Yale Psychiatry Grand Rounds: "Engaging Partners and Building Networks for Implementation in Health and Human Service Systems"

March 01, 2024

Information

March 1, 2024

"Engaging Partners and Building Networks for Implementation in Health and Human Service Systems"

Alicia C. Bunger, PhD, MSW, Professor, College of Medicine, Division of General Internal Medicine, The Ohio State University

ID11403

To CiteDCA Citation Guide

00:00Good morning. Thank you very
00:03much for that introduction, Dr.
00:05Teves, and thank you so much
00:07for inviting me to join you.
00:08This is really an honor to be able
00:11to share the system's work and
00:13the science that our teams and
00:15partners are doing here in Ohio.
00:16So thank you again for the invitation.
00:18And I'm really looking forward
00:20to getting to know many more of
00:22you at the end of the talk for AQ
00:24and A and the session afterwards.
00:25I'm getting to know more about
00:27the work that you do and exploring
00:30potential for collaboration there.
00:32And so I'm going to share my screen and
00:37get started.
00:46You can just swap displays.
00:49We can see the next slide there.
00:51So that's the presenter.
00:52So just do the display settings and say swap.
00:55Thank you. And we just did this,
00:57didn't we? It's OK.
01:00Perfect right there.
01:02Thank you and sorry about that.
01:04You know, who knows what technology,
01:07But all right, So again,
01:08let's get started.
01:10Good morning. All right.
01:12So I just wanted to disclose some
01:14funding sources in the beginning.
01:16Another thing that I wanted
01:17to point out was that I've,
01:19I've also been part of some
01:21research at Nationwide Children's
01:22Hospital here in Columbus,
01:24OH that was funded by Organon and Mark.
01:28All right.
01:28And before we get started,
01:30I really also want to
01:32thank my my collaborators,
01:34my partners in both the
01:37research and in the community,
01:39because without them this work is
01:41not possible and it just makes
01:43for a really rich scientific,
01:45scientific environment to
01:47be asking these questions.
01:50I thought I'd kick off and talk a
01:52little bit about how implementation
01:54of evidence based practices is
01:57really all about relationships.
01:59And then let's transition into
02:02some discussion about how there
02:04are potentially some strategies
02:06that we can use to build systems
02:08for implementation and then end
02:11with some discussion about engaging
02:14community partners in work in this area.
02:16So then we'll get into the good discussion.
02:21So the first thing that I
02:22wanted to to kick us off with is
02:25really about the importance of
02:26relationships for implementation.
02:28And these are relationships between
02:30individuals and organizations.
02:32And this work has really led to a
02:35couple important lessons learned
02:37for me that have sparked additional
02:40research questions along the way.
02:42So the first lesson that I learned
02:45unfortunately the hard way is that
02:47breakdowns in relationships have
02:49consequences for lots of things,
02:50including implementation.
02:53And I want to tell you a little
02:55bit about our gateway call study
02:56and this was a five year system
02:58demonstration that was funded
03:00through the US Children's Bureau and
03:02was conducted in partnership with
03:04Franklin County Children's Services.
03:05And I believe Connecticut also was one
03:08of the demonstration sites for these grants.
03:11And the purpose of this of this
03:13demonstration was to improve access
03:15to mental health services for kids
03:17who were involved in foster care.
03:19And we worked with Franklin
03:21County Children's Services,
03:22which is a county based child
03:25welfare agency here in Central Ohio
03:27to design and implement a service
03:30cascade that began with mental health
03:33screening and mental health assessment
03:35within the child's welfare system.
03:38And the assessments were then done by by
03:40Co located behavioral health clinicians.
03:43Based on the screening and the assessment,
03:45children were then supposed to be
03:47referred to treatment in the mental
03:50health system in the community.
03:51Service was for kids up to the age of
03:5418 and this was just for kids who are
03:56going into out of home placements.
03:59So sounds like it would make a lot of
04:01good sense considering the prevalence
04:03of mental health disorders among
04:05children in foster care so high and
04:07much higher than in the general population.
04:09And so our main question in
04:11the outcome evaluation was,
04:13well,
04:13did this care cascade improve
04:15access to mental health services
04:17and did kids get better?
04:19Did we see a reduction in symptoms
04:22and an improvement and functioning?
04:24And one of the things that we
04:26learned is that kids who received
04:28gateway call received three times
04:30the number of mental health visits
04:32compared to those in our comparison.
04:35We used a propensity score
04:37matched comparison condition.
04:39But when we took a look at whether or
04:42not children or the rate of mental
04:45health service access among kids,
04:47we saw that 46% in gateway call got
04:50access to mental health services.
04:52Of kids who screened positive
04:55in our screening process,
04:57only 46% got access to mental health.
05:00That means at least one specialty service
05:03visit compared to 33 in the comparison group.
05:06So even though there is a 13% difference
05:10in those in that percentage rate,
05:13it was not a statistically significant
05:16difference and 46% is pretty darn low,
05:19which raises questions about where,
05:23where did this intervention breakdown?
05:26On a surface,
05:27three times more services sounds
05:30wonderful and it is great.
05:32Why weren't we making the connection
05:34for more kids?
05:35So it's a $3,000,000 demonstration.
05:38So this was really troubling
05:40and despite the fact that 67% of the kids
05:43who were coming into our experimental
05:45condition were screening positive,
05:47so we were missing a lot
05:49of children along the way.
05:50So drawing on background
05:52and implementation science,
05:54we went back to take a closer look,
05:57component by component.
05:58And so this bar graph shows
06:00the distribution of children
06:02in our experimental condition.
06:04These were kids who came in through
06:06eight designated intake units in at the
06:11child welfare agency and the first bar,
06:14you'll want to follow the
06:15blue proportion of the bar.
06:17These are the percentage of kids that
06:20received the component of gateway call.
06:22So we screened almost 95% of kids
06:27who came into this intake unit.
06:29Of those kids,
06:3167 screened positive unless these
06:34were also with developmentally
06:36appropriate screeners or then we
06:38were able to complete assessments,
06:41full diagnostic assessments
06:43for 63% of these children,
06:45which is still really high.
06:47But then when we connected
06:49their child welfare records,
06:50screening and assessment records
06:52with Medicaid billing records,
06:54where we would be able to see did
06:57these kids get licensed mental
06:59health treatment in the community,
07:01We start to see exactly where
07:04this care cascade broke down.
07:06Which begs the question,
07:08what the heck happened in between
07:11assessment and getting served in the
07:13community in that period of time?
07:16One of we experienced over those
07:18five years a lot of disruptions
07:21in partnerships when we saw
07:24major collaboration breakdowns.
07:26There was a disruption in contracts
07:28with the Behavioral Health
07:29Assessment Team and other providers
07:31in the community who would have
07:33been providing evidence based
07:34mental health care for these kids.
07:36There was extraordinary worker
07:38turnover in both the child welfare
07:40and the behavioral health system,
07:43which disrupted those frontline
07:44partnerships that we rely on
07:46in practice to connect kids and
07:49families to care and to make
07:51sure that it's well coordinated.
