Webinar: Understanding Digital Health

Webinar: Understanding Digital Health

[MUSIC] Greetings from New York City and Columbia
Business School Executive Education. My name is Scott Gardner and
I’m very happy to be here with today Professor Stan Kachnowski for
the Understanding Digital Health Webinar. Before I introduce Stan, as always, I’d
like to go over a few quick logistics for the webinar. A recording will be made available,
and sent to you after the webinar. If you’d like to tweet about this webinar,
please do so at hashtag #CBSExecEd. And as always, we encourage you to upload
your questions throughout the webinar. And we’ll get to as many of those as
possible in the last ten minutes.>>Right. It’s my pleasure to introduce
Professor Stan Kachnowski. He’s an Oxford trained researcher and
an expert in the efficacy and diffusion of digital health,
including electronic health records, wearables, predictive algorithms,
and precision medicine. He is the chair of
the Digital Health Testing Center HITLAB. And the faculty director of our upcoming Digital Health Strategy
Executive Education Program. It’s great to be with you today.>>Thank you, Scott.
>>So I’m gonna leave it to you. And I will join you at about
20 minutes from now and we’ll go over some questions together. Sound good?
>>Perfect.>>All right, great.>>Well, thank you Scott and
thank you everyone for joining us today here at this program. Very excited to explain a little
bit about digital health and start with its definition. The learning objectives are really not
only about definitions, but also about insights and the use of digital health
today across many different stakeholders. Certainly hospitals and physicians are
where consumers sit on the front end but we wanna talk about more than that. And then at the end toward
the latter part of the 20 minutes, go into more detail around
opportunities and challenges that we’re seeing across the digital health spectrum,
really the health care system. Not just here in the United States,
but globally. Because digital healthcare is something
that, because of its definition and its breadth of its definition, encompasses
a lot in global health as well. That said, with those three key learning
objectives, the principal one is, the first learning objective is around
the definition of digital health. Which was first created
by Professor Eric Topol, who comes out of
the Scripps Institute in San Diego. After doing an enormous amount of
great work at the Cleveland Clinic, Professor Topol wrote a book in 2012
called The Creative Destruction of Medicine. And in that work,
he describes the digital human. And effectively what he does
is he begins to outline this use of binary data to
improve clinical medicine. And he takes this field
known as bioinformatics, which is the study of genomics,
proteomics, exosomics. And he merges it with this field
called biomedical informatics, which is the study of electronic health
records, personal health records. It looks at all the data that you would
pick up from your Apple watch and those sorts of data sets. And he combines them all by saying simply, what we need to think about is
a clinical record for a digital human that takes everything that we can
find from human beings that’s been turned into zeros and ones and
combines it into one human record. And with that definition,
set off an entirely new understanding of what digital health could
be, of what digital medicine could be. And it began this,
what is now becoming a construct, a rubric forming of something that
will help not only hospitals and physicians, but
also public health systems. And life science companies,
med-device companies, and even groups like family foundations, and multilateral
agencies, and national health systems. As it begins to improve the, not only the clinical efficacy, but
also the economic efficacy of how digital tools can increase access
to various medicines and various diagnostics, but
also reduce costs for on a population level for health care
as nations and as groups as whole. So this definition has
an importance of simplicity. We need to understand that ultimately
anything that comes from the human body or surrounds a human that can be
turned into zeros and ones. That is the definition of digital health
that we are going to use for today. And there are numerous definitions,
just to be clear. The most widely seen definition
is curated by Paul Sonnier, an author out of San Diego, California. And if you google digital health,
you will see his definition as well. And that definition discusses more of
the transformational nature of digital health that goes into more detail around
digital health’s disruptive powers. And that’s something, that’s another
definition that’s widely used. If you look at
the World Health Organization, and what the World Health Organization
uses as a definition. Their definition tends to be more
toward public health and global health, more towards electronic health records
systems and telemedicine systems. And so, what we have is an important
term in terms of encapsulating this change that we’re seeing in health
care and life sciences, this revolution. Again, not in the term of a revolution
like a one year revolution. This is a revolution in medicine and
in public health, and in health care in general, that is taking the term of
like a 40 year industrial revolution. And whether you started the clock at
the advent of the PC in the early 80s, or whether you start the clock at the advent
of the internet and web in the 1990s. Or if you want to start the clock at the
advent of and the invention of the iPhone. And really, the diffusion and the adoption
of smartphones across our society. We have this revolution that is
occurring that will be anywhere in neighborhood of 30 to 50 years. And the term and the definition
of digital health is one which is important to certainly grasp, understand. And glean insights from which we’ll
talk a little bit about now as well. So when you look at the definition of
digital health, it’s important to remember all of these different datasets, all of
these different components of human data, whether it’s the individual human or
the population health of humans. And when we look at what
insights we can glean from. And again, as a scientist and
as a researcher, having done hundreds of digital health studies, whether it was
