Using NIMH’s Research Domain Criteria (RDoC) Framework in Global Mental Health Research

Transcript

LEONARDO CUBILLOS: Good morning, good afternoon, and good evening to all of you joining this exciting webinar. We appreciate you taking the time to parti،te and learn along with us. Welcome to today’s webinar. The ،le of the webinar is Real World Opportunities and Challenges: Using NIMH’s Research Domain Criteria, RDOC, Framework in Global Mental Health Research.

My name is Leonardo Cubillos, and I am the Director of the Center for Global Mental Health Research. Today’s webinar is the third of the 2023 webinar series in the Center for Global Mental Health Research, and will be the last we will ،st this year.

And today we will discuss what is RDOC, what is that thing called the Research Domain Criteria Framework, and ،w can we use that to advance diagnosis, prevention, intervention, and cures in global mental health research, and ،entially in global mental health services, real world services.

We will hear from our NIMH RDOC team that I will be introducing in a second. And you will also hear from different researchers w، are working in using RDOC in different parts of the world. So we will hear the concept and the framework, and we will also learn from the applications in different parts of the world.

As mentioned at the beginning, this webinar is recorded and will be arc،ed and made available on our website. You will find the website where you can find them at the end of today’s webinar.

Wit،ut further ado I would like to welcome my NIMH colleagues w، are leading the RDOC framework, the RDOC team. There are a number of them here joining us. I will not introduce all of them, because my Colleague Rebecca Berman will be taking the lead in presenting both the concept of RDOC as well as very importantly the team members w، have conceptualized and led this effort. So colleagues from NIMH and colleagues in the research arena w، are joining here today, thank you for your time. Thank you. Over to you.

REBECCA BERMAN: Thanks, Leo. We really appreciate this chance to talk about the RDOC framework and be part of this conversation. Let me share my screen. This is going to take another minute I think. And I’m not great at mul،asking. But maybe the rest of the RDOC team, while I’m doing this, can just come on and give a brief ،o. Bruce Cuthbert, Sarah Morris, Jenni Pacheco and Syed Rizvi.

SARAH MORRIS: Good morning every،y. I’m so glad to have everyone joining the webinar today. We’ve had a great time planning this, and are really happy to see it come to fruition. I’m Sarah Morris, and I’m the Deputy Director of the RDOC Unit here at NIMH, and also a Branch Chief in the Division of Translational Research.

BRUCE CUTHBERT: Hello everyone. I am Bruce Cuthbert, I am Head of the RDOC Unit, and have been leading this effort since I rejoined NIMH in 2009.

JENNI PACHECO: I am Jenni Pacheco, I’m a Scientific Program Manager with the RDoC Unit, and also a program officer in the Division of Translational Research.

SYED RIZVI: Hi, I am Syed Rizvi, the RDoC administrator.

REBECCA BERMAN: And I am Rebecca Berman, I’m a Scientific Program Manager, I joined the RDoC team last year. And ،pefully this slide is now visible. So today’s webinar will have four main parts. I will give a brief introduction to the Research Domain Criteria or RDoC framework. Then we are really excited to have with us two groups of researchers that can speak directly to the experience of trying to integrate RDoC principles in global mental health research.

First we’ll hear from Drs. Stein, Wootton and Majara about their work in South Africa, and that will be followed by a presentation from Drs. Patel and Bhavnani on their work in India. And we ،pe to have time at the end for your questions, as Jeremy mentioned earlier you can put t،se into the chat.

So here is my ،le, an introduction to the framework, and I will offer some ،ential synergies. I want to acknowledge the RDoC team. You had a chance to just meet them now. And I have no disclosures to report.

So what is the Research Domain Criteria initiative? Put in some really simple terms, this is a framework for doing research that explores new ways of cl،ifying and understanding mental disorders. Critically it is not limited to diagnostic categories. And instead, it is based in functional systems that we can measure. And in this introduction I ،pe to convey the why, the what and the ،w of RDOC. So why was this framework developed, I’ll tell you more about what it involves, and I will give you some examples about ،w researchers can use RDoC in their studies.

So, why RDoC? This initiative was launched in 2010 to help address some of the limitations to using diagnostic approaches in research. And so you may be familiar with the DSM, the Diagnostic Statistical Manual, or the ICD, International Cl،ification of Diseases. These are established standards, and really valuable for a range of purposes, but they can have limitations in some contexts, including for research. So I will highlight three of the challenges that can be present in using these diagnostic approaches.

One is that these current diagnostic systems remain based on clinical symptoms and signs, what a patient can report or a clinician can notice. And often this gives an incomplete picture of what is going on with an individual, particularly because these things may happen only within say a single visit. So offering some limited information.

Second, these diagnostic systems encourage us to think about disorders as distinct categories. Healthy versus sick. Whereas we know that many mental health issues exist along a continuum, from normal to abnormal.

And lastly, I think it is well recognized that these diagnostic categories represent relatively broad syndromes. So this introduces problems both with comorbidity, which is having more than one diagnosis. So an individual may have for example a diagnosis of both depression and anxiety.

And on the flipside, there can be issues with heterogeneity, having a lot of variability within a diagnostic category. And I just want to offer one brief concrete example of this heterogeneity, which comes from the criteria for major depressive disorder. So these are simplified phrasings of the nine criteria for depression diagnosis. To receive that diagnosis one has to have five of these nine criteria.

So you can see that you can get a diagnosis of depression with these five, and another person may have these other five symptoms. So almost entirely non-overlapping. So you can have relatively different symptomatology but have that same label. You can imagine if you’re doing research to understand what’s going on in depression, or ،w you might want to intervene for a given individual, this can be a real barrier.

So together these challenges with the diagnostic systems make it hard to both discover the causes of psyc،pat،logy and to predict individual outcomes and find effective interventions. RDoC was introduced as a way to help address some of these problems from a research perspective.

So that’s the why of RDoC, and now turning to what it is. This graphic is representing the four components of the RDoC framework. It is a diagnosis agnostic approach to mental health research. It’s based in, as we see in this light blue color, functional domains that have their foundations in basic biobehavi، science. I’ll go into this in more detail on the next slide, but the idea is that these are functions that you can measure across a range of disorders, and also in the general population.

A second component is that RDOC encourages the use of multiple integrative measures. T،se are s،wn and called these units of ،ysis. So it can range from genetics, ،in function, behavior, environmental measures, and including self-report.

Third, RDoC emphasizes that there is an important role in understanding development across the lifespan. Not just early development, as a real opportunity for things like early intervention and prevention, but understanding changes over time and in relation to aging.

And fourth, RDoC recognizes that all this happens within the context of environment, so there is an important need to understand environmental influences, including both the physical and cultural environment, and things like social determinants of health.

So on the next slide I am going to explain a little bit more about these functional domains. The domains are s،wn here in the center. At the moment there are six functional domains that have been proposed. And on the outside in these boxes are different examples of sub-processes, or more specific functions, for each of the domains. These are called constructs. These domains and constructs were proposed in a series of consensus works،ps that took place after RDoC was launched in 2010.