07:53We saw a lot of variations in the way
07:56that frontline child welfare workers
07:58were referred and followed up on care,
08:01and we realized at the end that
08:04we did not engage a critical
08:06partner in the community.
08:08We were not engaging with some of
08:10our foster care providers in the
08:13community who were also responsible
08:15for making sure that kids were
08:18referred and followed up on service
08:21referrals to mental health care.
08:22So all around,
08:23this was a story of we were
08:25able to implement successfully
08:27within the child welfare system,
08:29but we didn't get to our ultimate
08:31desired outcome because our partnerships
08:33were not as strong as they needed to
08:36be between the child welfare system
08:38and the behavioral health system.
08:42So this goes back to what we know from
08:45the literature and what we know in theory
08:48about relationships and implementation.
08:50These social relationships,
08:52whether they are between individual
08:54clinicians working at the front lines,
08:57or organizations that are working in the
08:59community on behalf of a common population or
09:02among team members within a clinic setting.
09:05These relationships are critical because
09:07they defuse information that could be
09:11technical information about an innovation.
09:14They defuse expertise,
09:15and so information about how
09:18to use a new innovation,
09:21it diffuses social influence.
09:23We are, we are subject to
09:25influence by what our peers think.
09:27And when we hear about a successful
09:29example of a supervisor or a peer
09:32or someone else in our in our
09:34clinic about a successful use case,
09:37we might be more likely to
09:39adopt that later on.
09:40And then of course,
09:42especially for organizations,
09:43those relationships that are developed,
09:45they defuse resources,
09:47they're conduits for money,
09:49they're conduits for other types
09:51of of resources, influence,
09:53space, colocation.
09:55So these relationships are super important.
10:01So the next lesson that we learned
10:04is that our relationships not only
10:06does do breakdowns in collaborative
10:10relationships side sideline implementation,
10:12our relationships shape us
10:15in really important ways.
10:17And so this I'm going to share a
10:19little bit of information about
10:21a study that I that I did when I
10:24was a post doc back in the day.
10:27And this was with a group of 32
10:30children's mental health organizations
10:31that are in Saint Louis County,
10:34which is where I did my PhD before going
10:37over to North Carolina for my post doc.
10:40And these 32 organizations came together
10:42to implement trauma focused cognitive
10:44behavioral therapy in response to
10:46the rise of the number of kids that
10:49were coming into the mental health
10:51system with Post Traumatic Stress
10:53Disorder and really wanting to make
10:55a difference in terms of the quality
10:58and the amount of services that were
11:00available for kids and families.
11:01And so there the group of 32 organizations.
11:05This was led by the Missouri St.
11:08Louis County
11:11Children's Advocacy Center at the
11:13Missouri at University of Missouri.
11:14They, they organized a learning
11:17collaborative and they brought
11:18teams from these 32 organizations
11:20together because they thought
11:22that by bringing them together,
11:24they they'd be able to share
11:26learning and implement better.
11:28And so the first thing that we
11:30did together was just assess
11:31relationships at the beginning.
11:33So before you started implementing,
11:35before you started training on trauma Focus,
11:38CBT, exactly who do you go to for advice?
11:42And that was the first thing that we
11:44asked folks when they got started,
11:45because we know social influence matters.
11:48So this is an ego network analysis.
11:51And we asked each clinician from these
11:5432 organizations to nominate 5 people who
11:57they turn to for advice about serving
12:00kids with post traumatic stress disorder.
12:03And on average,
12:05clinicians nominated 3.9,
12:06so almost four people.
12:07So they had people to turn to for advice
12:10and 72% of the peers that they nominated
12:14for colleagues in their own own organization,
12:17which is not uncommon.
12:19But we also saw that 15%
12:22referred to other folks,
12:256% referred to colleagues
12:26in private practice.
12:27So outside of those those
12:30organizations that were part of
12:32this learning collaborative and
12:35about 5% referred to experts.
12:37So these were,
12:38these were kids in the child advocacy center.
12:41Now when you take a look at so a
12:45clinician might have this ego network
12:47of 3.9 or four other colleagues
12:50that they turn to for advice.
12:52But when you aggregate those
12:54findings at the network level,
12:56we can start to make some
12:58inferences about the structure
12:59of relationships in that network.
13:01And so using that data,
13:04we created a network diagram,
13:05which is the first step in a network
13:08analysis to visualize how this advice
13:11network looked in Saint Louis at the
13:14beginning of the Learning Collaborative.
13:16So through that nomination process,
13:18we identified 422 people who were
13:22providing advice to clinicians
13:24in Saint Louis about kids with
13:27post traumatic stress disorder.
13:29And this was after we cleaned up the
13:31data and removed references to Freud.
13:34So we we did a good job cleaning that data.
13:37And so each dot,
13:39each circle on these graphs
13:41represents a person, a clinician,
13:43and each line represents an
13:46advice seeking relationship.
13:48The triangle nodes,
13:49these are our experts,
13:51these are our child advocacy folks who
13:53are doing the training on trauma focus,
13:54CBT.
13:55And again this is before they began training.
13:59So what we can see here,
14:01we can use some network metrics.
14:06Here we go. So one of the things that
14:07we can start to observe just by taking
14:09a look at these graphs is that there is
14:12some disconnection within the network.
14:14We have a core group of folks that
14:16are sort of kind of connected with
14:17one another in very indirect ways.
14:19But then we have a lot of pockets
14:22out here on the periphery and we
14:24have a lot of folks who are not
14:25connected to anyone at all.
14:27And network analysis allows us to
14:29generate some whole network metrics
14:31that tell us just about the the level
14:33of connectivity within these systems.
14:36So density reflects the proportion
14:39of of actual ties to potential.
14:41So that we might read this as
14:45this is so you know,
14:47higher is better centralization,
14:49the degree to which the network is
14:52centralized around a few core individuals,
14:54the degree to which that it's clustered,
14:57the degree to which those advice
14:59seeking relationships are reciprocal.
15:01Because if I ask Doctor Teebs for
15:03advice and then Dr.
15:05Teebs asks me for advice,
15:06that's a much stronger relationship and
15:08that's a much stronger level of influence.
15:11And if it's, you know,
15:13I just ask Doctor Teebs for advice,
15:14which I totally would.
15:17And so agency homophily also reflects
15:20the degree to which that advice seeking
15:23those advice seeking patterns are governed
15:25by being in the same organization.
15:28So just at baseline just at the outset we
15:31got a sense for how this advice seeking,
15:34how this advice seeking network looked
15:37and and that we were already able to
15:39identify some patterns in terms of who
15:41is influential in this network which
15:43would be these faculty experts but
15:45that not everybody else was connected.
15:48The other lesson learned along the
15:49way is that just because folks are not
15:51connected at one minute doesn't mean
15:53that they won't be connected in another
15:55and that these relationships change.
15:56And so we followed up in with the
15:59same type of data collection with
16:01these clinicians at two other time
16:04points and we were able to take a
16:06look at how that network changed.
