ArticlesBlog

Applications of Network Science to Strengthen Systems of Care

Applications of Network Science to Strengthen Systems of Care


Linares: Good afternoon,
my name is Dr. Deborah Linares. I serve as a health scientist
and project officer in the Division of Research,
within the office of Epidemiology and Research
at the Maternal Child Bureau, Health Resources
and Services Administration. The Division of Research
provides ongoing support for Maternal and Child Health,
or MCH Extramural Research Activity,
including the engaging research innovations and challenges
or the EnRICH Webinar Series. You are joining a community of
more than 100 participants with an interest
in advancing MCH research. The EnRICH Webinar Series
provides technical assistance and methodological updates
aimed at stimulating interest in applied and translational
MCH research. Today’s webinar is entitled,
Applications of Network Science to Strengthen Systems of Care
for Children and Families. Before we start,
I would like to briefly introduce our speakers
for this afternoon Doctors Danielle Varda, and
and Ayelet Talmi. Dr. Danielle Varda is the CEO of
Visible Network Labs, an Associate Professor at the
School of Public Affairs at the University
of Colorado Denver, and the Director of the
Center on Network Science. She has developed tools
and methods to measure complex social systems,
advance measures of social connectedness,
and discovered applications for using these methods to
improve whole person care and clinical
and community settings. Dr. Ayelet Talmi is the
Professor of Psychiatry and Pediatrics at the University
of Colorado School of Medicine, Director of Integrated
Behavioral Health, and Co Director of
the Harris Program. She is engaged in clinical
research and policy efforts to build sustainable service
delivery systems for children and families including
behavioral health integration, and primary care settings,
early childhood, and children with
special healthcare needs. Lastly, before we begin the
presentation, we would like to let you know
that the nonfederal resources in this presentation have not
been formally approved or endorsed by HHS or HRSA. I will now turn the program over
to Doctors Varda and Talmi. Varda: Great. Good morning!
This is Danielle Varda. And I am here with my colleague
Ayelet Talmi. And we are so pleased to be
invited today to speak on this webinar series. It’s such an important one and
it’s fun for us to get all of these opportunities. So, today we’re going to talk
about Network Science and its applications
in systems of care for Babies and Young Children with
Special Healthcare Needs. We’re going to focus that on
thinking a little bit about concept of social
connectedness, and how social connectedness
can be a really grounding and framing type of way
to think about this work. One thing I am not going to try
and convince anyone here on this webinar is whether
social connectedness has an impact on our health
and well-being. I say that,
because I’m going to assume that most of us on this call,
who are doing this work in this space,
already know that. In fact, there are so many
different articles and research dating back
years and years that tell us what the impact of
social connectedness is on our health and well-being. So, the big question that
Ayelet and I are working on, is not just what we know,
that social connections influence mental, behavioral,
and health outcomes for both adults, families,
and also kids. But what we don’t know is
how we translate that knowledge into better care coordination
and treatments for people. So, the work that we do,
tries to take us to the next level of that thinking. When we talk about
social connectedness, we are not talking about
Facebook and LinkedIn and all the different ways that
we have to connect to people, but really what our social
supports look like — for example, those who help us
when we need help getting home from the doctor,
when we need someone to call at the end of a hard day, maybe
there are colleagues at work, at school or even at church
and other social places. And for some of us,
that is quite a lot of people who are connected to each other. For others it’s just one person,
and for some, it’s no one. And what’s really complicated
about this whole topic, is that while we know
this is true, and just about everyone on the
call I’m sure is embedded in some kind of social support
network that helps you or cannot help you at certain
times, it’s just really invisible. It looks different,
everybody has a preferred social connectedness to show up
in different ways. It’s invisible not just to
ourselves, but definitely to the people who are
trying to care for us. And that makes it really hard
for us to translate this knowledge that we know
about how important social connectedness is to care
and outcomes. And to realize how to integrate
that into other kinds of things. I think we all have a personal
story about social connectedness and I always like to share mine. One of mine. This is my mom. And when she was diagnosed with
breast cancer, she had several
chemo treatments. And I used to always go to the
oncologist’s office with her. And one time I was there with
them and he asked her how she was doing and,
actually for the first time, I heard her talking about what
we all know now, we call our
social determinants. She was talking about how it was
hard for her to get up the stairs and move
around the house and get to her appointments
and how she was having more and more trouble making
meals for herself. And [Clears throat] the
oncologist, not surprisingly, asked her,
“Well, do you have anyone to help you when you need it?” And she said, “Yes, I have three
adult daughters in town.” And I swear that oncologist
would’ve winked at me, if he could, because he was so
pleased with that answer. And we just kind of went on. And we talked about her chemo
care and didn’t really talk more about those factors. But what he didn’t learn that
day was that my mother was actually pretty socially
disconnected and her care coordination was
not happening. Two of us sisters are estranged
and another sister had had a stroke and her husband
was in the hospital, and this was the first time
that I was hearing my mom talk about these things. And this is all happening — I’m
a social – a network scientist. I think about these things. I had been working with Ayelet
for, I don’t know, six years at this point. I’m thinking about how to
coordinate care for people and I had no idea that this was
happening in my own family. And because of that lack of care
coordination for my mom, she made a lot
of poor decisions that affected
her health outcome. And you know,
this is just my example. You know I have slaved over
looking at all kinds of research over Social Connectedness as a
Social Determinant of Health. And I found this quote one day
and so it’s just a quote about the risks of
social connectedness for women with cancer. And you can see,
those who are socially isolated have more risk of recurrence, risk of dying from
breast cancer, or from any other cause as well. So, that makes it for me,
something that made sense. But for kids with special
healthcare needs and the
families that we’re working with,
this is the same. And so I don’t think I’m alone
in this. I think lots of that people have
their own stories of social connectedness
