provide a few examples, the study using 2015 NSDA data found that one in three sexual minority adults compared to one in five heterosexual adults reported a lifetime mental illness.  And data from the same year, from the national transgender survey, found that nearly 40% of transgender adults reported a serious past distress as compared to 5% of the general population.  And 40% also reported a lifetime suicide attempt, and 7% a past‑year attempt.  These disparities are really pretty staggering.

They've been documented as occurring across the life course.  And I think that's really important.  For instance, sexual minority youth have been found to have three times the odds of suicidal ideation compared to heterosexual peers.  And longitudinal research has shown how these disparities persist as adolescents move into young adulthood. 

As people age, we see similar patterns, actually.  For instance, middle‑aged and older gay men and bisexual women were found to have twice the odds of mental disorder and bisexual men, three and a half times the odds relative to heterosexual peers.

Unfortunately, much of the literature, particularly outside of adolescents and young adulthood, is cross‑sectional.  There's definitely a need for longitudinal study with adults. 

How do we begin to make sense of this disparity in poor mental health and suicide?  Well, increasingly we see from research that stigma processes from the structural level all the way to the individual level work in concert to drive risk for sexual and gender minority communities.

Probably the most common framework we used to unpack this is that of minority stress.  The framework conceptualizes gender and sexual minority stress processes that occur distally or external to the individual, such as experiences of discrimination driven by heteronormative social structures and also by cisnormative social structures and also structural stigma reflected in laws and policies.  And these distal processes really shape proximal processes in turn that are internal.  So, things like the internalization of stigma in anticipation of stigma by minoritized individuals. 

And there are critical psychological processes in response to these experiences such as coping and emotion regulation, how we engage in social interpersonal relationships and cognitive processes that ultimately shape if and how mental health is affected by these stressors.

I want to highlight a recent example of this in the literature of this framework being applied to understand mental health disparities for gender minority communities specifically.  Wilson and colleagues conducted a recent systematic review and were actually able to find studies among 47 different samples that looked at stigma in association with mental health disparities for gender diverse folks and found that both distal and proximal stressors were correlated with both depression and anxiety as would be expected in the minority stress framework, notably, with the proximal stressors generally showing stronger associations.  Though they did note a need for greater rigor and measurement of distal factors.  And notably as well few studies assessed suicidality, and those that did varied a great deal in how it was measured.  So, this really couldn't be analyzed across studies. 

To summarize, while there's great research out there documenting mental health disparities for SGM folks and understanding how minority stressors and stigma may drive these disparities, much of our knowledge, particularly among adults, is based on cross‑sectional studies.  A lack of longitudinal data really prohibits us from actually unpacking these effects as they relate to risk with attention to factors that are going to be modifiable by intervention.

Notably there's also some limited heterogeneity still on how we're engaging sexual and gender minority populations in research with relatively less research on transmasculine non‑binary folks and lesbian and bisexual women.  The current literature is also very risk focused, and there's a real need to understand protective factors to promote interventions that could actually foster positive mental health and work from a prevention framework.

So given that, I'd like to explain a bit more why we set out to do our longitudinal study among heterogenous STO sample, building upon early efforts at integrating minority stress and suicide theories and specifically why we're focusing on rural communities.

So, if we go back to our minority stress framework, there's been some great efforts at actually starting to unpack the right side of that framework I showed you earlier where we had suicide listed in that far right box.  This really is a complex outcome that involves ideation or thoughts about death or harming oneself or ending one's life.  But these thoughts only progress to action or an attempt to end one's life, thankfully, for a minority of people. 

Ideation to action frameworks have been developed to spell out this pathway in more detail and allow us to consider how predictors may actually vary throughout that process.

The most widely used intent to action framework is the interpersonal psychological theory of suicide, which highlights things like thwarted belongingness or an unmet need to belong, which we can pretty easily imagine how that might be driven by stigma and perceived burdensomeness. 

So that is really sort of how people understand themselves as being or not being a burden on others.  And that can reasonably be understood as a reaction to things like economic marginalization that could actually be the result of distal gender and sexual minority‑related stressors.

And these factors are hypothesized to be even more proximal to suicidal ideation than mental disorders themselves.  Capability for suicide lies on the pathway from ideation to action and may be affected by things like sensitization to loss or pain that can potentially be a result of exposure to trauma and experiencing trauma, but also could be affected by things like access to means to actually enact one's plan for suicide and impulsivity. 

And we're starting to see support of this in studies with sexual and gender minority populations, so Teston [phonetic] colleagues looked at theories in conjunction with one another to understand suicide disparities experienced by transgender folk and found that thwarted belongingness and perceived burdensomeness mediated the relationship between distal and proximal minority stressors and suicide ideation.

So, with that I want to turn it over to Kirsten to tell you a little bit more about why we want to base the study in rural areas specifically. 

KIRSTEN SIEBACH:  Thanks, Sarah.  To shift gears, we want to focus on why we're focusing specifically on rural SGM communities.  To date, research among SGM communities has focused almost exclusively…

SARAH MURRAY:  We may have lost Kirsten.  I think we may have some Internet issues.  I apologize.  I'm going to jump in and fill in for Kirsten until we can get her back because I might be a poor substitute for her, but bear with us while we get her back. 

I think what Kirsten was beginning to explain is that there are actually 46.1 million Americans who live in rural areas in the United States.

And approximately 3.8 million of those identify as sexual and gender minorities.  To date, research among sexual and gender minority folks has focused almost inclusively on those who live in urban settings, not totally but predominantly.  The studies in rural areas have actually tended to focus on youth, a very important population.  But we also just really don't know much about the experience of SGM adults living in rural parts of the country.  Yet, rural sexual and gender minority populations might have unique experiences that are important to understand so we can design interventions to improve outcomes.

We might infer that rural sexual and gender minority populations have limited social connectedness or capital because of their gender identity and/or sexual orientation as well as other intersectional identities.  They may have access to fewer affirming spaces and there may be higher religiosity that could be unaffirming of diverse identities and lower access to mental health care.  And all of these factors can really contribute to poor mental health outcomes.

Importantly, I think we may also find unique strengths and aspects of rural life that may promote mental well‑being among the rural SGP community ‑‑ sorry, among the rural sexual and gender minority community.  Close family ties and strong community networks potentially could provide opportunities for increased social support, if families are supportive, and could be an established protective factor against poor mental health.

And when we go back to our framework that we looked at before, I feel like there is sort of an urge to say we think rurality matters in our analyses.  We can put interaction to the term the risk is higher for rural versus urban individuals collected on this population.  And then we sort of check the intersectionality box, right? 

But I think the point we want to make today is that when we think about how and why rurality matters, we need to actually think across this whole process.  So, if we start over here in the distal minority stressors box, it may be that rural sexual and gender minority folks could experience a very different policy environment compared to urban folks.  So, what you're seeing here is a map from the Movement Advancement Project that documents more than 50 sexual and gender minority‑related laws and policies and creates a policy tally.  And we can immediately see that these regional differences may correlate with degree of rurality in terms of whether and how much people are actually exposed to negative or harmful policies and the lack of supportive policies.

If we move to the psychological processes box, we can think about overlaps as it relates to economic marginalization of these communities and economic change and hardship and job loss that may actually disproportionately impact minoritized communities and contribute to things like hopelessness and perceived burdensomeness. 

