The Importance of Public Health Surveillance in Responding to Overdoses

ROBERT JOHNSON:
This is Public Health Review. I'm Robert Joshnson.

On this episode: access to comprehensive, timely data takes on new importance as the COVID-19 virus spreads; public health laboratories always playing a critical role, whether the concern is novel coronavirus or, as we examine today, the opioid crisis.

DR. EWA KING:
In public health, reliable data is what we use to define the extent of the problem, it's what we use to monitor trends, and, ultimately, we need data to tell us that our public health approaches are working.

DR. MARK LEVINE:
Surveillance is critical to everything we do.

JOHNSON:
Welcome to Public Health Review, a podcast brought to you by the Association of State and Territorial Health Officials. With each episode, we explore what health departments are doing to tackle the most pressing public health issues facing our states and territories.

Today: public health labs and the role surveillance plays in every community health program.

Labs have attracted national attention recently as policy makers and healthcare officials discuss, with Americans, plans to test for COVID-19. Indeed, labs are vital to public health professionals as they try to understand myriad health trends and concerns in their communities—among them, the ongoing fight against opioid misuse. Even as labs begin to test patients thought to have the novel coronavirus, they continue to work on gathering and sharing mortality and morbidity data related to opioid-connected illness and death.

Our guests are leading efforts to improve and share opioid data.

Dr. Mark Levine is the commissioner of health for the Vermont Department of Health. He tells us about the state's opioid dashboard and previews a second public database currently in the works.

But first, we visit with Dr. Ewa King. She's the associate director of health at the Rhode Island Department of Health; also, director of the Rhode Island State Health Laboratories, and the past president of the Association of Public Health Laboratories.

KING:
In public health, reliable data is what we use to define the extent of the problem, it's what we use to monitor trends, and, ultimately, we need data to tell us that our public health approaches are working.

This is very much true in the case of the opioids epidemic: we need data to tell us what populations are effected; we need data to tell us what drugs are involved; we monitor trends; and, ultimately, we want to know that whatever we doing is helping to improve the situation.

JOHNSON:
Do the public health people fighting the opioids crisis have the information they need?

Are they getting real-time data to address these concerns across the country?

KING:
We have data available to monitor morbidity and mortality in the case of the opioids epidemic.

However, what we have seen is that the data used for mortality and morbidity have evolved very differently. Mortality is something that we measure very well—we know exactly how many people have died from opioids overdoses.

That's not necessarily the case for morbidity—it's much more difficult to measure how many people have experienced overdoses and have not died from them—and the reliability of those numbers is also very different at this point.

We're hoping to change this. We're hoping to improve the surveillance, the quality of surveillance, for non-fatal overdoses.

JOHNSON:
What are you trying to do along those lines to improve that information flow to the people who need it most?

KING:
So, I approach this, obviously, from the laboratory perspective; and what I was able to notice is that, for fatal overdoses, laboratory data is used extensively. Every case of a fatal overdose is being investigated thoroughly with the use of numerous laboratory tests.

That is not the case for non-fatal overdoses. The reason for this is that people who experienced symptoms of an overdose might show up at the emergency department at a hospital, they might not spend much time there—they get treated, sometimes they don't stay for very long. So, those fleeting encounters with healthcare are more difficult to capture. As a result, we don't necessarily know at all times how many people have experienced an overdose.

And what we're also missing is the laboratory component of this. Just because someone shows up with symptoms that resemble an overdose doesn't mean they actually are experiencing an overdose. So, we really do need laboratory data to get better quality data points.

JOHNSON:
How do you do that? How do you accomplish gathering all of this additional data from these cases that aren't being reported?

Is there a plan being worked up right now or a process that's going into effect?

KING:
So APHL—which is the Association of Public Health Laboratories—has been working on this issue for quite a few years. We are attempting to convince more laboratories to get involved in the program that what we call biosurveillance, opioids biosurveillance.

So, we're looking for public health laboratories to start testing specimens—urine samples from hospital emergency departments—that may have been characterized by a hospital laboratory, or maybe not, and forward those to the state departments for testing.

This will tell us exactly what drugs are being used by the population, it will tell us what combination of drugs are most frequently encountered, and will produce an entirely new, much more reliable, data stream than what we are dealing with right now.

JOHNSON:
Do they have the samples in hand already?

KING:
It varies. It depends on the standards of care used by hospitals in different states.

Not all hospitals treat overdoses the same way. In Rhode Island, all hospitals tend to test their patients that might appear with symptoms of an overdose. In other parts of the country, that's not necessarily the case.

However, that emergency department encounter exists, and there is a potential for collecting biological specimens for testing.

JOHNSON:
Do you have to pass a law in Rhode Island to get that data from the hospitals, or are they doing it voluntarily for you?

