 So, I am very excited to be here today, thanks for flying me in, and very excited to be talking about this project because this project I started when I was here at the end of the week started, when I was here at the NIH, and are working to write up right now, so this is a really good time to be presenting to everyone, and I'm very excited to hear thoughts and feedback that you have about it. So basically the crux of this project is that I'm interested in the difference between not knowing about something and not being all that interested in it, and the ways that that could be clinically useful to us to think about, and ways that we might measure it, and ways that we might develop interventions that are tailored to these different states that people may be in. And basically this interest started when I was at Georgetown where I was a part-time counselor, part-time postdoc, and working on a study, analyzing data from a study that was a decision aid, where the first paper came out and everything was pretty clean and nice about it. It was an effective decision aid, particularly for people who hadn't made decisions about treatment and management following BRCA 1 and 2 testing. It helped them reach a decision, it lowered their decisional conflict to increase their satisfaction. So that was great, and I was actually working to design the next decision aid that we would draw from this data, but analyzing some more data from this previous study and looking at the distress outcomes, and what we found was that the decision aid actually made people a little stressed, which at first glance doesn't sound like a very good thing about a decision aid, and what you can see, so BSI, just general distress, really there's no difference between the people who got the decision aid and didn't. But when you looked at genetic testing specific distress, which is what we were calling it at the time, I'll get back to that in a minute, in cancer specific distress, which is just how much they're worried about cancer, at the one month time period, the people who had the decision aid were a little bit more stressed than their counterparts who didn't get it, and this resolved by 12 months, so there was no issue. So we spent a lot of time thinking about, well, what is this? And I came to suspect that maybe it really wasn't distress after all. Maybe the fact that we were asking them to spend time thinking about cancer, thinking about their genetic test result, raised their thinking about it a little bit, and in fact a lot of the questions on these measures were like, how often do you think about it? How often do you think about cancer? How often, you know, these sorts of things. And maybe it wasn't so much distress, as much as just active processing of what was going on, and maybe in this case that's part of what made the decision aid useful. So this is conjecture, but maybe. So when I had the opportunity to come to the NIH, I was really interested in following up on this, and also very excited to be able to do it in a genome sequencing setting and worked with ClinSeq, so it follows on nicely from Katie's talk. So, oh yeah, there's the distress. So what I called this concept that I was looking at was engagement, and it's really sort of a cognitive engagement, and the way that I think about it is really how much you think and care about a particular thing. And I almost think about it sort of like real estate, like cognitive real estate, that there are a certain number of waking hours in a day, and you spend a certain amount of time in those waking hours thinking about specific things, and that the amount of time you spend thinking about those specific things may predict how you choose to act related to those things. So it's distinct from knowledge, because you can know about something, but not spend a lot of time thinking about it. There's a lot in the literature about the concept of arousal, which I think is sort of a related, which is the idea that you come to be interested in something or something comes to the forefront. But what I don't like about arousal is that it, to me, implies sort of a specific event that happens rather than a continued process. And I also think you see a lot in the literature things that are activities stemming from engagement, and this is really popular in the marketing literature and in IT, which is how much do people interact with a website, how many clicks do they have, how long do they spend on it, and a lot of times that gets called engagement. But what I would argue is that that's actually an outcome of engagement, that you're doing that because of a cognitive process that's coming in. So you could measure those things, but they're really proxies sort of for the cognitive process. So there's a model that I think really breaks this down nicely, and it's the precaution-adoption process model, which is a stage model, which means that it describes how people pass through various stages along the course, too, in this case, a decision. And there's the unaware stage where you just have never even heard of something, so obviously you're not to the point of thinking about it. Unengaged, which is just not, you know about it, but you're not spending a lot of time thinking about it, to active decision making, which then predicts action or inaction. And I think one of the really important things about this model is it breaks apart people who don't know about things, people who don't care about things, and people who actively decide not to do something. And I think in the clinic we conflate those a lot, and really often maybe not genetic counselors, but clinicians will assume it's lack of knowledge. And if you just talk to the patient longer, eventually they'll get it. So it's really nice model also because this whole decision part in the middle here, you could basically fit any health decision-making model into it and combine this model with other models. So I like that flexibility about it a lot. And I think it's really, well, it's been used, I'll say, also in like home radon testing. Why do people do radon testing or not? Have you heard of radon testing? Do you care about radon testing? Do you think radon testing is a bad idea? And those are three very different reasons for not doing radon testing. I think it fits really well in the genomics environment along lines of many of the themes we've talked about today that people may be getting results for which they don't have much saliency. They may be able to understand that there's a variant, but they just don't have the background or the reason to be interested in it. And that may predict the way that they subsequently respond or not to that result. And that's very different from not understanding and actively deciding not to do something. So what we did was bring this idea into ClinSeq and think about how can we really do some concept exploration of engagement within the ClinSeq population with the studies that we're doing there. And Katie described ClinSeq to you, so I won't go through that. But the first analyses of engagement that we're doing are a part of the ClinSeq baseline survey. Which also is something that was started about the time both Katie and I came to the NIH and a whole group of us said we all want to get measures from the population before people get results. So we can get a sense of what the impact of results may be in a number of different settings, knowing that they're going to get all kinds of results. So let's make a huge survey and then let's have a lot of