We can also think about a greater degree of potential social isolation in small communities where there might not be established networks or affirming spaces like we mentioned.  But also, I think something that's been really overlooked is the many ways in which religion and religiosity may actually look very different in rural America.

For instance, both religiosity and attendance of religious services has been found to be higher in rural versus urban areas.  And in theories of suicide, religiosity in the study as a protective factor.  And there's actually support for this in the general population.

But conversely, a greater importance of religion has actually been found to be associated with greater recent ideation and attempts for young sexual and minority adults for suicide.

And things like sexual‑orientation‑change efforts, 80% of which are implemented by religious leaders, have been associated with twice the odds of suicide ideation and greater attempts by sexual minorities. 

And you might imagine, in rural areas, if your religious institution is supporting and engaging in those kinds of harmful practices and endorsing stigmatizing attitudes, there may not be many or any other options to engage a religion or foster spirituality like you may have in an urban space that might be affirming.

And if we move to think about capability, one thing we know that's incredibly important and related to access to means in the United States for suicide is gun ownership.  You can see from the map on the left, states with a high level of gun ownership also have a high level of rurality, meaning that that's something you really need to consider in thinking about whether the research from more urban spaces is generalizable. 

Given the range of risk factors that could be present in the rural sexual and gender minority communities is the fact that there's really limited research among sexual and gender minority adults in rural areas.  We felt it was really important to focus in on this population. 

So, with that, I'm going to pass it over to Dr. Shoshanna Fine, who I'm hoping whose Internet is working, to give an overview of the REALM study that has grown out of this concern and this need. 

SHOSHANNA FINE:  Thank you so much, Sarah, and hopefully everybody can hear me, and my Internet stays good. 

We're going to switch gears now, and we're going to talk about the aims of the REALM study and specifically about how it fills the key gaps and evidence that was just covered so well by Sarah. 

So, REALM stands for Rural Exploration and Approaches for LGBTQ+ Mental Health.  As Sarah mentioned, REALM it's a collaboration between Johns Hopkins University and Emory University, which builds on Emory's really strong prior work recruiting and collecting working with SGM persons online.  And it's funded by an R01 through NIMH. 

The overall goal of the study is to build longitudinal evidence within rural SGM populations on the complex and often quite nuanced interrelationships between minority stressors, such as stigma; mental health conditions, such as depression and suicidal behaviors.  Specifically, we're focusing on around 2500 SGM adults living in rural areas of the U.S.  And we're also intentionally sampling a diverse group that includes both sexual minorities and gender minorities.

Next slide.  I'm now going to briefly walk through the three aims for the REALM study alongside their associated methods. 

So, the first aim is to determine whether classes of stigma, discrimination and traumatic experiences vary across subgroups of rural SGM adults and whether these classes are associated with an increased prevalence of depression, suicidal ideation and suicide attempts.

In order to carry out this aim, we're using online processes to recruit and enroll a diverse sample of up to 2500 SGM adults, aged 18 plus, again, living in rural counties and small metropolitan areas of the U.S.  This sample includes five distinct subgroups.  So we have 500 cisgender women who identify as sexual minorities; 500 cisgender men who identify as sexual minorities; 500 persons who were assigned male at birth and identify as a woman, female and/or transfeminine; 500 persons who were assigned female at birth and identify as a man, male and/or transmasculine; and finally 500 persons who identify as sub‑other gender, including nonbinary, gender nonconforming and/or agender, regardless of sex assigned at birth.

Also, in order to explore differences by race and ethnicity across these subgroups, we aim to enroll racial and ethnic minority persons at a level that's, at a minimum, representative of the national demographics of the rural U.S. 

I'll leave a more detailed description of our online recruitment processes to my colleagues, but I'll mention that all potential participants go through screening, a very rigorous process of identity validation and informed consent prior to filling out the baseline questionnaire.

So, to carry out aim one, we'll be using baseline data from the study to conduct a latent class analysis, or LCA, exploring how clusters of distal minority stressors may be differentially associated with mental health‑related outcomes.

Specifically, we'll be using LCA to identify clusters or classes of participants based on their patterns of endorsement across indicators related to stigma, to discrimination and to traumatic experiences.  And we'll then use a latent class regression approach to assess associations between class membership, on one hand, and prevalent depression, suicidal ideation and suicide attempts on the other.

Initially, these analyses will be conducted separately and compared among the five SGM subgroups.  And we'll explore variations by age and by racial and ethnic identities.

The second aim is to determine whether the classes identified in aim one is associated with an increased incidence of depression, suicidal ideation and suicide attempts among rural SGM adults.  You'll notice that first aim focused on cross‑sectional prevalence, and this aim focuses on longitudinal incidence. 

In order to carry out this aim, we're following the baseline cohort over a 12‑month period with participants asked to complete follow‑up questionnaires every 3 months, including 3‑month, 6‑month, 9‑month, and finally 12‑month questionnaires.

To carry out aim two, we'll be using this longitudinal data to conduct Kaplan‑Meier and Cox proportional regression models to describe the associations between the latent classes of distal minority stressors from the previous aim and then the incidents of depression, suicidal ideation and suicide attempts over the 12‑month follow‑up period. 

As before, these analyses will initially be conducted separately and compared among the five SGM subgroups and subsequent analysis will also explore stratification by age and by racial/ethnic identities as before.

Also, guided by the theoretical framework that Sarah presented earlier, we'll also be exploring potential mediators and the relationships between our distal minority stressors and mental health‑related outcomes.

So, for example, we'll be exploring the role of psychological risk and resilience processes, such as hopelessness on the risk side and coping skills on the resilience side; proximal minority stressors, such as internalized homophobia and transphobia; and additional mental health conditions, such as substance abuse, anxiety and PTSD.

The third aim moves beyond simply evaluating the relationships between minority stressors, mental health conditions and suicidal behaviors to really better understand how to actually intervene upon these relationships.

And specifically, we aim to measure preferences for technology‑delivered preventive interventions for depression and suicide tailored to rural SGM persons actual experiences of stigma, discrimination and trauma.

Aim three will be accomplished through a discrete choice experiment, or DCE, with REALM participants.  Without going into too much depth here, I want to describe DCEs in slightly more detail, as I know this isn't a familiar method for many folks.

So, at its essence, DCE is a health economics method that can be used to elicit and quantify stated preferences from people, i.e., what people say that they want, in the absence of revealed preferences or what people actually do.

And this is really useful because it allows us to explore and place a value on products or services that don't actually exist yet.  One of the basic assumptions of DCEs is that people make tradeoffs when engaging in a decision by choosing an alternative of a product or service that offers them greatest utility or benefit.

So, during a DCE, participants are presented with a series of hypothetical scenarios in which they have to choose between two or more alternatives of a product or service, with each alternative consisting of a specific set of components or attributes, as they're known in the DCE lingo.  And the structure of this experiment allows us to actually quantify the relative utility of each of these components.

In our case, we'll be using a DCE to quantify the relative acceptability of theoretical components for a technology‑delivered intervention targeting depression and suicide among rural SGM adults.  So, the types of components we'll be assessing in this case will likely include things like cost, social networking capabilities, content moderation approaches and specific therapeutic strategies.

In order to develop the theoretical components for our DCE, we'll start by looking at preliminary findings from aims one and two to understand which risk factors may be most important to target in this population.

We'll also carry out consultation with stakeholders who have specific expertise in technology‑delivered mental health interventions.  And after the 6‑month follow‑up, we'll conduct a series of qualitative interviews with selected participants, first to directly elicit their preferences around technology‑delivered interventions, and then to help refine the content of the experiment itself.