and how it affects our health and wellness. And the crazy thing about this
is we are constantly trying to figure out how to
integrate that into care, without really good tools to do that. So, what we’re doing with our
work is asking the question, “How do we prevent this?” How do we prevent moms like mine
from not looking like high risk, risk of adverse social
connectedness. How do we connect care for
people like that? And what are the tools that we
can do it with? And Ayelet and I have had
the great pleasure of working together for years to
answer this question using system science methods. And one we used called
Network Analysis. When I say that,
I refer to networks. And I just wanted to
clarify this real early on. What is a network? We are talking about basically
people and resources and things as these kinds of nodes
and lines. In a network like this the nodes
can be people, places, they can be the resources that
people access, all kinds of things. And then the lines between them
are the relationships between those nodes. And so that can be anything. It can be client referrals,
care coordination, who you go to
for help and support, however we conceptualize that. So basically, we use this
method, we refer to it as social network analysis
and we use that method to — in combination with something
called Network Science. Network Science is a really
simple science, actually, it’s the science of how things
are connected. So some folks are using it to
look at how things in the natural science world
are connected. We often use it to look at how
people or organizations are connected to each other. And it can provide this really
unique lens to some of these problems
that we’re trying to solve. And one that is very basic
that it’s kind of grounding for I know all of the work
that I do and definitely the work
that we’re gonna show you as an example here, is this
Network Science principle that more is not always better. When we talk about social
connectedness and care for people,
we often narrow that to a problem of social
isolation. And social isolation is
definitely a problem, but it’s only one type of
adverse social connectedness. Other types of adverse social
connectedness are when people are,
for example, the mother connected to a lot of
people at home and family, but perhaps, has one or more
children with special needs that she’s working with and a
mother that she’s looking after. So she could be surrounded by
family with a bunch of people that depend on her
and she’s really suffering from some adverse
social connectedness. So this idea
that more is not always better is pretty grounding. So, just real quick,
these are just some examples, I’m not going to
go into depth like I would in some presentations,
but what we know is that this is you — that little
orange node there. When those of us, everyone on
the call is connected in our little social
sub-networks here. Those connect to each other
and we tend to think that that’s kind of the best way
to be embedded in networks. And there’s a theory called the
strength of weak ties. In Network Science a guy named
Granovetter in the ’70s did his dissertation work on
this and asked how people got
jobs. And found that people got jobs
when they reached across these strong connections of
their social support networks into these weak ties to other
social networks. Where people who they are not
connected to tend to know things they don’t know about
or have access to resources they don’t have access to. And this is really, though,
a more is better concept and kind of the way
that we think we should be connecting people
today, but it’s really overwhelming
and it’s not, necessarily, the best way. There’s all these alternatives
in Network Science that show us that perhaps we
could get the same advantages of connecting to these
social sub-networks, without having so many
connections, and instead that’s what might
allow us to actually have some new advantages and
successful relationships where we can branch out to other
new social sub networks. So, for people who are
managing large, interorganizational networks
of resources in a community, or you’re looking at a person’s
personal support network, there’s just all these really
interesting applications of Network Science that can
inform new ways of thinking and almost test our assumptions. And so help us
understand these variations that people are presenting when
they come to the clinic, or when they come to be screened
that will allow us to better understand whether
someone is embedded in a strong network of coordinated
ties or not. So we really do believe
we can manage these kinds of relationships,
for our impact, for effectiveness, for outcomes,
but that we need data to do it. And that data is actually —
I’m gonna skip this slide — comes from this method
that I talked about called social network analysis,
it’s just a method that collects data on who’s
connected to whom, how those connections vary
and change. With focus on patterns of
relationships, we can do things like look at
the type, extent, and the quality of connections. And I did mention those nodes
can be almost anything. I’ve worked on terraced networks
at the RAND Corporation where those were beliefs in
people and sermons, but it doesn’t —
the node is just the object, and then we’re looking at the
lines between them and how they are connected and
the variations that presents. So we have taken all of that
thinking and sat around and said, “What if we could
create tools for people? Tools for providers, and tools
for families and their children that could help use
this kind of science and use this kind of method
in their care? How would we use these kinds of
innovations in ways that could let people,
in a quick way, understand how different a
person in front of you is, in terms of whether they have
trusted people in their lives, whether they can depend on them
and if those people are connected to one another. So Ayelet and I worked through
that thinking we’d created something called the
Person Centered Network App, which is a tool that can
assess the strengths and gaps in a person’s
personal support network, letting providers use the
information to make personalized assessments of
their needs for connection to community
and systems resources. And we want to show you that
and tell you about that, but it wouldn’t really make much
sense if we told you all about that without telling
you a little bit of the story of how we got there and the
research that has led us to believe that this is a really
validated and reliable way for people to do this. We had the amazing opportunity,
one of those synergistic moments where someone in both of our
lives said, “I think you two should meet the
a power of a network.” And one day Ayelet and I
were connected. And the woman
who connected us said, “I think you’re both doing
something pretty interesting, and if you came together,
maybe you could do something that’s really impactful.” And so I met Ayelet who was, at
that time, running WONDERbabies, a Partnerships
for Health Initiative that funded by
the Colorado Trust. and this was an initiative that
had 12 partners at the time. More partners than 12,
and they were working to coordinate the system of care
for babies and young children with special healthcare and
developmental needs. And Ayelet said to me,