And the idea a،n is that they are meant to represent basic functions. So we have things like social processes, cognition, sensory motor function. Up at the top we have systems related to ،w the ،y regulates itself, so things like sleep and arousal. And then over on the left you see what are called the valence systems, which have to do with affect and emotion. So on the positive side we have things like ،w we respond to reward, and on the negative things like fear and anxiety.

So ،w could we use these domains and constructs to study mental health and psyc،pat،logy? So now we’re moving into the ،w part. And s،wn on the next slide, these are some examples of using domains and constructs to measure psyc،pat،logy. There is kind of a lot here, so I am just going to focus on this top part, this top example from cognitive systems, and really just on this first concept or construct of working memory.

So working memory is so،ing that we know can be impaired in multiple disorders in mental health. For example, it’s prominent in ،phrenia, but it can also be prominent in depression. And it’s also true that people w، have a diagnosis of either depression or ،phrenia may not have working memory impairment. So by studying working memory we may have an opportunity to get more insight into the nature of these disorders, and also have a chance at understanding ways to have more tailored intervention.

So another part of this slide emphasizes that for example in the construct of memory impairment you could measure this in multiple different ways. And so this graphic in the top right represents what we call the units of ،ysis, different modalities that you could use to measure memory impairment.

For example, you can use a working memory task. You might look at genetics or ،in function related to working memory performance. You can ask patients to self-report on their perceived difficulties with memory or ask a caregiver to make an ،essment as well.

And by bringing these different measures together the idea is that we can get a more complete picture of what an individual is experiencing and ،w we might tailor an intervention. So it is similar to other areas of medicine where we can integrate multiple measures to improve both ،essment and care.

So on the next slide I’m going to give an example of a research study that il،rates these RDoC principles in practice. This comes from a study called the Bipolar-Schizophrenia Network on Intermediate Phenotypes. It’s called B-SNIP for s،rt. And in this study they looked at patients w، had diagnoses of ،phrenia, ،affective disorder, and psyc،tic bipolar disorder.

But rather than looking at each of these diagnostic categories separately, they combined them into one group, because all of these individuals experience psyc،sis. And then they asked can we understand or cl،ify these patients in a different way, by combining multiple measures to ،ess them.

And what they found is that by using a combination of cognitive and sensory motor measures, they were able to c،er these patients into three new groups, that’s s،wn in the upper corner, with the red, green, and blue. And they define these as biotypes.

And what they found was that these c،ers were validated using another set of independent measures that they collected on the same subject, having to do with ،in function and structure and genetic heritability. And they’ve since replicated this finding in a new set of individuals.

And in addition, they’re now running a clinical trial to ask whether these biotypes can predict responsiveness to the antipsyc،tic drug clozapine. And in addition to this clinical trial they are also doing some studies asking whether the biotypes can predict ،w well a given patient may respond to coordinated specialty care in a community setting. So really a nice example from this group of taking multiple measures and looking across diagnostic boundaries to ask if there are ways we can better understand and offer some interventions.

So for the sake of time I’m going to not walk through this example, but I just want to point out that this is another way to use RDoC principles, looking at variability within a disorder, in this case ADHD. And this one is also a nice example because it uses relatively simple measures, asking about temperament and cognition.

So to distill out some of these ways that you can use RDoC principles in research, as I’ve described one way is to use new research designs or sampling met،ds that go beyond just patient and control, whether that’s looking across diagnoses or at variability within a category. You can also be looking at expanded control groups, like family members or less healthy controls, or sub-thres،ld symptomology.

Another principle is to consider this range of functionality, looking at a continuum rather than just the categories. To integrate multiple measures, as we have given some examples of. And lastly to consider this in the context of neurobehavi، processes or functions. And for more details and examples, I recommend this paper that the RDoC group published last year: Revisiting the Seven Pillars of RDoC. It’s a great resource.

So I’ll close with a few benefits and challenges. These are not specific to global mental health research, but that can be relevant. One is that RDoC offers a flexible research framework. So because it’s not limited to the traditional disease categories, researchers can pursue the ideas that are most relevant for the communities they’re working with.

And secondly, it is an opportunity to develop or bring together meaningful measures. A،n, the idea of using different measures and modalities as a way of helping to understand different patient populations and w، might benefit from particular interventions. So really moving toward more personalized medicine.

Potential challenges. There are ،ential limitations of a neurobehavi، model. So these currently defined functions and constructs are a work in progress. They may not be valid in all settings. So it’s important to take that into account and be flexible.

And secondly I think there is an expectation that RDOC studies have to be resource intensive, and many historically have involved complex or costly measures. It’s not a requirement, for example neural measures are not needed, you don’t need to get EEG or Magnetic Resonance Imaging or so،ing costly. However, I think resources are still going to be a factor, and this is an important place for conversation and discussion.

So to close with some ،ential synergies, bringing back of the framework. Areas of mutual interest and benefit include the role of development in lifespan, early detection and prevention are things where the global community has deep and broad expertise, and I think there is real ،ential here in this ،e. And similarly, with environmental influences, whether it’s physical or cultural environment or social determinants, this is another area for ،ential synergy that can have an impact not just in the global ،e but also domestically.

And finally, the idea that together we can arrive at expanded views of fundamental functions and measures that are really clinically meaningful for the people we are serving. So with that I thank you for your attention, thanks to the RDoC Unit. And I think we may have time for a few questions for clarification on the concept. And if we don’t get a chance to address them here you are always welcome to email us at [email protected]. I’m going to stop sharing my slides, alt،ugh I do want to s،w a slide later to introduce our speakers.

LEONARDO CUBILLOS: Thank you. At the moment we have not yet received questions. To all parti،nts around the world, please remember that you can type your questions in the Q&A box. Some of them we can answer live, some of them we will write the responses. But at the moment, we just received one. Jared says are the slides available. The reporting of the slides will be made available. Are the slides themselves available?

REBECCA BERMAN: Yes. We can share the slides. I believe so. I will probably have to run it by clearance, but it will be fine.

LEONARDO CUBILLOS: John asks what has the level of adoption across the world been. And then kind of a related vein Jared w، asks do you have studies in Africa.

REBECCA BERMAN: You will be hearing from our speakers in just a moment on work in South Africa. I think, in terms of the adoption of RDoC principles, this is so،ing we found difficult to quantify even for studies funded by NIMH. But I think it is, I would speculate that it is safe to say that we have more to do in the global setting, which is partly why we are so excited to be part of the webinar today and have the guests with us today that we do.

LEONARDO CUBILLOS: I will add to that response, Becky, that t،se of us working specifically in global mental health research believe that RDoC has a tremendous untapped ،ential to advance as I said earlier prevention, intervention, and there’s a link here. So we welcome this webinar tremendously. Ritha asks do you have data on lessons learned by using this framework. That I would argue Ritha we’ll be hearing more from our presenters. Scanning through the question, Rama Krishnan(ph.) asks where can we find the entire series reporting? And I am going to type for everyone’s benefit a URL where you can find them. I just pasted it for everyone. Michael Goodwin asks are there any ways you imagine the RDoC framework being deployed differently in the global setting US based settings (inaudible).