16:09So for the background,
16:10the Learning Collaborative Model that
16:12was used to help support implementation
16:15across these 32 organizations is
16:17based on the Institute for Healthcare
16:19Improvements Breakthrough Series and
16:21Learning Collaborative Model and it is
16:24steeped in quality improvement practices.
16:26It brings together teams
16:28from multiple agencies.
16:30It emphasizes shared learning and influence.
16:33So theoretically,
16:34we would be not only building networks,
16:36but also activating them in
16:39support of implementation.
16:40And it was intended,
16:41this model was intended to not only
16:44build relationships within teams among
16:46clinicians from the same organization,
16:49but also build relationships
16:51among clinicians across teams
16:53and across organizations.
16:55And so there's some preparatory
16:57work that agencies do before they
17:00begin the learning collaborative.
17:02Then they come together for three
17:04in person learning sessions over
17:06the duration of about a year or so.
17:07And in between those in person
17:10learning sessions, they take what
17:11they learn and begin implementing
17:13it using plan to study at cycles,
17:16which is classic quality improvement work.
17:18And in between those teams are supported
17:22using coaching calls and team calls.
17:25There's web support and other
17:26types of support to use quality
17:28improvement techniques.
17:29So it's a very intensive model
17:32for supporting implementation and
17:34really scaling across the region.
17:36There's some questions about
17:38how they work though.
17:39And so our theory was that they work
17:42potentially by building networks.
17:44So this bar on the left reflects
17:46that breakdown of those ego
17:48network partnerships at Time 1.
17:50So 72% of folks that were being
17:54asked for advice about treating
17:56post traumatic stress disorder
17:58among children at time at the end
18:00of the Learning Collaborative,
18:01that was still the most dominant
18:04source of advice.
18:05But we've got to see that there
18:07were some significant changes in the
18:10proportion of these ego networks.
18:12So we saw a significant reduction
18:15of outside individuals of the other
18:17folks that were being asked for advice.
18:20We saw a significant reduction in
18:22the number of private practitioners
18:24who were being asked for advice
18:26about treating Trauma focus or about
18:28implementing trauma focus CBT.
18:31And we saw a significant increase
18:34in the number of times that folks
18:36were asking the experts,
18:38the learning collaborative leaders for
18:40advice about post traumatic stress disorder.
18:43So we did see a shift in those ego networks.
18:46Now this doesn't look like,
18:47you know,
18:4872% to 66%.
18:50This does not look like a drastic
18:53difference at the individual level,
18:55but when we extrapolate and bring
18:57it out to the systems level,
18:59these minor changes resulted in some
19:02pretty major changes at the network level.
19:06So this compares the network
19:08metrics between learning session
19:09one and learning session three.
19:12We're still dealing with the same 422
19:15individuals who had been nominated.
19:17Over time we saw an increase in the
19:19isolates where the people who are
19:22not connected and these happened
19:23to be people who were not part
19:26of the learning collaborative.
19:28We did see in we cite we saw a slight
19:30decrease in density which means
19:33that the overall connectedness,
19:35but we saw an increase in the centralization.
19:38We saw that there was increasing
19:40clustering among folks in the
19:42learning collaborative who are
19:44asking one another for advice,
19:46who are leaning on one another
19:48for support to treat kids with
19:50post traumatic stress disorder.
19:51We saw an increase in reciprocity.
19:53So a strengthening of these advice,
19:55sharing relationships.
20:01So the key takeaways here is
20:02that there was more isolates,
20:04it was more centralized and we
20:06saw these stronger relationships.
20:08So these relationships can
20:09change over time and they can
20:12change during implementation.
20:14We did a deeper dive into trying
20:17to explain why we would see some
20:19of these changes and this is using
20:22exponential random graph modeling to
20:25understand predictors of tie formation.
20:27So what would predict someone forming
20:30a new advice seeking tie over the
20:33course of the learning collaborative
20:35and some of our key takeaways.
20:37And this is framed in terms
20:39of transactive memory systems,
20:40which is an information seeking
20:43or communication theory that
20:45expertise quality is important.
20:47We saw that being a faculty expert or
20:50supervisor was associated with forming
20:52new ties or being a source of information,
20:55a source of advice.
20:57So we know that expertise
20:58quality is important.
21:00We saw that being in the same organization
21:03and also being from the same discipline,
21:06social workers asking advice
21:07of other social workers,
21:08psychologists asking advice of
21:10other psychologists that that was
21:12associated with forming a new tie or
21:15maintaining an advice seeking tie,
21:17which reinforces how important
21:19accessibility both physical proximity
21:21but also social similarity might be
21:24for shaping the development of these
21:27relationships for implementation.
21:30We also saw that having a prior
21:32tie or being connected in some
21:35way either directly or indirectly
21:37to others is also important for
21:40shaping and ties that we build.
21:41We build on prior relationships.
21:44It's much easier to build a tie with
21:47someone that's already there to either
21:50to either increase the frequency or
21:53to reciprocate advice seeking and
21:55sharing or to connect through triads.
21:58So that's the concept that if I'm going
22:01to pick on Doctor Elizabeth Connors here.
22:03So if Elizabeth asks Jack for
22:06advice and I ask Jack for advice,
22:10that theory would suggest that
22:12over time by us both virtually
22:14being connected to Doctor Teves,
22:16that we would then ask each other for advice.
22:18And that's exactly what we saw happening.
22:20So we that's how that explains this
22:24clustering phenomenon that we saw over time.
22:26So we really do build on these existing
22:29relationships when we are looking to
22:32build new ones over time and this has
22:35important implications for what we
22:37might expect later and when we are trying to.
22:39Very deliberately build these networks and
22:43build these partnerships for implementation.
22:46So this led to a series of questions about
22:50how exactly could we build networks,
22:53Can we be more thoughtful,
22:55planful,
22:56strategic in the field of implementation?
22:59We're all about building and advancing
23:02the science of implementation strategies,
23:05these deliberate efforts to integrate
23:08and sustain evidence based practice
23:11into routine care systems.
23:12And if we know relationships are
23:15important for implementation,
23:17are there strategies that we can use
23:20to deliberately build the system in a
23:22way that will support implementation
23:24over the long home?
23:25And so with a with my colleague, Dr.
23:28Reza Yousafi Murai,
23:29who's at the University of Rochester,
23:31we started thinking about
23:33network building interventions.
23:35And Tom Valenti has done a
23:36lot of work in this area.
23:38He has a new book, I believe,
23:40about network interventions.
23:42These are purposeful efforts to use
23:45social networks or social network data.
23:47And he characterized 4 different
23:50types of network interventions.
23:52These are individuals where we
23:54might identify a champion to lead
23:56our implementation efforts or other
23:58types of system change efforts.
24:00Maybe we use a segmentation approach
24:03where we would I use network data to
24:06identify groups where we might drop in
24:09an intervention or prioritize for an
24:12experiment or a change of some type.
24:15Induction that refers to using networks to
24:19simulate interactions and promote diffusion,
24:22so relying on those existing
24:24networks to spread messages.
24:26But the one that we thought was most
24:30related to implementation strategies were
24:33these alteration types of interventions.