“I think you are doing some mapping stuff
that would help us.” And together we started doing
this project together and I’m going to hand it over to
Ayelet right now, so she can tell you
more about that. Talmi: Hello, everyone
and thank you, Danielle. So, as Danielle was saying,
we had the opportunity to have our fields of study
collide basically. And have spent many hours,
and now years and years over a decade of collaboration,
because WONDERbabies Partnerships for Health
Initiative was really interested in the
networks of support around babies and young children
with special healthcare and developmental needs. And this is just some
information about the impact of children with special
healthcare needs, thinking about social
connectedness and what we knew about social connectedness. And so,
we’re talking about a population of more than
13.5 million children. And of the children who are
in their early years, only 17% are actually getting
the services that they require for the delays
that have been identified. So, we know that families are
telling us 30% of the families do not have access to
coordinated Comprehensive Care. Thirty-six percent of children
don’t have adequate insurance or benefits to cover
the costs of all of their
required services. Twenty-eight percent of children
and families have difficulty getting a referral. And 18% have unmet needs. And what we also know about
these figures is that they are likely a
dramatic under reporting of what the actual needs
and circumstances are. The partnership for the WONDERbabies Partnership for
Health was tasked with creating the ideal system
of care, and coordinating services
in Colorado, really focused on the metro
Denver area. And one of our first exercises
with the grant was to draw the map. What does the current system of
care look like? What will WONDERbabies do? And then
what is the ideal system? And so by the time
Danielle and I met, we had three beautiful maps. Unfortunately the maps had
no data behind them. And this is where the
partnership and collaboration with Danielle was so amazing,
because we were able to actually measure
the system of care for babies and young children with special
developmental needs. And this slide just talks about
the outcomes if we don’t have a system of care that is
comprehensive, coordinated, and culturally responsive. And we know they are factors
outside of the system of care, that dramatically impacts
health outcomes and well-being, and family outcomes. So, I am a child development
researcher, and have been
very interested in systems, as they impact child
and family well-being. This is Bronfenbrenner’s
“Human Ecology theory. Depicted with some pictures
we can look at with the child in the center,
and all of the surrounding, formal and informal,
sources of support impacting the child and
family’s well-being. Danielle came with this
tremendous expertise about Network Science,
and what we call, in child development research,
systems of care, Danielle calls it networks. And so it really provided us
with an opportunity to empirically look at what the
system of care was. And this work is being done
in Colorado, the context of our early
childhood Colorado framework, which again, you will see here
on the right. The concentric circles and these
systems that are networks. And so in the context of our
early childhood framework, we were really interested in
understanding for a particular population of
young children and families, who required additional access
to services, resources, and support, what was really
happening in the system of care. And WONDERbabies was
a five-year initiative, and one of the things that we
did, was hosted annual meetings, where we invited people who
worked with babies and young children with special
healthcare and developmental needs to
come together and help us understand the
system of care. And at one of the very first
annual meetings, we had a room full of about
200 people. We had construction paper strips
on the table. And the tables were mixed,
lots of different disciplines lots of different service
sectors represented, lots of different perspectives
on how we have a system of care for children with special
healthcare needs. And what you can see here,
is that each table created a different system of care, even
though everybody in the room was working in the system of
care in Colorado. And we found that that was the
anecdotal experience that we were getting
from providers and professionals in the system. Is that it looks different,
depending on whether I’m in early intervention
or I’m in healthcare or I’m specialty care. We also, at every meeting,
and every opportunity, engaged with families. Because as a clinician
and provider myself, I was every day,
hearing stories and experiencing and sharing in the lives of
families whose children have special healthcare needs
and helping them navigate what to me as a provider,
inside the system, was a really complicated
system of care. And so we were always interested
in and bringing and sharing the stories that
families had, and again in the literature,
this was deemed as anecdotal evidence,
even though this is a clinical reality for families. And the systemic reality —
we did this a number of ways. Sometimes we had families come
and present, which was very powerful. We also started looking to see
what families were telling us about their system of care. And so here you see
a family map. And this Danielle and I
found online. And we think it really
represents and depicts what families, who have children
with complex healthcare needs, among other complexities,
have to navigate in terms of the system of care. And so you see here
different sectors, different types of services,
and the lines that Danielle was showing you
with the network earlier, are also lines that are here. And what we think of about
is the tremendous burden that falls on the family
to be the connection, and really the core of this very
complicated network. So, Danielle introduced us
to a tool that she created called Partner. And one of the very first things
we did together, was to survey the WONDERbabies
partners, and say to them,
who helps you in the system work with you on helping to care
for babies and young children with special healthcare
and developmental needs? What you see here is
the result of the big survey. So what we got,
it was a self nominating, so I could say,
if I worked with 10 people, I would nominate those 10
people, or organizations. And those people would then be
sent a survey, and they could report on who
they work with, and then the different
dimensions of the survey which we’ll talk about. And what you see here
are the nodes. And in the center is really
what we think about as our Metro Denver hub of
organizations and partnerships. So, what’s represented here is
450 different organizations, programs, services, or agencies. And 723 connections,
relationships, or partnerships. Well, as you know, when people
give you information, they’re really interested in
where they are in the network. And so everybody wants to know,
which node am I, when they responded to the
survey? We also were able to, because of
the way we collect data, code the data for who in the
network the different partners and agencies represented. So what you see here are the
mental health agencies, and entities in Colorado. In the large blue nodes,
