 I think we're going to get started and I would like to welcome everyone to our, I guess the first in our 2016 series, the Social and Behavioral Research Branch hosts scientists who are at the intersection of social and behavioral research, genomics and health. Today we have a wonderful person coming on to give a talk, Dr. Megan Lewis. In 1992, Dr. Lewis received her doctorate in social ecology from the University of California, Irvine. She is currently the director of the Patient and Family Engagement Research Program in the Center for Communication Science at RTI International. In addition, Dr. Lewis holds an adjunct faculty position at the University of North Carolina, Chapel Hill in the Gilling School of Global Public Health. Dr. Lewis has a unique expertise in health-related interpersonal communication. That's where I know her work. She's been really formative in terms of the science that goes on in my group. Her research aims to leverage this expertise in applying social and behavioral science theory and health theory in the development of basic science as well as the development of interventions focusing on health promotion and disease management. Her research is distinguished by the use of dyadic and interpersonal theories as well as complex analytic methods to understand how couples and family functioning affect a wide array of health behavior and disease processes, including decisions related to genomics. So today, Dr. Lewis will share aspects of one of her current research efforts, the NC NEXIS study in which she is examining parental decision-making processes that are related to genomic sequencing in newborns and children. Thanks, Laura. Thank you. Can everyone hear me? Thanks so much. And I'll try to speak to the entire room, although you are a little disadvantaged over here on the left side. I want to say to everyone, it's great to be here. I feel like I said this before, I've kind of come home to a tribe that I didn't know that I was missing, but it's really been great. The NC NEXIS study, which is one of the NSITE centers funded by NICHD and NHGRI, has really, really spawned my interest in genomic communication, and it's not something I had done before this particular project, but these are really challenging issues to address. Very sticky because not only is it difficult, as you know, to communicate genomic sequencing information in the context of a family related to a newborn or young child, it gets even stickier in terms of the ethical, legal, social implication issues that revolve around this area. So I'm going to be talking about some of my work that's related to the NC NEXIS study, and people have asked me, hey, are you in the field? And we actually are not in the field quite yet. We just finished a year-long process of review with the FDA to get an investigational device exemption review completed, and we're granted that in last December, which we didn't really plan on having to do. That just kind of presented itself, and then we just have IRB. So I'm going to be talking about the formative work that's leading up to the trial that we're going to be launching in a couple of months. So to start, I want to acknowledge my project team members. These are really the core team members at RTI and UNC. And besides these people, there are many, many other people who are involved in this project, and Cindy Powell and Jonathan Berg are the center co-PIs for the NC NEXIS center. And we have these combined areas of expertise, along with many others, because this really takes a village to build a decision aid to help support parents in this area of work. And I'm really not going to be talking about some of my more basic science, theoretical hypothesis testing work. This is really all about translational work. So how do we take the information that we know that might be evidence-based and translate it and use it in a way that's very accessible to parents when they're making decisions like this? So that is, I'm coming from that perspective today. So I'm going to be talking about the NC NEXIS aims and why it's important to look at these things. The challenges that we face as part of the work of the center, what we're doing to address those challenges, and I'm going to be talking about results and some formative qualitative interviews that we did with couples. This is the dyadic part, and the discrete choice experiment that we did with parents was an online study. And then show you a little bit of the screenshots from the decision aid that we've developed based on health communication best practices and principles as one of the formative work and experimental work. And then give you a little bit of a clue what's the next step, which is the RCT. So the overarching center aims are here is really to evaluate how next gen sequencing is applied to the newborn screening can extend the current utility in newborn screening and to devise and evaluate a clinically oriented framework for analysis of NGS-NBS and then develop best practices for incorporating it into clinical care. And so the center is made of three projects as most centers are. And the first project is a sequencing project, and that project is led by Jonathan Berg. And they're developing a pipeline of whole exome sequencing data to study NGS-NBS and developing an analytic strategy for what we're calling, and I'm saying it's incidental findings here, but this isn't really incidental findings in the kind of context that we typically think of like a diagnostic context. So here parents are actually making decisions about a whole wealth of information, so nothing is truly incidental in this context. And then they're developing and evaluating novel bioinformatics approaches for NGS-NBS. The second project is the clinical implications project, and they're looking at whole exome sequencing as a diagnostic tool that could be used in newborn screening and determining the capacity and utility of whole exome sequencing to extend the range and current newborn screening technologies. So there's emerging evidence that newborn sequencing and whole exome sequencing, whole genome sequencing could be applied to newborn screening, which is typically not genomic sequencing at this point. So there's emerging evidence that it would be applied in this area, but the full range of its applicability is still to be determined. So in this clinical study, three groups of children will be involved in cohort study. So