All of this groundwork will lead up to the actual DCE, which will be carried out with all study participants who complete the 12‑month follow‑up.  And the ultimate goal here, of course, is to inform the development of a novel technology‑delivered depression support and suicide prevention intervention with findings from the DCE used to help ensure that the intervention's content and delivery strategies are tailored, acceptable and relevant to SGM adults living in rural areas. 

So now I want to move us into the next section of our presentation, which is focused on where things stand to date in the REALM study.

To start, we'll be talking about considerations related to our measure’s selection, including how we have built strength‑based perspectives into this work.  We'll talk about our community advisory board, staff training and safety procedures.  And finally, we'll update you on our ongoing recruitment efforts, including both baseline recruitment, as well as our 3‑month and 6‑month follow‑ups.

I'm not going to spend a lot of time talking about our measures, but what I do want to emphasize is that, in measured selection, we prioritize the selection of validated tools that have previously been used with SGM populations.  Our mental health‑related measures are listed on this slide with our primary outcomes in gold. 

And as you can see, we're using the PHQ‑9 to measure depression and the Electronic Columbia Suicidality Severity Rating Scale, or eC‑SSRS, to measure suicidal ideation and attempts.  And for both of these measures part of the appeal is that they're relatively brief and can be scaled to measure both lifetime experiences as well as experiences since last assessment.

Again, I want to just briefly mention our minority stress related measures here.  You can see that we have two primary measures, including the LGBTQ‑related stigma and discrimination scale, which was developed and psychometrically evaluated by our team among SGM populations, both domestically and internationally, and which measures perceived anticipated and enacted stigma as a result of sexual orientation and/or gender identity.  And we have the intersectional day‑to‑day discrimination index, which assesses general experiences of discrimination, microaggressions and harassment.

Okay.  So finally, you can see here some of our measures on psychological processes, including hopelessness, perceived burdensomeness and religiosity.  These measures are intended to tap into our theoretical framework and specifically to understand some of the important mediators in the relationship between minority stressors and mental health outcomes. 

And now I think to talk about some of our strength‑based approaches and measures I'll turn things back to Sarah. 

SARAH MURRAY:  Thanks, Shoshanna.  In addition to our process of measurement selection refinement, we also early on, as a team, decided it was really important to bolster our assessment of protective factors and positive mental health in a more strength‑based approach.  To that end, we developed several additional aims we hope to accomplish that will repeat some of our person‑centered approaches to identifying patterns, but in this case of protective factors like social support and access to gender‑affirming care across our cohorts.

Then we want to look how changes in these patterns over time predict flourishing or what in positive psychology might also be thought of as thriving or a broad conceptualization of human well‑being.  And potentially we think this might happen through reducing internalized stigma. 

We will also look at these protective patterns as potential moderators of the effects of distal minority stress on depression and suicidality over time.

And this is just a list of some of those protective factors that we will be measuring and that we are measuring, and the scales that we are using.  For the outcome of flourishing, we'll use the secure flourish scale, which looks at things like happiness and life satisfaction, health, meaning and purpose, character and virtue, close relationships and financial/material stability, which really come from these domains of flourishing or the proposed (indiscernible) view.

And I'm going to turn it over to Savannah. 

SAVANNAH WINTER:  Hi.  This is a video.  Hi.  So, one of the major priorities of our work was ensuring that we had a community‑steered, community guide set of processes and implementation of our actual survey and just the recruitment efforts.

So we set forth from the beginning to recruit from across the country, including Puerto Rico, members of community organizations, individuals and so forth who would broadly represent, in terms of the diversity of everyone in the country that we're trying to reach, so looking for not just those five subgroups but also ensuring we have racial and ethnic diversity, having age diversity, diversity of experience both in terms of working in the community as well as, say, people who have come out later in life as opposed to people who came out earlier, things like that.

We have rotating membership.  Our CAB members are paid.  We want to make sure they are able to provide us guidance while also ensuring their time and involvement are respected and properly compensated.

They've provided for us a lot of really important suggestions, one of which I think I saw in the QAs asking about Internet connectivity.  And that's something that the CAB very much pushed us to address.  And one of their suggestions was working with community libraries and other organizations which have ‑‑ that serve as pillars of their community that will be accessible to people who may or may not have otherwise have access.  They helped us broaden our ideas of how we recruit people from just simply using the Internet.

So, kind of hybridizing our recruitment measures.  They also imparted upon us importance in having transparent language because there is a lot of justifiable distrust towards research community at large, as well as just, since we are government research, understandable concerns with that.  So being transparent, being open with our participants about compensation, confidentiality and things like that.

And they also imparted upon us the importance of having advertisements that didn't just come off as hokey or fake but making sure we were using real human beings in our ads, not just some corporate‑looking stock photo that gets the idea across but doesn't actually humanize what we're trying to do.

I think that's all I have.  Oh, I'm so sorry.  But just in general, they let us know that we can't just rely on some of the same tactics we've used previously because, I mean, previously these people have been overlooked.  And we have to understand why we've been doing that.  And so, they've been absolutely instrumental in our work.  Thank you. 

SHOSHANNA FINE:  All right.  So going into some of the prelaunch training and building of safety procedures that we incorporated into the study, so given the focus of our study and given what we know about existing mental health burden among our population, in particular, we wanted to make sure that, one, we had a robust safety protocol in place before we started the study; and, two, that all of our staff that were going to be interacting with study participants had adequate training and access to suicide‑prevention resources if they needed it.

We have two study clinicians who implement our safety protocol, and they have advanced degrees and training in counseling with LGBTQ populations.  They have found that motivational interviewing as well as community resiliency model training have been particularly helpful in guiding the phone sessions that they do with phone participants. 

Just to give a little bit of background, motivational interviewing is a counseling approach designed to help people find intrinsic motivation to make positive change in their lives.  And this style of interviewing really guides the phone interviews that our nurse practitioner has regularly with our participants.  And then community resiliency model training focuses on understanding the impact of trauma and chronic stress on the nervous system.  And so, it teaches how to regulate your nervous system in tangible ways and how to teach that to others so that folks are better able to withstand stress.  And it ideally hopes to promote resiliency. 

So, some of the other trainings that we had staff complete, QPR was a big one.  It's a suicide prevention training that trains what the program refers to as gatekeepers to be able to recognize the warning signs of suicide, and in particular to do what they call question, persuade and refer people for help who are at risk for suicide.

We had all staff trained for QPR, graduate research assistants, clinicians, full time coordinators.  Anybody who would be interacting with study participants, we wanted to make sure that they were prepared for the possibility that some sort of escalation could happen, not solely in counseling sessions but even on the phone or in a social media comment on our ad, that type of thing.

We had clinical staff complete counseling on access to legal means which was a course that provides skills to reduce access to lethal means, in particular, access to medication and firearms.  We had clinical staff trained in Stanley‑Brown Safety training, which is a brief intervention to help those experiencing self-harm and suicidal thoughts with concrete ways to mitigate risk and increase safety.  And those strategies are documented in collaboration with a potential participant or client on this safety planning form that you see here.  It can be used virtually or over the phone as well. 

And then lastly, we made sure to have all staff familiarize themselves with the 988 Suicide & Crisis Lifeline, which is the national hotline for suicide prevention.  And that's our primary referral org.  So, we had staff familiarize themselves with this service by having them call and text the hotline individually before we launched to see the process that participants would go through and how much time it takes to get connected. 