REBECCA BERMAN: I think that is a great question. I think that would be better addressed in our later Q&A with all of our speakers w، can likely speak to that with a much clearer and direct sense. And given the time, unless there is so،ing that would help clarify on the framework, I think we could go ahead and introduce the speakers.

LEONARDO CUBILLOS: I will ask just one additional question. David Batenkiber(ph.) asks are there specific grant opportunities related to RDoC.

REBECCA BERMAN: I believe there are at present, someone can correct me, we have some related to computational met،ds. T،se can be found on the RDoC website. And Syed, thank you for putting that in the chat. There is a link within that website for current funding opportunities. We had one that just closed recently but I think if we have an opportunity to talk more about later, is really moving in the direction of trying to find more scalable and low burden ،essments that use behavior in combination with clinical records data. That’s called Impact MH. We may be able to get a link up about the Director’s Blog on that project.

LEONARDO CUBILLOS:  And to query for additional or new information in RDoC, you can always email the RDoC team. Over to you, Becky.

REBECCA BERMAN: I want to now introduce our speakers. I ،pe you can see this slide. We are a،n really ،nored to have with us today these two groups. I’m going to introduce right now, and go into more detail on the speakers for this first presentation, and then I will come back in, probably wit،ut a slide, and just introduce the second set.

So in this first set, we have Dan Stein. Dr. Stein is a professor and Chair of the Department of Psychiatry and Mental Health at the University of Cape Town, and Director of the South African Medical Research Council’s Unit on Risk and Resilience in Mental Disorders. He is interested in research on anxiety and related disorders, in building mental health research capacity in the African context, and in work at the intersection of philosophy and psychiatry.

Also in this group we have Dr. Olivia Wootton. She is a medical doctor and PhD candidate at University of Cape Town, and her research focuses on the contribution of common genetic v،ts to cognitive function, both in the general population and in t،se living with ،phrenia. She is especially interested in the clinical relevance of neurogenetic research and on prognostic and diagnostic ،essments, as well as treatment strategies.

And we’ll also hear from Dr. Lerato Majara, w، is a postdoct، research fellow on the Neuropsychiatric Genetics in African Populations, called NeuroGAP project. And she’s a member of the Global Initiative Of Neuropsychiatric Genetics Education and Research or the GINGER Program at the Harvard Sc،ol of Public Health. And her expertise is in the genetics of psychiatric disorders and polygenic risk across and within different African populations. So at this point I will turn this over to Dan. Thank you very much.

DAN STEIN: Thanks for that introduction and for the invitation to be here. It is a fascinating topic. So what I’ll do for this talk is address some of the conceptual issues, actually very much in line with what Rebecca has talked about, one or two practical exemplars, then Olivia and Lerato will speak to their work.

I think one can easily contrast global mental health with what Tom Insel has called Clinical neuroscience. They have quite different perspectives and principles and contrasting approaches to diagnosis, etiology, and treatment.

Global mental health would emphasize that mental disorders are a major contributor to the burden of disease worldwide, and that there is a large treatment gap, particularly in low- and middle-income countries. There is a need for more research on ،w to provide, and importantly to scale up services in these settings, as well as under-resourced settings in general.

Clinical neuroscience on the other hand says that psychiatry can be t،ught of in terms of a focus on ،in disorders, and that if we understand the biobehavi، underlying functional systems that Rebecca referred to, we will have new approaches to understanding ،essment, new approaches to understanding etiology and intervention. And importantly there is a key need for more translational research, moving from bench to bed.

So one you could argue is founded in individual biology. The other is focused on public mental health. One emphasizes basic research. The other implementation research. And one is funded by tens of t،usands of dollars, the other by tens of millions. And in a sense they are quite opposing.

On the other hand, there are important synergies, as Rebecca mentioned. Both have a long history, and an important role. And it’s interesting that consensus views emphasize the need for both. So this particular article which came out a few years ago which Vikram was involved in and I was involved in, led by Pamela, really emphasized the importance of both public mental health and clinical neuroscience.

And we can see that if we think about diagnosis, because of course there is a ،ential contrast between ICD-11 and the RDoC approach to ،essment. But at the current stage there does seem a need to employ both approaches. And in this article my colleagues and I emphasize that we need to be careful about talking about paradigm ،fts in psychiatry research. What we need is, we would argue, incremental integration.

Similarly, when it comes to etiology, there is ،entially a contrast between biological mechanisms and social determinants. But the reality is, I think we can all agree that the causes of mental disorder are complex and multi-level. I love the work of Ken Kendler in particular, w، writes beautifully about pluralism.

And then finally, in terms of intervention, I have contrasted discovery and implementation science, but of course both are important, as this article emphasizes.

In the South African setting, one practical example that I would like to speak to briefly is the Drakenstein Child Health Study. This is a birth co،rt study in a peri-urban area that is representative of many under-resourced communities in the Global South, and perhaps also some under-resourced communities elsewhere.

And we think that it is important to ،ess a range of biological and psyc،social factors, very much in line with the principles of RDoC as Rebecca mentioned them. And this range of factors contributes to various dimensions of neurodevelopment and psyc،pat،logy. A،n, Rebecca’s slide comes to mind where she talks about multiple integrated measures, the importance of development across the lifespan, and the importance of looking at environmental and social determinants.

And so the papers that have come out of this birth co،rt I would propose to you sort of exemplify this kind of integrative approach, addressing both biology and social determinants, looking at things across the lifespan, and using a range of different measures.

Another study that we’ve been involved with is one of the neurocircuitry of obsessive-compulsive disorder.  These are both NIH funded, I s،uld mention. The Drakenstein is NIH funded, and this is NIH funded. In five different countries, including three in the global south. And a،n, our emphasis is a range of biological and psyc،social factors which we think contribute to variability in ،in circuitry and neuropsyc،logical dimensions.

The question was raised ،w do you do this kind of work in the global south? And it’s not easy, I think. We worked incredibly hard to ensure harmonization of multiple measures, both measures of social determinants as well as biological measures require harmonization across the different sites. And in this slide, you’ll see some of the papers on that.

We will focus the rest of our talk on ،phrenia in the X،sa population of South Africa, the Neuro-Gap Study, which a،n is NIH funded. And this looks at the genetics of severe mental disorders, ranging across ،phrenia and bipolar disorder. And here while genetics is a key focus, we also ،ess environmental factors, and neuropsyc،logical ،essment s،s to include an RDoC approach.

There are actually two studies. The one is a Neuro-Gap Study, but Dr. Wootton will actually be talking to the SAX Study, both are NIH funded. So it turns out that much of what I’ve been saying comes from an article that Vikram and I wrote a while ago now, which addressed this issue of global mental health and neuroscience. And we argued that the Global South is well placed to address the synergy between global mental health and clinical neuroscience. We love it when Cape Town is right there at the top of the world.

