CII - Harlan Krumholz, MD, SM

August 28, 2023

Information

Harlan Krumholz, MD, SM, Harold H. Hines, Jr. Professor of Medicine, Cardiology,

Director, Center for Outcomes Research and Evaluation (CORE)

“Digital, Decentralized and Democratized: Next Generation Research at Scale”

ID10661

To CiteDCA Citation Guide

00:00So our next speaker is
00:01Doctor Harlan Kremoltz.
00:03He's a cardiologist here and Harold
00:06H Hines junior professor of Medicine.
00:09Doctor Kremoltz went to IS from Ohio.
00:12He went to Yale for undergrad
00:15HMS or Harvard Med School,
00:18and then went on to the School of
00:20Public Health at Harvard as well.
00:22He's a distinguished scientist of
00:24the American Heart Association and
00:26a member of the National Academy of
00:28Medicine and has served as a member of
00:30the advisory committee for for the NIH.
00:34So I'll give you the floor. Yeah.
00:45Thank you. And Nicole just told
00:46me I don't have to end it too.
00:48So I appreciate that.
00:50But maybe that's not merciful
00:52for the audience. I'm not sure.
00:55So I'm so happy to be here today.
00:56I'm so happy to be here at the inaugural
00:58of the Center for Infection and Immunity.
01:00I'm so privileged to work with Akiko Iwasaki,
01:03who is just an extraordinary individual and
01:06both as a scientist and as a collaborator.
01:09And I also want to call
01:11out Vernali Bhattachary,
01:12who has been just a key individual in
01:15anything that I've been able to do with
01:18the Iwasaki lab and and for whom I'm.
01:20I'm really grateful.
01:21And she embodies really everything
01:22that's good about science and
01:24collaboration and generosity.
01:26And it's really my privilege and
01:27honor to be able to work with her.
01:29And thank you,
01:30Bernali for for being in that position.
01:33I want to thank you for the swag.
01:34I think that's so cool.
01:35I have.
01:36This thing isn't like the Taylor
01:37Swift concert where you get to
01:39wear the the bracelets.
01:40So thank you.
01:40I'm going to wear that the rest of
01:42the day and to be very proud of that.
01:44The.
01:45So Bernali and I are kind of doing
01:47a little dance around the talk.
01:49So she's going to talk a little
01:50bit about some of the studies
01:52in the mechanics of the studies
01:53that we're doing together.
01:54And I changed my title because to
01:56yield to a little bit of what she's
01:58going to talk about it and I'm just
02:00going to talk about a collaboration on
02:02the way to impact and to share some
02:04thoughts with you a little bit about.
02:06So I also Lisa said this,
02:08I feel a little bit like a
02:10stranger in a strange land.
02:11This isn't a natural thing for me.
02:13I'm a cardiologist and also
02:15I do very applied research,
02:16very different than the typical things
02:18that are going on in the tack building.
02:20I have a a group where Center for
02:23Outcomes research and evaluation are.
02:26You may not be as familiar
02:27with what outcomes research is.
02:29So I just want to frame for a
02:31little bit what what this is.
02:32So outcomes research evaluates and
02:36optimize health outcomes for individual
02:38patients and for populations.
02:39It's not so much about a particular
02:42technique as it is a driven approach
02:43to say that we want our work to
02:45be consequential and to have
02:47tangible impact on people's lives.
02:49It emphasizes what's important to
02:51patients and people, the end result.
02:54So we're interested in surrogates
02:56and markers on the way to actually,
02:58what does it matter for each individual?
03:00We don't want to say we've
03:02improved your lab test.
03:02We don't want to say we've
03:04improved your profile.
03:04We want to know that you feel better.
03:06We want to know that your life is better.
03:07We don't want to know whether
03:09people are living longer and better.
03:10It's a it's a orientation we
03:12seek to understand mechanisms,
03:14identify targets, test strategies.
03:16That may sound familiar to you,
03:18but but it's on the way to actually
03:21knowing that people are better.
03:23It involves discovery,
03:24accountability and improvement,
03:25discovery of new approaches,
03:27accountability for what we're actually doing.
03:31You know, what is it that we can say
03:32the current level of performances,
03:33What is it that's actually happening?
03:36An improvement in an
03:37aspiration to do ever better.
03:39And we're very results oriented.
03:41I mean, in the end of the day,
03:43I want to know like what actually happened.
03:45And if I'm going to try to help Lisa,
03:47I don't want to tell her about
03:49papers published or grants obtained.
03:50I want to talk to her about we've
03:53actually in enhanced her ability to
03:55help people and not just her ability,
03:58but that people actually are better,
04:00people are better as a result.
04:03So some of the key questions that we address,
04:05we may ask how do we improve
04:07healthcare performance?
04:07You know,
04:08there's this large gap between what we
04:09actually know today and what's actually
04:11being delivered on the front lines and
04:13what's being achieved for patients.