We did this for both the English and Spanish phone and text lines.  And the crisis line operators were very amenable to answering questions.  Of course, we kept everything very brief so as not to take time away from urgent calls.  But they do encourage researchers and folks who have questions can also reach out through the hotline.

Great.  And then in terms of recruitment and retention, our baseline recruitment is ongoing.  As of last week, we had enrolled 1,492 participants.  So, we're about 60% of the way to our total goal.

As mentioned, we're aiming to recruit 500 participants per cohort.  And so, we have enrolled around 66% of our cis male cohorts, 52% of our cis female, 47% of our transmasc cohort, 68% of our transfemme cohort and 65% of our nonbinary cohort. 

Our 3‑ and 6‑month follow‑ups are also ongoing.  A quick note about our retention estimates that we have here.  So, each of our follow‑up surveys have a target date and then a window period for participants to complete.  The best‑case scenario listed here is an estimate of when all participants who are within their window period complete the survey.  And then worst‑case retention scenario is an estimate of when all participants who are in the window period do not complete the survey. 

So best‑case retention estimates are attainable if participants in that window are contacted frequently encouraged to complete.  We do have a process where we do what's called last‑chance reminders for people who are within 25 days of their window closing and then 9 days of their window closing.

So, for our 3‑month follow‑up, we have 898 participants that have reached this follow‑up point with a worst‑case of 91% and a best‑case of 94.  And then for 6‑month we have 594 participants that have reached this point, with a worst‑case of 75 and a best‑case of 96.  And our 9‑month follow‑up will begin in about a month, at the end of September. 

SARAH MURRAY:  Great.  So, we would like to share with you now some very initial patterns that we're seeing in our first interim dataset of our baseline sample.  So, this contains 770 of the nearly 1500 participants that Mariah just told you about that we've recruited so far, with the caveat that we expect that these may shift as we continue recruitment and finalize our baseline analyses.

We see fairly similar distributions of age, race, ethnicity education across our cohorts and overall see our sample in the beginning has trended young, with just under three‑quarters of our participants under age 35 across the cohorts.  This is something we've been working to address in recruitment.  And this will be discussed more by my colleagues later.

We also see that about 75 to 85 percent of our sample by cohort identifies as white.  So, in particular we're looking to improve recruitment of Black and Hispanic LatinX participants.  And, again, you'll hear more about that in a few minutes.  And that's so we can be in line with what we see overall as a distribution of race/ethnicity in the rural U.S. population at least, if not more diverse.

And we see some differences in education across cohorts, but overall, in general about half our sample has a college degree or beyond. 

On the left of the slide, you'll see a respondent by state for this sample.  And this is something we're tracking to ensure that we don't have one state dominating our population in a region and that states that are less populous, that have a large percentage overall rural population are being fully engaged.

Our regional distribution is on the right.  And we had disproportionately higher recruitment in the south Atlantic region, and actually relatively lower in the Mid‑Atlantic region.  And that's really helping us target our recruitment going forward.  This is something that's shifted a bit over the continued recruitment, it's happened ‑‑ that's not represented in the sample yet.

We've also been concerned that in trying to recruit a rural sample based on ZIP Code and the U.S. Census Bureau's Urban Rural Classification Scheme, we might end up recruiting from communities that mostly fall in small metropolitan category, which is a statistical area with an urban center that has a population under 250,000 people but over 50,000 people. 

But what we see here is that only about 40 to 50%, roughly of participants across cohorts are coming from those areas.  About 30 to 40% of our participants are coming from micropolitan areas that have an urban center under 50,000 individuals but more than 10,000.  And around a fifth of our sample is coming from noncore areas where there's no urban cluster of 10,000 people or more.  So that's been encouraging.

For our primary outcome, one of our primary outcomes, depression, for descriptive purposes, we have classified depression symptoms measured via the PHQ‑9 as none or low is a score 4 under, mild with a score 5 to 9, moderate with a score of 10 to 14, moderately severe with a score of 15 to 19, or severe with a score of 20 to 27. 

Across our sample, most participants fall in the none or minimal category, fortunately, so over 50%.  However, across our whole sample we see that nearly a quarter actually fall into the moderately severe or severe categories, unfortunately.

And when we look at this distribution by cohort, we see that the percentage is relatively greater falling in those categories in our gender minority cohorts than in our sexual minority cohorts.

When we look across regions, we see that in particular the East/North Central region has the highest depression scores, average depression scores, relative to other regions.  And ultimately, we see, interestingly, a similar distribution of depression scores across our noncore micropolitan and small metropolitan areas. 

Moving on to suicidality, when we look at responses to the C‑SSRS, our suicidality measure question, about ever having thoughts of harming oneself or killing oneself, we note that lifetime ideation overall is quite common.  It ranges from 57% of the cisgender male sexual minority cohort to just over 80% of the transmasculine cohort.  Again, this distribution is very similar across rurality.  And it's actually more consistent than depression was in its regional distribution.

When we look at intention and attempts assessed via the C‑SSRS question ‑‑ have you ever done anything, started to do anything or prepared to do anything to end your life; I want to note here, this is a little bit broader than just suicide attempt and really is a broader suicide behavior question in that it also captures preparatory action ‑‑ we see that this, again, is more common among our gender minority than our sexual minority participants, generally speaking.  But it's particularly high among our nonbinary cohort.  That's 40% of the sample have endorsed ever having those behaviors.  And our transfeminine cohort was 35%.  But all cohorts had a prevalence of lifetime endorsement of this greater than 25%.

And there's some differences across rurality and geography with a higher prevalence in noncore areas and then the mountain region of the United States.

Looking at scores, switching to sort of our main exposure of interest being stigma, in this case we're giving you an example of daily discrimination, or intersectional daily discrimination.

Importantly, the scale, as a higher score, indicates greater frequency of these experiences.  And the total score can range from 0 to 27.

We see overall more daily discrimination being reported by gender minority participants than our sexual minority participants, with relatively similar scores by rurality ‑‑ slightly lower for small metropolitan areas, but not so much.  And the greatest frequency of daily discrimination experiences is being reported in the West/South Central region, so that's Texas, Louisiana, Oklahoma, Arkansas.

When we look at flourishing, please note here that a higher average scale score indicates a greater degree of flourishing, and the range is from 0 to 10.

There's a similar distribution for all of our population cohorts, but we do see the scores are just a bit lower for our gender minority cohorts than our sexual minority cohorts.  So overall I would say these differences may not be as stark as what we saw for the negative mental health outcomes for depression and suicide.  And while scores are relatively similar across rurality, with participants in small metropolitan areas fairing slightly better, we see participants in New England having higher scores than those in other regions.  So good on you, New England.  I think this is a really encouraging thing to be seeing given some of our other findings, that flourishing is what we're seeing here.

Just some summary of some early patterns that we're seeing.  Cisgender sexual minority.  Cohorts are tending to report a lower level of stigma compared to gender minority cohort participants, but stigma experiences overall are quite common. 

Other distal stressors, like adverse childhood events and things we didn't show here have tended to follow the same pattern.  Levels of poor mental health and suicide ideation are not too dissimilar from what we generally see in the literature and are much higher than what we see in the general population.  We see worse outcomes here for gender minority cohort participants relative to sexual minority cohort participants and slightly better positive mental health outcomes among our cisgender sexual minority cohort. 

