INTRO

PIPER

Hi, this is Piper Hale, and you’re listening to Inform Me, Informatics!

Like many other informatics nerds all around the country, I’m currently gearing up for National Health IT week, a 13-year-old tradition that dedicates a week at the end of September to action for improving the state of—well, health IT! This year, the National Association of City and County Health Officials (or NACCHO) is trying something new in honor of Health IT Week, and will be hosting a virtual conference on September 25 to connect local health department staff working in health IT with each other, and to jointly tackle some of the biggest current issues in health IT.

In anticipation of this conference and its focus on health IT innovations in local public health, I wanted to talk to someone working in a role this conference was designed to serve, who’s currently doing something innovative in this space. I was very fortunate to get the opportunity to chat with Dr. Alyson Shupe, who manages informatics and epidemiology planning for Tri-County Health Department in Colorado. Alyson told me all about some really cool localized work coming out of her agency to stem the tide of the opioid epidemic using data. In fact, Tri-County just won a Model Health Practice award from NACCHO in recognition of the outstanding work it’s doing in its jurisdiction, which is how I first heard about this project.

When I spoke with Alyson, I wanted to understand why and how this project got started, so I asked her what the opioid epidemic looked like in her corner of the country in Colorado.

ALYSON

So in Tri-County in Colorado, which is like all over the nation really, as the opioid crisis kind of started to show up, we were tracking it through our regular surveillance data, and then we found that we were getting increasing inquiries from community residents, from partner organizations, and probably most influential was one of the county commissioners was really encouraging our executive director to really help understand the nature of the issue in Colorado and in the tri-county region. And so once we were really finding that people were needing more information, we had the idea to build kind of a one-stop shop to help people understand how many cases, what were the demographics, so geographically where there are hot spots in our communities, and to map some other resources, such as treatment centers and naloxone distribution and the likes.

PIPER

The final form of this information one-stop-shop ended up being an interactive opioid mapping website. It has dashboard views that map neighborhood-level data like opioid overdose deaths, heroin overdose deaths, the location of Naloxone retailers in the area, and locations for drug take-back programs. It’s all visually very clean and engaging, and easy to understand, mostly using heat maps that immediately draw the eye to key hotspots. We’ll have a link in the shownotes if you want to take a closer look for yourself—and you should! It’s really interesting.

Putting these kinds of maps together means unifying some complex and diverse data sources in one place to be digestible to the layperson. Alyson walked me through some of these data sources and how Tri-County pulled them together.

ALYSON

Already, we have a pretty rich data environment, I would say, in Colorado, and just as part of our routine surveillance, you know, we have access to the identified birth and death records, the vital statistics from the state health department. We get our hospitalization data, we have both emergency department data and discharge data, which we get from the Colorado Hospital Association. We have Behavioral Risk Factor Surveillance data for the state and have a relatively robust sample size, and so we’re able to get some of those data elements from that survey. We have our version of the Youth Risk Behavior Survey in Colorado, we call it the Healthy Kids Colorado Survey, which asks questions of high school students, and we have county-level data from that survey. So you know, just really looking at the range of sources that we had, we began to build, you know, put these resources together to try to, you know, clarify and tell the story of what it looks like in our community as we were going through it.

And at that time where opioid was just really kind of being understood as the, you know, crippling crisis that it is, new data were coming on the scene as a result.

We also have syndromic surveillance, and so that’s gathering real-time data from hospitals to see kind of what’s going on with regard to the emergency department and also just kind of the diagnosis of the folks who are coming in.

SAMHSA produces regularly a dataset with treatment centers and treatment programs, you know, down to the address of the center, the name, the address. And we were just able to pull the data off of their website and use it in our mapping product. So you can go to an icon on a map that has, you know, you click on that particular icon and information about the treatment center will pop up that’s very, very rich and robust, not only the contact information but who they serve, what kind of insurance they take, you know, just a wealth of information. So once we really started kind of scouring the universe for what’s out there that would be helpful to partners, the public, our staff, our various constituent groups that we serve. We felt like we were able to discover some new data resources and kind of use some of our traditional data resources in a more unique way to really use both the mapping and other features to display these data.