KING:
In Rhode Island, as is the case in most states, the Department of Health has broad authority to investigate health problems and outbreaks.

So, we happen to have a very precise line in our regulations for reportable diseases that allow us to request biological specimens in addition to the actual counts of non-fatal overdoses from the hospitals. That may not be the case for all the states; however, most health departments do have some authority to investigate problems and that's what we have used.

JOHNSON:
So, a state that doesn't do that right now may already have the authority—maybe choosing not to do it, maybe needs to know how?

KING:
I think the how is very much what APHL is trying to accomplish, in collaboration with the CDC; to work on components of the program that needs to be in place.

It's a fairly complicated situation because many components that exist for infectious diseases, other types of reportable diseases—that has been going on for years, therefore people are used to it. This is a new program where we tell people you can utilize the same mechanisms for non-infectious diseases such as an opioids overdose problem.

So, the states need to have the authority. They need to determine what their legal authority is. They might need to clarify their regulatory authority, but they also need to know how to do it and which partners they need to engage to move this forward.

JOHNSON:
And the Association of Laboratories is attempting to teach them or provide them instruction for moving ahead if they're not doing so already?

KING:
Right. I wouldn't call them instructions—we call them guidance.

I'm here for a meeting at the APHL headquarters that will finalize our Opioids Biosurveillance task force model, biosurveillance strategy, so that is a resource that we are looking to make available to state laboratories and state health departments to figure out what components of the surveillance program they would like to put in place.

It's not meant to be a prescription. It's not a set of instructions because we understand states have different needs and they are different sizes. Not everyone is as small as a Rhode Island where we can pick up the phone and call all of our hospitals in an hour. However, we do feel that our experience will help inform the model program.

JOHNSON:
We are talking about getting hospital laboratories to do this testing.

There are lots of other kinds of labs—are they involved at all?

KING:
Yes.

That's another difficulty and difference from the system that we have in place for the infectious disease surveillance and the concept of reportable diseases because, until a few years ago, drug use and overdoses were not really considered a public health problem.</p

The majority of the laboratories involved in testing for drug use are not public health laboratories. Public health laboratories traditionally work with clinical laboratories in the space of infectious diseases. This is new space for us.

Rhode Island State Health Laboratories are in a unique position because we have both the laboratory that is a clinical toxicology laboratory, and we also have a forensic toxicology laboratory. So, forensic laboratories or crime laboratories are another side of partners that we needed to engage to make the system work.

JOHNSON:
Included in that group would be these labs that employees are sent to for testing for drug use and other things before they get a job.

KING:
Absolutely.

So, those are toxicology laboratories. They might be clinical laboratories or specialized clinical toxicology laboratories.

There's a lot of drug testing—that actually is part of the problem. Because there is so much drug testing for different purposes, hospital laboratories have to develop a system to pull out the specimens related to overdoses. So, just because of a drug test is requested, that does not mean that it was an overdose.

So, it's somewhat difficult for hospital labs that just get tests requests and urine specimens to sort out which ones belong to the emergency department and which ones were involved in non-fatal overdoses. So, that was part of the dilemma on how to put this in place.

JOHNSON:
What do you say to them about the need to get this done as soon as possible?

KING:
Well, one of the selling points that we have for our laboratories is the fact that we have all, about 15 years ago, developed a network of laboratories that were designed to deal with what we call chemical threats—those are our laboratory response network laboratories. They were developed for dealing with issues such as chemical terrorism, acts of chemical terrorism, or chemical spills.

However, as it happens, the technology, the laboratory technology that we have put in place, help us develop tests for opioids testing. Most states already have in place the expertise, the personnel, and the equipment that it's needed to put this program in place.

JOHNSON:
What can we learn from all of that extra data if we have in our hands?

KING:
We don't know if what we are investigating—for fatal overdoses—we don't know if the drugs responsible for fatal overdoses are exactly the same drugs people overdose on and survive. So, that's one very basic question that no one has the answer to, and that's one thing then we are hoping to learn from this experience.

We also are looking to see what population, what are the kinds of people who overdose? Do they come back again and again—there is some anecdotal evidence to this effect—and, again, are there other combinations of drugs that we might not know about that cause people to overdose?

We tend to talk a lot about opioid overdoses; but what we are learning is that we really no longer have an opioid-only problem. What we have is a poly substance problem, meaning people don't just use opioids—people use amphetamines, people use stimulants, they use cocaine a lot—so, there are a combination. So, we do think there are combinations of drugs that are most likely to cause people to overdose, but we don't really know which ones they are unless they overdose fatally.

JOHNSON:
The goal, then, with the data is to drive better outcomes, to give public health professionals the information they need so they can perhaps intervene, maybe develop a strategy that starts to bring these numbers down?