24:36These are deliberate attempts to
24:38change the networks either by adding
24:41or deleting people or organizations
24:43from the network,
24:45adding or deleting links that
24:46are in the network,
24:47those pair wise relationships.
24:49We know that a few tweaks in those
24:52relationships from the learning
24:53collaborative study can go a long way
24:56in terms of shifting the structure
24:58or potentially rewiring the links
25:00in terms of increasing the strength
25:03or decreasing the strength or
25:05changing the type of resource or
25:08the content of those relationships.
25:10So one of our in some of our early meetings,
25:14so Rosa and I met at the dissemination
25:18and implementation Research
25:20conference back in 2016 as like blown
25:23network social network analysts and
25:26implementation scientists and both
25:28really excited about this area.
25:30And so we started by looking at the
25:33Eric taxonomy of implementation strategies.
25:36The expert recommended implementation change.
25:39I I just what's your bad acronym and
25:42Eric is an acronym not a person but
25:44the the taxonomy of implementation
25:46strategies we were thinking about
25:48think about all the implementation
25:50strategies that are in the taxonomy that
25:53reference relationships in some way we do.
25:57We advocate that folks use
25:59strategies like build a coalition.
26:01Well that's building relationships
26:03among entities for the purposes of
26:06of implementation. Create a new team.
26:10Promote network weaving.
26:11Can you explain what that means to me?
26:14I would appreciate it.
26:15I'm still a little confused.
26:16Create a learning collaborative,
26:18which we knew from the Saint
26:20Louis study was an implementation
26:22strategy and we knew that those
26:24partnerships changed over time.
26:26Develop academic partnerships between
26:28community partners and researchers.
26:30Develop resource sharing agreements.
26:33So these relationship building
26:35approaches were inherent in our Eric
26:38taxonomy and they connected with this
26:40idea of a of a network alteration
26:43intervention where we were deliberately
26:45trying to build new relationships or
26:49alter those relationships in some way.
26:52So we embarked on a systematic
26:54scoping review and I will not share
26:57how many years that it took us to
26:59do the systematic scoping review,
27:02but we come the literature not just
27:04in healthcare but across a variety
27:07of disciplines to identify studies
27:09that tested or examined network
27:13alteration interventions.
27:14Now not all of those studies characterize
27:16them as network alteration interventions.
27:19So we wanted to be very thoughtful
27:21about including only those studies that
27:24looked at network change over time.
27:26So at least two data points and where
27:28there was a sufficient discussion
27:30about the type of intervention that was
27:33being used to change those networks.
27:36And through an iterative synthesis process,
27:38we developed a typology of eight
27:41strategies or eight types of network
27:44building interventions that we
27:46think could be potentially useful
27:48in implementation and elsewhere.
27:50So we found three types of context
27:54based strategies.
27:55So these are interventions that try to
27:58change the larger environment of the network,
28:01including creating a group like
28:04creating a coalition,
28:05changing the environment.
28:07So this might be a natural
28:09shift in community needs,
28:11maybe this is a shift in the
28:15regulatory environment or change the
28:17composition like introduce a mentor.
28:19These environmental changes,
28:20prompt shifts in the way that
28:23we connect with one another.
28:26And so it is an indirect intervention
28:28that's intended to leverage sort
28:31of those naturally occurring
28:33partnerships that we have.
28:34And our natural tendency is to shift
28:36the way that we partner with one
28:37another in those environmental.
28:39In those environmental changes,
28:42we identified 4 actor based strategies.
28:44So these are strategies that target
28:47individuals within the network.
28:49These might involve changing skills,
28:51so training folks on social skills or
28:54or how to how to be a better partner,
28:57or how to be or how to ask for advice,
29:00or how to ask or or how not to be lonely.
29:04We found a lot of literature in
29:07the loneliness field,
29:08changing knowledge so helping
29:10folks understand what their larger
29:12network looks like,
29:13where there are gaps and where
29:15there are strengths,
29:16under the premise that when we're
29:18more aware of our environment,
29:19we might work within it a little differently.
29:22Changing prominence.
29:23So elevating an individual as a champion
29:26or some other type of leader within
29:29the network can automatically shift
29:31the way that others partner with that person.
29:34Alternatively,
29:34in the bullying literature,
29:36we also see the same type of strategy
29:38work in the opposite direction,
29:41where we try to deemphasize an individual
29:44that might be diffusing negative
29:47negative behavior or negative types
29:49of resources throughout the network.
29:52And then the 4th actor base strategy
29:54was around changing the motivation,
29:56so dropping an incentive in
29:59there for partnerships.
30:01Whether it was very deliberate or you know,
30:04sometimes the you know,
30:05you might have a motivation to write
30:08a paper together that's a motivation
30:10that might stimulate a new partnership
30:13between individuals in the network.
30:15And then the 8th strategy that we
30:18identified were these tie based strategies.
30:20These are the ones where it's
30:22a it's almost a very,
30:26I'm going in with a scalpel almost.
30:28I am not a surgeon, I'm not a I'm not
30:30even a a physician even though I'm
30:32in general attorney for medicine.
30:34You know this is going in and
30:36operating on a particular tie,
30:39one tie or a specific type of
30:41tie in the network and this is
30:43by changing the relationship.
30:45So maybe it is dissolving
30:48that pair wise relationship,
30:50maybe it is specifically brokering that
30:52relationship or maybe it's building
30:55on that relationship or changing the
30:57way that that relationship works.
30:59So we can think about this in the
31:01context of service delivery systems.
31:03There are two organizations that
31:05maybe refer clients to one another
31:08and we go in and we say can't
31:10refer clients to one another.
31:12What if you developed a
31:13joint program together.
31:14So that would that's an example of a tie
31:17based strategy or a tie based intervention.
31:19And so we just published that last
31:22year and we're very proud of it
31:24and we're looking for examples and
31:26opportunities to begin bringing
31:28this into the implementation field.
31:30And the reason that we did this
31:32was to help further specify some
31:35of those more relationally focused
31:38implementation strategies.
31:40I want to share a little bit more.
31:41So the third,
31:42third set of studies that I wanted
31:45to share about were building on this
31:48theme about tie based relationships.
31:50We knew from our learning collaborative
31:53study that you know small changes
31:55and a few relationships can have
31:58resounding effects on the larger system.
32:01And so at the organizational level,
32:04we want to be able to understand
32:06exactly what kinds of tie based
32:08relationships might we want to build
32:10or might be want to target and in
32:13particular what can we do to help
32:16strengthen those relationships.
32:19There's a history of large
32:21demonstration projects and network
32:23building studies that have been done
32:26in mental health service delivery
32:27over the years by Keith Proband,
32:29Brent Millward and Public Administration,
32:32Joe Morrissey,
32:33Bob Rosenheck and Mental
32:35Health Services Research.
32:36And these series of demonstration
32:39projects throughout the 80s,
32:41the 90s and the early 2000s demonstrated
32:43to us that just building relationships
32:46is not necessarily going to make
32:49a difference for client outcomes.