But these are actually more similar across cohorts.  And this pattern follows some things that we didn't show you like resilience and social support across cohorts as well.  And, again, I think in general these flourishing measures are great to see and are hopeful. 

Ultimately, there may be some unique coping mechanisms that are being used or some unique strengths among gender minority participants in these rural spaces that support a comparable level of thriving despite reporting more stigma and poor mental health as well.  That's something we really want to dig into. 

And lastly, we see some interesting regional variation that we're going to keep an eye on to see how it plays out in our full sample, but also relatively similar distributions in our study across rurality that we're going to keep an eye on.

And I'm going to turn it over to Kirsten who is hopefully here with functional Internet.  We're just re‑enacting the role, the reality of unreliable Internet. 

KIRSTEN SIEBACH:  Yeah, sorry about that, everyone.  I'm going to keep my video off to try to help the Internet. 

I'll be talking about our participant safety protocol.  So as Mariah mentioned earlier, participant safety is a really high priority for this study given the topics that we're addressing.  And we noticed that there hasn't been a lot of discussion within the research community about mental health safety protocols generally.  So, we wanted to share what our protocol looks like, both for transparency purposes but also in case it might be helpful for future online research studies. 

The safety protocol is initiated based on survey responses to the C‑SSRS and also if the parent requests a study staff to contact them for support.  Based on the C‑SSRS, if a participant answers yes to current thoughts of suicide and yes to any one of the following questions ‑‑ the intention to act, an existing plan to carry out an attempt, steps taken toward an attempt in the last three months, and an actual suicide attempt in the last three months ‑‑ an automatic email is sent to study clinicians notifying them of this participant's survey.  That email includes the participant's study ID with obviously no personal information, the responses to the C‑SSRS questions and the ZIP Code of the participant so that clinicians can look up geographically relevant resources before reaching out to the participant. 

We also have the option for participants to opt into an outreach call in case they're feeling the stress from participation but don't meet the safety threshold from the C‑SSRS.  Within one business day, clinicians reach out to the participants to conduct a safety assessment.  This could be through a phone call, a text message or an email depending on the preferred contact method selected by the participant in the baseline survey. 

If the participant can't be reached on the first attempt, contact attempts continue once a day for up to seven business days.

And thankfully this has never happened yet, but if we can't contact the participant within seven business days, we will reach out to their emergency contact for up to three business days.  Again, we haven't had to do that, but that is in place in case.

It's often been the case that the initial contact is made, but then clinicians and participants schedule another time to talk further about how the participant is doing.  And during that call clinicians assess the level of risk and provide resources as needed.

Each contact attempt of these outreach calls is documented and kept on file.

So, here's an overview of how much outreach has been conducted as of August 8th, 2024.  You can see on the left the follow‑up period, then next to the number of surveys that have been taken.

Then you see the number of safety calls based on the C‑SSRS and how many were requests for contact. 

If we're looking at the safety calls based on the C‑SSRS we see that between 5.3 and 6.7% of the surveys initiate the safety protocol.  If we look at the total of 157 safety calls based on the C‑SSRS.  138 of that 157 were with unique participants.  So, this shows us that the calls are not with the same participants across different time points but with unique individuals.

Here is some of our lessons learned in conducting this participant safety protocol.  First, there's been a higher‑than‑expected awareness of and connection to mental health resources.  Participants often talk about having a therapist or going to see a therapist soon, having social support in their area. 

And they have knowledge of and in some cases have used mental health resources such as the Trevor Project or the suicide project hotline, 988. 

That leads me to number two.  A majority of safety calls have focused more on providing ongoing supportive contact rather than crisis work.  Next, and, again, thankfully, we have not had to utilize more formal safety planning and access to lethal means.

Next, participants have noted that the Internet is both a source of community, which is understandable given the reality of the population, but also that the Internet is a source of distressing news from the media.  There's certainly intention there between seeking social support on the Internet but then also being exposed to potentially harmful messages.

Lastly, we've been really struck by the use of creative outlets as sources of resilience and coping.  Participants have talked about doing art, puzzles, Legos, graffiti, a range of artistic activities as ways of calming mental distress.  I'll now turn it over to Mariah and Savannah to talk more about recruitment.

SAVANNAH WINTER:  Thanks so much, Kirsten.  So before or as we start talking about recruitment, I just want to preface by saying that rural America is incredibly diverse, and strategies that we would use to reach, for example, Black rural Americans, 80% of whom live in the rural South, that strategy needs to be very different from how we recruit folks living in Appalachia or Hispanic, foreign‑born rural folks living in the Midwest, near meat‑packing plants, for example.  I just want to acknowledge how important and nuanced these intersections of rurality, race, ethnicity, sexuality, gender, language are.  And we definitely think about that and talk about that constantly in trying to revise our recruitment approaches. 

And these methods are definitely imperfect.  But so far, looking back retrospectively, we can think about our recruitment strategies sort of in three phases.  So, phase one encompassed our more traditional strategies of using online social media advertising to recruit participants.  And that was more broad, national targeting because of the constraints that social media sites put on advertisers in terms of targeting. 

And then phase two focused on more targeted strategies that were smaller in volume and allowed us more control for this space.  We primarily relied on contact lists of previous participants from other studies. 

And phase three involved more innovative, nontraditional strategies, specifically using dating apps in a capacity we hadn't before.  So, we only had ever used dating app advertising for sexual health study recruitment, never for mental health.  And then influencer marketing was a large part of phase three as well. 

Going into more detail here, for phase one we launched on Instagram and Reddit last November.  And that was during the holiday season which definitely helped bolster the response.  We started fairly small with just these two ads on Instagram.  And then we did a sort of humorous meme‑type ad on Reddit.  And we had huge success. 

During this first phase we enrolled around 600 people from December through February, and we had thousands of respondents.  Some strategies that led to success were, number one, doing broad demographic and geographic targeting on Meta in particular.  So, we didn't narrow our recruitment area to rural ZIP codes, for example.  We didn't use keyword targeting approaches which we often do to reach queer and trans people on Meta.

Number two, we ensured that CAB members were very closely involved and had seen and were able to provide feedback on all of our ad copy and creative, prior to launch.

And then lastly, we ensured that comments on all of our ads were hidden throughout the recruitment period, which is something that Meta does not allow natively.  But we use an external site called NapoleonCat that auto moderates and allows us to completely hide all comments automatically.  It has a bunch of other features that are super helpful as well.

I really recommend something like that over manual moderation, not only because of the time and effort it takes to do that manually, but also and more importantly because of the emotional burden it places on staff to have to sit there and manually sift through thousands of hate comments, transphobic, homophobic, racist comments, just the vitriol that you see on social media sometimes.

Despite our success or maybe in spite of our success during this phase, the volume quickly became a bit unmanageable for us.  So, we didn't foresee the response being so great.

But this also came with very high rates of fraud.  So, both scammers that were real people and bots and ‑‑ very committed scammers, might I add.  So, folks who went so far as to sign up for a phone session with our clinicians, who spoke to clinicians and staff on the phone, giving phony responses.

And so, this all led to us having to iterate and change up our fraud‑prevention processes multiple times.  And it also made it so that we needed to hire additional staff to do all the validation work and the kind of manual data entry that was required for this.

And then we quickly noticed that despite the large volume coming in from social media, which was filling up the nonbinary cohort very quickly, it was overwhelmingly white and trended much younger.