represented here. We were also able to,
because of the way the data were coded to look at
different networks by county. And you can already see the
distinctions between one very large,
more suburban county, and then a more rural county. So Adams/Arapahoe/Douglas
counties comprised a very large area
in the metro Denver area. And then Weld County is a more
rural community. And we looked and said,
“Wow, this is the system of care for babies and young children
with special healthcare and developmental needs
in Colorado, look how different it looks when
we pull out mental health. Look how different it looks when
we pull it out by county. We continued our work,
we presented the network, and there’s a lot more data
about the overall system of care
network findings. But we presented it
many, many times with different
stakeholder groups. One of the things
that we wanted to do, because it is one of the goals
of the WONDERbabies initiative, was to have our collaborators,
all the people involved in the system of care,
tell us what the ideal system would look like. So what you see here,
is a group of providers, and agency representatives,
and program heads coming together to create
the ideal system of care. And every one of these boards
looks a little different, just like every one of the
construction paper systems of care looked
different. And so that was a really
interesting exercise, because then we were tasked
with going from if this is the system of care
as captured by Danielle’s network survey,
how do we move the system in the direction of the ideal
system of care? This was very interesting,
because one of the things that we did was present both
the network survey results, and these ideal systems to
families, and families said to us, “That’s
all great,” and, “Look at how fancy your system
is and how connected it looks and trusted and related it is.” But that’s not the reality
for us. Our reality is that
our systems of care, as families,
look different. They look like
this giant rubber ball we construct it as needed
and we have a connection here that the survey that you are
doing don’t capture. And what it showed us,
as researchers, is that there is this
fundamental disconnect between what we think of as
the system of care, and what families,
who are getting services from the system experience as a
system of care. And we knew this from the data
that our families have given us. So, this is yet another example
of a family map. And I always say,
this is the one slide, if I were on a desert island,
never mind that I won’t have A PowerPoint projector or
computer, but if I were on a desert
island, this is the one slide that I would take with me
because it is so emblematic of the struggle that we face. And what you see here is
the family identifying their support resources,
and then what you don’t see are all of the arrows,
some of them bidirectional, some of them unidirectional,
that have to be present for this system of care to
function in a coordinated and comprehensive way. So Danielle and I were struck
by this finding that the systems that families
were telling us about that we knew about clinically
and anecdotally were not represented in this
amazing network survey that we had done, and we’re
really proud of that work, too. And what we did was, we said,
“Okay, we need to figure out what’s going on with families.” And we developed a survey
for families, and what you see here
is the methodology for that. We have three phases, one was
that development of the survey, and I know this from working
with Danielle for so long, these are called egocentric
network surveys, because we wanted to get
the ego-ness of families. What is the families’ perceived
network – social support network, around taking care of
their child with special healthcare needs? So, the first thing is
developing the question. The questions were actually
similar to the questions that we were asking,
the systems-level people, only it was
the family was referring to their own network
about their child. The second phase,
we pilot tested the surveys with families and got feedback. And then the third phase was
administering the survey to to 21 families, English- and
Spanish-speaking families. We collected a lot of
information, in all three phases,
but what we have here are the results
from phase three. So you can see that
these are families, who clearly have children with
special healthcare needs. Even though it was only 21
families, there were 208
relationships reported, and each family reported an
average of six relationships, which is a lot to navigate when
you’re also trying to care for your child
with special healthcare and developmental needs. This is a breakdown,
next of the relationships. Many with community
organizations about one quarter, medical specialty providers were
very prevalent in the responses, family members,
school-based resources, primary care providers,
and then to a lesser extent, home visitors, friends,
and early interventionists. There was this other node, was really what we think of
and label as informal sources of support. And what you see here are
the types of support. In this network,
that a family said, “We are getting support,”
43% was emotional support. Only 25% was therapy or
intervention support, 21% healthcare,
and then 14% financial. What was striking to us,
was that 55%, more than half of the families,
when they had to answer, “how effective this system of
care, this network is,” 55% said it was either somewhat
or not successful in helping them address
the needs of their children. And almost 60% reported that
people in their networks, only worked together
a fair or small amount. So this told us
that families relied primarily on their
informal supports. So, there wasn’t a particular
health system, or other system node that was
really represented and said, “Oh, this is really what
families rely on.” It was really, mostly
informal support. And that families,
while they reported, and what you see here
in the figure is that families are reporting
that the source of the support are trusted and important,
but that there are many uncoordinated relationships
in their networks, even in these smaller,
self constructed, and often by necessity
constructed networks, that there was not enough
coordination. And what was striking to us, was
that when at the end of the day when we looked at
the system networks, that are here on the left,
compared to the family networks, the networks were really
different in appearance, which tells us there are fewer
relationships, fewer resources, and a lot more complexity. And the families were
intentionally at the center of the family networks,
but there also wasn’t that much overlap across
what families were saying. Families were having to
basically almost individually construct their
networks of care. We know that families
from this work, families have a preference
for tapping into their informal supports for help
first, and, when that’s not enough,
they certainly use the system, or when the system provides them
with things that their informal network
cannot, we also know from the vast
literature that Danielle mentioned,
on social connectedness that people want to
help themselves. They want to be resilient,
and they are working really hard to create
successful networks of support that provide comprehensive care. Right, so