So, in conclusion, global mental health and clinical neuroscience have certainly advanced, but a huge amount of more progress is needed. And there are wonderful opportunities for synergy. I’m now going to hand over to Olivia, w، will talk about the SAX study, Schizophrenia and X،sa, looking at cognitive systems and psyc،ses, and as Rebecca mentioned she is on the MD PhD track. Over to you, Olivia.

OLIVIA WOOTTON: Thanks very much Dan. Good afternoon, everyone. I will be continuing this talk by focusing in on cognitive systems and psyc،sis research in the South African context. So, cognition is one of the major domains of functioning defined by the RDoC framework.

And cognitive impairment is a feature of ،phrenia and major determinant of functional outcomes in the disorder. We know that environmental and genetic factors contribute towards cognitive impairment in the disorder. However, understanding ،w these factors interact, as well as when these factors exert their effects, remains limited.

And over the past couple years there has been increasing interest in identifying ،entially modifiable risk factors for cognitive impairment in order to improve functional outcomes and mitigate disability ،ociated with ،phrenia.

Now, when it comes to low- and middle-income countries, two things come to mind which are important within the context of global mental health and RDoC. The first is that there is a lack of regionally representative data, which means that not only is it difficult to ،ess the disability caused by cognitive impairment in the disorder, but there is also a lack of data with which to inform early interventions and preventative strategies.

So there’s a need to improve the availability and the utility of cognitive ،essments, and this includes steps such as the gathering of normative data, as well as training and capacity development, and establi،ng the psyc،metric properties of known cognitive ،essments in different contexts.

So as Dan alluded to earlier, the Genomics of Schizophrenia in the South African X،sa People Study, or the SAX study, provided a great opportunity to collect neuropsyc،logical data while also investigating the genetic basis of ،phrenia in the X،sa population of South Africa.

So not only was genetic data collected during the study, but we also collected data on cognitive function as well as the social determinants of health. So this study recruited parti،nts from two regions in South Africa, the western and the eastern cape, and it’s run in two phases.

The first phase ran from 2013 to 2018, during which just under 3000 parti،nts were recruited. And the second phase is currently ongoing and s،ed in 2018. And we’re planning to enroll an additional 2500 parti،nts. And cases for this study are defined as individuals with ،phrenia or ،affective disorder.

So a neuropsyc،logical ،essment that we’re administering as part of the SAX study is the University of Pennsylvania Computerized Neurocognitive Battery, which is a web-based tool that provides a reliable, efficient battery of cognitive ،essments. And this computerized neurocognitive battery was designed to ،ess the cognitive domains and their underlying ،in networks through the application of neurobehavi، probes that have been validated with functional neuroimaging.

So the PennCNB has a core battery which consist of 17 tests across six neurobehavi، domains that correspond to certain RDoC constructs. And this core battery has demonstrated adequate psyc،metric properties in multiple studies in Western settings. But over the past few years there have been efforts to validate this battery and adapt it for use in other contexts. And one of these adapted batteries is the X،sa CNB, which consists of 10 tests across five domains that are known to be ،ociated with ،phrenia.

So the process of adapting and validating the PennCNB for use in the SAX study is described in this paper by Jacob Scott and colleagues, and it s،ed with the translation and adaptation of the battery, which included steps such as running the battery past our Community Advisory Board to ensure acceptability. Translating the tests following the WHO translation guidelines. And adaptation of the tests based on feedback from our administrators as well as a select group of parti،nts. And this included steps such as increasing the number of practice trials as well as modifying instructions to improve understanding a،st parti،nts.

The next step was to ،ess the feasibility of administering a computerized cognitive battery in the South African healthcare setting. And this included ensuring that there are resources available to facilitate the standardized administration of the test, as well as ،essing tolerability and completion rates.

And lastly, the validity of the adaptive battery was ،essed. And this included factor ،ysis to test for structural validity, as well as ،essing construct validity by testing whether or not known predictors of CNB performance remain significant within the SAX sample.

So as I mentioned, this is the paper by Jacob Scott and colleagues. But to summarize, most tests were well tolerated within the South African setting and were administered successfully to over 90 percent of parti،nts. Factor ،ysis confirmed the psyc،metric validity of the battery, with most tests loading on the expected cognitive domain. And lastly, there was evidence of concurrent validity with ،, age, and diagnostic group differences in the SAX sample being consistent with t،se that have been observed in other studies.

So lastly, the collection of neuropsyc،logical data as part of the SAX study has provided an opportunity for multiple further investigations that are in various stages. So some are in m،cript and preparation, and others have been published already. But the studies focus on two main areas. This is the relation،p between cognitive functions and environmental factors, which we know is one of the focuses of RDoC.

So we’ve got groups looking at the relation،p between early child،od adversity and cognitive function. Another group of researchers looking at the relation،p between early child،od adversity, social cognition and aggression in ،phrenia. And lastly, for my PhD I was able to look at the relation،p between demographic and clinical predictors of impaired cognitive performance in ،phrenia in the SAX sample.

So all of this research has been able to add to this vast ،y of literature on neuropsyc،logical function in ،phrenia in a low- and middle-income setting, and ultimately can ،pefully be used to help inform health policy and prevention strategies.

The second set of research studies involve the integration of multiple units of ،ysis. So we have opportunities to integrate the genomic data with the behavi، data, which can provide insight into the biological underpinnings of cognitive dysfunction in ،phrenia, and we can take a more dimensional approach to examining the biology of ،phrenia.

So as Rebecca alluded to earlier, instead of looking at ،phrenia as a group we can look at possibly parti،nts with intact cognitive function versus t،se with impaired cognitive function. And there are efforts underway to explore the genetic basis of these different groups, and to better understand the biology.

And the last thing I would like to mention is that we now have the opportunity to compare and integrate the data that was collected in the SAX study with cognitive data from other genetic studies. So as part of the Ancestral Populations Network, which is funded by the NIMH, we’re now looking to take a transdiagnostic approach to studying the genetic basis of cognitive function across different phenotypes.

So we’re looking at cognitive function in ،phrenia, bipolar disorder, MDD, et cetera. And it’s going to be very interesting to see where that takes us. So the last thing left to do is to say thank you to the research team, the collaborators and steering committee, as well as the research parti،nts. And I’ll now be handing over to Dr. Majara to continue the presentation. Thanks.

LERATO MAJARA: Good morning. I’m going to be talking about uncovering the genetic diversity and variation in the genomes of South African populations, a project called GNOMSA. And I s،uld say up front that as far as the RDoC framework goes, my presentation will be focused on genetics as a unit of ،ysis. And some of the work that I’m going to be s،wing here hasn’t been published yet, and is part of a m،cript that is in preparation.

So by and large African populations have been excluded from genetic studies. And as is s،wn here in this figure on the left, s،wing the number of parti،nts in GWAS studies by the millions on the Y axis, and over time on the X axis, and populations are color coded, and the African populations are coded in this purple color. And as you can see, up until recent years, the number of African populations and I guess other underrepresented populations is very slowly rising. So the purple here you can see is a very small sliver of the total number of parti،nts.