So, we went into phase two knowing this and we also reached saturation on social media pretty quickly.  So, by March, our ads were not really being shown or engaged with nearly to the same extent as when we launched.

Next slide.  So then that leads us to phase two, where we paused our social media campaigns, focused on solely more targeted lower volume strategies to allow us to catch our breath but to address the problems we had during phase one.  We gathered participants from other studies AMIS and TWIST research truants they're annual cross‑sectional studies.  AMIS for men and TWIST for transgender women's.  We sent out an email for future contact and allowed us to focus on the specific cohorts where we needed to bolster numbers and also to have the peace of mind that folks from these lists had already been vetted to some extent, they were unlikely to be fraudulent.  We knew that AMIS participants also tended to be older which we wanted because at this point around 90% of our REALM sample was between 18 and 36.  We wanted 36‑plus and AMIS prevented help with that age discrimination.

MARIAH VALENTINE‑GRAVES:  As we mentioned, we had to start cultivating nontraditional strategies using data apps, which we were initially hesitant to do.  Did I turn on my video?  Initially hesitant to do given that this is not a sexual health survey, but rather a mental health survey.

We found some very strong success with A4A, also known as Adam4Adam, which is primarily a dating/hookup site for cis men, who have sex with men.  But they also do have a sizable transmasc and transfemme user base.  They were very helpful in getting us not just people from across the cohorts.  They were also more racially and ethnically diverse than what we were receiving on social media ‑‑ through self‑service social media advertisement, so more diverse in terms of age, which is something we very definitely need.

We had less luck using Grindr, which is supposedly the premier dating and hookup app for gay men, but also, like A4A, also has a sizable trans contingent.  It, unfortunately, proved to be too expensive for us to use long-term considering the somewhat anemic response rate we received.

And Tinder, lastly, was extremely unsuccessful, super expensive.  Tinder ‑‑ so something we've seen across the board is a lot of places do not want to provide any measures for reaching out to specific subsets of their user base.

Ostensibly this is to protect against harassment, but anecdotally, from what I've heard, that doesn't necessarily turn out to be the case.  It has a somewhat chilling effect on actual recruitment through those sites.  So, we cannot necessarily use them.  Say, Tinder, which has ‑‑ it's the number one dating site ‑‑ but we can't just try and advertise towards LGBTQ people, meaning that most of the money ends up just spread across the entire audience, which is just not really practical for the work we're doing.

Next.  So, we also stepped into influencer marketing.  It's something that is attested to by sales marketers and things like that.  But it's kind of a different audience.  Your sales versus recruitment, especially recruitment of marginalized communities who have reason to be distrusting.

So, it's something we have slowly worked our way into.  Our first partnership was with Erin Reed, a transfemme journalist who does a lot of reporting on anti‑trans and, more broadly, anti‑LGBTQ legislation across the country.  And since there is some overlap, as we've established, between rural spaces and anti‑LGBTQ legislation, we found a pretty enthusiastic response, sizable number of participants, that she managed to bring us.

There were downsides.  It was pretty racially/ethnically homogenous.  But it kind of helped us shape our strategy moving forward, giving us an idea of what is possible with influencer marketing so long as we identify the right people on the right platforms at the right time.

And their big benefit of influencers being their followers are more likely to be engaged as opposed to any random person on a social media site.  They are more invested in the actual findings of the work.  And they're more likely to find feedback.  That's something that's super useful to us.  It's just a lot of comments, emails and so forth that have really helped us address some of the shortcomings that we have in our work as it is currently ongoing.

We also did find issue, as mentioned before, since it did have a disproportionately white response, we've had to implement quota‑based strategy.  And it’s a pro and a con, but because the increased engagement we did get a better understanding about issues with our definition of rurality versus what the public view ‑‑ what the public view rurality as. 

Next, please.  Some of the challenges that we've had across the board, we've had to deal with quite a bit of attempted fraud.  Because it's an online survey, because it is compensated, and because ideally since we're trying to reach people, we don't want to have several barriers to access, which honestly don't even necessarily help with bots and things like that, but only really serve to inhibit the actual completion of our surveys.

We've struggled a lot dealing with bad faith actors who are in it for the incentive.  A lot of the measures that we have traditionally used, all the stuff that's attested to in the literature is out of date.  And it gets out of date extremely quickly.

With all the AI, tech and things like that hitting the ecosystem now, it's something that we constantly have to iterate upon to make sure that not only do we not just completely run out of money to advertise and recruit legitimate participants and pay them, but also just makes it bugbear in terms of keeping our ‑‑  validating, processing all these entries and making sure that the data we have is genuine, authentic and reflects the population we're actually trying to benefit. 

Regarding the influencer marketing, it's something that ‑‑ it's growing pains.  We are not necessarily the ‑‑ we don't have all of the strategies in place, though we are developing them.  So, it is good but has been difficult. 

We have struggled reaching Spanish speakers or at least monolingual Spanish speakers.  We know there are cultural barriers we can't just broach by having an ad translated or having the survey translated.  We looked to the literature for suggestions on this as well as reaching out to members of not just this community advisory board but Spanish speakers elsewhere and tried to see if we could identify those.  But we have had trouble addressing intersecting marginalizations of potentially monolingual Spanish speakers who live in rural areas who may or may not be alienated from the broader community.  We also ran ads in Puerto Rico, but we found that participants often just opted to take the surveys in English. 

And pardon me, something ‑‑ so something I mentioned before, we had issues with regards to our measure of rurality versus what we have heard from participants.  We're using the NCHS Urban Rural Classification Scheme which bases it off of core ‑‑ just totally lost the word ‑‑ core population areas. 

And that leads to cases where people who live in ostensibly rural areas fall through the cracks and raises the question to us, how do we recruit without misleading and how do we make sure we're reaching who we're trying to reach. 

And the last issue we've had, especially with regards to traditional social media, recruitment is getting exceedingly difficult, especially in an election year, to advertise any SGM, LGBTQ topics without it being immediately shut down, blocked for being political content, which is frustrating and something that is hard to work out with them at times. 

SHOSHANNA FINE:  In terms of our current recruitment phase, we're continuing to try to innovate and use nontraditional strategies.  We are prioritizing increasing numbers in specific cohorts where we are lacking in numbers.  And, so, specifically that's our transmasc and cis female cohorts.  And we're also prioritizing increasing racial and ethnic diversity across all of our cohorts. 

With these goals in mind, we have partnered with the Association for Rural & Small Libraries which is a CAB suggestion.  ARSL is a network of rural librarians and library workers.  They have about 3,000 members and a larger network of around 8,000.  And they also have an LGBTQ networking group.  So, we worked with them to create a set of flyers that can be posted in computer labs or physical spaces in libraries, as well as could be sent out electronically.  We hope that will increase access for folks who aren't seeing our advertisements online. 

And second, we want to prioritize outreach to monolingual Spanish speakers, as Savannah mentioned, and Hispanic participants more generally, which has definitely been more challenging than anticipated.  And likely this is because recruitment strategies we would normally use to recruit Hispanic and Latino participants generally don't work for those living in rural areas.  We can't just run statewide campaigns in California or Florida because these states are overwhelmingly metropolitan, for example.  So, we really have to shift our idea of prioritizing the West and Southwest and instead look at where rural Hispanic folks are geographically, which is increasingly clustered around meat‑packing and poultry industries in the South and Midwest, places like North Georgia or Ford and Finney County in Kansas.