we’re switching back over. So this is
all really interesting, but Ayelet and I have
a habit of saying, great research,
let’s publish it, but let’s, more importantly,
see how we can use this to help people. So we’re always asking these
questions to kind of test these assumptions and
push us forward in our thinking. Towards the end of this project
we started wondering, are we even measuring
the right outcome? We have been looking at
systems of care, organizations
and how they work together, but isn’t it true
that we actually are wondering how people fit into that system. So we started asking
why do we expect people to fit into the system,
why don’t we expect the system to fit around people
and become adaptable to these variations
that they’re presenting? And then we said,
if the system is coordinated, let’s change our
frame of thinking, it should be reflected in the
personal networks of families, so that our new outcome measure
is when 100% of families say their personal networks
are coordinated, then we know
the system is working. And this is a huge shift
from how, even to this day, that we think we tend to measure
systems working well, when we think
the system is coordinated. But, we’re really pushing
ourselves to say, let’s try to measure a different
level of analysis which is the families,
and not be satisfied that these systems are working
until all of our families say that’s working them. So what we did with that
information, is we said, okay, there are two
things that have to happen here. What I went through in the
beginning was saying, these are invisible, we just do
not know what these look like, how was my mom’s doctor supposed
to know to ask, okay, great, three daughters,
do they coordinate care? Do you trust them to help you
when you need it? Can you depend on them? How do we make visible,
the personal support network, the needs of people,
and then, following that, how to use that information to
build adaptive systems? So, we know
we can’t do both at once. So we said let’s try to figure
out how we can better understand these personal support networks. So, that’s where we
turned that work into a tool. And we used to end this talk
here and say, one day we’re going to build
this tool that providers can use
to do this, and luckily,
we’re invited into the innovator society with the
community first foundation, who said, “you can pitch to an
audience and get funding,” and we actually won funding to
build this tool and we did it, and so the tool actually is an
app that we can put in the hands of a provider
who can then — and I actually refer to
these providers as the helping professionals,
so sometimes this happens within a clinic, it can happen with
front-line staff, this can be teachers or
counselors in the school. It can be people in home
visitation settings. It’s usually some helping
professional whose job or hope is to
help support families. And so with this tool, what
we’ve done is we’ve created — oop, this was my slide on that,
on who could use it — actually our first uses were
firefighters in Tulsa, Oklahoma, who were using it to look at the
social connectedness of high utilizers
and their 911 system. So what we did is we created
another kind of survey, a set of questions
that someone can answer to really understand
these things. The way the tool is built
is that you can put in any screener you want into it. In fact, Ayelet’s clinic
uses a different screener than some of the other clinics
that use these kinds of things. And it’s adaptable
for all of that, but the default one says
basically, something like how many people
do you live with? What are your
most pressing concerns? Which we list,
of social determinants like transportation, childcare,
employment and so forth. How often do you feel
lonely or isolated from those around you? We ask people to rate their own
health, and then we ask them, how open are you to receiving
help? because Ayelet has taught me
that this is probably the first intervention with
families is if they don’t want help,
then it’s really difficult to move forward,
so we need to figure out what’s blocking them there. But then we
ask these questions like, who helps you
with the things you need? The kind of questions my mom’s
doctor asked, but then it goes forward and a
little screen like this appears, and the person with their finger
gets to connect those that are
coordinating care for them. They’re answer questions like,
to what degree do you depend on this person for care,
to do what degree do you trust them to help you? So, in just a minute or two you
can quickly get a visual of whether a person
is embedded in a coordinated network of care,
or not and a little bit of information
about that. And this is where we see so much
variation. You almost cannot predict what
that person is going to put on that screen,
and the degree to which they will rate these others as
coordinated, trusted, or dependent or not. And then on the last screen,
it takes from their list of things they said they need,
it puts them, lays them out on the screen,
and then we ask a really crucial question. We say, who can help you with
the things you need? This is a real deviation from
saying, great, what are the three things you
need? Let’s say you say child-related
issues, employment, and food, typically we say, great,
here are three referrals. But what we’ve failed to do and what this research
taught us to do, was to honor the fact that
people want to help themselves. And so we asked them
the crucial question, who helps you with the things
that you need? And they can show us. And again they’re swiping with
their fingers to these, and really quickly we can see
their most pressing need with the least amount
of social support. So at this point,
we’ve both communicate to them that we believe
they can help themselves, we actually want to integrate
that kind of personal resilience
into their care, and we want to help focus on
their most pressing needs that they don’t have
other help with. And what we’re finding,
by using this method, is that people are more likely
to follow through on their referrals
that you give them. Because we’ve done
all of those things. And it’s very simple, the whole
tool is really quick, but it’s about a philosophy
of treating people like they want to help
themselves, it’s about making a system
adaptable to the different kinds of
characteristics that they’re presenting. And not assuming just because
they said food, that they should be sent to
a food bank. And then it’s connected to
a resource database, and the way we’ve designed it
now, is any resource list can be put in there. So you can connect to a 211
or you can put your spreadsheet of resources
right in. So, what we are doing with that
on the backend, though, so that helps one person
at a time, right. We can screen one person and
help one person or one family at a time. But on the backend,
we can take that information and can hold it so we can watch
how that changes over time. We can screen people
at all visits, or one, we can look at
the population level. And how our whole community of resources and people
are connected, and we can start to
divvy that out by demographics. You know, gender, neighborhood,
need, disease that they present, and try to better understand,
from a research lens how social connectedness and