04:14While Covid's a different thing
04:15because actually we know nothing
04:17about how to make people better.
04:18But there are lots of areas of
04:20medicine where we know things but
04:22but it's not being translated.
04:23We ask how can we identify target
04:25and address factors that can be
04:27transformative for health and
04:29and we're very much interested
04:30in multimodal data coming in.
04:32So biological data has great utility
04:34but but social context can actually
04:36have more powerful influences.
04:38You may understand them as
04:39epigenetic influences,
04:40but we understand them as,
04:42you know,
04:43where people live and the exposures
04:44that they have and the lives and
04:47the stressors that they experience
04:48can have profound impacts on on
04:51modifying the disease process that
04:52they're that they're encountering
04:54or facing or recovering from.
04:56How do we give more voice to patients?
04:58How do we put them in a more
04:59powerful position?
05:00How do we beat the information
05:01asymmetry that has been historic and
05:03hierarchical paternalistic profession
05:05that where people have tended to
05:07walk in the room and just tell
05:09people what to do in a modern era,
05:12we're going to break down those barriers
05:14and make sure that even people with
05:15various levels of health literacy
05:16can have a basic understanding of
05:18what the tradeoffs are between the
05:19options that are available to them.
05:21Based on how we stream forward.
05:22How do we best promote HealthEquity?
05:24We're doing a horrific job
05:25of this in this country.
05:26We have an unjust system.
05:28We have people who are disadvantaged
05:29merely by the color of their skin or
05:31the circumstances of their birth.
05:33And they're they're in a category
05:35where they are are are largely
05:37going to live shorter lives and
05:39and at with higher comorbidity
05:42and worse function as they age
05:43because of the station that they
05:45are in their lives having nothing
05:46to do with intrinsic biology and
05:48everything to do with the social
05:50context of their lives.
05:52And it's something that demands our
05:54attention and how can we improve
05:56the knowledge generation pipelines.
05:57Our research is slow and sluggish.
05:59It's expensive and it's often not
06:01responsive to the needs of what
06:03people are are asking us for.
06:05And so I'm all for the very basic science
06:07of unlocking secrets of the universe,
06:09helping us understand some of the basic
06:12beautiful ways in which biology unfolds.
06:14But actually I'm more focused on how
06:16do we make sure that that's in the
06:18service of actually promoting better,
06:20better life, better better humanity.
06:23You know what are we doing
06:25to actually improve things.
06:26So and I want to be able to do
06:28this quickly test things faster.
06:30You look at long COVID,
06:31the the knowledge gaps are so profound,
06:33part of what Bernalli will mention to us,
06:35we're trying to build platforms
06:36so that we can reload.
06:37Not we're not building bespoke
06:39research projects, not huntergatherer.
06:40How do we go out, have an idea,
06:43build something up only to break
06:44it down when we're done?
06:46How can we do industrial farming
06:47where we build it,
06:48we've got the platforms and we just
06:50can continue to ask new questions
06:52rapidly and efficiently and be
06:54able to keep cycling through.
06:56And people who are working with
06:57us know that they're going to
06:59be honored and respected.
07:00We're going to listen.
07:01We're going to try to make the
07:02studies in a way that delights them,
07:04that they feel that they would do
07:06it again because of the way that
07:08we interact with them and be able
07:10to create an entirely new approach
07:12from the very hierarchical where
07:13people are subjects.
07:14We don't use the word subjects anymore.
07:16We we think that that's entirely
07:17different construct king and a
07:19queen and a subjects and they just
07:20follow directions and what they
07:21don't do what you tell them they're
07:23lost the following I got and chase
07:24them and bring them back.
07:25They're in studies are supposed to
07:27help them and they they leave in
07:29droves because they're so alienated
07:30by the way in which we do research
07:33we we do it in ways and makes it
07:35difficult for them to to participate
07:36and and doesn't ennoble them in any
07:39way and and I participate in research
07:41that I always say I'm ashamed to say
07:43where I finished the studies and I
07:45didn't tell people what we found.
07:47I mean that's the ultimate disrespect.
07:49People are in these studies and they
07:51didn't we just told them it's over.
07:53But we we didn't even do them.
07:54The honor of saying by the way here's
07:56what your efforts helped us to learn.
07:59They they we were exploiting them.
08:01They were just working for us.
08:02They we were they we weren't bringing
08:05them in as partners and refused to
08:07participate in that kind of research anymore.
08:09And I was socialized about that.
08:12I mean that's how we were taught.
08:14So
08:16and I'm just saying ultimately you know
08:18it's about the people and and we have
08:20to approach our research with humility.
08:22You know, it's not like we're the
08:23smartest people in the room and
08:25and everyone just should listen to
08:26us until we say it's a matter of
08:28us having that humility about our
08:29ideas and wanting to test them,
08:31figuring out what we can do to help
08:32and knowing if we're successful.
08:34It's because we've worked synergistically
08:35and in a complementary way with those
08:38people who are experts in their own
08:40disease because they live it every day.