the first one is children who are zero to five, who've already been diagnosed with one of four of these genetically based health conditions. The second cohort is a cohort of children with rare diseases, and these were added because there's a concern that will reach the end that's required, so the range of potential conditions were expanded. And then the third cohort is a well-child group, and these are newborns from birth to three months old who will also be as part of the study with their parents. And this is the cohort where people sometimes when I say this go, what are you doing? Like how are you ever going to do this in an ethically responsible way where you're getting parents consent and approval to do genomic sequencing in their newborns? And that's one of the challenges that we're trying to address here. So the ELSI project is the third project, the Ethical Legal Social Implications Project. And in this project, we're trying to refine and describe in lay terms meaningful bins of genes and information that would enable parents to make informed decisions about what they want to learn, what results they want to get back about their child or newborn to develop and evaluate the effectiveness of a decision aid. And so much of what I'm going to be talking about today is preparatory to that decision aid and then apply the decision aid and it's mostly a randomized trial. I say controlled as a little bit of an overstatement here. And look at parents' decision to participate in the study and do when they, if they, do they feel more informed and more comfortable consenting to have this type of sequencing done for their child, their willingness to accept genetic sequencing information for their child, the choices of the kinds of information that they want to learn as part of the study and then the factors, the social and psychological factors associated with the parental choice and then are there any consequences of the return of these kinds of results for children and families that are part of the study? So the children consequences is something we can't really study in the, because it's a really short term follow up, but certainly for families, we're going to look at things like anxiety and those other kinds of short term consequences. So why is this important? So newborn screening is arguably is the most successful public health program that has ever existed. It identifies children with treatable conditions at birth and improves their health and survival. It's rapidly expanding and changing as new technologies are available and new genetic discoveries come on board. But the utility and as I said, it's starting to emerge that there is some utility to the application of next generation sequencing to newborn screening. But the real kind of nail in the coffin has still yet to be determined and there's a lot of resistance in the newborn screening community to use new technologies beyond what is currently being used. And there's ethical issues as well. Like what do we do with all of the information that is generated from sequencing if it was whole exome sequencing or whole genome sequencing? What do we do with that information when we're only focusing on a subset of conditions that are treatable? And so that's really at the core of what we're trying to understand in nexus. So we really don't know about the consequence of parents for learning this kind of information. And so that's one of the things that we are trying to understand. And when I say again, incidental information, that's kind of in quotes here. And so I think this is also an important issue because not only can you can get your child's genome sequence online now. You can go online, pay some money, probably a lot of money, but that cost is declining and you can get that information. And not only that, I was at a recent meeting at the Hastings Center last year and there was a representative of 23andMe and they said, well, we're not going to do newborn screening tomorrow, which made it think, well, maybe the day after tomorrow that might be going on. And so there's a real strong momentum for industry pushing, pushing, pushing related to newborn screening and other kinds of conditions that might be treatable. So to me, it's incumbent on us as social behavioral scientists to try to understand what are the consequences and what can we do to support families. So really, I'm focusing here on the translational challenges. So we know that the application of next generation sequencing in the pediatric population really requires a special examination of these ethical challenges and the clinical benefits. And really how do we provide support for families that are making these decisions now? And then how do we handle the disclosure of other information that is available to them once they learn these things like carrier status, genetic susceptibility to childhood onset conditions that are not treatable or genetic susceptibility to adult onset diseases that are treatable or not. So all of these are kinds of information that people could learn if they started to embark down this road. And so another issue that's come up a lot when I present this study is that how can you let parents make these choices when really you are undermining the child's autonomy and the child should be able to make these decisions when they get older. And I think these are real concerns. Biologists are very concerned about undermining autonomy and increasing potential risk. And of course, I'm speaking to the choir here when I say the amount of information is absolutely huge. How do we communicate that information in a way that at least in the baby steps when we're doing this, and I didn't mean that as a pun, maybe I'm getting like the psychological influence of this little baby here. But when we go that we can really reduce this information in a meaningful way that people can make informed decisions. This information is uncertain and conditional. And parents must make mutual joint decisions about this information. Our IRB has ruled in previous studies and in this study that if both parents are reasonably available, they must be part of the consenting process and they must be part of the joint decision making process about what types of information that they want to learn. Now if a parent is deployed to Iraq or has some other is not available then the remaining parent can make those decisions by themselves. So, you know, it depends. And