before the term was even actually created. Or today, you can see that whether
it’s in the healthcare e-commerce or whether it’s in wearables and
biosensors, there are tremendous opportunities to glean insights on
how either individual health or population health can be significantly
transformed by using these digital tools. And what I want to do in this
segment in the next five or seven minutes is to talk a little
bit about some of the research that I’ve done over the last
almost three decades now. And just hopefully share with you,
some of what I’ve learned and how it’s certainly changed a lot of
how medicine is delivered as well. First of all, in a study that was done,
this is going back almost two decades now. In England on a tool called Advoy],
which still is being used, a hemophilia tracking software. I don’t know if you remember
the old palm pilot, hemophilia patients would actually track
their prophylaxis, and their hemophilia, their bleeds, their exercise on
a palm pilot, take it home at night. Talk it to the PC and
upload their data to their physician. And that study involved 1,200 subjects
over the course of a few years. And revealed that those subjects,
those patients, that actually utilized the palm pilot and
tracked their their diet, their exercise and their prophylaxis
use more aggressively, were seeing significantly better outcomes and
significantly lower hospital admissions. And so that was an early study of
digital health that showed, and again, the variables could be confounding, maybe not, but
it certainly was hypothesis-generating. That this use of digital health had some sort of correlation
with improved patient outcomes. And again, that was in England,
and that was 2000 to 2005, and that was with
a significantly chronically ill patient population of hemophilia patients,
so that was a good start. And then fast-forward now to the Jawbone. Again, a wrist based device, I don’t know
if any of you own those devices or not, but it tracked activity,
both daytime and nighttime. The Jawbone study that we did was 565
subjects over the course of six months. And it showed whether or
not patient populations and subjects would actually improve their
BMI and improve their sleep quality while they saw the feedback loop of their
daily usage of an activity tracker. So they could actually look at the Jawbone
app on their phone and see whether or not they were exercising as much as they
wanted to, their 10,000 steps a day. That was the daily recommended course
back then, now it’s 15,000 steps a day. And also their sleep, the Jawbone was
able to track sleep during the night, and detect sleep quality. And what we found in that study
was actually interesting. It really didn’t have a substantial impact
on either sleep quality, or BMI reduction. Except for a small population,
within one of the strata of population that we studied,
we found that a specific group of women were more likely to change their behavior
based upon those natural feedback loops. And so again, that was significantly
different than the hypothesis that we had generated in submitting
the study to the IRB and in actually formulating the protocol. So again, digital health had efficacy. Digital health showed, or at least we
thought it would show a significant efficacy for the subjects in that study,
and yet we were shown that it was really not efficacious from
either an economic standpoint or a clinical standpoint for
the clinical endpoints of BMI. And we also used a standard form from
the NIH to measure patient well-being. Another study to look at,
in terms of actual wearables, was a study that we did with an actograph. That actually was able to track
congestive heart failure patients, we put a wrist based device that tracked
their activity again, in daytime, and nighttime to determine if
activity during the day, i.e. If they are more active during the day,
would they actually have a reduced rate of deterioration in their
congestive heart failure symptoms? And again,
our hypothesis was off on that as well. We had, and went and collecting these
data on these 61 subjects over 9 months, continuous 24/7 monitoring, we found that
it was actually the nighttime activity. That predicted the rapid rate or the increased rate of deterioration
of their congestive heart failure. What does that mean? That meant, and again, this was published
in the Journal of Cardiac Care in 2009, that meant that patients that weren’t
sleeping well, were the ones that were seeing the worsening of their outcomes,
the worsening of their health effects. And so that was, again, another study
that indicated our hypothesis originally expected, deterioration of heart
disease because of daytime activity, were off and were significantly
varied than what the outcome was. The point here is that digital health, while oftentimes we think
we see the intuitive. We think we know what to expect when
we put together a device or an app, or some other tool digitally with patient
population, we feel like we’re confident in knowing what those outcomes will
look like, that’s actually not the case. Study after study what we’re finding
is that the outcomes are much different than we predicted, and that the hypothesis
that we had first written about, were actually not ones that had
come out at the end of the day. The last point I want to make, the last anecdote I wanna share is
about a study that was done in Africa. Again, digital health is very
much a part of a global movement. It’s part of a global
infrastructure effort. And it certainly,
if you look at countries like Ghana or India, other nations around the world, you’ll see that digital health oftentimes
is trying to leapfrog current medical systems that even here in the United
States we’re accustomed to using. So in Ghana, the study was,
again, it was facilitated by, Funding from the Gates Foundation,
and it was a technology developed by the Grameen Foundation, called MOTACH, the
Mobile Technology for Community Health. And it was a nurse-based voice messaging
system that would be delivered to new moms that were 0 to
12 months after birth. And it allowed the nurses to utilize these messages to vast numbers
of new moms in Ghana. We looked at the efficacy of 96 voice
messages being delivered to these new moms to see if it increased vaccination rates. If it increased breast feeding rates,
malaria kits and clean water tools. And what we found was, from the beginning, from our hypothesis, we assumed that it