08:42And so we've got to be able to
08:45understand how we create that synergism.
08:47So you know it.
08:50I think a very fortuitous thing occurred
08:53when I met Akiko and it occurred to
08:55me that that maybe there was some
08:56opportunities for us to come together.
08:58And she's been such a gracious and
09:00generous collaborator who immediately
09:02embraced all of these ideas,
09:03moving from mouse models now to
09:05working with people and immediately
09:07wanting to to work in the way with
09:10people that would honor and respect.
09:11Of course that's part of her,
09:13the way she operates in every different
09:15direction of of her life and in her science.
09:17And and so we thought you know that
09:19this would be good to bring the
09:20groups together and work together.
09:22And in our approaches that were one team,
09:25even though we have different
09:26areas of expertise.
09:27Our goal is to make a difference
09:29the how matters the how means it.
09:32It's not like just get the results
09:34with and doesn't matter how people
09:37feel when they're participating
09:38or or or you know how we,
09:40you know what happens as a result
09:42collateral damage because you
09:44know we're just trying to pound on
09:45people to get get all the work done.
09:47It has to be in a way that we
09:49are even among the researchers,
09:51recognizing that everyone deserves
09:54to be respected,
09:55that people are working hard and how
09:57do we create the conditions where
09:59people can excel to the greatest
10:01extent while at the same time,
10:02you know,
10:03we need to make progress together so
10:05that the how we approach this is important.
10:08It's a sensibility within Akiko's lab.
10:10I've seen that from the very beginning
10:12bring together the best lab and applied
10:14science and and see through its see
10:16it through the application benefit.
10:18Often times the very strongest
10:20scientific groups aren't necessarily
10:21working with strongest clinical groups
10:23and the clinical groups are just
10:25like trying to get tests in the in
10:26the basic science groups you're just
10:28trying to get subjects participants.
10:30But you know when we're trying to do
10:32this in a way that's a true partnership.
10:35We're also trying to partner
10:37this this aspect of it.
10:39If I can figure out how to go
10:41forward that's a metaphor problem.
10:45So the, so we've launched 2 studies.
10:48Again, Bernali will go into more detail.
10:51I'm just going to mention one is in
10:52a digital observational study.
10:54So these are digital and decentralized
10:56and we call them democratized.
10:57Democratized in this context
10:59really means full access.
11:01We're we're trying to let people select
11:02themselves to be part of the studies.
11:04You may say,
11:05well doesn't bring very high selection.
11:06You know what happens normally
11:07in clinical practice?
11:08The doctor walks in and decides,
11:11Gee, are you.
11:11I wonder if you'd be a good person
11:12for statement and all the people
11:13who might be eligible for study it.
11:15They're in a busy day.
11:16You know,
11:17they may look at the person and say,
11:19you know, I don't know, do we have time?
11:20Is it going to be hard to
11:22explain to this person?
11:23I mean, that's why we get this selection.
11:24And who gets into studies.
11:26Maybe they look at people with
11:27lower health literacy and think
11:28this is going to be take too long.
11:30You know that they're it's
11:31not equipped for this.
11:32We're trying to figure out can
11:33we create the means by which
11:35we can reach out to people,
11:36make them aware of these kind of studies,
11:38make it easy for them to join,
11:39make it so they don't have
11:40to take time off work.
11:41If you're an hourly worker,
11:42taking any time off work to participate
11:44in study can be a great burden.
11:45Can we make it so that the people
11:47can join us and can we make it so we
11:49can ship drugs to people's houses?
11:50Can we make it so we can collect
11:52bloods at their homes?
11:53Can we make it so that people can
11:54do this so that we minimize the
11:57burden and enhance the experience?
11:59So we have the listen studies
12:01and observational study,
12:01the Paxil C trial as a as a phase
12:06two investigation on new drug
12:08randomized trial Paxil for 15 days.
12:10But we're also in the course of this
12:12trying to build and test new ways of
12:14doing this knowledge generation pipelines.
12:16If we've got large numbers of people
12:18with these conditions now we can
12:19quickly enroll them in trials that we
12:21can quickly get them into studies.
12:22They're they're eager and and it's
12:25a readiness cohort that's in.
12:28But it's important for us again
12:29not to be exploitive,
12:31but to be participatory in
12:32partnering in ways that they feel
12:34that they want to stay with us,
12:36they can leave at any time.
12:37And then how do we use all the digital
12:39strategies to move the data and collect it?
12:41Another thing that we've done together,
12:44which has been a remarkable
12:46experience for me,
12:48is actually give people in our studies
12:51direct access to the investigators.
12:53So this is like something
12:55people generally thought,
12:56well,
12:56that seems like a bad idea.
12:57Aren't you going to contaminate
12:58the study or what does this do?
12:59Well it turns out if you have a town
13:01hall where you invite people who
13:03are in your studies to come and and
13:05you just pick a time lots of people
13:07show up. We've had you know at
13:09a random time we pick because we
13:10we can't schedule with everyone.