32:52And that more relationships by trying
32:54to improve the density of these
32:56relationships by hoping that everybody
32:58comes together and everybody works
33:00with one another and everybody refers
33:02that more is not necessarily better.
33:05That really what has been tied to
33:08improvements in client outcomes is this
33:11idea that a very strong partnerships
33:13among small groups of organizations
33:15that networks that are organized
33:18around very intensely collaborating
33:20small groups of organizations that
33:22in a sense are really integrating
33:25their work together both at the
33:27front line and also operationally
33:29administratively that these are the
33:31types of groups that are likely to help
33:35produce better outcomes rather than
33:36trying to make everybody work with everyone,
33:38which is really hard.
33:40We only have 24 hours a day.
33:42So this motivated a next series of studies
33:46and we're developing a tool kit.
33:47Well, we did develop a tool kit
33:50called Collaborating across Systems
33:52for Program Implementation CASPI.
33:54And this work was funded by the
33:55National Institute on Drug Abuse
33:57and the Robert Wood Johnson
33:59Foundation System for Action program.
34:00And our goals were to examine and
34:03specify the specific ways that
34:05agencies were working with one another,
34:07the specific time based relationships
34:09that were being developed to
34:12implement an intervention at the
34:14intersection of child welfare and
34:16substance use treatment systems.
34:18And it was intended to inform a toolkit.
34:21We leveraged a naturally occurring
34:22roll out of an evidence based practice
34:25here in Ohio and we use mixed methods,
34:27multiple case study and so this
34:30is this project is being carried
34:32out in the context of Ohio START.
34:35Start is Sobriety Treatment and
34:37Recovery teams which is an evidence
34:40based model for child welfare agencies
34:44to help identify parents who are
34:46coming into the system because of
34:48parental substance use disorder.
34:50In Ohio,
34:51especially at the rise of the opiate crisis,
34:53we saw a dramatic increase in the
34:55number of kids going into foster
34:58care because their parents had some
35:00type of substance use disorder.
35:02And we also know from the literature
35:04and from our experience experiences
35:06that because of the difficulty in
35:09treating substance use disorder,
35:11these families were less likely to reunify.
35:14These were families that, you know,
35:16ended up staying separated and
35:18their children ended up going,
35:20you know, persisting in foster care.
35:22Termination of parental rights,
35:24that is incredibly traumatic for families.
35:26And so START was developed by Tina
35:29Willauer at Children and Family Futures
35:31originally in Cuyahoga County which
35:34is the Cleveland area here in Ohio.
35:37It was originally developed for these
35:41families and we we know there was a
35:44three county demonstration in Kentucky
35:47before it was adopted here in Ohio.
35:50And if they were able to demonstrate
35:52that this particular model expedites
35:54parents access to treatment and
35:56improved treatment retention,
35:58parents were able to demonstrate
36:00higher levels of sobriety.
36:01And most it's important to me it
36:03kept parents and kids together
36:05before and after the intervention
36:07start is a multi component model.
36:09And so the way that it works here in Ohio
36:12is that when a child welfare case opens,
36:15initially,
36:15there's a screening that's done.
36:19And I did.
36:22We are, we're wrestling a little bit right
36:24now about whether to do universal screening,
36:26whether there's capacity
36:28to do universal screening,
36:29which as a researcher I'm like,
36:30yes, we should do this.
36:31But as a practitioner,
36:32I also understand some of the resource
36:35constraints that we've got in the field.
36:37But the idea is that when a
36:39child welfare case opens,
36:40so when there is sufficient
36:42information to warrant opening a child,
36:44significant information about a risk
36:46to the safety of children in the home,
36:49that it warrants opening the case.
36:51Ideally,
36:52A Screener for substance use
36:54disorder is completed.
36:55And if the Screener is positive,
36:57ideally those parents are
36:59referred to the START program in
37:01their county within 14 days.
37:03Within four days they have a shared
37:06decision making meeting which involves
37:08child welfare caseworkers, supervisors,
37:10behavioral HealthPartners from the
37:13organization and most importantly,
37:16parents and parent advocates get together
37:19around the table to talk about the
37:22case plan and make decisions together
37:24about the next course of action.
37:26Within seven days of that
37:28shared decision making meeting,
37:30parents are connected with
37:31a family peer mentor.
37:32This is someone with lived experience
37:35of both recovery and child welfare
37:38experience and I personally think
37:40is the magic of this model.
37:42And so they have intensive
37:44contact with this family, peer,
37:45mentor, At the same time,
37:47they're getting referred for a mental health
37:50or substance use disorder assessment.
37:52They're getting into treatment and they
37:55are expected to complete 4 treatments.
37:57All of this is supposed to happen in 38 days,
38:01Doesn't always happen.
38:03It very rarely happens in 38 days
38:05because it's very difficult to implement
38:08all of these components to align them
38:11across child welfare and substance
38:13use treatment systems and make sure
38:16that they're happening so quickly.
38:18Like the gateway call study that
38:20I began with START reminds me
38:23so much of that because it is,
38:25it is a service cascade,
38:26it's a clinical pathway that
38:29requires really tight alignment not
38:31only within these systems and the
38:33components that are delivered within
38:36child welfare and substance use,
38:38but such close alignment across
38:41those systems.
38:42In fact,
38:43sometimes the family peer mentor
38:45is employed by the substance
38:47use treatment organization.
38:48You know,
38:49the substance use treatment
38:50clinicians need to be part of the
38:52shared decision making meetings.
38:53So there's a lot of points of integration
38:56across these systems with this model.
38:58And thinking about that first hard
39:01lesson learned about how relationships
39:03can disrupt implementation,
39:05we had lots of very,
39:07very early conversations with our partners.
39:10And our partner for this particular
39:12work is the Public Children's
39:14Services Association of Ohio, TCSAO.
39:16And they're a nonprofit organization
39:19that represents all 85 county child
39:22welfare agencies here in the state.
39:24And they have been,
39:26they've taken the leadership
39:28of implementing Ohio Start,
39:29so they provide implementation support
39:33training for all of these counties.
39:35And we had some very early conversations
39:38at first the model first pulled out
39:40with 17 counties in Southeastern Ohio,
39:42which is predominantly the Appalachian
39:44region of the state where we saw
39:46those the highest rates of foster
39:49care entry because of parental Sud.
39:51We had some very,
39:52very early conversations about the
39:54lessons that we learned in gateway
39:56call and how we need to really
39:58focus on these relationships.
40:00And so on their end,
40:02their implementation support,
40:03their technical assistance
40:05providers do help agencies.
40:07They have been coaching agencies to develop
40:10those partnerships early on across systems.
40:13But at the same time,
40:14we developed this research,
40:16these research studies together
40:19thinking like, oh,
40:19this is a good opportunity.
40:21We begin to understand these
40:23phenomenon a little bit more.
40:25And so our studies rely or
40:27we draw on a variety of data sources.
40:29We drew. We collected contracts
40:32and memorandums of understanding
40:33between child welfare and substance
40:36use treatment organizations.
40:37We learned a lot about contracting
40:40capacity in public child welfare agencies.