really tense discussions about what actually gets to stay on the survey and make it bearable for participants. And our definition of bearable was about 45 minutes, which is a pretty long survey. And so this was then offered to all ClinSeq participants who had not yet received results and were more than a month out from enrollment. They were given the option of doing it online or in paper. And Katie and her group did an amazing job of getting the end up to, I don't have that slide yet, but I'll get to that in, 551. So they got a lot of people to fill out the survey for a 45 minute survey is pretty good, but they contacted them multiple times to get this in. So the engagement part of the baseline survey was to look at the constructs of knowledge, engagement, and communication. And specifically at the baseline, they haven't gotten results yet. So we're talking about communication about being in ClinSeq and to see how these things relate. Largely with the goal of exploring the concept of engagement. Looking at some of these relationships predicted in the model, the precaution adoption model between knowledge and communication. And then also looking at some other things that we would expect might be related to engagement as we see it. So as I said, Katie largely did an amazing job getting people to fill out this survey and there were 551 survey responses for analysis. They were pretty comparable to the general ClinSeq population, although slightly more likely to have a high level of education and slightly more likely to be white or not Hispanic or Latino. But these are pretty small differences, less than 10%. Okay, so the question I asked to get an engagement in the baseline was just how often do you think about your participation in ClinSeq? And there's a normal distribution ranging from never, which probably isn't true because they filled out the survey. But anyway, two yearly, all the way down to one person who thinks about ClinSeq daily. And yearly makes sense because that's about the frequency with which the study team contacts participants. So that's sort of a logical answer for them. So then we looked at these constructs from the model and they were related. So knowledge was related to engagement and engagement was related to frequency of communication. So we asked, how often do you talk to your doctor about ClinSeq? How often do you talk to your family about ClinSeq? How often do you talk to your friends about ClinSeq? And those were related as they would be predicted by the model. And then we also, so we looked at whether engagement was related to who they tell and it wasn't really. But just in the process of looking at who they tell, we saw that communication just about being in a genome sequencing study, which you imagine might have some relevance for family members, was fairly high with almost everybody telling their spouse they were enrolled. Three quarters of people told their children it gets smaller as you get down to parents and siblings and there was a statistically significant difference between whether they told their sister versus whether they told their brother with more people telling their sisters that they were in a sequencing study than their brothers. And three quarters of them told their physicians. Other correlates of engagement we looked at and I think some of this is consistent with some of what Katie was saying about how people viewed their results that they got, but just how people would think about a study like this in general is that people who are in better health were more likely to report thinking about ClinSeq more often. People who saw the doctor more often, so perhaps they have more comorbidities, thought about ClinSeq less often. People who are health information seekers thought about ClinSeq more often. People who've had genetic testing in the past thought about ClinSeq more often. People had a graduate degree thought about ClinSeq more often and people who had enrolled more recently in ClinSeq thought about ClinSeq more often. So the longer it had been since enrollment they were less likely to think about it. So to summarize, most people do think about their participation in ClinSeq but to varying degrees as predicted by the model. Engagement is related to knowledge and communication and healthier information seekers with history of graduate degrees and genetic testing are more likely to be engaged. So a couple limitations, one you would imagine people who are more engaged might be more likely to fill out a survey about ClinSeq. In this analysis we just used that single item measure of engagement and it is cross-sectional data so we can't really infer anything about directional relationships beyond the theoretical background. So future directions to address some of those limitations is that we've put together several items, four items scale with four potential other items we might include that could potentially expand the validity of this measure that all try to get at again this cognitive real estate idea of how often people are thinking about it. And some of these others that I've listed below I think fall into the category of proxies of engagement or outcomes of engagement but I'll be interested to see how they would perform. I think that there are a few testable hypotheses that could play out in longitudinal studies and one that we've spent the most time talking about would be looking at ClinSeq as returning carrier results to participants and looking at one month at their engagement to see if that's predictive long-term at six months or 12 months of whether or not they communicated to family members and we're still kind of guessing a little at the time frame on which this might happen but going back to the first study I mentioned from Georgetown that was about the time frame we saw so we have some reason to think that might work. And then where I think we can make an even bigger impact to this is looking at actionable risk associated variants and if engagement predicts uptake of behaviors in response to those sort of risk management behaviors. So if you returned a cancer risk variant to somebody and they were more engaged would they be more likely to tell their physician would they be more likely to get screening would they be more likely to consider other management decisions or prevention if it was available for whatever that variant was. I think that's that's the hypothesis I'd really like to test someday and then the long term wildly speculative hypothesis is that if we could design interventions aimed at the disengaged or you know stage tailored interventions that divide people by not knowing not being interested and actively deciding against and then deliver them accordingly perhaps we could achieve positive health outcomes in a more effective way. So one of these I mentioned already but I had to get a few tangential plugs in here just really quick. I apologize to Barb for using my time for this but the practice guidelines committee is in great need of genetic counselors with research training. This is a great group of people who have really good research training and they're going to be looking for members in the next month or so. So they really need you guys. The research SIG 2 as the research SIG chair right now we're hoping to increase activity invisibility engagement potentially within the research SIGs. So you'll have me know if you're interested. So thanks to my co authors on the project as well as the entire ClinSeq social science team and the ClinSeq participants.