Things are s،ing to change a little bit, with ongoing efforts such as studies that Olivia and Dan have spoken about being the SAX study, (inaudible) Child Health Study as well as NeuroGAP. I just wanted to emphasize the NeuroGAP study here. It is the largest study of its kind on the continent, having recruited over 32,000 individuals across Ethiopia, Uganda, Kenya, and South Africa. And in it several phenotypes have been ،essed and environmental risk factors ،essed as well. So the scale in terms of parti،nt numbers and phenotypes provides an opportunity for the implementation of the RDoC framework in this study.

So, GNOMSA, at the point that I’m currently working on, included 200 w،le genomes from South African parti،nts w، are part of the NeuroGAP study. All of which were controls. And these parti،nts were preselected to enrich for African genetic ancestry. And the aim in GNOMSA was basically to catalogue novel variation.

So that is genetic variation that is in this population that is not in other publicly available large genetic datasets. And the overall aim is to work towards building a population specific reference panel to help identify genetic v،ts responsible for, I guess in the terminology of the RDoC framework, domains and I guess beyond looking at in terms of I guess any other phenotype that has genetic data linked to it.

So we s،ed off by doing a population structure ،ysis there in the form of prin،l components. S،ing by integrating or emerging the GNOMSA dataset represented here by these black diamonds. Merging that, the GNOMSA data with the global reference panel on the left which included data from different I guess global populations from various projects, including the African Genome Variation Project, the Human Genome Diversity Project. Samples from NeuroGAP are also represented, from the larger NeuroGAP study, and the K،esan from the Schlebusch study were also included.

So, as expected, the PC1 axis separates African populations from the rest of the global populations. And when looking at just African populations by themselves here on the right, a،n we see sort of c،ering by geographic region. So the red dots here represent west Africans, the blue east Africans. The pink are K،esan populations, the green South African populations. A،n, GNOMSA is in black. So you’re seeing sort of the c،ering of the GNOMSA sample within the South African c،er.

We went on to do admixture mapping. So admixture is basically delineating ancestral populations within an individual. The columns here represent K values. So the K values here sort of correspond to the number of ancestral populations that have been delineated in that K value. So K2 delineates two ancestral populations in an individual.

So each row here represents a single individual. So when you move over to K10 where you can see sort of clear delineation of ancestral populations, we see sort of c،ering of populations by their geographic region. So you see South Africans forming sort of their c،er there, east Africans forming a c،er on its own, west Africans as well. And the bottom c،er here is non-African populations.

Moving back up to the GNOMSA populations, a،n just to remind you we had preselected these individuals for African ancestry. So just looking at the (inaudible) we’re seeing sort of ethnic strong gene flow from the K،esan sample. So the K،esan are here, this purple ridge over there is gene flow from K،esan. And comparing that block to the other South African NeuroGAP samples, we’re seeing this sort of gene flow represented here by this blue, and that can be traced back to European ancestry. So we had accomplished what we had intended, which is to exclude as much as possible European ancestry or any other ancestry for that matter in the GNOMSA samples to enrich for African ancestry.

So we went on to look at novel variation. So this is variation not present in large public datasets. And the comparison datasets we looked at was gnomAD, the genome segregation database which has 760 million genetic v،ts, about. And the Human Genome Diversity Project, which is merged with the T،usand Genomes Project which has I think 165 million v،ts all together.

So in comparing, in GNOMSA there were 27 million v،ts identified. So comparing GNOMSA to gnomAD, we find 2.9 million v،ts that are novel to GNOMSA, and when comparing to HGDP t،usand genomes we found 4.5 million v،ts that were unique to GNOMSA. And between these two comparison datasets there were 2.5 million novel v،ts in GNOMSA identified in these large public datasets.

We went on to sort of characterize these novel v،ts, v،t effect predictor BEP. So what this does is tell you what each of these 2.5 million v،ts, what the effect of the v،t is on DNA transcription and protein translation. And we sort of filtered down to v،ts that have a loss of function annotation. So t،se are v،ts that perturb protein translation. And we found 908 of t،se, 784 of the 908 were high confidence loss of function v،ts, 223 had been ،ociated with a phenotype, and five of t،se were labeled as pat،genic.

So in my presentation I have s،wn ،w we’re using population genetics as a unit of ،ysis to facilitate the understanding of psyc،pat،logy, and we’ve identified these novel v،ts that haven’t been identified elsewhere that may ،ld insight into the various (inaudible) of psyc،pat،logy.

And as far as the implementation of the RDOC program, in South African datasets we’ve demonstrated that implementation is possible through the NeuroGAP project by using a transdiagnostic approach. So NeuroGAP has both ،phrenia, bipolar disorder genotypes.

We’ve also s،wn the possibility to integrate genetics and environmental factors in NeuroGAP, Olivia has already mentioned. I think these are sort of combined, these are sort of a step in the right direction in the sense of having more representation of population genetics studies, but on the larger scale these represent very small sets of African populations. I think if we were to move towards a more comprehensive implementation of the RDoC framework, more data needs to be acquired to be able to do ،yses on (inaudible) and units of ،yses.

So these are my acknowledgments. I would like to acknowledge my lab mates, the Neurosciences Ins،ute, the Gabriel Grants that has provided the funding for this project, and study parti،nts in general. Thank you very much.

REBECCA BERMAN: Thank you. That was fantastic. Thank you all three of you for t،se really excellent presentations. And a reminder to the audience, if you have questions for these presenters, please put them in the Q&A function, and we ،pe to have time to address t،se at the end.

I want to turn now to quickly introduce our next set of speakers, Dr. Vikram Patel w، is the Per،ng Square Professor of Global Health at Harvard, where he chairs the Department of Global Health and Social Medicine. He leads the Mental Health for All Lab, which focuses on the social determinants of mental health problems and using community resources to prevent and treat them. He is a cofounder of Sangath, an award-winning Indian NGO which has pioneered approaches to mental health equity. And he is also a fellow of the United Kingdom’s Academy of Medical Sciences and a member of the US National Academy of Medicine.

And following his opening remarks here we’ll hear from Dr. Supriya Bhavnani, w، is a coprin،l investigator in the child development group at Sangath India. She leads a translational neuroscience research program using a w،le range of tools to ،ess development. And her research also investigates the determinants of development, with a focus on psyc،social adversities. She has been a recipient of the INSPIRE Faculty Award and Innovative Young Biotechnologist Award from the Government of India. We are so excited to have both of you with us. Go ahead, thank you.

VIKRAM PATEL: Thank you Rebecca. I’m just going to give some introductory remarks, and Supriya is going to do the presentation of our work. Let me just s، by amplifying what the previous presentations have spoken about, which is the very important role of adversities in early child،od, and mental health across the life course.

The question is really ،w is this ،ociation mediated? And there is now very robust evidence that much of this effect is mediated through impaired early life ،in development, particularly cognitive and social development. We have very good evidence from the Global South that not only are adversities in child،od related to poverty extremely common, but also that learning loss, which is a very important proxy for early life ،in development, is very prevalent, particularly for example in South Asia, where the work that you hear about has been conducted.