adverse social connectedness of people influence their
outcomes. So, it’s letting us
both prioritize, and build personalized referrals
for people. But also on the backend,
from a population health plan, a community health plan,
better know how to invest in these types of networks
and where and when. So the providers who use this can use these kinds of systems
then. And the very last thing I’ll
present on, I know we’re getting to the very
end of our time here, is — you can probably tell
I like to tell stories — so this is the last story. This is my daughter, Olivia. On this slide she’s eight,
she’s now 10. And actually, when I bring home
technology and things like this, my kids think this is wonderful. I have three little girls. And one day she asked me if she
could actually use the screener. And I let her,
and she answered all those questions
I just showed you. Like, what her most pressing
needs were, she rated her own health, which,
asking kids to rate their health is a very amazing practice,
if she felt lonely or isolated, who helps her,
she created her network, showed me who she trusted,
who she was dependent on, and it blew my mind. Because I am her mother
and I know her well, and I think about this stuff
all the time. And I had no idea that she would
rate things in the way she did. And the way she constructed
that support networks. She uses it often now,
especially when she’s having trouble talking
about something. We have revised the list
so that things that kids have to pick from are
things like, getting angry, being hungry, being bullied, having trouble talking to
others. And then what we’re doing is,
building an evidence based, age-appropriate
resource database, so that if a kid picks for
example, being angry, they can click on a node
and open up calming down techniques
right inside the app. What we’re finding is that it’s
great for kids, but they lean in, actually when
someone gives you a tablet, I don’t care
if you’re an older adult, a young person,
people lean into this approach, they love interacting with it
kids feel like it is a game. But all the while it’s giving
them a new voice, for their mental and behavioral
health needs that they don’t find easy,
that they can easily talk about. So we’ve just been really amazed
at the way kids in particular will use this kind of tool to
send up red flags. Smoke signals, I always say
for what’s going on. And if they’re addicted to
their technology, this idea that we can use it
for their mental and behavioral health needs is
just fascinating. And again, on the backend,
if the school’s using this, they can look at what all the
fourth-graders are saying they’re struggling with,
where they have social support or not, and if we can get anyone
to fund mental and behavioral health in schools,
which I am still hoping happens more and more, they can
target their intervention, to the greatest need
that those kids have, based on an evidence base that
they have built up over time, to be the most effective. So, I think that’s it. Those are just a couple of
quotes. You know, what Ayelet and I are
thinking about is how this creates a network of
care plans for people. This is a back of the napkin
drawing of how this is going to influence at our state
policy levels all kinds of different levels. You can imagine the way that we
dream about how to solve big problems using
Network Science approaches. We hope we have shared a little
bit of that with your today, in interesting ways. We’d love to answer questions. And just thank you, again,
having us on this webinar today. Linares: Thank you so much,
Doctors Varda, and Talmi. What an informative and
interesting presentation. We really appreciate you taking
the time to share your expertise with the MCH community. We have our first question. They ask, “How do you access
the app?” Varda: Oh, sure. It’s actually
on the Play Store. Or the, uh I forget what that,
I’m an Android user. You can download it the app from
those stores. And when you do, it’ll ask you
to register an account, and then you can get access to,
and keep in mind, it thinks you are a provider
so it doesn’t think you’re going to go in
and put all your info on your social support network. It thinks you’re a provider so
it lets you put in basically three patients that you can load
in there. And then you’ll get assigned a
survey that’s called The Social Determinants of
Health Survey. One thing I will say, because
it’s probably going to be a question in a moment,
the kid’s version of the survey, you won’t see the load. And the reason for that is,
there is so much embedded in asking kids these kinds of
questions. And this is why I need Ayelet. I am no expert on how to address
the issues that kids present. So we’ve only opened the kid’s
version up to organizations that have programs, where they
work with kids on those issues. And so if someone wants to be a
part of that, we’ve just been really doing it
in a pilot-testing way, to learn more about that
and deal with both the legal aspects, and that
ethical aspects of who gets that data. So, if you are interested in the
kid’s version, e-mail me. We’re actually always looking
for people who are willing to try this out and practice so
that we can build that evidence-base
and knowledge-base out. Linares: Great, thank you. The next question is about
the discrepancies that you found in terms of between
a system network and an individual
family network, how did you analyze those? And could you share a little bit
more information with dealing with that? Talmi: So that was one of the
most fascinating findings that really inspired us to do
the family network surveys. And we didn’t find families
in the larger system of care survey,
which was the first clue that there was
a huge discrepancy. And then what we did was
we looked at the different sources
of support in the two networks. So we we did a comparison one. We had the family data, we looked at
the sources of support, so, who are the nodes. We looked at the types of
support provided, so those findings that I was
talking about earlier, the levels of trust,
the what this particular entity in the network does,
were really different in the family
and network level surveys. We have a couple of papers
that describe this work. And that’s where you can see
more of the discrepancies and they’re very clearly
articulated there. We also know that it is
the anecdotal piece, so what we have been trying to
do is gather additional data, both from families and from the
system of care, to more clearly articulate
the discrepancies. I think one of the challenging
things in doing this research, is that there’s also
huge discrepancies and I would say disparities
in how things are funded. So the way we fund services
and supports and systems is that the money goes to the
agencies and organizations. I work for an agency and
organization that gets funding to the work that I do
in support of families. Families, despite having their
own personal networks, don’t get the resources,
typically, directly to them. And so that is another
discrepancy. And when Danielle
was talking about, the, where do we take this from
here?” we have the opportunity to do things,
obviously, at the child and family level,
at the agency and organization level,
at the systems level, but we also have to take this
information and use it at the policy level. Linares: Great, thank you.