40:43And as a side note, when I went to
40:45social work school as a master student,
40:47I never thought, you know, 20 years
40:50later I'd be so fascinated by contracts.
40:52So go figure.
40:53But they're really interesting.
40:55We also did small group interviews
40:59across 17 counties,
41:00ended up being 48 small group interviews.
41:02We interviewed folks on the
41:04child welfare side,
41:05folks on the behavioral health side.
41:08We this also happened.
41:10This also happened in between COVID,
41:13we started and then COVID happened
41:15and then we had to take a break
41:17because of COVID mitigation
41:20measures and and shutdowns.
41:23And so keep that in mind as you're
41:25as you're listening to our results
41:27then we brought everyone together.
41:28So after we did the interviews
41:30and the contract analysis,
41:31we brought our partners together.
41:34This involved our PCSAO partners.
41:37It involved the model purveyors.
41:39We also had partners from
41:40the Behavioral health system,
41:41from the Ohio Association of County
41:44Behavioral Health Authorities and the Ohio
41:46Council of Behavioral Health Providers.
41:48We brought them all together to try
41:51to make sense of our results and to
41:54identify and begin to specify the
41:56specific ways agencies were collaborating.
41:59In the meantime,
42:00we were also relying on case records,
42:02worker surveys and we integrated all of
42:06these data using coincidence analysis.
42:08And there's some preliminary findings
42:10that are under review right now,
42:12but that's the the findings that
42:14I wanted to share with you because
42:16they relate back to this idea about
42:18exactly how do we build these
42:20tie based relationships,
42:21our qualitative work and our
42:23expert panel together,
42:24we identified 8 or 7 strategies
42:26that these agencies were using to
42:29develop partnerships between child
42:31welfare and substance use treatment.
42:34Oops,
42:35sorry.
42:35And so we had three strategies that
42:38they were using to staff the program.
42:41So these were occurring at the
42:43executive or administrative level.
42:44These were agency leaders talking
42:47with one another about developing
42:50contracts with one another,
42:52Family peer mentors with lived experience.
42:55Many child welfare agencies,
42:57which are governmental entities governed
43:00by public agency workforce regulations,
43:03have a lot of difficulty
43:06establishing a new position,
43:08a new specialized position,
43:10and often involved negotiations
43:11with their labor unions.
43:13If a position was able to be
43:16established that those individuals
43:17needed to pass a background check in
43:19order to be employed and people with
43:22lived experience might have had a
43:23more recent felony experience that
43:25will preclude them from being the
43:27type of experience that we want to
43:29bring to this to this intervention.
43:32But because of regulations they couldn't
43:34be hired within the child welfare system.
43:36So there were a lot of contracting
43:38arrangements with behavioral health
43:40organizations to bring the family
43:42peer mentor into the behavioral
43:44Health Organization and work then
43:45in a Co located
43:47way which was the 2nd strategy
43:49within the child welfare system.
43:51So we've got administrators hammering
43:53out contracts and memorandums of
43:55understanding around this position.
43:57We had them working together
43:59to figure out how they were.
44:01We're going to Co locate the family
44:03peer mentor of behavioral health
44:06service provider employee within
44:08the child welfare system which is
44:10a public agency to make sure that
44:13they've got access to records,
44:14supervision resources and then
44:16of course how they were going to
44:19supervise these teams across systems.
44:22So these were very much administrative
44:25level strategies that that we were
44:27seeing and they were not as common
44:29as we would have expected to see.
44:30That's what it is a bit of thought.
44:33Then we also saw agencies,
44:35their administrators and then their
44:37supervisors or clinical program
44:39directors working together to
44:40try to figure out about how will
44:43we help bridge the systems for
44:44the parents that are coming in.
44:46And so we saw two strategies that agencies
44:49were using to promote service access.
44:52So again there might have been a
44:54contract or memorandum of understanding
44:56to expedite service access.
44:58Behavioral health services
45:00notoriously have wait lists.
45:01Especially in our rural and
45:04Appalachian communities where we have,
45:06you know, service deserts,
45:07there could be 6 month or a year wait list.
45:11And we know from substance abuse
45:13if for substance use disorder,
45:15if someone's ready to go to
45:17treatment that that could be,
45:18you know that that's a real that's
45:20a huge barrier if if they're
45:22not able to get treatment,
45:23when they're ready to go to treatment.
45:25Also in the context of
45:27the child welfare system,
45:28parents have 12 months to work a
45:31case plan and be able to demonstrate
45:34progress in in their substance use
45:36disorder treatment before going to
45:38a judge where the judge makes the
45:40determination about what's going
45:41to happen with their children.
45:43And so if especially considering
45:45that return to use is common in
45:48substance use disorder treatment,
45:50you know those wait lists have
45:53resounding consequences for families.
45:55And so in order to implement this
45:57model and make sure that parents
45:59could get treatment,
46:00to get treatment quickly,
46:02demonstrate progress,
46:03and ultimately reunite with their
46:05children before that 12 month time is up,
46:09they really needed to work together
46:11to see if they could circumvent
46:12those wait lists,
46:14which introduces a whole other
46:15set of ethical issues.
46:17But you know,
46:19they really work together to try to
46:21make these workarounds within the
46:23system and the resources that they had.
46:25And we also saw instances
46:27of referral protocol.
46:28So really being thoughtful
46:29and streamlined about,
46:31you know,
46:31how exactly what information do we
46:33need to send to this provider in
46:35order to get our parent connected.
46:38Who do I need to talk to?
46:39I need to get this parent at
46:41at the front door,
46:42which was much more of a
46:45frontline collaboration strategy.
46:46Then we also saw two examples
46:49of how frontline workers were
46:51working together to collaborate
46:53across systems for implementation.
46:55So those shared decision making meetings
46:57was a platform for so much case plan
47:00work and collaboration around the case
47:02plan and making sure that those services
47:05that were being offered through the
47:07child welfare system like parenting,
47:10parenting classes and other types of
47:12concrete resources that were being
47:13provided were able to be aligned with the
47:15use disorder treatment that they were
47:17getting in the behavioral health system.
47:19And then of course data sharing
47:22which is never easy.
47:23So there was a lot of time and
47:25attention focused to how we share
47:27information across these systems in
47:29a way that still respects the dignity
47:32and confidentiality of the parents.
47:33But we're able to continue to do our work
47:37together and make informed decisions
47:39about child safety and parent progress.
47:41So we recently, just two weeks ago,
47:45that paper got published
47:47in Implementation Science.
47:48We also,
47:48if you're interested in this kind of work,
47:51we also specified the causal mechanisms
47:54that at least we believe predict or
47:57explain how these types of strategies
48:01align these organizations across systems,
48:03why we think it leads to better
48:06implementation outcomes and why
48:07we think it leads to better client
48:09outcomes in the long run.
48:10And so if you're interested in
48:12testing hypotheses in this area,
48:14come talk to me,
48:15I would love,
48:16I would love for folks to to join me
48:18and and begin looking at this in a
48:21little bit more of a granular level.
48:23So that's where we are.