We kind of built this site from the grassroots level. So these inquiries that were coming in about, you know, what’s going on, what does this look like, who’s at risk, where is this happening in our community, all these kind of really intriguing questions also led us to exploring further and further into what data exists to answer these questions that people had.

So our criteria was, again, that we needed to make sure that the data were from a reliable source and that we could use them at least at the county level, and we were really looking, obviously, for mapping. Mapping at the county level was not really very useful, so we were intent on grabbing data that could be displayed at the census tract level. And that introduces a range of technical and scientific concerns in terms of protecting confidentiality. So like without a records data in our state, there is a better records policy that dictates what level of granularity you can show in order to still protect the privacy of the individual from whom the data is provided. So in straightforward kind of regular, not thinking about mapping, you know, we have those parameters, they’re set, we know them and abide by them, of course. But with mapping, you have an opportunity to use a range of procedures, processes, tools to display data but still maintain that confidentiality. So the maps that we have on our site are kernel density maps. And I won’t get to any of the technical aspect of how those are actually created, but you know, we’re not displaying point level data. In other words, you can’t, other than the treatment center data which you can click on and it gives you address, phone number, etc., but for the, like, death data, obviously, we don’t want people to have those data exposed in a way that could identify an individual.

So if you look at our map, the kernel density map, you know, you’ll see areas that are clearly lit up or, you know, some people call them hotspots, that not a technical term, but areas where this crisis was taking a toll and also being able to see how that is moving over time. So for our data on the map, we’re able to show some much earlier further back using that data, you know, like what was the nature of the crisis then, how has it grown through time, and how has it moved or has it moved geographically throughout our community. And the reason that’s so important is being able to make sure that any response and efforts, be they prevention or a policy or access to services, knowing where folks are is really important to make sure you’re getting the word out and building access in those particular communities. So using data, being able to use data that we could visualize at the census tract level but still protecting confidentiality was a unique challenge, and it really drove some of the things that we could map and some of the things that we could not.

PIPER

One thing I wanted to understand was the vision for how these maps and the stories they tell about the opioid epidemic in the Tri-County area might influence public health practice or individual action. For Alyson, a big part of the benefit of these maps is empowering individual community members to protect their own health and the health of their families.

ALYSON

So for instance, you know, one of the main features of this is getting naloxone in the hands of people who need it, such as emergency responders, but also parents, you know, who are struggling with maybe a child who suffers from addiction issues, or spouses of someone who’s struggling with that, community providers, people themselves who wanna carry naloxone. So really, that’s aspect of how we try to combat this crisis. So by putting as much information as we could without making it just be garbage pail of everything we possibly might know and everything you might possibly do, really trying to curate the website so that there was enough that could appeal to our various audiences

And we also did not want to, even though we were thinking kind of one-stop shop for information, we weren’t trying to take on any kind of role of, you know, triaging individual calls, because that’s not what we do, that’s not Tri-County Health Department’s role in this.

But we know for a fact that a lot of our staff, internally, in the organization use it, and they use it in presentations. They advertise the site to their partners. We get feedback sometimes from citizens that will say, “Hey, I saw this. I don’t understand this,” or “This is great. Where can I get more information?”

But another thing that happens is you start seeing maybe a screenshot of the map or a graph or data that was downloaded through a Tableau presentation of data on the website that’s showing up in other people’s report, presentations, news, clippings, those kinds of things so that you know that people are grabbing the information they need and hopefully properly sourcing it and using it to spread the word and describe the situation through various other mediums.

I think one of the mantras of public health is, you know, what gets measured gets done, and so bringing, putting a really bright spotlight on the opioid crisis on our website like this and having so much data about what was going on really kind of reflected a commitment that we were making as a health department to really address this issue.