some of the really formative qualitative data that we have is shedding light on the kinds of mutual and joint decisions that parents are comfortable making. So what are we doing to address these challenges? So in the first project, project one, Jonathan Berg is adapting some of his previous work developing an age-based metric system that will help us classify the really huge amount of genetic information into smaller chunks or bins. We're developing a decision aid to support parental decision making. And of course there's the hypothesis in the RCT. Will this help people? Will it make them feel more informed? Will they make choices that match their values and beliefs or not? That's still to be determined. And we're conducting formative work and user-centered work and experimental work to develop a decision aid that should, if all the evidence about decision aids is true, help them make more informed decisions about their choices to participate in the study and about the types of information they want to learn. And then of course we're going to be evaluating this decision aid in the RCT. So this is the age-based modified metric system. So on the y-axis is medical actionability. And on the x-axis is the onset of a genetic condition. And so in this quadrant here is next-generation newborn sequencing. And so, and then there's childhood onset, non-medically actionable, adult onset, medically actionable, and adult onset, non-medically actionable conditions. So what happens for a gene variant to get into one of these bins or buckets is a extensive review process, very systematic, is undertaken. And a binning committee meets and discusses the evidence, treatments that are available. When is the age of onset, which can be very murky because for many conditions some of the symptoms are not well known because we don't have a knowledge base about some of them because it's not caught early enough. And this group process then scores each of these genes in a systematic process and they're placed into one of these categories. So in the NC NEXIS study, parents will have available to them next-generation newborn screening conditions and others like them. So childhood onset conditions that have some medical treatment or actionability, they will have available to them depending on their study group. Childhood onset, non-medically actionable conditions if they would like to make that choice and they're in that particular study group. They will have adult onset medically actionable conditions as well as a choice point and then also carrier status. They will not be able to learn adult onset non-medically actionable conditions for their child and that was a decision that was made for a lot of different reasons, primarily ethical issues which is what benefit does that provide for the family at this very young point in time and maybe 20, 30, 40, 50 years from now that things will be actionable. So I want to tell you a little bit about the formative qualitative studies we did with couples. And we did this really in a very tested, we tested materials and asked open-ended questions about all aspects of the study and about messaging we were going to be using in the decision aid and we really wanted to understand how they communicated and made decisions about these kinds of screening and testing results for their child and newborn and what characteristics of the conditions parents find important to learn this information about their child. So the sample was 33 couples who were married or in a committed relationship and we had some nice balance across race. We also were interested, would there be differences based on pregnancy status? So is someone pregnant? Did they have a recent birth or do they already have some previous experience with genetic testing because of some childhood condition that they were grappling with with their child? And so really the end here is 66 because we're working with dyads here. But I can tell you off the bat there's no difference between pregnant, recent birth or childhood experience when we looked at the themes that were generated from the qualitative study. So we had some balance by race and gender. The average age of participants was 35. Most were more highly educated even distribution across income categories and most people were working for pay. People completed a brief questionnaire and then they engaged in a pretty in-depth 90-minute in-person interview with both couple members present and we asked about their general knowledge and opinions. We gave them definitions of the categories of information we would be using and asked for their feedback. We asked why are they or why not they would not choose this kind of information what they would want to know and then how they would work together to make these kinds of decisions should they choose to learn them. And then we also had them fill out decision worksheets separately. So we asked each of them to fill out categories of information of the kinds that I just told you about. So childhood actionable conditions and newborn screening things, childhood non-medically actionable conditions, adult actionable conditions and then carrier status. And we weren't going to provide the adult onset non-treatable or non-actionable conditions but we were curious what parents and how they would react to this information. So we did include these in the interviews although they won't have that information available to them in the next study. So they completed these things separately and then they shared their answers and part of what I'm going to be talking about today is related to the coding of these communications and then we have two papers that are coming from this that are going to be under review quite soon that look at kind of the communication dynamics among them and then some of the information I'm going to be presenting about their collaboration and their agreement about these categories of information. So we audio recorded and transcribed the interviews used in vivo, we developed code book, we refined, we coded, we refined. Anyone who does qualitative work knows that you code and refined a lot. We looked at reliability and did all the kinds of things you would do when you want to have a nice systematic qualitative study and then we used a framework analysis approach to look at themes that emerged through the course of these interviews. So the first thing I want to present is the decision results