We have another question. About if it’s possible
for other researchers to use this app
in their research? And if you’re planning to build
any more functions on the app? Varda: Yeah, and you know,
I saw the question, too, about the name of it. So I’m just going to put up the
slide up that has the name. Yes, whomever is asking that, is
definitely on my wavelength. Because when we have build
tools, that visible network lab, we always build the tool so that
it is adaptable. And I what I mean by that is, because we actually work in
service for communities, and we know that communities
are different. So we don’t hardcode
almost anything. What I just showed you is
the context in which I presented the tool, which is
on the social determinants of health, in this setting. It’s actually just a tool to
collect egocentric data so it has a survey builder
on the backend that you can put in anything. I mean, we’ve had people use it
for egocentric studies that have nothing to do with
health and wellness. It’s a tool
that collects that data. So you can put basically any
questions in there. You can put any of the little
network questions in there. The little features we have
that for example, connect people to resources is a feature
that you can turn on or off. And on the backend,
what happens is you can download all your data,
it’s very clean. We are data scientists. And our goal is to take complex
data sets and make it usable for people in practice, which
isn’t usually a researcher. We are talking about people
who don’t know data. And then we have the dashboard
to link up all that stuff. We are always building little
features and stuff, it’s usually in the context
of a project. Actually Ayelet and I and some
partners were just funded to use the app in some integrated
pediatric care clinics. And it’s primarily
a design grant. What we’re going to be able to
do with that is really work on how it works in the workflow and
what the UX/UI needs to do to change to really fit
families and kids. And so I anticipate a lot of
features will come out of that. If you are a researcher,
you can contact us. We’re researchers, too,
so that’s how we built it. And then we put it
in context of use cases to put it into the world. Linares: Great, thank you.
We have another question. Do you see applicability of
using the PCN and addressing communication between parents
and healthcare providers? Talmi: This is a phenomenal area
that we are so interested. I work in
a primary care setting. And we ask families to give us
so much information. And then,
one of the things that we do, is train our providers to
actually look at the information
that families give us. So this app as a communication
tool, is really critical. I treat it as any other
information that we’re asking of families. It makes visible things that we
might make assumptions about, as in the story that Danielle
shared about her mother. And it allows us to have
conversations, not only about present state,
which I think is great and really important to have,
but about what is the desired state,
and how do we get there. And how do I, as a provider, and
professional, partnering with families
help them move into their desired
direction. that is going to promote the
optimal health and well-being and development. We absolutely see this
as a communication tool it is really critical to make
the invisible visible, and up till now,
unless we ask, we do not know as providers
and professionals working with families. And there are so many questions
that we have to ask, that we don’t always focus on
something as important as social connectedness
and social networks. Linares: Great, thank you. Our next question is asking
about the screening question. So there several screening
questions that you use at the beginning of the app. Has that been
a barrier for some people when they are using the app? Talmi:
Is the question asking about
whether people are reluctant to answer the screening questions?
Linares: Yes. Talmi: I do a lot of screening
in primary care, that’s our signature thing
that we do. We ask families
all sorts of questions. I think
there’s some belief that, if we don’t ask, the things
don’t exist and aren’t challenges. And what we’ve found is that
when we ask, everybody, every time, it feels like we
really care about the answers, and we’re not necessarily
targeting people and saying, we are asking about depression
because you look a little sad. Right? Then I think people feel
much more reluctant to answer. But our policy in our clinic
is that we ask every time, every person, it’s a universal
screening approach. And what we find is that
people are very appreciative to be able to tell somebody,
in a healthcare setting, about other things that are
going on, that are impacting either their
healthcare, or their child’s healthcare. And again, back to the
communication question, it’s a really important tool
and mechanism for letting the system of care
know what is happening, and where you’re struggling. You can have the best medical
care in the world, but if you can’t make it to
appointments because transportation
is a barrier, it doesn’t actually matter what
type of care you have. Varda: I might just add one
quick thing. When I read the question,
it also might be a little bit about whether it is a barrier
for people to answer a series of questions,
just any questions, but two things on that,
we have seen when we put a call out
for pilots for this, we did not target
clinical settings, because the runway to get new
technology, new innovations, and new cultural
ways of doing something into a clinical setting
is very hard, as probably everyone
on the call knows. Getting it into a health and
human service organization, where just about nobody is
paying much attention to them, nobody’s trying to
build technology and make their lives better and
providers ways easier. They adopted it very quickly. And so we’ve had a lot more use
outside of a clinical setting or like these firefighters,
in home visitation settings, in schools. And because there isn’t the
burden of the quick visit, the 10 minute visit, we haven’t
had too much complaint about it. In fact, it does feel kind of
game-y and so people do look forward,
instead of someone — I know my doctor sits and enters
stuff on a computer while they ask me
if I feel depressed and instead we hand then
something and they get to interact with the questions. And so we imagine
in a clinical setting, in this new funding
that we have, we’ll have to deal with this in
a lot more detail. And it may be a frontline staff
type of thing where the provider
gets a report. The app does, it creates a