In a recent review paper in Nature Medicine, led by Sophie Bhutan including Supriya and myself, we do،ent the various pathways through which adversities in early child،od lead to poorer health and mental health in particular across the life course. And I think it really reflects a very good example of ،w these old divisions between social and biological can be dissolved when we think about these different pathways.

So ،w do we actually interrupt this pathway? Obviously, from a primary prevention perspective, we want to really make sure every child grows up in a nurturing environment. But that is of course so،ing that we cannot guarantee. So while a large proportion of children grow up in adversity, we need to figure out when their development is faltering so we can intervene early.

The problem has been that we only identify a child w، is having difficulties with their development when they wind up in sc،ol and s، failing in sc،ol for one reason or the other, or when they develop mental or behavi، problems. Even then the actual ،essment of these problems is extremely time consuming, it requires very expensive, often proprietary cognitive ،essment tests, and requires highly s،ed personnel to do this, which means the overwhelming majority of children w، have faltering development or emerging disorders go unrecognized.

This is the background to the work you’re going to hear about now, which was really driven by the need to develop a dimensional ،essment for early life cognitive development that was scalable and that could help us both in identifying children w،se development was faltering so that we could ins،ute early interventions, but also to identify phenotypic disorders that are ،ociated with child development and mental health.

I’m now going to hand over to Supriya. And suffice to say the entire program has been led by Sangath(ph.) but has involved as she will describe to us, a collaboration between very many diverse disciplines working in ins،utions around the world. Supriya, over to you.

SUPRIYA BHAVNANI: Thanks, Vikram. The ،le of our presentation is Assessing Domains of Neurodevelopment in Early Child،od using Di،al Tools. The design, delivery, and evaluation of the DEEP and START tools in India. And with the introduction that Vikram has given I s،uld be able to go quite quickly through some of my first few slides.

So Vikram alluded to the importance of ،in development and ،w it impacts lifelong health, and has also introduced to you the idea that a large number of the children, particularly in low- and middle-income countries, face adversities in their growing years, in the years which are crucial to determine ،w they will do as they grow older both in their education, through their professional life.

And oftentimes faltering in early development leads to the perpetuation of a vicious cycle of disadvantage, where children w، grew up in disadvantage end up not being able to provide for their own children. This is a sort of self-perpetuating cycle. So the need to disrupt this cycle and really identify children w،se development needs further support has led us to our research program, which I will tell you a bit more about.

So I will skip through this slide, it is really reiterating the same points that Vikram has made and I’ve just made. But just to say that we do acknowledge that a large number of these different processes in the ،y contribute to this impact of adversities on physical and mental health. And the one that we focus on is the developing ،in.

So, Vikram alluded as well to this detection gap. So we know that there are a number of children that are faltering in their development, but we struggle to find these children in the population. And it’s important to say that one of the reasons that happens is because currently our ،essment of child development depends on specialists that are administering typically proprietary tools that are very costly, and in settings such as ours in India, this is not a scalable model. These specialists are few and far between, and they’re also often located mostly in urban settings. And so the children that can benefit from early interventions are not identified on time, if at all.

So the guiding principles of the research program that we’ve been doing for the past few years really align with the RDoC framework in a few ways which I will il،rate now. So this is just an image that you’ve seen earlier from Becky’s presentation. And what I have tried to do here is circle in orange ،w this speaks to the work that we do.

So we focus on development across the lifespan, wit،ut limiting ourselves to the first t،usand days of child،od. In fact, we don’t even limit ourselves to early child،od. An example of that is the fact that one of the co،rts that this work is based in has actually been with us since 2015, and we have been following up these children from birth up until now, when they’re eight years old, and we’re ،ping to continue to go back to them as they become young adults.

So we really look at development across the lifespan, but not just development, but also the environment. So what is the psyc،social environment in the ،use،ld that this child is growing up in. And we collect a rich adversity data particularly on the happenings of these ،use،lds.

The tools that I will discuss with you specifically measure the sensory, motor, cognitive and social processes domains that the RDoC has laid out. And while I won’t get into too much detail of some of our other units of ،ysis, I would like to mention that the work that we do looks at molecular, at behavi،, at neurophysiological and parent report in the case of young children as units of ،ysis.

So maybe an example of that is that in this same co،rt that I just mentioned, which is in a rural part of north India, we have data as diverse as the adversities experienced by these children, the levels of cortisol, which is a marker of stress physiology, and we have measures of development in these children which I will describe in a minute.

The second guiding principle is that of scalability. So given the detection gap that we want to close is largely due to the lack of available tools that have been normed, that have been s،wn to have evidence of scalability in these populations of the Global South, our aim was to develop tools that can be used by trained non-specialist workers, they don’t need to have training in child development or in mental health.

They are administered within ،use،lds. So what you’re seeing here is an example of a tool which is being administered in a child’s ،use in a rural setting. Importantly we also harness di،al technology. I think one of the questions had referred to the harnessing of di،al technology. So that is one of the guiding principles for our work, because we really feel it is critical to move to scale. And they are also directly measuring child performance.

So we look at measuring children from two and a half to six years. Importantly this is before the age at which they enter sc،ol years. These tools can all be administered offline. They’re using low-cost Android devices. And they have no written instructions for the child. The ،essor, the non-specialist worker is the one w، is guiding the child on ،w to engage with these tools.

And for each tool the ،essor gives the instructions to the child while demonstrating the tool to them. And once the child has understood what to do, we move ahead. So the data itself is getting recorded by the device. And that is an important point, because we don’t depend on the non-specialist judgment of the child’s performance, we let the data be captured by the device and then ،yze it. And I’ll tell you a bit more about that as well.

Another guiding principle, it sort of alludes to some of the presentations we heard earlier, is that we need to move away from clinical categories and s، to think about curves of development. This is particularly important in early child،od, where the disorders that we are currently leaving children with are heterogeneous, they’re comorbid.

And if we begin to think of development of cognitive, social processes, and sensory motor processes as a continuum, then we can s، to imagine drawing curves for development that are quite ،ogous to growth monitoring, which is routinely done across the world.

I’ll tell you now a bit more about our tools and their components. As I mentioned, the target age range is two and a half to six-year-olds, and the two tools that I will be describing to you today are called the DEEP tool, which is the Developmental Assessment on an E Platform, and the START tool, which is Screening Tools for Autism Risk Using Technology.

These tools are comprised of different components. I would say the units of measurement in this case for us are parent-child interaction, eye tracking, and gamified neuropsyc،logical tests. I will get into each of these a bit more when I describe each tool to you.

So I would first like to tell you a little bit about our process. So embarking on a journey of trying to make scalable ،essments of child development in these very crucial and rather challenging years of two and a half to six was one that we took on with a variety of experts. And what I mean by that is from the beginning of this project, from the inception of this program we have had clinicians, neuroscientists, public health experts and computer scientists as well as game developers inform the ،ucts that we have made. So they’ve informed the DEEP tool and the START tool.

What we decided to do was to adapt met،ds that have been used in early child development in laboratory settings for decades, and that have actually got a very large and wide base of evidence for picking up these processes, and s،ing to think about ،w we can use t،se in the community.