because to me this is a really interesting part of the study. So we had 33 couples, so most of them, yes, they would participate in a study like NCNXIS with the kinds of information that we provided in our informed decision consenting brochure which is part of a long set of decision aids that are part of this study, but not everyone did. They said they would need more information and many of the times that information they wanted was like logistics and nuts and bolts and very commonly like how is this information going to be stored as a private? Is it going to be an electronic medical record? Some of these people were very savvy, as you can imagine some parents who had some experience working with their child who has a genetic disorder. So most said, you know, they would want traditional newborn screening information in similar conditions, although there was some that they could not arrive at a mutual decision about that kind of information even though these were kinds of things that parents get back in a routine way. Would they want childhood onset conditions with no medical treatment? And so here's where we start to see a little bit of breakdown that they might say no, they don't know no mutual decision or there's just not enough information to make decisions about that. Adult onset conditions with medical treatment, here's again where it starts to break down just a little bit where they'd say don't know or they couldn't arrive at a mutual decision in the course of these interviews to be able to make a simulated decision in this context. Adult onset conditions with no treatment, here's again where there was a very big difference where there was a lot of couples who said we couldn't reach a mutual decision or it's no or we don't know, there was a lot more in decision in comparison to the other categories and carrier status as well was a place where it started to break down and there was, you know, do we need to know this information and why would we find it beneficial? And so one of the contributions of this, just even this table is when we've started to work in this area is many times the question is do you want this information, genomic sequencing information yes or no? And we find even in adults and sometimes in parental decision making, people uniformly say yes, I want to know this information. But when you start to break it down to categories of information based on actionability and age of onset, the conversation starts to change a bit and people start to refine their answers a little bit more. So that's one of the contributions that we're finding with our qualitative work. So we also looked at the themes about the decision to receive this kind of information and we looked at reasons for and reasons against and so these are overall the categories and we found that there's benefits to the child, there's benefits to the parents, you know, that they liked that the information would be confidential but there was some concern that they wanted the information to be used to help other children and there was practical implications about the study that they liked, that it's free and they get, you know, I get to get my child's genome sequenced. And then there's pros and our example of cons and risks. There was aspects of the sequencing that really disturbed them when we described it like are these tests accurate? How do we know what's really happening in terms of false positives or false negatives that, you know, there's their conditional kinds of information that are being presented to them. So we don't really know how something serious would be in the future. They're concerned about psychological harm, stress, breach of confidentiality. There was a variety of values of beliefs that parents discussed as either being in some cases against doing something like this. You know, why should we do this now unless we have a reason to know this information? So there was a variety of kinds of information that emerged the seams related to parental decision making. When we looked across the different categories of information we found a kind of, I would say a similar pattern in this qualitative data which is when you look at medically actionable childhood conditions, there's a few things this is great. We would do this. We can't really think of a huge number of risks to doing this. But as we get to these kinds of things that are like a non-medically actionable adult onset or carrier status, there was more uncertainty. There was more concern. And what we're doing with these, we call these values and beliefs. The verbatim values and beliefs that we learned in the course of these interviews are actually fed into the decision aid. And so there's a values sorting exercise that parents go through and they are able to sort their values and beliefs that come from this formative work and then input their own into the decision aid as a way to help their decision making process. So our conclusions from this work and some other themes that I'm not presenting right now is that couples really desire making collaborative decisions. It was mentioned more than once I would hate to make this decision by myself but in fact we will have single parents who want to participate in this study making this decision by themselves and so one hypothesis could be is do parents when they make these decisions outside the social context experience more distress or more anxiety as a result of that decision making. Agreement declines with greater uncertainty and there's many many different values or perceptions of risks and benefits that we can use in the context of a decision aid to help people clarify what might be important to them to help them make more informed decisions. So the other I want to call formative work all this was a pretty large formative study is that we launched the discrete choice experiment which was an online study to greater understand parental preferences in this context. And so discrete choice experiments are valuable methods for understanding information preferences and how people trade offers and benefits in a kind of comparative approach. It's an economic method that's used a lot by economists like how much are you willing to pay for this drug is a very common application of this method. It reveals how people value various attributes or characteristics or issues related to a problem relative to other characteristics and attributes. It's useful for studying highly