48:25Our next steps in this in this program
48:29of work is to begin trialling some
48:31of these strategies or trialling and
48:34begin testing programs where we can
48:37support agency leaders to use these
48:40strategies to use them consistently
48:42and well to build partnerships
48:44across systems and understand the
48:47impact on implementation.
48:48And so we,
48:50we've been working together with our
48:53community partners and our research teams.
48:56We've got a proposal that's was
48:58under review that we're going to
49:00have to resubmit if that's OK,
49:02always more work to do.
49:04So the last thing that I wanted
49:07to chat about is your partners and
49:09you so are the relationships that
49:12are developed in implementation.
49:14You know,
49:15we've been talking about the
49:17relationships among clinicians
49:18or relationships among providers
49:20and organizations.
49:21There is another critical relationship
49:23here that we need to talk about and
49:25that is the relationship that we
49:27have with our community partners.
49:28Implementation is is something,
49:30you know,
49:31the purpose of implementation science
49:33is to understand strategies that are
49:35going to work in the real world.
49:37And so that means that for a lot of
49:39the work that we do in this field,
49:41we need to actually work with
49:42organizations in the real world.
49:44We have to develop those community
49:46partnerships and you know, that's not easy.
49:49You know,
49:50not every community partner wants to
49:53partner on research. They take time.
49:56They're very resource intensive.
49:58Sometimes it doesn't work out
50:00according to plan and that can
50:02be problematic if you know,
50:03if your early career on the 10
50:04year clock you're thinking about
50:06like can I get publications,
50:07can I get a grant proposal,
50:09can I get grant funding within,
50:11you know the five or six years that
50:12I have in order to demonstrate this.
50:14And you know partnership development
50:16does not happen overnight and it
50:19could take five or six years just
50:20to be at a point when you are ready
50:23to collaboratively design a study
50:25that you submit for federal funding.
50:27So these are complicated relationships
50:30and it bears discussion because we need
50:33to have them in order to advance this field.
50:36And so this part is,
50:37is more about my experience and my,
50:40my thoughts which I welcome to be
50:43and disagreement about you know,
50:45but the way that we work together,
50:47some of the things that I've observed
50:49over time is that the way that we
50:51work together changes depending
50:52upon how long we work together and
50:54and where where our projects are.
50:56You know,
50:57you might have those initial
50:59conversations and I was really
51:01trying to reflect on what it was
51:03like when I first moved to Ohio and
51:06and wanted to develop partnerships
51:08with local organizations.
51:09It's very consistent with Ohio
51:12State's land grant mission.
51:14And so thinking about what that looked
51:16like and how those conversations
51:18played out and how, you know,
51:20a nervous 30 something who's an
51:23introvert approached, you know,
51:24cold calling organizations about like,
51:26hey, you know,
51:27do you want to do something together?
51:29Do you want to partner together?
51:31And so you might have those initial
51:34conversations with folks to try
51:36to test out the water.
51:37And that looks different than when
51:39you have a partner and you're ready
51:41to begin planning a study and then
51:44when you've got to study concept,
51:45when you're really like, OK,
51:47we're going to do this now,
51:48How do we move our partnership
51:50from just these plans to actually
51:52getting this off the ground?
51:54And then when you get that,
51:55those projects off the ground together,
51:57there's a whole other set of
51:59challenges that some you could have
52:01predicted but many that you can't.
52:03And it doesn't matter if you
52:04predicted them or not.
52:05They're challenges that you're
52:07going to have to address over time.
52:09And so I just wanted to.
52:12One of my partners from PCSAO,
52:14Fawn Goodell,
52:15she and I went out to Saint Louis
52:18for the Implementation Research
52:20Institute back in 2019, I believe.
52:23And we did a session about working
52:25with your community partner.
52:27And so in preparation for that,
52:30I was like,
52:31you know,
52:32I think we should talk about like
52:33what our partnership looked like
52:35and like all the silly questions
52:36I asked you along the way and
52:38all the questions you wish I had
52:40asked you along the way or that
52:42you wish you had asked me.
52:44And so we,
52:45we put together a list of
52:47questions that's on our project
52:48website and I'll share the link
52:50with you at the end of the presentation.
52:52But I think that each step
52:54of that relationship,
52:55each stage of that partnership, you know,
52:58there's some really intensive questions
53:00that I think we need to ask ourselves
53:03as scholars and that our partners
53:05also need to be asking themselves.
53:08You know, who cares about this work?
53:10You know when you're first,
53:11when you first have an idea or
53:13you're first interested in working
53:14with a community partner, like who?
53:16Who cares? Who cares?
53:18And not everybody's going to
53:19care about your research idea.
53:21And not everything that your partner
53:22cares about is going to be something that
53:24you want to invest your time and study.
53:26So I think being really honest
53:28about who cares about your work
53:30is important for directing you to
53:32the types of community partners.
53:34When you're going to have this really
53:37supportive and productive partnership,
53:39will you partner and when might
53:41you not partner?
53:42Not every question needs to
53:44have a community partner.
53:45You know,
53:46sometimes our theory building questions
53:49or if we need to develop evidence around
53:52a relationship between constructs,
53:54if we can do that using secondary data,
53:57that can be a much more efficient
54:00approach to establishing that
54:01evidence base than putting the time
54:03into developing those partnerships.
54:05Likewise, you know, working with a partner,
54:08you also have to recognize that they
54:10are under political pressures that we
54:13don't necessarily have as scholars.
54:15And so even though you want to
54:17be the subjective observer,
54:19sometimes that creeps into the conversations.
54:21And so if there needs to be this
54:24very objective outside view where
54:26that is not accounting,
54:28I'm not advocating at all for adjusting
54:31your relationships or adjusting your
54:33findings based on what your partner wants.
54:35You know,
54:36the science is science.
54:36The data or the data you got to report them.
54:39But there are ways You know you are.
54:41You become more sensitive to
54:43what your partner needs and the
54:45pressures that your partner is under
54:47through your partnerships.
54:48And so if there is a situation where
54:51you need uncontrovertible evidence or
54:53you need a decision about something
54:55that's objective and external,
54:56a partnership is not necessarily
54:58going to get you there.
54:59So I think you need to be thoughtful
55:03about purpose of your partnership
55:05and whether you can ask your
55:08research question without them.
55:10So I'm a big advocate for partnerships,
55:12but I think you need to be
55:14thoughtful and make sure that it
55:15is that it really is necessary.
55:17What are you looking for in a partner
55:19and how can you be a good partner?
55:21They're the field is rife with stories
55:24about academics who come in and do
55:26research on community partners or on
55:28communities and not with communities.
55:31And yes,
55:32we have publisher parish and funding
55:35pressures and we sort of have to put
55:37them on the side for a second and
55:39make sure that we're we're being good
55:41humans and respectful of our partners.
55:43How do you want to partner?
55:45You know you might not want to Do you know.
55:48If you are in a point in your career
55:50where you need to get publications out,
55:52you might not be able to be full
55:54on a community based participatory
55:56research approach and see control of
55:59your research agenda to the partners.
56:02You might not be able to be in that place,
56:05but if you are, that might be
56:06the way that you want to partner.