So I think it helps us as a health department really say, “Okay, how do we focus our resources in meaningful ways to combat this problem in our community?” And so really honing in on focusing on this effort or this issue in particular.

Most public health issues are not really isolated, like, “Oh, it’s just because of one thing or another.” And particularly with substance abuse, of course, or substance use disorder, it’s very complex and connected to a range of different things. And one of the things we learned from the opioid crisis that I think was new in terms of both the addiction world and the drug environment was with so much of the opioid addiction coming from prescription drug use and prescription drug misuse. It’s not just about, of course, the street drugs, it’s really about the legally prescribed drugs. And so, how do we understand, you know, those two components of it and make sure that we’re addressing, you know, the right issues from the right lens. So I’m sure the whole audience is aware that there’s many efforts going on, from suing pharmaceutical companies and legislation around that, to prescription drug monitoring programs to make sure that prescribers have accountability for how much they’re prescribing and to whom, and then not forgetting that a lot of the crisis also, there is that fight in the medical system, and it’s connected to folks who maybe even started on prescription opioids and then maybe ended up using heroin.

I think one of the things that we have also found is there’s an obvious connection between suicide and the drug prices. And you know, some of the things, it’s not as easy to ferret out, is an overdose an accidental overdose or an intended overdose? And you know, in some ways, that is a completely irrelevant question, because the outcomes is the same. But where it’s important in terms of prevention is not forgetting that part of what’s behind, at least for some people, in substance use disorder, is the mental health component and depression and those kinds of mental health conditions that feed into substance use disorder. So it’s a very complex problem at the community level and the family level and the individual level.

So public health really addressing, trying to address issues from that population basis, you know, we have our lane that we stay in, but we also, through this work, I think really made a lot more important connections with folks in the community who are providing services for those who have substance use disorder, and as that relate to mental health conditions as well. And so, you know, really understanding and expanding our connection with that more complicated fabric of what’s available in the community. I would like to mention, too, interestingly enough, during this time we were building this site, we were also conducting our community health assessment. And part of our data gathering there was input from residents, and certainly, we were hearing residents speak about drugs and their impact. Some spoke directly to prescription misuse, others to more street drugs. But you know, it was really coming up as a problem that people were seeing and could articulate. Likewise, we got a lot of input from our residents as well about the importance of social connections, and as you delve into more of the psychological literature behind addictions and suicide and whatnot, mental health problems, you know, we’re seeing this concept of isolation and the lack of social connections that can exacerbate people who are having problems with maybe some addiction issues or some mental health stuff. A lot of that is at the core.

And so understanding, too, the importance of building that or making conditions right for people to connect to communities in ways that we have kinda moved away from in our culture, both in terms of how our, you know, built environment, how it looked, and whatnot. So, to me, it spans everything from the beginning of life and supporting adequate prenatal care and whatnot, and then how people are connecting to their neighborhoods, their families, their faith communities, the schools, to organizations and resources, and then the availability issue and what that looks like, the role of the criminal justice system, the role of the health care system. This is so multifaceted, and we all have to be kind of in step working together and thinking about how we put resources together in ways that support people in ways that they have not been supported before.

PIPER

Given Tri-County Health Department’s focus on collaboration, especially in the context of addressing complex intersections of the social determinants of health, I wanted to get Alyson’s thoughts on NACCHO’s upcoming virtual conference and what she sees as the value of convening these local health IT folks from all over the country to interact with one another.

ALYSON

I think what happens a lot of times in local health departments, and state health departments for that matter as well, is that if you don’t have a strong informatics presence and haven’t demonstrated the link between data and managing information and the systems and infrastructure that supports it all, you’re not really seeing the full picture, and people tend to, and health departments, go from, you know, anything related to the computing environment, that must be an IT issue or an IT problem. And I think this conversation, and what the conference will help us do, is begin to really think through how we help our staff and organizations understand that IT is very much responsible for the infrastructure pieces, the very important pieces of setting up and protecting our computing environment. But they’re not gonna tell you, you know, how to necessarily use data or, you know, how anything related to information is not necessarily an information technology issue. And I think the beauty of informatics, for me, is really helping to explain that there’s a continuum and information is collected and consumed and analyzed and stored and disseminated in systems that are built to attend to privacy and protection and security.