correlated attributes which is probably most of the social world and we use this really to engage parents as a way to understand these characteristics of genetic conditions that might help drive their decision making. So we identified attributes of genetic health conditions that might be related to parental preferences based on a systematic review that was actually done around newborn sequencing and then we were going to use some of this information to inform the development of our decision aid and develop messaging around it and then just to build the evidence base on parental decision making for next generation sequencing because there's not a lot of information in this area and we specifically focused in our study on non-treatable childhood conditions because there's not a lot of experience for parents making those kinds of conditions so we defined these attributes and attribute levels for the course of this experiment and again part of this was based on the systematic review part was based on clinical input from the genetic counselors and geneticists that are part of this team and we did a lot of formative work in cognitive testing of these attributes and levels as well to make sure that people clearly understood what we were talking about so we looked at the chance that something would develop and define these attributes they would 90% chance 75, 50 and 10, the age of onset would it be less than one year all the way to adulthood the speed that the condition might get worse rapid, moderate, slow and stable and so on and so we manipulated all of these attributes and levels in our discrete choice experiment so we used a deoptimal design please don't ask me about the logistics of this because I was working with a really well-trained economist to do the design and some of the analysis and the wonderful Ryan Pequen who many of you know was also involved in this analysis I feel so fortunate to have him as part of our team and so people were randomly assigned to complete 8 choice questions which showed two different profiles of a genetic condition profile A or profile B and we randomly vary the attributes and characteristic levels across these different profiles and across the sample of the people who are in the study and then use the conditional loge at regression to estimate the implicit decision weights or preference weights that might be consistent with a pattern of choices again across the sample because if we were going to ask people to make all possible comparisons those people would still be doing the study I mean they'd still be and we'd still be slogging through so there's nice ways that we can look at balance across the design so this again was an online sample of over 1200 parents who had a child under age 5 we were very interested in race and gender differences and in this study design my understanding is if you want to look at subgroup differences you basically double your sample so we had a really big hefty sample anticipating that we would find race and gender differences but I can give you a peek under the hood which is we didn't find big race and gender differences and I think it could be a result of using the online panel versus a community based panel or cohort but that's something we still are interested in exploring we didn't find any differences based on some of the kinds of measures that we took for medical mistrust or numeracy and literacy and some other kinds of things you typically might expect to find so we're still exploring that as well so parents had to be 18 to 40 they had to have a child under 5 and speak English and again stratified by race and gender and so our sample turned out to be mostly married not surprising but also a large component of single parents as well there was representation across educational levels which I think is good and employment most people were employed so people entered the survey and then they're given information and characteristics about the attributes and levels and so again we did really extensive cognitive testing of this survey before we ever launched it and then when people come into the survey there is a period of training basically to help them understand what we're going to ask them to do and so they said okay what's a risk array so we told them what a risk array was and we explained it and you said what do we mean by a condition that has the age of onset we explained that and then we give them a practice question so by the time they get to the actual experiment they've been in the survey for quite a while and that training helps them make those comparisons in a pretty and I think it worked very well in our case so again we focus on non-medically actionable genetic conditions which we define for them and we looked at those seven attributes with the varying levels that I presented in a previous slide and then they were presented with the eight tradeoff questions which said which genetic test report would you be most important would be most important to you to learn if you had this type of information to choose from and then they answered these psychosocial and demographic questions so here's an example of one of the profiles that they might have as part of this study and they answered eight of these and so we varied again the different attributes along this along this column here and then the age of onset you could see five or five to twelve in this example and they were given you know here's a chance that it would develop and so these types of things they made eight of these choices as the part of the experiment and then the dependent variable that I'm going to talk about today is which would be more important for you to know profile A or profile B and so then we looked at these comparisons across the sample to figure out what is the most important thing that parents want to know based on the kinds of profiles that they were viewing this slide came out this is one of the slides I had a little slide crisis this morning that came out all jumbled like a scrambled egg but thankfully it looks good here so this is just overall the condition attributes just overall without regard to the levels of those attributes and we found that the likelihood that something would occur was by far the most important thing that parents wanted to know if they had these profiles to choose from and so that clearly was most important for their decision making quality of life was not important of all on