report at the end of it. So the actual workflow
of integrating this it is challenging and it’s a
really good question on both of those dimensions. Linares: Great, our next
question is asking about how this information can be
shared with local communities without compromising
confidentiality, so you can bring about
system change. Varda: This came up when we were
doing the WONDERbabies work. And whether we show people’s
names or not. So you can turn the labels off
and on. But this is actually
a Network Science and network analysis and if
there are other analysts on the phone,
you can feel the pain. How we share this information,
because when we can label those nodes, it brings
this enormous amount of value. And when we turn them off you
lose some of that. but you don’t lose all of it. And so, the best thing,
the way I would answer this, is it’s different across sites. You can aggregate the data,
you can definitely communicate this data back
in ways that can be beneficial to the community. So at the system level,
there’s lots of ways to do that without labeling the nodes. At the personal network level,
of course, we wouldn’t share
the personal data, but you can share these trends. You can say most kids
are struggling with talking to adults when,
go figure, we thought they were all hungry. You know the thing that they’re
really struggling with is this. And so we can use the data
in that kind of way. So, there’s a lot behind that. Talmi: And our experience,
has been, you know, Danielle has done this with we
100s, 1000s of communities, at the systems level, and
there’s a way of aggregating and de-identifying the data that
makes people very comfortable. Our experience with WONDERbabies
people were looking for themselves in the network
and really wanted to know. and we didn’t and we rolled up
to mental health, or public health,
or at the level that we felt comfortable
protecting anonymity, what we will say is that in more
rural communities, and in networks
that are less dense, you have to be really careful
about the way the data are aggregated and presented,
because it could be identifying. There’s not as many
providers. Public health in the metro
Denver area is really different than public health
in Weld County. Varda: Yeah. Linares: Great, thank you.
We have one last question. How can individuals and families
help improve their social connectedness,
what would you recommend? Varda: Oh, I’m so sorry,
can you repeat the question? Linares: Sure. How can individuals and families
help improve their social connectedness,
what would you recommend? Varda: Oh. [ Laughs ] Talmi: I think that that’s
a great question. And it is something that
we in healthcare, and in systems of care, really
need to be thinking about. And one of the very first ways
to do it, is to actually use
a tool like the app, to identify what is
a network of care, and where are the gaps,
and where are the areas that an individual or a family
doesn’t feel supported? And where are the areas
that they do feel supported? And have rich resources. We don’t often do it,
and Danielle talks about and when we present together,
sometimes in a teaching environment we
actually bring those boards. And we have people create their
own networks, they tell us
what their networks look like, not the ideal networks that
we were talking about, but what is your network,
and when you start as a person or family, really articulating
that for yourself, and articulating it in the
context of support resource in your network,
it then becomes clear, the areas where you could use
additional support. And then you can work with the
supported elements of your network,
to shore up those resources and in those areas. And can identify the gaps
and the needs. And then move toward that. And we, as health professionals
and providers, have a responsibility
to help mobilize, because we know how important
social connectedness is to health and well-being
and outcome, our responsibility
is really to mobilize that. And to help families
articulate it. I think one of the most
fascinating opportunities that we have with the tool
like this, is to look at change over time. We have so many programs
and services, and supports in the system of care,
that target improving health and well-being,
but don’t ever measure the impact
on social connectedness and social networks. And if we looked
at network change over time, we would have yet another way
of measuring the impact of our intervention. Varda: And I’ll just add would
like to add one quick thing, from the Network Science
perspective, I can’t say this enough times. But people really want to
help themselves. And so,
I think what’s really crucial, is that moment
in a screener like this or the conversation. And I’ve been looking at all of
the research and theories on suicidology,
and creating a feeling of burdensomeness for people
by basically saying, what are your needs,
let me give you the resources. It could create it could create
a sense of burdensomeness for them, and we know from this
depression research and suicide research,
that when people feel like more of a burden on a system,
they tend to become more withdrawn
and more depressed. And so if we could take a moment
and say, who helps you, how do you help yourself? And we honor that about them,
and we make them believe that we think they can help
themselves, we already moved them towards better feelings
of belongingness, better feelings
of social connectedness. And so that kind of way
of treating people, to me, like outside of
the clinical setting, I’m just like when can we get
providers to treat families, and patients like they want to
help themselves? And ask them how they do
in their care. I truly believe that
that one little way of working can mediate
all kinds of ways in which people believe
that they can help themselves to increase their
social connectedness. Linares: Thank you so much,
again, for a great presentation and for sharing your expertise. If you did not get a chance
to ask your question for the speaker,
please still feel free to submit your questions to the
Q & A fields. We’ll try to respond to your
questions after the webinar. We are now almost at the end of
our program. After this webinar,
you will receive a request to complete an evaluation. We hope that you will fill this
out and provide the MCH Division of Research with
feedback on today’s events. Your response will help us plan
future webinars in the EnRICH series. Thank you all for your
attendance and participation. I also want to thank
Rebecca Harnik, Jenn Rogers, and Jim Wetherill at Altarum
for helping to organize this event.

Comment here