would increase the rates of all of these sorts of interventions for
reducing infant mortality. What we found was, a lot like you, when
you get messages from your pharmacy to pick up your prescription
at your local pharmacy, a lot of times you’ll just either delete
them or you won’t answer your phone. So we found that the mother’s behavior and
the interventions that actually worked, were simply breastfeeding and
the reminders to breastfeed. And again, that showed that
infrastructure, and any kind of disruption in their daily activities,
any kind of activities of daily living. Whether it was their working, or their
child rearing, or whatever it might be. They were very unlikely to
disrupt those sorts of behaviors. Those sorts of practices, those sorts
of habits, to do something like walk to a clinic that would be an hour away,
take a day off from work and then have to pursue a vaccination that
they weren’t quite sure or confident in. So these are the sorts of studies that,
when looking at digital health, we begin to see that it
is a very complex field. Digital health is not predictive. It’s not something that
can be easily seen as, if we develop this tool, they will use it. In fact, what we’ve seen over the last ten
years, particularly, given the ramp up of investments, and just to give you an idea
of what people are investing these days, Infi digital health, it’s in
the neighborhood of anywhere from a $1 billion to $2 billion a year, and
that ‘s just your US based statistics. We’re finding that these sorts of
investments are something that are having significant issues actually making
their way into either patients’ hands, or into hospitals, or into health plans,
or life science companies. And what we try to do when we’re looking
at the diffusion of digital health, we try to find markers, if you will. That will help predict
the more likely diffusion and uptake of an adoption of
a digital health tool. And I’ve got a few minutes left here, that’s kind of the opportunity
section of the webinar today. And that really is, as an investor,
or as a major institution, whether it’s a hospital, a health plan,
or a life science company. Where are the opportunities for
you and your institution? Whether you’re an executive in
the C Suite, a VP level executive or you’re managing a department,
Digital has a number of opportunities. It has the short-term, in terms of being
able to do what you’re doing today but do it for far less money, or
at least somewhat less money. It has the opportunity to allow
you to work smarter, not harder. So can you do more work and have more productivity by using
the same number of labor units? And can it have an impact
on your quality ratings? So if you’re a hospital and you’re looking
at digital tools, where’s the opportunity for you to bring that into your system,
to integrate that into your system? I think the number one
word here is integration. We’re seeing significantly high barriers
of taking a digital health tool. Whether it’s something as
simple an Apple watch and trying to integrate it into either
an electronic health record system or an electronic data capture system or
care management system at a health plan. These sorts of integration points
are turning out to be the vast barriers of the diffusion of digital health into and across the majority of large
enterprises and institutions. And that is really to
say that on a B2B level, there are significant opportunities. If you look at Keytruda from Merck,
again a nice digital health play. Merck invested carefully. They very nicely crafted their clinical
trial with patient populations that met specific genetic requirements. And they created a genetic tool to help
actually stabilize the drug usage within patient populations that were going
to actually meet the criteria, and actually have that drug be
used in an effective way. And if you look at ResMed, for instance, it’s a sleep apnea device being used
at home, again, another great case of digital health diffusion happening because
it was integrated into a pay market. And it didn’t require significant
enterprise level integration into care management systems or
electronic health record systems. Finally, if you look at a live core and
you look at eye rhythm and what they developed in terms of
an actual heart attack detector. You look at now something that is
significantly reducing costs for health plans. So why would health plans all of a sudden
want to run toward these sorts of heart attack detection devices for
patients suffering from AFib or DFib? Well, of course, they’re going
to reduce hospital admissions, they’re going to reduce length of stay. And they’re going to reduce the
readmissions that occur after an event. And so, the simple fact of integration
with that health plan system, with the health plan in general,
based upon the return on investment, the integration costs there were much,
much higher, were much, much actually higher than you might see
in another app or another device but the cost savings were also higher. So it’s important to see that integration
as a focal point for digital health invention, adoption, and eventually
diffusion is a core component here. And it’s absolutely
critical to remember that of all the works that’s being
done in digital health, it’s important to know that patients have
to be able to utilize these technologies. And you’ve had in past webinars from
Columbia Business School lectures and talks on design thinking, and design thinking is a big part
of digital health diffusion. It’s certainly one to take careful
notes on and to certainly understand the patient centricity in digital
health as the adoption cycle widens. And as it improves and ultimately think
about platform solutions when you’re investing or when you’re looking to bring
in new technologies, new point solutions. Try to stay away from
single-point solutions and think more about how does this point
solution fit on to another platform that my enterprise is already using? And I see that my time for the individual
part of this webinar is over, and I’m going to invite Scott
back into the frame here. So we can continue our conversation and
take some questions from the audience.>>Thank you very much, and it’s so interesting, such great information,
it’s the future. And so
I’m excited to be part of this with you.>>Right.