the other end of the spectrum the level of physical and mental disabilities was also important as was the age of onset and the lifespan the impact on lifespan of this genetic condition so it was interesting that quality of life didn't work here at first we thought maybe there's like some kind of subgroup differences we looked at all kinds of things I just don't think that attribute might have worked but it wasn't important in this study then we looked not just at overall but we looked at the levels of the attributes and I have to say this is the most orderly data set I have ever emerged from and so what we found is that there were significant differences between each of the levels that we looked at except in a couple of cases so when the age of onset was less than a year or one to five there was no significant differences there or when it was the rate of progression was stable or slow I'm sorry I'm not very good with this little pointer and then quality of life available treatments for that was not significant but in all other cases there was statistically significant differences and these preference rates across the different levels that we examined we also looked and we as a kind of a manipulation check we looked at distress levels so we asked people to rate the distress levels of these varying levels and variables that we looked at and so shortened life span was clearly on average there we go much more distressing than a normal life span and so on and so that provided to us kind of a nice check that actually people were perceiving these very attributes and levels that we had defined kind of a priori in a way that we thought was important for their decision making so clearly people want to learn more distressing information that is more likely to occur which was surprising to a group of genetic counselors that I presented this information at last fall they predicted the exact opposite so it was really an interesting discussion we had around that so differences by gender and race we examined these because we were very interested in would we find differences by gender and race based on differences you see in the literature and it really didn't change the main pattern of results that we were finding there was a magnitude of preferences as outlined here but really the main pattern of results was the same across all of these subgroups so compared to mothers fathers cared less about learning information that had a smaller chance of developing or was less likely to decrease lifespan or was characterized with severe physical disability and compared to white participants African American participants cared more about learning information that had a smaller chance of developing was unlikely to affect a child's lifespan or could result in a severe mental disability and I present these but really the main difference is our story but here these are small little differences but the magnitude of results is all the same so our conclusion really is that parents when it's something about a non-treatable childhood disorder parents want to learn that information even if it's distressing and it's highly likely to occur and we believe that is the case based on our qualitative data which shows that parents are very interested in things that have personal utility even if there is no clinical or medical action ability related to those conditions so here we go the conclusions I've already I think gone over most of those with you so all this information is preparatory to the development of a decision aid and so the problem is and this is a problem that's facing most genomic sequencing studies is that when we consent people into studies typical consent forms are cumbersome they're confusing they're long they complain terminology and jargon they're at a very high literacy level they require lengthy individual interactions to complete and they really don't provide knowledge or information about the study and this isn't just I mean this is pervasive in all areas of science not just here but it's especially acute with genomic sequencing studies because it's not an an idea or a methodology that's really accessible to people but we know it will be as precision medicine rolls out as genomic sequencing information becomes more prevalent and available in the society so traditional consent models are going to be problematic when next generation sequencing is more pervasive in research and in practice in population studies and so part of what we're trying to do in the NC NEXAS study and as part of the insight consortium is to look at can we use a decision aid to help ease the burden in the consent process so our approach is the decision aid so it explains genomic information at as the lay level as possible and I have to say as lay level as possible because there's a lot of tension between what I would like to be a lay level and the precision that clinicians really want because they see you know they really believe that people need to know really detailed information to make an informed decision and so this is a ongoing debate and it's one thing I think we're going to find out in the study is what kind of information do people really need to know to make informed decision so we applied communication science strategies and best principles for clear communication, plain language and how literacy as much as we possibly could we used text, graphics and audio to convey challenging concepts and used a little bit of animation of how things move on to the slides that people see as part of the decision aid and we included a value clarification exercise because many studies show that when people engage in a process of values clarification they come to a decision sometimes where they feel comfortable with it and experience less distress now they may not consent to the study and in fact in one of our previous studies on newborn screening related to fragile X we had lower consent rates to screening for fragile X once parents completed a decision aid that was related more to like an informed decision making approach rather than just a typical consenting and we were okay with that so we think it meets participants needs better because it facilitates this informed decision it could make enrollment easier and it just reduces burden all the way around and it will reduce the burden for study staff as well which tend to be in many of these studies genetic counselors who are on the front line consenting