56:08So just giving some thought to how
56:10do you want to partner with these
56:12community agencies is very important.
56:14And then what challenges do you expect?
56:16You can't anticipate them all,
56:18and there will be some doozies that you
56:20encounter along the way that you're like,
56:21oh, I couldn't have predicted that one that
56:23should go in a book somewhere authored
56:25by student eventually let long term.
56:29But the more that you can anticipate
56:31some of those challenges,
56:33the more that you can put some protections
56:35in place and address them up front.
56:38Some lessons along the way do no harm.
56:41We are not independent observers
56:43when we work with community agencies.
56:45We are actively part of their system.
56:48Even if you have a contract to
56:51be an external evaluator,
56:52you are still producing information that
56:55your partners are going to use to shape
56:58their services and program delivery.
56:59You are not an independent observer.
57:02You are in the system in some
57:04way shape or form.
57:05And so I think it's we have to
57:07make sure that we are not doing
57:09anything that's going to cause harm.
57:11And so thinking long term and
57:13strategically and having upfront
57:15conversations to be to be aware of that
57:20prioritize research questions that matter.
57:23You know one of the best examples
57:25that I have of this is early on
57:28in the Start in the Start project,
57:30we had a partner from the county,
57:33the Ohio Association of County
57:35Behavioral Health Authorities and
57:37they are a membership organization.
57:39We have these regional behavioral
57:41health coalitions that get together
57:44and help do service planning and
57:46priority setting in the regions.
57:48When you were start on the
57:50child welfare side,
57:51they were having some difficulty
57:53identifying behavioral HealthPartners
57:54in each of the stark counties to
57:56work with the child welfare agencies.
57:58And Sherry Walter,
57:59who's the executive director of the
58:01Ohio Association of County Behavioral
58:02Health have already said they should
58:04be talking to their coalitions.
58:06And I bet you that the ones,
58:08the counties that talk to their coalitions,
58:11they're able to help broker
58:13those relationships.
58:14And sure enough,
58:15those are the preliminary results that
58:17we have under review right now that
58:19in those counties that worked with
58:21their behavioral health coalitions,
58:22they were able to implement start with
58:24higher fidelity because they were able
58:26to put those partnerships in place.
58:28So that might not have been
58:30my #1 research question,
58:32but our partners felt very strongly that
58:34this is what was going on in the system.
58:38And so that's the question
58:39that we decided to pursue.
58:41That was the hypothesis we decided
58:43to test together because it
58:44was important to our partners,
58:46Check your ego.
58:47You don't know all the things.
58:49You have a PhD,
58:50you're very smart,
58:51but you don't know all the things and
58:53you don't know what it's like day in
58:55and day out to be these clinicians
58:56and these leaders in the setting and
58:59the types of pressures that they
59:00that they that they experience.
59:02And so I think listen more than
59:05talk and and you know, you know,
59:08pay attention and respect what
59:10you're hearing from your partners.
59:12This one is controversial and definitely
59:14perhaps not endorsed by my institution,
59:17but sometimes you got to do
59:19some stuff for free.
59:20You know our college,
59:21I was at the College of Social
59:22Work before coming to the College
59:24of Medicine and that's our team
59:26there at the College of Social Work
59:27got involved in start initially.
59:29And we did that by saying,
59:31hey,
59:31I know you don't have any money
59:33set aside for evaluation,
59:34but we want to be able to help
59:36you in some way.
59:37So maybe there are some surveys
59:38that we could do or maybe do some
59:40literature reviewing for you.
59:42So some things that are part and parcel,
59:44things that we can do very easily fully
59:46within the context of our institutions.
59:49Those are things that can help
59:51your partners get started in ways
59:53that are low cost but potentially
59:55impactful and helps you get your
59:56foot in the door and demonstrate
59:58that you care about your partners.
59:59So you want to put some skin in the game?
01:00:02Compensate time and effort.
01:00:04Likewise,
01:00:04you might not have any money at the
01:00:07beginning to offer to your partner
01:00:09to work with you, but when you do,
01:00:12that is an opportunity.
01:00:13We can build partners into
01:00:15grant applications.
01:00:16Yes, it's going to take a chunk of
01:00:17your budget and you're going to
01:00:19have to work with them to set up a
01:00:20subcontract with your big institution.
01:00:22But it goes further in terms of
01:00:24building your partnership and create
01:00:26a safe environment for disagreement.
01:00:28The last thing is about challenges
01:00:31and Deepa Gopalan,
01:00:32Byron Powell and I wrote a paper
01:00:34a couple years ago about sort of
01:00:37the challenges of working in a
01:00:38with community based organizations
01:00:41in implementation research.
01:00:43And at the time,
01:00:44we were very early career scholars
01:00:46and thinking about the pressures
01:00:48and how to balance, you know,
01:00:50the needs of your partner and also the
01:00:52needs of assistant professors who were like,
01:00:55Oh my gosh,
01:00:56I I need to build my dossier
01:00:58so that I can keep my job.
01:01:00And so,
01:01:01you know,
01:01:02we outlined several different
01:01:03challenges and some strategies for
01:01:05addressing them in this paper.
01:01:07This was an administration policy
01:01:10and mental health services and
01:01:12so you know this is it's a it's a
01:01:15big lift and it it takes a lot of
01:01:17work and it's kind of unsung work,
01:01:19but really important to to push those
01:01:22partnerships and the science further.
01:01:25And the last thing I know that we're,
01:01:27we're getting a little close on time.
01:01:29So you know,
01:01:30you're probably wondering
01:01:32why is there rainbow,
01:01:33you know in a scholarly presentation.
01:01:36But I think that partnerships,
01:01:38community based partnerships,
01:01:39you know the good ones where you
01:01:43feel like you can be creative and
01:01:45you're asking and answering new
01:01:47questions that are so important
01:01:49to the field and your partners.
01:01:51Feel free to come with to you with
01:01:54questions and where you can see
01:01:56the results of your work informing
01:01:58practice and you can see that
01:02:00practice informing the research.
01:02:01Where your partners feel free to come
01:02:03to you and say I think you're wrong
01:02:05and you're on the wrong track and here's why.
01:02:08Or your interpretation is wrong.
01:02:10Those partnerships are rainbow unicorns.
01:02:13They are rare.
01:02:15They are gift.
01:02:17And so you know,
01:02:18those are the ones that I think are
01:02:21going to take our field really into
01:02:23the next chapter of implementation
01:02:25science and service delivery.
01:02:27And so I will also leave you with
01:02:29here a link to some of our work we've
01:02:31been using to develop this toolkit,
01:02:34collaborating across systems
01:02:35for program implementation.
01:02:37We have APDF version.
01:02:38We also have an online module with
01:02:41the same content that sort of laid
01:02:44out in a more user friendly platform
01:02:46that's available on our website.
01:02:49It's me managing the website.
01:02:51So it's not beautiful,
01:02:52but there's content there in case you're
01:02:54interested in learning a little more.
01:02:57So I think I went over and I apologize.
01:03:00So maybe
01:03:03discussion. I think you're fine, Alicia.