>>We have a lot of good questions that came in.>>Good, good.
>>So let’s get to those, all right?>>Okay.
>>So the first one is and I like this one because it’s
kind of when we have our people, the people that are watching
right now we really want to say, well what do they do
with this information? So the question came from Amina, how should I be using digital
in my institutions today?>>That’s a great question, Amina. And a lot of it depends on which
insitution you’re coming from. In healthcare, generally we look at about
nine major institutions that affect the global health system. And they range from certainly providers to
health plans to life science companies and med device companies, to family
foundations, national health systems, multilateral agencies,
certainly patient advocacy groups. And depending on the organization,
and this is where the complexity and oftentimes the contradictory nature
of digital health presents issues for institutions, is that it is
significantly dependent upon the population you’re serving because
every population uses digital differently. A population of moms in Ghana, they have
an entirely different digital profile than a population of moms in Manhattan. And I think when we
look at an institution, we’re really trying to
understand what the enterprise legacy systems are looking like, what
populations are they trying to manage and trying to serve and just fundamentally,
what are their success metrics. So great question, but
absolutely that one is filled with more questions on top of them.>>Next question is from Abishek], how important is patient
data in digital healthcare?>>Yeah, patient data is
the core of digital health. Again, if you look at either Paul
Sonnier’s definition on Google when you Google digital health you’ll see
Paul’s definition out there. Or if you go back to Professor Topple’s
original definition, you’ll see that at the base of both of those we’re looking
at this element of zeroes and ones. And so patient data and how it’s
generated, the quality of the data, there are seven variables for
patient data quality to be tracked, things like timeliness,
consistency, cleanliness, other misfit kinda alpha
numerics in the data field. Those sorts of components of
patient data are critical. Because as you’re seeing
with machine learning, and some of the technologies that have been
invented around cognitive computing, machine learning,
AI as some people like to call it. You’ll see that, in fact, a lot of
the issues in having positive results from those emerging technology have been raised
because the data quality is not high. And anyone who’s worked with healthcare
data knows these a generally not easy to work with and it needs a lot of cleansing,
and the integration’s also very difficult. And so big patient data is at the root
of any digital health project.>>This is an interesting question,
I think, it goes to culture, creating a culture with this digital
revolution, which is how do get, this is from Shona, how do you get
over resistance from clinicians, those afraid technology is replacing them? That’s a great question thank you for
asking it. Because for
almost 28 years of my teaching career, I’ve been hearing and reading that
physicians are technophobes and I would pose it that, that is
exactly the opposite from the truth. If you think about clinicians and
physicians, they have to re-certify by
taking a board every five years. Much of which is about the newest
technologies that are out there, whether it is molecular or device based. Physicians are using medical
technology everyday. The definition of medical technology is
the application to society from science. And they are using science
everyday in what they do. Whether it’s a new drug, whether it’s a diagnostic tool, they
are constantly learning new technology. When it comes to digital technologies, physicians are working with things
like electronic health record systems, personal health record systems, they’re
now seeing new streams of data come into those systems that are coming from
the algorithms in the live course. They’re coming from the Apple Watches,
and so they’re adjusting their practices to
these new data profile’s genetics. Again, every time you go to the doctor
now, you’ll see a new genetic diagnostic, or a new genetic tool for
you to use to help manage your conditions. And I think the physician interaction with
digital health is at the core of what Professor Topel set up in his 2012 book. I think that’s something that,
really, when we look at this on the population health basis,
that physician interaction with digital, as electronic health records
move to swipe screens. As they move to voice integration and
voice interaction, so they can actually make voice calls to their health records
like you make voice calls to your Alexa or your Google Assistant, or your Siri. That’s when you’ll see this kind of
digital revolution that we’re in the midst of really begin to take off.>>That right? Great, so Amritha has asked,
looking five to ten years forward, what can PARE organizations
do to invest in today? To make sure needed digital health
innovations can be democratized, or made accessible, or
adaptable to reduce healthcare costs?>>Another great question and