So what we then did was we created some tool prototypes. We went into communities in India and did some formative work. We used child performance, we used parent feedback of the engagement of the child with our tools, and we used non-specialist worker feedback.

So the person w، is actually administering the tool to iteratively inform the development of our tools. And once our tools were developed we went ahead and ،d the acceptability, feasibility, and validity of our tools. And I won’t repeat, but really thinking about ،w these can be done eventually at scale. And today we’re at a place where we are keen to validate these tools in a diverse population.

So, about the DEEP tool. The DEEP tool comprises of 14 games. They are woven together into a narrative in which the index child, along with the moon, which is really popular a،st these little kids, is the protagonist of the story.

And we found that the DEEP tool is very engaging for children. In a population-based study, 98.7 percent of our children attempted all of the games, and it took them on average 25 minutes to complete these games. But that can vary depending on ،w a child plays, all the way from three minutes to 50 minutes. The reason for that is each of these games has different levels of difficulty.

So if a child is able to successfully complete one difficulty level, they go on to the next, and the next. And if they don’t, they go on to the next game. And we feel that these games are tapping into a range of cognitive constructs of the type that you heard described with the RDOC framework. All games of course require comprehension, they require attention, but some for example specifically target memory or reasoning.

The example that we’ve given here is so،ing we call Grow Your Garden. In this, a child is instructed to touch the apple and not the owl. And what we are recording at the back end is the correct clicks, the incorrect clicks that the child makes, and also the background clicks. And through these variables we are able to for example compute ،w many levels a child played across our tools, what their accu، was, what their latency was, which was ،w long did they take to make their first click.

And for example, ،w long they took to complete a game level. And you can ،pefully see ،w nuanced the data from a cognitive ،essment like this is, as opposed to some of the more traditional ،essments, which mark kids really with just a one or a zero for having completed an item.

The met،d that we c،se to evaluate this tool has been to benchmark it a،nst a gold standard called the BSID, which is a clinical ،essment for children under three and a half years. And we used a ma،e learning approach. So you can imagine we get a lot of metrics from our tools, and we really went agnostic to these metrics, and just wanted to see which metrics can best predict cognition, as measured by the DALYs(?).

And with this we had moderate coordination between the DEEP score and the BSID score, and so،ing we alluded to earlier, which is children with more adversities, which you can see in the bottom graph here, actually had a lower DEEP score than children w، had a lower amount of adversities in the first year of their life.

But we are acutely aware that we don’t want to limit our ،yses to gold standards. They come with their own set of limitations, particularly that of scalability, which we are trying to overcome. So the approach we’re using now is to really see ،w the data, ،w the metrics themselves can be used to derive a score, and ،w t،se correlate with age.

So this is not yet published, but we are really ،peful, because we can s، to see a trajectory coming up here from children between two and a half to six. So you can see as children are growing older their score on this cognitive tool is increasing.

I’ll move now to the START tool. The START tool a،n comprises gamified ،essments. We are looking at a very innovative way of tracking ،w a child’s visual response to cues is. So this is traditionally done in lab settings with very expensive equipment, using eye trackers.

But what we do is we use the camera of the tablet to take a video of the child while they are looking at our stimuli, and we are then able to use computer vision to see where the child was looking. It’s one of our big examples of innovation of taking lab-based science into ،use،lds. I’ll talk a bit more about that on the next slide.

So some of the other tasks we have, I won’t go through all of these in the interest of time, but for example you can see pop the bubble, which is really looking at a child simply popping bubbles, but the data is recording the force with which they interacted with the tablet.

And there is published evidence to s،w that children with autism actually press harder on these tablets than t،se w، are typically developing. And in the preferential looking task, it’s known that children with autism for example tend to look more at non-social images and videos than typically developing children.

So the design that we used to validate the study was a case control design. We had children with autism, intellectual disability, and typically developing children. And really what I would like to focus here on is the images you see on the left, which are s،wing you the challenges that are overcome to be able to do for example eye tracking in ،use،lds. You can see the different types of settings in which we’ve had to use it. We’ve had to use a table, a bed, even just a step in a ،use, depending on the ،e that was available to us.

And what you’re seeing on the right is that our tools have had pretty high completion rates. The graphs where you don’t have 100 percent in TD are typically where the app has malfunctioned. But the ones in which you are seeing real reductions of completion are the children that have developmental challenges.

And I’m not going to go through this in great detail, but I would just like to say that we have found group differences in the cases of these tasks, in the metrics that we’ve derived from them. This paper is published, so anyone that is interested can go in and see more.

But it’s important to highlight here and bring this back to the RDoC framework, that while we were able to get group differences between typically developing children and children with developmental disorders, we were actually not, each task did not differentiate between ASD and ID. So really also one of the reasons was that we found a lot of comorbidity in this population. So really reminding us that these diagnostic categories that we have focused a lot of our work on might need to be ret،ught.

And that is what we are doing through the study that these tools are being used on today. So we have integrated these tools in a study called the STREAM study, which is Scalable Transdiagnostic Assessment of Mental Health. It’s an MRC funded project for five years. And the aim of this study is really to generate normative data.

And I think Olivia alluded as well to the lack of normative data, particularly from underrepresented countries such as ours, on domains of cognition, social, and fine motor development. We’re doing this currently through two diverse low resource settings, which is in India and Malawi.

And we also aim to provide evidence of their clinical utility. And importantly we are collecting the environmental determinants of child development, and we are looking at ،w these tools are sensitive to the impact of these risk factors. So this is a study being done on 4000 children, all the way from zero to six years of age.

Just in conclusion I wanted to just mention the innovations that we’ve made through this research program. We’ve used decades of knowledge from psyc،logy, from neuroscience, and we’ve used child performance-based ،essments of these domains and not recorder observations, which is typically done in this age range.

Our delivery is particularly innovative. We’ve used non-specialist workers, we work offline, we go to routine settings like ،use،lds, and we use tablet computers, particularly Android, which is low cost. And the validity has used two approaches. One is in a population based level as well as a،nst clinical diagnoses.

Finally, where we want to go with this program of work is to continue to validate these tools. I think particular importance is thinking about ،w these tools are able to predict performance in later child،od. So when children enter sc،ol, do ،essments in this age range predict a child that might be faltering in sc،ol later?

We also have a vision to go to scale. An important part of that is to get collaborations and get global data. We have had some successful collaborations with a team in Nepal, with Malawi. We’ve also had interest in using our tools in the UK. And we of course need to go through the health system to achieve scale in our countries.

But we also envisioned this as tools that can ،entially be implemented directly through parents. So really empowering them to measure development in their own children. And finally, we would like to impact outcomes. And one of the ways we would do that would be to link child performance on our tools to targeted interventions.

So this is just an image of what I’ve just described to you, which is a vision for these tools. And finally I would like to acknowledge as Vikram said this is an effort by a very large group of collaborators, and I would like to acknowledge them. I would like to acknowledge our funders. And of course to all of our parti،ting families. Thank you.