13:12We have 2000 people in the listen study
13:15now and you know 10% of people show
13:18up and they we we're still working
13:22on this how to optimize this they
13:24love Akiko that there's no surprise
13:26it's like and it's it's thrilling
13:28for them to have an opportunity.
13:30We we're we're careful about things
13:32we can and can't say what we can and
13:35can't disclose about what's going on.
13:37But we're we're telling them as
13:39much as we can and and we're trying
13:41to listen to their suggestions and
13:42what are their concerns and and
13:44what are the questions and how can
13:45we be a resource to them.
13:46But it it ties us to them in ways
13:48that has never been possible before.
13:50And I've found one of the best
13:53experiences I've had my entire academic
13:55career is to attend these these.
13:58And you know there was one where Akiko
14:01really presented the entire time and the
14:03number of hearts and claps at the end.
14:05You know,
14:06I found it so touching really
14:08honestly that we were in the same
14:10virtual room with people who are in
14:12our studies and we were also able
14:14to express directly our appreciation
14:16for their involvement in the studies.
14:18It it was,
14:19I think it's an innovation that way.
14:21So some findings,
14:23this isn't meant to be anything
14:25more than just giving you a sense.
14:27So one thing is we use validated assays, so.
14:29So we're trying to triangulate different
14:31information, clinical information,
14:32testing information.
14:33Ultimately we're we're going to link
14:36to wearables so we can get information
14:38coming from sensors that people are wearing.
14:40So a lot of real world information but
14:42also patient reported outcome measures.
14:44So for people in the lab,
14:45you know you're thinking about
14:46assays all the time.
14:47They've got different
14:49characteristics and properties.
14:50You want them to be reproducible,
14:51you want them to be reflective of the
14:53whatever it is you think you're measuring.
14:55The analytic validity is important
14:57and you know in the in clinical
14:59research there are tools that are
15:01about patients reporting their
15:02experience that have undergone quite
15:04a lot of testing and validation.
15:06Their psychometric properties are
15:07quite strong that we think we can
15:10rely on them and we have a lot of
15:11reference populations to compare them to.
15:13This is just an example and you
15:15look at these and go like what?
15:17What's so special about this?
15:18In the past seven days I felt worthless
15:20about helpless, felt depressed.
15:21And you can put never, rarely,
15:22sometimes, often, always.
15:23But but these have been through so many
15:27rounds of testing for understandability,
15:29context, validity,
15:30a whole range of criteria to
15:34be able to determine that,
15:35yeah, we can use them.
15:35They produce results that that
15:38can compare across populations
15:39and have some meaning.
15:41And so you know,
15:42we that for example the Promise
15:4529 which we're using in
15:47the Paxil C trial can be put translated
15:49into a scale from zero to 100.
15:52And for example, the promise cut
15:54points of that can correlate,
15:56have some interpretability with regard
15:57to what the person's overall health is.
16:00But because there are a lot of
16:02specific questions you can dig
16:03into what's driving their results.
16:05What is it that led to the findings
16:07that that we have And it gives a,
16:09we can, we can do computational
16:12phenotyping on their clinical data.
16:14But we can also when we have a lot
16:15of data about their experience,
16:17their symptoms their their
16:21how their lives are led,
16:22then it also gives us a chance
16:24to to to phenotype based on that.
16:26And it can be as simple as this.
16:28I mean this is the EQ5D visual analog
16:31scale where again you think this is gosh,
16:34this is so simple why you know some
16:35of you must have done this overnight.
16:37But but you know this is a tool that
16:39has been tested in in millions of
16:42individuals in different kinds of
16:44populations for interpretability,
16:46meaning and so forth about just saying,
16:48you know we would like you to indicate
16:50on this scale how good or bad is
16:52your health today in your opinion.
16:54And you you say like the thing about
16:56it is when people report how they feel,
16:59they're intrinsically correct because
17:01that's that's their impression
17:03of how they feel that day.
17:05And of course lots of things can affect
17:07it but we're trying to get some sense
17:10of a draw a line you know across this
17:12scale that represents where you are.
17:14The zero is the worst imaginable
17:16health state and 100 is the
17:18best imaginable health state.
17:20And if you look like in surveys
17:22of the United States adults,
17:24of course it varies a little
17:26bit by age you may have this is
17:29showing you different scales,
17:30but the the the black line is the
17:35EQ online vast that would be like
17:37equivalent to what we're doing.
17:39And you can see that you know in
17:41the younger group it's about 80,
17:42it can dip down that there are
17:43different ones.
17:44Sometimes it suggests for the United
17:46States population maybe around 80
17:48overall and it can correlate to how
17:50people say excellent, very good,
17:52good, fair or poor.
17:53But you can see you're like 70 to 80
17:56EQ vast for for the general population,
17:58which by the way that's not
18:00the healthy population.
18:01That's just the general population all
18:03things considered including people
18:05who who have health issues going on.