And we're looking into partnering with migrant farm workers associations, Black and Hispanic churches in the South and art collectives, among other things. 

And then, third, we're considering partnering with additional influencers, but we want to be intentional about recruiting participants that we need the most and that are the most diverse.  And so, one challenging aspect of influencer partnerships is that an influencer being of a particular demographic doesn't necessarily mean that their audience is of that demographic.  This just complicates the process a bit.  But it really just means that we need to be really intentional about who we're partnering with and who their audience is. 

Lastly, we're constantly evaluating and revising our recruitment procedures to respond to this ever‑changing landscape online.  And so, for better or worse online recruitment is, it now really requires an ongoing iterative approach with robust fraud‑prevention strategies that are implemented by actively engaged human staff.  And I say that because a lot of this, in our experience, can't really be automated and it requires a great deal of manual human attention.

SARAH MURRAY:  Because of the time I want to move us into Q&A.  I just want to say we're finishing our baseline recruitment this year.  We're diving into all of our analyses and our discrete choice experiment design.  And we're going to be looking more and more through this study at structural stigma and mental health and tell those shapes experiences for our participants. 

Okay, with that, we just want to thank you all for your time and attention and thank the Office for Disparities Research and Workforce Diversity and Dr. Davis and others for the opportunity to present today, as well as our CAB, people that have been instrumental in getting this study going, and most importantly all of our participants whose time and effort we really hope to honor through taking this work and turning it into public health action.

I'm going to stop sharing.

SHAUN DAVIS: That's totally okay.  And just to be mindful of the fact that we have a little bit of time for an abridged Q&A.  We have extended it a little bit.  We can extend it to 3:05, I think, per our contract.  So, we'll have a few more moments to discuss some of the questions that were presented in the chat.  And there are a lot of really good questions that will lead to really good discussion.

So, I'd like to invite all of the presenters to turn on your cameras if you would like, based off of broadband and all of that, and we can go ahead and discuss some of these questions.

First off, there are a lot of suggestions in the chat, in the comments, about how to recruit in places that might be really helpful.  I just want to assure everyone that we have those, and we will share them with the team.  I can't address all the "did they recruit here; did they recruit there," right now because of time.  But you'll have access to the resources.  And I like to think of this as a great idea.  We're all a community of practice and research and resources, and so it's really great you all shared that.  I think there's a lot of really good suggestions in chat about places you may or may not be recruiting.  Yes, I'll slow down.  I'm talking a little fast because I'm a little caffeinated.  My bad.

To start, one question that was asked that is a simple one, is there a way we can share to invite potential participants.  Are there links or websites that would be helpful for the subsets of people that you're looking to recruit? 

>> Yes, definitely.  I don't know the best way, but, yes, we have QR Codes, we have a link, we have flyers ‑‑ however it's easiest to share.

SHAUN DAVIS: Could you put the link in the chat?  And I think we can have the event host share that for everyone broadly.  Because I think we can't share it.  They have to be the ones to share.  
>> Sarah, I'm assuming you're comfortable with that.  That's fine? 

SAVANNAH WINTER:  Okay. 
>> Thanks, Savannah.

SHAUN DAVIS:  All right, the next question gets into this idea by fraudulent responses.  And that's something that you've talked extensively about.  One thing that was asked is have you been thinking about publishing some of those experiences, like sussing out who is fraudulent, who is not, who is truly of the sample you want to recruit.  Could you talk more about that? 

MARIAH VALENTINE‑GRAVES:  We've submitted two abstracts.  Savannah wrote one and I wrote one.  And they were both rejected.  So, we are working on it, but as this process goes, you know.  But, yeah, we definitely ‑‑ we have documented formally all of our procedures and definitely will continue trying to publish.  Savannah, I don't know if you want to add anything. 

SAVANNAH WINTER:  I think you covered it.  It's just something that we have kind of had to come to terms with, and it's a constant process. 

SHAUN DAVIS: That makes a lot of sense.  I think just hearing about the recruitment procedures, I think this is a unique challenge based off the nature of the work that you're doing that I think many folks could learn from.  Especially with recruiting in the social media landscape, which is ever changing, as you already mentioned.  I think that's something that the field definitely would be responsive to in some way.

All right.  For the sake of time, I'll move on to the next question.  There were a few questions about sort of racialized identities and how that might intersect with rural identity.  So how do you think about sort of some of the risk and protective factors for suicidality and depression for those with racialized identities? 

SARAH MURRAY:  I think that's such a great question.  I think one of the myths that we hope to really dispel for our work is this idea that rurality means white or means homogenous, which is just not the case. 

Rural areas are increasingly diverse.  This is something that we hope to look at through some really stratified analyses as we go through and really trying to take a more of an intersectional lens.  It's also why we have some measures in our data that don't just ask about stigma and discrimination experience because of your gender identity or because of your sexual orientation but because of who you are, which we can recognize can encompass a wide variety of things, including race/ethnicity, including your immigration status, including the fact you're from a rural area, depending on where you're living. 

And it is something that we really ‑‑ it's why we're really striving to increase the diversity of our sample, to be consistent with the rural population and actually exceed it so we can actually do some more stratified analyses. 

And we also ‑‑ I really appreciate the comments about the diversity or the lack of diversity of our panel in terms of race/ethnicity.  It's something we strive to have in our team.  It's a well‑taken point about us being on this presentation today.  But it is something we strive and value on our team, on our CAB, and that we really hope to use the study as an opportunity to keep people bringing in of a wide variety of backgrounds into this research processes because it makes the research better.  And we want this to be reflective of the wide variety of experiences that we know exist.  So, we appreciate the point.  It's well taken.

SHAUN DAVIS: The office that I'm in also talks about research work, and it's a huge area.  I think that we're trying to make efforts to increase the diversity of the workforce, in particular within the SGM research workforce.  But we're still in the early stages and we're doing our best at this time, right? 

So, speaking of sort of stratifying analysis, there were some other questions about how rurality is not homogenous.  Like, Appalachia might look different from the Midwest, from the Deep South, all of that.  How do you think about some of those subgroup analyses?  Like, is that on the horizon for you as well?  Could you tell us a little bit more about that? 

SARAH MURRAY:  Yes.  So, we're hoping to really have the ability to look within region and look specifically within region as we really do recognize that point.  We also really want to look ‑‑ and I didn't get to talk about this too much, but actually led by Kirsten as a part of her doctoral work ‑‑ will really also be thinking about heterogeneity in terms of policy environment across rural areas.  And that might be region‑specific, but that also might be sort of some places having more protective policies.  Some having much more harmful policies in place and really thinking how that is shaping experiences in diverse ways and really recognizing that really important heterogeneity.  And so that is work that we'll be engaging in, led by Kirsten.  And I'm sure she'll be happy to tell you more about it.

SHAUN DAVIS: Certainly.  Kirsten, do you want to comment on that?  Any other thoughts you want to add to that. 

KIRSTEN SIEBACH:  Yeah, I can talk a little about my research, looking at structural stigma, looking at the policy environment, both kind of legislatively, but I'm also planning as part of my research to actually go and interview folks living in the rural U.S. to kind of understand how their policy environments really impact their mental health, how they think about kind of their identities, their intersecting identities, including rurality, and how they experience those intersecting identities in this place geographically. 

It's still in the works.  I haven't started quite yet but hoping to get underway soon.  But, yeah, I'm very excited to undertake this kind of important area of research. 