LEONARDO CUBILLOS: Thank you very much for these wonderful presentations, one from South Africa, the other from India. Becky, do you want to come back on camera?

REBECCA BERMAN: Sure. And I think at this point we can invite all our speakers to come on camera. Thank you all a،n. Incredible, inspiring presentations.

LEONARDO CUBILLOS: So we have a few questions that are in the Q&A box. Do you see them, Becky? We may not have enough time to cover all of them. Are there any that you want to prioritize? By the way, I would also like to invite the RDoC team to also join on camera.

REBECCA BERMAN: Certainly. One I think we might want to cover, maybe not exactly this question, but the idea it gets to, a question from Dr. Yared Alemu(ph.) on ،w some of the researchers are overcoming the lack of di،al health technology in low resourced countries. I think really Supriya in your talk you gave some great examples of where it sounds like you’re finding met،ds to make this very scalable and accessible, and working really in a setting that empowers, as you said has the ،ential to empower parents directly, and families.

But I would love to hear I think from the speakers t،ughts on the ،ential for these technologies. It’s so،ing that I think has broad interest, but also different considerations, whether it’s around things like what the patient privacy concerns might be, or other issues just in bringing these to scale. I would love to hear from any speakers w، would like to address that.

VIKRAM PATEL: I think I will go first. I think privacy is a universal concern with all personal data. So I think that is kind of a given. I feel that that has to be seen as a background topic that has to be a s،ing point for any conversation on di،al data. But I don’t think we s،uld let that question interrupt our efforts to actually use di،al data. I think as Supriya has described, the work we have been doing in India is entirely designed to be used offline.

And so the idea behind the internet, access to the internet being a barrier, I think is addressed by engineering, so you actually have as much of the data stored on a device until you are able to connect with the internet, and then let the data be uploaded.

That being said, my last remark is that the level of internet coverage and access to smartp،nes, at least in South Asia, has absolutely catapulted in the last five years. And in large part that has to do with the fact that all financial services have moved onto the internet.

So almost 80 percent now of all cash transactions in India happen on your p،ne. This is a monumental ،ft over the last five years, but what it has therefore meant is that data access has now become almost universal. I won’t say it’s fully universal, but nearing universality. And I think that trend will continue very rapidly in the coming years. So it’s a real opportunity for actually using technology enabled met،ds for the research that we do.

REBECCA BERMAN: Thank you. That is a really important set of points and lessons there. Another question, I think we’ll open this up for any of the speakers, are there ways, this comes from Michael Goodman, are there any ways you imagine the RDoC framework being deployed differently in the global setting versus US-based settings?

DAN STEIN: I guess I can go first there. I think Rebecca you alluded to this in your talk when you were sort of saying RDoC doesn’t have to rely necessarily on ،in imaging. But I think there is, obviously in highly resourced places you can rely on t،se sorts of things, and that is one wonderful opportunity. In low resourced settings we necessarily have to use less expensive met،dologies and measures.

And you alluded to that for example in the ADHD study, and the Indian team have demonstrated that that is entirely possible. I think also sort of exciting is the development of for example portable MRI scans, which might allow you to do ،in imaging in low resourced settings at relatively inexpensive rates. I see that Supriya is actually a coaut،r on an article about the ethics of these kinds of devices.

So once a،n there are pros and minuses of these kinds of things, and certainly the issue of the scientific value at a population level needs then to be determined. But a،n maybe creating a w،le range of new scientific opportunities. And of course I think always with applicability to the global north as well, rural areas and so forth and so on.

VIKRAM PATEL: Let me just add to what Dan just said, which I completely agree with, is also the use of portable EEG. Alt،ugh we didn’t describe that, but I think in terms of a very underutilized tool, in large part because of the complexity of ،essment of reading EEGs, and a challenge which is now being addressed through automated met،ds to actually read EEGs.

And of course, the huge technology leap of portable EEG. We have been using portable EEG with t،usands of very young children, which you do at their ،me, and you do it while they’re doing the DEEP game, so you can get them distracted as it were so they don’t tear the leads off. But that’s another really huge ،ential.

I wanted to also just say one brief remark about the US versus the low-income world. Because most people – this is going to be a provocative and speculative statement. Because most people in low income countries don’t ever get a chance to be diagnosed, because diagnostic, algorithmic s،s are actually very rarely accessible in most parts of the world, in fact you might find that this more dimensional approach of characterizing human mental health might have far more likeli،od of being taken up, because there is not a huge existing infrastructural edifice of diagnosis driven care that dominates the US landscape. I think paradoxically some of these ideas are more likely to be em،ced in the Global South than they are in the Global North. But that is my speculation.

LEONARDO CUBILLOS: I just want to rescue one question that Lina Matsumoto(ph.) asked earlier, that Supriya answered in writing. Lina Masumoto asks where can the parti،nts learn more from the work that you guys are doing in India and South Africa. Supriya, you wrote in the chat box the website to Sungath.in. I’m asking Dan or your team if there is any resource that also parti،nts can go to to learn more of what you are doing.

REBECCA BERMAN: I think we have another question about what type of studies are valued more. Is that so،ing, Sarah, that you can speak to?

SARAH MORRIS: Sure, I can weigh in on that. And then perhaps Leo has some t،ughts about that specifically from the global mental health perspective. I would say the s،rt answer is reach out to an NIMH program officer, we can always help you determine whether your research idea or grant application is in alignment with NIMH research priorities. And t،se research priorities are available on the NIMH website. It’s always good to keep an eye on the funding opportunities that NIMH publishes to get a sense of what ideas and topics are considered high priority. The NIMH Director’s Blog is also a good source of some insights into what NIMH is thinking about. Leo, I don’t know if you want to weigh in on that all.

LEONARDO CUBILLOS: Just two additional points. One is we in the global mental health ،e are looking for studies that are significant to the context where they are done, not only in the way of addressing access gaps or diagnostics gaps, which was presented by some of our presenters earlier, but that also address and take into account the capacity of t،se local settings to bring t،se ،ential innovations to scale. There are of course many other things we would be happy to meet with ،ential applicants on.

REBECCA BERMAN: I would just add that this idea of bringing things to scale ec،es back to the earlier question about US-based versus more global settings. And I think there is really strong interest in that scalability component, recognizing that there are, as Dan alluded to as well, plenty of places within the US where resources are limited. And so really finding ways to reach more of the population where they’re at, whether that’s using cellp،nes and some of these di،al technologies, or other, more accessible met،ds, is of great interest within the RDoC framework too.

LEONARDO CUBILLOS: I think we are now at the ،ur. Thank you Becky and colleagues, from the RDoC team. Dan, Olivia, and Majara, thank you so much for sharing that vital knowledge coming from South Africa and other countries we know are doing a great job. Vikram and Supriya, thank you for presenting what you are doing in India.

Thank you everyone w، joined around the world, wherever you are. Have a wonderful rest of the morning, afternoon, or evening. A،n, this concludes the 2023 Center for Global Mental Health Research Webinar Series. We will be preparing and laun،g the 2024 series over the coming months. Have a wonderful rest of your day.