18:08And then if you look at us,
18:11the people who are in the listen
18:13study who are reporting long COVID,
18:15these are the distributions.
18:16So you can see in the far left this
18:18is just the overall distribution with
18:20a line sort of going down at 50 and
18:22and and maybe if we had more people
18:24that would be fully a normal distribution.
18:26It's got a little bit of this notch.
18:27I think it you know maybe
18:29more people fills it in.
18:30I think it's probably normally distributed.
18:32But you know, we're we're down 50
18:34or less and a lot of people much
18:37lower that that's very poor health.
18:40It's it's fair or poor health.
18:42And we by the way,
18:43just compared men and women,
18:44young and old.
18:45We looked at different,
18:47different waves of the of the virus.
18:51We were able to look from people's reports
18:54of 25 most prevalent symptoms in our group.
18:56Others have reported this.
18:58This isn't necessarily a breakthrough stuff,
19:00but I'm just giving you an idea
19:02of the kind of data we have 99
19:04symptoms that are collected.
19:06These are the most common ones,
19:07but we're able to in pretty clear
19:10detail characterized these people's
19:11experience and begin to look at
19:13not just calling it long COVID
19:15and everybody's got everything,
19:17but that actually there are some
19:19specific clusters within this where
19:20there are not only do we think
19:22they're underlying mechanisms,
19:24long COVID may be different,
19:25but it's it's being reflected in
19:27different ways that that people
19:29are manifesting it.
19:30They're not all the same,
19:31but again,
19:32if you're practitioner right now
19:33you you you're just keep seeing
19:35people with lots of symptoms.
19:36We need tools that are helping to
19:38take an inventory of symptoms and then
19:40to help in multidimensional space,
19:42sort of locate you who's who are your
19:44neighbors versus some other neighbors
19:45who are you like We got to start
19:47building a taxonomy that helps us
19:49understand this with greater nuance
19:51than just calling everyone long COVID.
19:55This is the frequency of treatments
19:56tried among long COVID participants.
19:58And by the way it goes down to,
20:00I mean these are categories.
20:01But when you actually within each
20:03of these categories and these
20:04people are trying like I think
20:05the average number of people have
20:07tried like 88 different things.
20:08You know,
20:09it's like,
20:10so it's just a remarkable amount
20:12of and of 1 trying things without
20:15really any systematic collection of
20:17information about what it brings.
20:19But it shows you the level of desperation
20:22that exists within this this group,
20:25that that they're trying everything.
20:27Because they're the,
20:27you know,
20:28they they they feel that their
20:30current life is untenable and are so
20:33desperate to find relief that that
20:35they're willing to go after everything
20:36and anything comes up on Facebook.
20:38And and I've seen this stuff
20:40like gambling block stuff.
20:41I haven't don't know what
20:42to make of it either.
20:43They had on the national news,
20:44one person went to Cleveland Clinic,
20:45had it done and then the
20:46patient goes like it,
20:47you know,
20:47it's like the people had water thrown
20:49on them and then they would stand
20:50up from their wheelchair in the in,
20:51you know, and they can walk again.
20:53You know it's like I don't know
20:56like maybe the water is magic.
20:58I don't know.
20:59But you know it's like I I I think we
21:03at in the Academy have an obligation
21:06to be able to help people
21:08understand what what happens and
21:09and what can make people better.
21:11I'm I'm happy if that person smelled
21:13the coffee but I do before I begin
21:15to start doing that on hundreds of
21:17thousands of people would like to
21:19have some basis to to to believe
21:21that it's it's actually working.
21:23So we you know we we we're looking at
21:26these in different ways this is symptom
21:28severity it's a different thing which
21:30is on in this case 100 unlike the vast
21:32score is like my symptoms are unbearable.
21:35So we're asking people on the worst days
21:37how bad are your symptoms and we're
21:40up around 80 or even higher for some people.
21:43I mean, people are saying not only,
21:45I mean these symptoms are coming and going,
21:47but so that's the other thing When
21:49you captured in any given moment,
21:50they could be feeling better in that moment.
21:52But if you ask them overall in the last
21:54two weeks how how bad is the worst day?
21:57They're saying it's horrific, right?
21:59So we have to be thinking,
22:00how do we capture this because it's
22:02about capturing periods of time.
22:03It's also about understanding
22:05how things track over time.
22:07We also ask a whole bunch
22:09of psychosocial questions,
22:10but this is just some for example,
22:12how many felt fearful and a lot
22:16of people feel that way and 16%
22:20that feel often feel anxious.
22:23A lot of people feel anxious, feel worried.
22:24I mean these people are are in
22:26terrible shape with regard to that.
22:27We asked about transportation challenges,
22:29we asked about insecurity about food,
22:31we asked about insecurity about housing.
22:33I mean these people in a very tenuous
22:35position with regard to their lives.
22:37Their incomes have often been been
22:39cut off because they're unable to
22:41work and they're they don't have
22:42a a big safety net behind them.