people into studies so decision support tools that people haven't used them tend to be evidence based educational approaches that provide a very personalized focus for people to weigh options and outcomes when they're making decisions and so it's typically applied in context when there's no best course right so for some medical treatments and decision aids are typically applied when there's no best medical treatment like what should I do should I get this treatment or that treatment or no treatment and so decision aids are used in a individual or shared decision making context to help people make those decisions and so we thought these could be this framework could be applied in this context because whether people participate in the study really is an individual choice or a dyadic choice you know there's really no really hardcore benefit for them to participate in except maybe they just want to so there's an evidence based literature that supports their use and it really engages people in the process of decision making people have to be more actively engaged rather than just like oh my gosh here's another page of medical terminology where do I sign boom boom boom not even knowing what they've signed and so it really focuses on kind of this deliberative healthcare choice that people can make and in the context of next they would affirm their decisions in the decision aid but then they would go through a process and meet with a genetic counselor before they're actually consented completely into the study so how do we gather parental input for the decision aid so we use these best practices we use the international IPDAS standards which I have to say was something we proposed initially and became actually very challenging to apply in our case because those IPDAS standards were developed with very specific one gene or one kind of treatment in mind we're talking about massive categories of information of uncertain information and so it was very challenging to apply those standards we followed an iterative development process of testing and refining we applied these plain language principles as I mentioned we did user testing with the decision aid once there were we had it developed to test areas of confusion related to content and navigation and we allow people to go at their own pace they can repeat and review information as needed so it's not like there's a time to kind of you know you have 10 minutes to do this and they do it at home or they can do it in the clinic on laptops or tablets that we provide so the decision aid actually contains four sections and this is a screenshot of one of the sections there's traditional newborn screening conditions and similar other conditions that are treatable in childhood then there's a and you know what the NC NEXUS studies about is also included in that section there's non-medically actionable childhood conditions that are described and people provided education about that there's medically actionable adult onset conditions and carrier status so each of these sections is parallel to each other there's some preparatory knowledge and education there's risk rates that are presented people kind of we ask them which way they're leaning to get at their intentions there's values clarification and then they're asked to basically make a decision do you want to learn this type of information about your child and there's also in the very first step in the decision aid there's a tutorial and so this is a screenshot of the tutorial about how you would drag and drop things we talk in this specific examples about newborn screening and genomic sequencing what the study is about as I said reasons for and against and the decision in next steps and so we describe in pretty detailed ways what would happen if people participate in the study when they're in this first step of the decision aid so what we think makes this tool unique is that it's iterative and can be applied applied across multiple platforms so laptops tablets, desktops on some phones but not iPhones but iPads it got to be a little bit cumbersome in terms of this testing but apparently Apple has not agreed to use something on iPhones which makes it very problematic for the use of that it's theory driven in the extent that it accounts for risk communication and other kinds of communication from health communication research it's input from both users and clinicians playing language as much as possible it has the iterative value sorting exercise that's shown here on this screenshot and it has had an iterative user centered design and testing process and I think this is really important because this is the translational part if I was really doing a theory just totally theory driven hypothesis driven study I probably or maybe would have used a different approach but we find it again and again and again with all the digital tools that we're developing that we absolutely have to have people's input on the tool and it makes it better it makes it more accessible to them it's just all around better but I have to say it's challenging sometimes to convince people that this is absolutely the way to go so here are screenshots about condition category description and a risk array so this is medically actionable childhood condition and so what happens is all of these little bars are not on the screen at once so there's narration that happens and the narration almost exactly maps to things that are shown on the screen and these bars come down just one at a time in sync with the narration and so here's an example of a risk array related to newborn screening and it this is the progress bar as it goes around and they can hit going to the next screen so they have to move forward themselves they can't stop and it's they have to to push it forward which is another way to engage people in the process here's difficult this actually when I saw it is actually not the final one but we've tried to use graphics to communicate genomic sequencing information that might be challenging because of the platform that we use we couldn't use animation which of course would be way more useful in this particular context we also use hover overs so anything that has a blue they can click on and then they get a more what I would call a more lay language definition if they would like it here is again as an example where we talk about more about the logistics of the study and then in addition to the value