SHAUN DAVIS: That's wonderful.  Maybe we can invite to you the next webinar, and you'll be the one presenting on your research.  That would be so exciting.

All right.  So, a couple of other questions came up around trust for sort of LGBTQ+ rural folks.  How do we ‑‑ how are you all thinking about trust in terms of trusting institutions?  Just how do you think about that, in terms of getting people to engage when there's a history of mistrust for valid reasons, of course?  And how did that factor into how you think about recruitment? 

MARIAH VALENTINE‑GRAVES:  We think about it and talk about it constantly.  We respond certainly to feedback from participants that express that.  And then we try and address it either by having set language that we send, communicating certain things by email usually or automated responses on social media. 

But we also ‑‑ we've done things like create an FAQ document that can be embedded on our landing page of our survey that talks about ‑‑ a big thing is data protection and confidentiality related to that.  And, so, having language there for folks that they can refer to and we can refer to is often helpful. 

Also, we've also had discussions about whether or not it's helpful to have the Emory and JHU logos on our ads, or if that's a deterrent for some folks.  Actually, I think we've done A/B testing to try both to see what the response was.  We're always trying to iterate on this and respond to folks' concerns.

Savannah, did I miss anything?  I'm sure I did. 

SAVANNAH WINTER:  No, I think you did good.  It's just making sure that we, I think, just being earnest, being honest and acknowledging.  I think a big thing is acknowledging that people's concerns are legitimate because considering who we are, who we represent, so on and so forth, there are general harms that have been perpetrated.  And it's something that we have to work with our communities to work past and be, I guess, be humble around that fact. 

MARIAH VALENTINE‑GRAVES:  And we're asking participants to be very vulnerable.  And they've divulged a lot of information to us, and they don't know us.  So just acknowledging that their concerns are legitimate.  And communicating, I think working with influencers, we've seen that sometimes one person's comment about distrust can really blow up into something big.  And so just communicating with folks often will quell some of the concerns.  People's kind of fear or anger is really just about not having all of the information.  Oftentimes when we provide information and communicate with them, then they feel like, okay, this is a real person I'm talking to, and it alleviates some of it.

But it's imperfect. 

SHOSHANNA FINE:  I know we're running short on time, but if I could jump in briefly on a related comment to trust.  I think actually one thing that I think is worth mentioning here, we weren't sure how recruitment would go for this study.  Our team has a lot of experience reaching SGM folks but not a lot of experience reaching SGM folks in rural areas, specifically in a study focused on mental health. 

I'd say one of the things that's been amazing in this study and in our recruitment, efforts are actually the level of interest we've found from people in participating.  I think we've had a huge response to the study.  I think it goes along with trust, but it's a testament to the fact that there hasn't been a lot of research traditionally done with these communities.  And there hasn't been a lot of opportunity for folks to have their voices heard within these types of studies. 

And so, I think we have to be sure, obviously, to establish trust.  But I also think the flipside of that is people are really interested in sharing and being a part of something like this. 

SHAUN DAVIS: For so long, we talked about how rural communities have not, rural LGBTQ folks have not been included in research.  I can see how there might be a lot of sorts of interest ‑‑ like, you all want to hear from me now, right?  It probably feels good to be represented and have folks take an interest in their studies.

All right.  For the sake of time, I'll wrap up with the last question about flourishing and the resilience factor and protective factors you all talked about.  Could you talk a little bit more about sort of things that might have surprised you with the flourishing findings, or just the flourishing findings in general?  I think there's a lot of really interesting sort of gems in that section of the presentation. 

SARAH MURRAY:  This is one of the places that we were most excited to dive into.  I think we sometimes have this idea that these things are fully in contrast to one another, that mental distress and flourishing are, like, two ends of the same spectrum, but that people can experience distrust and actually flourish in different domains of life. 

And I think, to be honest, when we proposed these aims and we sat down and we looked at our measures as we started to get ready for this study, we kind of had this moment as a team where we were, like, whoa, this is so negative, and risk focused.  And of course, it is really important that we understand risk, but we did not feel good about putting a survey out there for people to answer ‑‑ and putting results out there (indiscernible) study that also didn't consider unique strengths and positive mental health outcomes. 

And so, this is something that we don't see as much in the literature, not to say that it's not there.  But it's not currently seen as much as ‑‑ and can people be experiencing a lot of risk but still flourishing?  And, if so, sort of what is happening there?  How is that happening?  What are the strengths that people are relying on?  What is the sort of unique resources. 

And because our goal is ultimately to go through these discrete choice experiments and really develop support, some kind of mobile or sort of app‑based support, we really wanted to understand how to build upon those strengths and promote those strengths.  And ultimately the outcome we want isn't just the lack of distress.

Like, of course, we'd like to ameliorate distress.  We'd like to keep people safe, but we want more than that.  We want people to flourish.  We want people to thrive.  We want to support people in that.  And that requires a bit of a broader perspective.

So, seeing those numbers coming back in has been so encouraging, I think, and reinforcing of that.  And I would like to encourage others to do that as well because I think we'll be continually surprised by what we find. 

SHAUN DAVIS: Mariah, were you going to say something else?

MARIAH VALENTINE‑GRAVES:  Could I rapid fire two responses?  You all have the best questions, and I want to respond to all of them. 

SHAUN DAVIS: Before you do that, just want to say we just pasted a link in the chat for the next webinar in the series.  I know we're about to wrap up.  I just want folks to see that.  It's coming up.  Sorry, go ahead, Mariah. 
MARIAH VALENTINE‑GRAVES:  I've seen a few questions about specifically recruiting Asian Americans in rural areas.  Great question.  And we actually haven't needed to do that yet because the recruitment methods that we've been using have brought in ‑‑ around 4% of our current sample identifies Asian, Native Hawaiian or another Pacific Islander.  And that's compared to around 1% of what the census says is the Asian American rural population.

We've been sort of more diverse in that area than we anticipated.  And so, for that reason we haven't done specific outreach to that community.  But it is super important and we're definitely looking at that.

And then the other question I saw was about neurodiversity, I think, and disability and whether we've looked at those intersections in relation to our research.  We definitely have.  We've had discussions about that.  And we've had comments from participants as well about whether or not we're studying that.  But also, we've had people respond with certain gender identities that are kind of neurodivergent/queer identities, so how to talk about classify that, like birdgender, which is a zerogender that's neurodivergent, and those types of things.

So, yeah, we've definitely seen comments and we've talked about it as a team.  And I don't think it's in this research plan, but it may be for the future.  I'm not sure.

SHAUN DAVIS: Thank you for attacking those questions rapid fire.  Because I was, like, there's so much that we could get to, and we could probably have an hour‑long Q&A truthfully with some of the questions in the chat.  I appreciate you addressing those really briefly.

With that, we are at time.  So I really just want to thank you all for presenting today on this really sort of important work that you're doing and all the sort of challenges that are associated with it because I think there's so much that people can learn from all the lessons that you've learned along the way about how to recruit the sample and to really give voice to rural LGBTQ communities and their needs.

So, thank you so much.  And also thank you everyone for attending and being so active in the Q&A.  I know there's some questions we didn't get to, but hopefully we can have more discussions like this in the future.  So, thank you all so much.  And with that, we will wrap up.  Have a wonderful day. 

SAVANNAH WINTER:  Thanks for having us. 

SARAH MURRAY:  Thank you, everybody, and have a great day.