22:44And then a lot of them are socially
22:46isolated and depressed and these
22:48weren't preexisting conditions but
22:50these have come up as a result
22:52of what they're experiencing.
22:53We also are looking at people we're calling,
22:55I'm calling so far,
22:56I don't know if I'm still keep going
22:58on this yet post vaccination syndrome.
23:00But you know,
23:01the idea that there are some people
23:03who started developing a bunch of
23:06symptoms in in a period that was very
23:09short after they got their vaccination,
23:11maybe within six days after they got
23:13their vaccination that's had a long tail.
23:15Now people have said they've got a lot,
23:16they sounds like long COVID.
23:17It's true.
23:18They have a large number of
23:20different symptoms.
23:20But in some of the research we're doing,
23:23we can actually differentiate the
23:24pattern of the symptoms and people
23:26if if I just give you a bunch of
23:28people with a bunch of symptoms,
23:29then I say predict which ones have
23:31long COVID and which ones have
23:33this post vaccination syndrome.
23:34When you look at it, you might say,
23:36well,
23:36let's just look like a bunch
23:37of people got a bunch of symptoms.
23:38But if you actually do,
23:40you know, if you look at it
23:42mathematically and prediction wise,
23:43you actually can predict which ones of
23:45them have post vaccination syndrome,
23:46which ones have long COVID,
23:48which I think gives some credence
23:49to the fact that these may
23:51have some overlapping features,
23:52but they're actually distinctive.
23:53And I think we're going to be able
23:55to show for the first time this,
23:57this distinctive nature of it.
24:00And Andrew Wangston doing some
24:02I think very great work kind of
24:04working with these individuals
24:05that again just like if if people
24:07with long COVID get dismissed,
24:09these people get dismissed doubly because
24:11they fall into the political maelstrom.
24:13And and nobody wants to talk about
24:14it and I think it can it can be true
24:16that the vaccines were a miracle
24:18and saved millions of lives and
24:19that there were a number of people
24:21who were adversely affected.
24:22Both things can be true and if we're
24:24truly scientists we're not going to
24:26to shy away from investigation of
24:28this even though we know that we may
24:30put ourselves in a position where
24:31what we talk about maybe weaponized
24:33by others who have agendas that are
24:35different than ours but we we have
24:37to keep pushing forward with but
24:39what we think is the right thing
24:40to do and and and try to do this
24:42and again Kiko at every step has
24:44been I think so strong about this
24:46too and I'm so appreciate that.
24:48So the top lines of what we've
24:49done so far and you look at these
24:51groups highly symptomatic group
24:52that we've been able to assemble.
24:53I do think it's a subset of the
24:55people with long COVID but but this
24:56is the where we're going to find
24:58clues where we're going to find
24:59clues is where there's the most
25:02manifestation of what people have.
25:04Right.
25:04Let's start there.
25:05And so we're able to assemble I think
25:08large numbers of people who are highly
25:10symptomatic A diversity of symptom
25:11profiles that were beginning to be
25:13able to differentiate and characterize
25:15so that that that they're different.
25:18Tianna Joe,
25:19medical student has been doing some
25:20great work looking at people who are
25:22complaining of internal vibrations
25:24and and tremors and how they're
25:25different from people who don't
25:27have that as a prominent symptom.
25:28And and actually again you can
25:30based on the pattern of symptoms
25:32outside of that symptom you can
25:34differentiate them and and so we can,
25:36we can begin to start to understand
25:38these are clues we're on a search now.
25:40We're looking for any clues that
25:41help us begin to understand and
25:43more that we can look at this
25:44better off we are people have tried
25:46many treatments without relief.
25:48They have substantial psychosocial
25:50burden and extensive opportunities
25:51for their extensive opportunities
25:53for impact for all of us.
25:54And and I'll say one more thing that
25:56there are a lot of people in this group
25:57who were completely healthy before.
26:00So that's not just a group who was
26:03struggling with their health before.
26:05And that's again what makes me
26:06think there were marathon runners,
26:07there were people who are highly active.
26:09Now. It's not that we're just
26:11interested in those people,
26:12but again that those people may give us
26:14the opportunity to really look at clues
26:16because the where we find contrast,
26:18where we find things that that puzzle us,
26:21why did that happen then?
26:22That's where I think we can
26:24find rich opportunities.
26:25And our interest is looking at
26:26contrast by age, sex, race,
26:28ethnicity differences, timing differences.
26:30You know, when did when did it occur?
26:31How long did it last?
26:33What what, what did it change by waves,
26:35trajectory differences?
26:36Can we start to plot these out
26:38with latent class analyses and see,
26:40you know,
26:40this person is going up and down like this,
26:42this person's starting to get better.
26:44You know, what are the different?
26:45How do we begin to differentiate people
26:47based on trajectories, syndrome,
26:49contrast like what Tianna's doing with
26:51with the vibrations or we look pots Yes no,
26:53we look tinnitus.