sorting there's questions that we prompt them to consider and then they can put in yes and no yes and no yes and no depending on what they want to provide as a way to either spark discussion between the couple or to have someone who's an individual fillings out think about some of these questions so are you prepared to learn this information that might be medically actionable information about childhood conditions yes or no with the idea and then these answers are populated on the next screen and fed back to them as a way to prompt their awareness engagement and decision making here is a screenshot that's actually not the best screenshot but it shows a summary of the values clarification exercises so reasons for and a reasons against and what happens is they sort through values and then a summary is shown to them as a way to summarize for them and to think about what is important to them is to either get the sequencing or to choose certain kinds of information so this is how they record their decisions and as I said prior to this prior to the values sorting exercise we have a same slider scale so this arrow measures will can slide across this bar and it actually has numbers that show when it slides and so even those either really categorical decisions that people are making we did this as a way to see if we could somehow capture finer grained intentions and decisions like yes I definitely want to do that I'm going to put it all the way over here on this side or yeah I kind of want it over here and so we're going to look at kind of pre-post values clarification as a way to see do people's intentions change with their decision so the next step is the NC NEXUS randomized trial where we'll have 400 parents who participate and use the decision aid as well as work with a genetic counselor to make these decisions so all consenting parents will be able to make decisions about newborn screening conditions that others that are treatable in childhood and then about two thirds we did an uneven randomization because this is a pilot study and will be randomly assigned to additional decision categories that I've talked about so carrier status non medically actionable and adult actionable and they can choose one none or all of those so they don't have to choose any like permutation of those categories of decisions to be a part of the study we're looking at decision confidence conflict distress and other outcomes that are monitored over very short follow-up period and again we're hoping that we can get more funding to look at people longer over time and so this will really allow us to evaluate if a decision aid helps support informed decision making in this context I mean it could be a useful tool not only to support parents but genetic counselors and others who are going to be involved in these large population-based studies as we move forward so I'm going to end there and thank you for your attention I'm happy to answer any questions that people have yes no so the question was will there be people who will use the decision aid and not have access to a genetic counselor which I think is a good decision that would be in some ways the full design right to look at that comparison and we thought for ethical reasons we don't know the effectiveness of the decision aid at this point we don't know the needs of parents so that we thought we wouldn't do that that could be a really interesting next step is now if we have some we know this helps what happens to people consent without the access to that interpersonal communication do they have that same level of confidence maybe or informed choice making that we may learn about but that's a really really good question but we're not in this study for mostly ethical reasons yes and so it's a pretty complicated design and so really it's basically four months overall but when you there's a really long lag for some of the return of these results for the processing of the information so it's actually probably a little bit longer this is the goal I don't know if we'll be able to achieve that goal I mean there's a lot of information this to me is a lot of people to be processing that kind of information on so we will see how long that lag tends to be but that's the standard that we're trying to get at and it's mostly we're backing up against now we're in year three so we're also backing up against the end of the five year grant period yes exactly right so that's why we decided that everyone should be able to have access to and I should make it clear everyone who has a newborn in this study will get traditional newborn screening and then there's lots of methodological issues that they're interested in examining when they compare traditional newborn screening with the next generation sequencing that will be done that's why we wanted everyone to have access to that decision-making category almost like it was a traditional newborn screening kind of thing everyone has access to making decisions about medically actionable conditions so that's why we wanted to be able to have access to the child if I understand your question correctly now people could get to the point where they opt out so at any point in this study people can drop out they can even do the sequencing which will be done by a swab in the baby's mouth or the child's mouth I think so that's both cases with the swab but they can even have the sequencing done and choose not to learn it and opt out of the study but we wanted it to be kind of like newborn screening might be which is everyone has that option but we couldn't really force them to have that I doubt the FDA would allow that and I'm sure our IRB would not allow that either but you're right it's it's really interesting the kinds of debates that we have in the Insight Consortium about the public health significance of the application of this technology to newborn screening because it brings up so many issues about opting in and opting out and so we're doing some other work that's separate from this looking at elective panels of genomic sequencing but elective panels that would supplement newborn screening information and some of you I all know Holly PA and Holly and I are working with Don Bailey on that project which is basically to have a Tier 2 screening system that we're hoping to implement in North Carolina other questions great thank you so much I appreciate it and happy to talk with anyone individually thanks so much