I think that the health plan community, in the United States,
is not a singular community. It really is made up of, maybe 20 or
25 major health plans and integrated delivery networks or IDNs
with our really hospital based systems. And then there are a number
that are intermediate size and there are a number of health
plans that are much smaller. And it really depends on
the size of your institution. If you’re a larger health plan, a larger
IDN, then making those investments and taking some of that overhead capital
that might be around to test and learn with some new digital interventions,
that’s a great plan. But if your not, if your either
a mid-sized health plan or a smaller sized health plan, it’s important to understand
exactly what the success metrics are. It’s important to see evidence from other
health plans and what they’re doing. And so really, digital becomes less
of an innovation and more of a kind of try to be a fast follower, and not
like a lost leader or an early pioneer.>>Great, Javier asks, what is your opinion about the future on
privacy over all this health information?>>Privacy is absolutely critical,
and it’s not just a US issue. If you look at what’s
happening in the EU with GDPR. And again, you can Google that. I won’t go into detail there,
since I have four minutes left.>>[LAUGH]
>>But if you look at what’s happening in India. In India right now,
they have a national ID number. And there’s a lot of uproar and
controversy, because a lot of folks don’t want to be tracked by governments and
that sort of thing. So privacy of your healthcare
is quintessential. And if you look at what was begun under
the Bush administration, Bush two, in the mid-2000s started a part of
HIPPA and started implementing and enforcing a part of the Health Insurance
Portability and Accountability Act. That started looking at data breaches and
patient data breaches. And if we look at, and again these are all
publicly available on the Department of Justice’s Office of
Civil Liberties website. You can see that there are enormous
numbers of patient data breaches, not just happening across the smaller
brands, but all brands of hospitals and health plans etc. So patient privacy is critical and,
I think, in this era of we feel comfortable
with Google reading our emails and sending us ads based upon that, we feel
comfortable, maybe a little bit less than it was a year ago, with Facebook
reading our stuff and sending us ads. When we look at patient privacy, that is
one of those walls where people start to, at least most people start to put
a barrier and say, look, I want my medical records protected and there’s
a little bit differentiation there. It’s very interesting to
see the difference of patient data in the US versus Europe. Where Europe it is mostly publicly
funded health care systems. There is very little line to cross between
commercial and public payer systems, very little integration there. Patients don’t want their data going
to private organizations at all. In the United States, it’s very different. There’s a lot more openness,
because we’re kind of half corporate and half government healthcare system, there’s
a lot more openness with US patients to have their data being used
by private organizations. As long as it’s being done for research
purposes, and not for marketing purposes.>>Sanjeev is asking,
how critical will it be to unsilo patient data across institutions
to move digital healthcare forward?>>That’s fundamental, and again, it
goes towards that integration component. If we cannot find faster, better, cheaper
ways of being able to take data from one department, say,
your emergency medicine department and have it become an integrated component
of an overall institutional record. You’re not gonna have that
complete kind of digital human that Professor Topol talked about. And you’re certainly not gonna have
the ability to manage population health issues and
to understand the institutional issues. And so that is gonna be fundamental and I think we’ve seen a significant amount
of progress over the last ten years, around data standardization
efforts in the United States. And now there’s something called fire,
again, I would recommend you
Google that term as well. [COUGH] And the fire standards
are something that both HHS and CMS are promoting and utilizing. And then you’re looking at as well as
NIH funding a lot of research grants that are using those
standards to help improve that sort of data integration
effort that you’re talking about.>>All right, this was a wealth
of very good information. Thank you so much for being with us today. And thank all of you for
your wonderful questions that came in. I look forward to your new program,
Digital Health Strategy in early 2019, if I’m correct,
>>That’s correct.>>Executive Education, we’re providing
that, we’d be great to work with you.>>Thank you.
>>Thank you very much>>Thanks for having me>>And thanks for coming.

Daniel Yohans

1 thought on “Webinar: Understanding Digital Health

  1. Der Kompetenzentwickler Alexander Riegler says:

    Thanks from Austria for being part of this webinar. It was very informative.

Leave a Reply

Your email address will not be published. Required fields are marked *