26:54Yes no, Like.
26:55Are there clues here about people who
26:58primarily have one driving symptom?
27:00I think the the problem with the
27:01taxonomy that came up in JAMA that
27:03came out of the project recover
27:04was that they were wanted us to
27:06count symptoms and when you got
27:07to a certain number count,
27:07they were some of them were weighted,
27:09you got a score and they said bang,
27:10that's that's social.
27:11With long COVID,
27:12you know there's some people just
27:14have one symptom but it's intense
27:16and and I think it still can be post
27:18infectious in nature or post vaccination,
27:20computational,
27:20clinical and lab phenotype correlation.
27:22Of course, this is the Holy Grail
27:23of what we want to be able to do.
27:25Take all the information coming
27:26in clinically,
27:27take the information coming in from the
27:28lab and see where do we see the overlaps.
27:31I mean, where are the correlations,
27:32what can we learn together from this?
27:34I think ultimately we want to
27:36be doing taxonomy development,
27:38strategy testing and much more.
27:40I think the goals are,
27:41I think we understand it is what
27:43we want to do.
27:44We understand it,
27:45we can treat and mitigate or cure it
27:48and ultimately we can prevent it,
27:50that this is what our marching orders are.
27:52This is where we want to be in 10 years,
27:54five years, 6-6 months.
27:57If somebody in here is really smart,
28:00I'm looking for that person.
28:01Now, who's that going to be?
28:04Progress requires teamwork,
28:06trust and tenacity and the courage
28:08to believe that anything is possible
28:09If we work together,
28:10that that's the spirit that we're
28:12trying to bring to this effort.
28:14The goal is better outcomes.
28:16Let's work together to make it so.
28:17This is a poem.
28:19I want to say just that one member
28:21of Listen gave us which,
28:22which I'm not saying everyone's
28:24had this experience but
28:26but I'll let you read it. But
28:51I thought she was far too kind to us.
28:54But it did show me that maybe we're
28:56making some success in trying to build
28:58a study that people feel truly part of.
29:01And as Lisa was saying,
29:02what she's trying to clinic, listening,
29:04acknowledging and engaging and earnestly
29:07is the same thing that we're trying to
29:10do within the the research side as well.
29:13And and we have work to do
29:16this this isn't an end.
29:18This is just encouraging us
29:19to continue along with that.
29:21There's so many people to thank who've
29:23been part of this and I'm sure left it
29:25out because I did this really quickly.
29:27But it's been just a remarkable
29:28team effort and anyone who wants
29:30to join us is very welcome.
29:32Thank you.
29:54Okay, right. Thanks. Great talk.
29:58So I just have a question about the
30:01tax Levid long COVID, any potential,
30:06you know, predictions on when
30:07you might have some readouts.
30:08I think you might have said that
30:10it's a randomized phase two study.
30:12So I'm just curious about, you know,
30:14when you might anticipate some
30:16readouts for that for that study.
30:17So I'm hoping after the first of the year,
30:21you know it's actually actively
30:22enrolling anyone who's knows people
30:24who have long COVID who would
30:25like to participate in the trial.
30:26We'd love to include them.
30:28It it, it includes people who were
30:30had good or excellent health before
30:32and now have fair or poor health
30:35attributed to long COVID now.
30:37And I think one of the most
30:39interesting parts, it's decentralized.
30:40I like the the platform,
30:42but also that the fact that Akiko's
30:45lab is doing deep immunophenotyping
30:49before treatment and and then after
30:52treatment And because it may give us
30:55clues about who are responders and who fits,
30:57because if we do think that
30:58there are multiple mechanisms,
30:59we don't want to just say what
31:00was the average result.
31:01I was just pointing average result.
31:02But we don't want to actually
31:04dig in deeper to say, you know,
31:06even if the average result isn't there,
31:09maybe there were some people who
31:11did have remarkable improvement.
31:12And is there any clues to that
31:14within the the immuno phenotyping?
31:16So,
31:16but I'm hoping after the first of the year.
31:23Yeah, so I'm wondering about this question.
31:26So if you consider a plot with on Y axis
31:32health state and then there's
31:34somewhere threshold where we
31:36anyone below threshold is healthy,
31:38above threshold is not healthy and
31:40has some symptoms with some names.
31:42But below threshold the healthy range is
31:45not all the same though it varies a lot,
31:49some much closer to the
31:51threshold than others.
31:52And the question is and and for
31:55different reasons some somebody's
31:57kidneys at not that 100% but at
32:0019 somebody's liver and so forth.
32:03And then the question is when
32:06something like this infection or
32:09vaccine or or whatever challenge
32:12hits people who close up to
32:14thresholds as the ones that maybe
32:17will develop a particular illness.
32:19And in the case of long COVID,
32:24I wonder whether these the common symptoms
32:28that people develop have some relation
32:31to what where these people are on on
32:34their health status below threshold.
32:36They all were healthy before and
32:38then something happened and then
32:40this strange relations with very
32:42young athletic women suddenly
32:44developing something that way off.