But the IT side, doing that very important job, kind of ends with, “Okay, then what do you do with this information?” And I kind of see then that’s where the informaticians and epidemiologists kind of play their role in making sure that we’re getting the right information, that we’re analyzing and manipulating it appropriately, determining what can be shared and what can’t, looking at linkages within different datasets within the organization, and really providing them the data and information to our program folks who are the ones who are really acting on that information. And so understanding, I think, the links, but also the discrete beginning and end points of the various functions needed to really make the best use of information and data in the public health arena. I think these local health IT conversations will really help us as a community across the country understand better what those roles are and how to manage that resource within an organization.

PIPER

And finally, that million-dollar question we ask all our guests: I asked Alyson how she personally defines public health informatics.

ALYSON

Boy, that’s always kind of a struggle, and I think we’re evolving in our understanding as a learning community about what that term actually means and how to make it really more tangible than kind of the Webster type definition of informatics. I think in its early days in public health, it was really looked more as, you know, electronic medical records. That’s public health informatics. And I think it’s well beyond that, at least the way we use that term in our organization, and I kinda already said it before, but really, the focus on bringing data into the organization, how it lives in our information systems, how the IT professionals work to make sure that systems are up and running, effective, protect the data, so that informaticians can manipulate and use those data against relinkages with, possibly, datasets within the organization, sharing data out with other partners in the community. So really focusing on that data management and information management piece is, I think, what I think of is the core of informatics as it connects program folks and epidemiologists to the tools and systems that make the work possible. And having a strong, secure IT infrastructure is critical to the whole thing.

I think this is an extremely exciting time to be in the business of information, whether you’re more on the information technology side or more on the epidemiologic side. But you know, with having so much more data available, the technology, increasing our ability to do so much more with information, the possibilities are just mind-blowing. And at the same time, you know, we have to be very judicious about what we’re doing. And you know, everybody talks about big data and, you know, “Oh, wow, we capture all this, you know, commercial data, or whatnot.” And I think we have to be really circumspect about making sure that we’re not just grabbing data and pushing it out for the purpose of, you know, because it’s there and we can, to really make sure that we stay true to the scientific principles of data and epidemiology and curate these data in ways that really do serve a very important purpose in identifying and pointing out solutions necessary for combatting various public health issues.

PIPER

Thank you again to Dr. Alyson Shupe for taking the time to come on the show and share Tri-County’s success stories! Thanks also go to Angie McPherson at NACCHO for connecting me with Alyson.

If you’re interested in attending the NACCHO virtual conference on September 25, “Inform Me, Informatics” listeners can get $20 off registration. Just go to lhitcon.org to register and enter the code INFORM2019 at checkout. That’s INFORM2019. One of my public health heroes, Dr. Karen DeSalvo, formerly acting assistant secretary for health and national coordinator for Health IT, will be keynoting, so you won’t want to miss it!

This podcast is a project of the Public Health Informatics Institute, which is a program of The Task Force for Global Health. Visit phii.org to learn more about all of our informatics work! You can also find us on Facebook and follow us on Twitter @PHInformatics. The music used in this episode was composed by Kevin MacLeod.

Do you know of an innovative or interesting public health informatics project or story? Reach out and let us know at podcast@phii.org!

Also, if you want to learn more about Tri-County’s Opiate Mapping website, head to phii.org, where we have links to the map dashboards in the notes for this episode.

I’m Piper Hale, and you’ve been informed.

BUTTON

ALYSON

Got it, okay. Wow. Well, great. This is a really fun experience

PIPER

Good.