32:47And the reason I'm thinking that
32:48is I got like terrible tinnitus
32:51I think after my booster shot,
32:54but I had very mild tonight as before.
32:56And that's what made me think that
32:58if you have some predisposition to
33:01something you just slightly off,
33:03then something hits some information
33:05or whatever and that's what you're
33:07going to develop.
33:08Will you be able to capture these
33:10types of relations in your,
33:12I mean do you you probably will
33:13have all the data to to find that
33:15if there is such a relation?
33:17I think it's a really good question
33:19and you know we have the chance
33:21to collect more data and be more
33:23specific about what we want.
33:24Obviously there's some recall
33:25bias that some people can have,
33:26but but to report whether you had
33:28prior tinnitus, most people should be
33:30able to tell us if it in that case.
33:34Yeah, I think I mean the beauty
33:37of this is it's an active live
33:39community that we're interacting with.
33:41So as we come up,
33:42it may be that we haven't collected
33:44enough information to be able to do that.
33:45But because you have that idea,
33:46we should say we we should go back
33:49and and collect that information.
33:51There's another feature here though,
33:52I think that's important that you just said,
33:54like I really don't like dichotomous
33:57variables.
33:57I I I really think that they are
33:59reductionists in nature and they
34:01obscure important relationships.
34:03So the degree to which we can
34:04collect spectrum, you know,
34:05not just saying did you have it
34:07before and someone says no because
34:08it was really just a little bit and
34:10it didn't didn't really bother them.
34:12You know, we want to say no,
34:13no from zero to 100.
34:140 means you absolutely had none.
34:17And then you know how how much
34:18did you have before?
34:19We need to start moving towards,
34:21I think, higher dimensional information.
34:23So that just like you said,
34:24it's not just saying, oh,
34:25you were healthy and not healthy,
34:27but like,
34:27how healthy were you and what
34:28and what did that mean before?
34:30And I totally agree with you.
34:31I do know that there's some people
34:33in here who had low levels,
34:34something that was totally
34:36amplified by and maybe, you know,
34:38you talk a lot of homeostasis.
34:39It's something was keeping things in check.
34:42And also by the way,
34:43their own perceptions of their body,
34:45how they react to their feelings
34:47within their body is changing too.
34:49But there's a whole science of
34:51sort of people's perception.
34:52Some people can really feel
34:53their heartbeats all the time.
34:54You know, it's like is it amplifying our
34:56sensitivity to things within our body?
34:58And then how do we that may be a
35:00whole different group that that's
35:01a whole thing for that we need
35:03to understand and not dismisses.
35:04That's all in your head,
35:05yet it's in your head, it's in your body too.
35:08We're going to take one more
35:09question and then we're
35:10going to go on break.
35:13Really. So fantastic. Thank you.
35:20But so I totally agree.
35:23This is a story that I hear so often
35:25in the patients who come to see me
35:27is that I had this thing and several
35:29people said this, it was annoying.
35:31I had COVID it kicked it up
35:33to being a very big problem.
35:35So I see that a lot.
35:37So I think that's an interesting thing and
35:39I'd love to see you start to measure that.
35:41My question has to do with the amount of
35:44suffering in that graph that you showed.
35:47Is there a way,
35:48or have you considered a way to find
35:51out how much the uncertainty about what
35:54they have and how long it's going to
35:56last and whether they'll ever get better,
35:59how much that plays a role?
36:00Because I think yes,
36:02people are physically suffering,
36:03but there's a lot of psychic suffering.
36:05I mean, we don't know what it
36:07is and we don't even have it.
36:08So is are you measuring that?
36:11Yeah. I I think that in some of this
36:13measurement of people feeling fearful
36:15and anxious and uncertain and losing
36:16hope and we we have some of those
36:18dimensions of course they there can
36:20be reverse causation and people
36:22who you know it's like is that a
36:24modifying factor or is it a consequence
36:26of what they're they're feeling.
36:28I think if we continue to
36:30follow people longitudinally,
36:31maybe we'll get a better sense that
36:33certainly the way the the reception
36:35people have had within the medical
36:36system has tend to be in be an
36:39exacerbating factor obviously with
36:41regard to their their desperation
36:43feeling that no one believes them
36:45or understands them or or wants to.
36:47You know, people are just lost patience.
36:48With me it's enough already like you know,
36:50it's like it's been enough time.
36:52You know, you should be getting better
36:53and people just don't feel better.
36:55But I think it's a really good point.
36:56Lisa, what's something we should
36:58be looking into?
36:59Thank you.
37:00I know we're a little bit.
37:01Thank
37:06you.
37:07So we'll we'll only break for 5
37:09minutes to set up for the next session,
37:11so grab some coffee every minute.
37:13It was worth it. Harlan.
37:14I'm glad. I'm glad.
37:15And the discussion and the questions?
37:17We need to get you all in the same room
37:19to discuss these topics at length.