KING:
Absolutely.

If we know people are overdosing in a specific location, we can make sure that we put out alerts, that we have some kind of intervention that helps people learn about, for example, a dangerous combination of drugs that's circulating.

Obviously, those interventions are not predefined, they're not always obvious. But the better data we have, the better interventions we can devise.

JOHNSON:
There is an appetite for this, obviously. People who are consuming this data need and want more of it.

KING:
I hope so. That's certainly our goal. We definitely do see a lot of interest at the federal level.

I am very encouraged by how many states have applied for funds for laboratory programs under the most recent round of opioids grants from the CDC called Overdose Data to Action. So, that kind of title captures it all.

We're looking for a better data, but we are really looking to drive action that would reduce the incidence of non-fatal overdoses and reduce the prevalence of drug use overall.

JOHNSON:
There's funding available to help states improve their data on this topic?

KING:
CDC has made funds available for states to apply to a variety of approaches, one of which—and they call it an innovative approach—is the laboratory-based surveillance programs, including biosurveillance.

There are other types of laboratory approaches that one can use: we can monitor and track drugs that have been seized by the police, for example; some sort of surveillance of what circulating on the street—that is something that we are doing in Rhode Island also.

But we find the closer we get to the people who actually consume the drugs, the more accurate the data will be.

JOHNSON:
Do you have any projection at all as to how long it might take to achieve that data set you're looking for?

How long do you think it will be before you're able to say, "We have the information. We need to do something about these issues."?

KING:
Well, that's very difficult to predict.

In Rhode Island, we are doing it right now—we have under our belt six months’ worth of data collection. So, we are at the point of looking at what this is telling us, looking at the differences, because we're also collecting tests for results from the hospitals—which are different types of tests, to be clear. What hospitals do, they usually use screening tests, which are meant to quickly determine the general class of drugs that a person might have been using.

When these specimens come to our state laboratories, we use much more sophisticated—more expensive—equipment to tell us precisely which drugs and at what quantity was used or concentration was used. The big differences—we can see at much lower levels than those screening tests also, so that's another benefit of using our laboratories.

JOHNSON:
Wrapping up, what should public health professionals know if they're interested in pursuing this effort?

KING:
Public health professionals in states that have an opioid overdose problem, and there are many of us, need to think about the different approaches for non-fatal overdose surveillance.

We know that's a difficult problem to resolve. The data points are not as easily available as some other data streams. However, we do know that we get better data if we involve laboratories.

If those are laboratory-confirmed cases that are consistent with federal guidelines and the CSTE [Ed. note: Council of State and Territorial Epidemiologists] case definition for non-fatal overdoses, I do firmly believe that this will result in much better surveillance streams, and therefore better data and better public health approaches.

JOHNSON:
And the best place for them to get more information?

KING:
Your state public health laboratory already has the people, already has the expertise, and already has the equipment to get started. They might need more resources, but they can approach APHL for guidance on how to get started and we will be happy to work with them.

JOHNSON:
Dr. Mark Levine shares Vermont's opioid data with public health professionals and their stakeholders, using this information to drive better outcomes for his constituents. He's commissioner of health at the Vermont Department of Health and a big fan of the power of using data to help heal populations.

How has surveillance helped you and your team in the state of Vermont deal with substance misuse cases?

LEVINE:
Surveillance is critical to everything we do.

When I think about when I turn my comprehensive public health response to the opioid issues or any substance misuse issues, I think there are three critical, basic elements that every state has to do.

First one has to do with diminishing the supply, if you will; and if we agree that most people who get into trouble with intravenous opioids or other drugs began with prescription drugs, it's really reducing the circulating burden of prescription drugs in the state.

The second area is really a harm reduction area, making sure that you keep people alive long enough so that if they decide eventually to access treatment, they're around to do that. So, those are harm reduction strategies like having naloxone on the streets, having drug disposal options for people, and making sure that syringe service programs are up and running.

And then, the third element is really having capacity for treatment and making sure that you have a treatment system that people have access to so that they don't have to wait a certain amount of time, during which they could succumb to an unintentional overdose, which would not be a desired outcome.

JOHNSON:
You have a lot of this information online in your opioids scorecard.

Can you take us through that? Tell us what's included and how current that information is to the user.

LEVINE:
Yeah, so we clearly have listings of our accidental overdoses, whether they be fatal or nonfatal, and the ability to compare them to prior time points.

We obviously get fatal data from our chief medical examiner, from vital statistics. We get nonfatal data from emergency department, from a system called ESSENCE—which is the Electronic Surveillance Systems for the Early Notification of Community-based Epidemics—and that really is what the first responders use. And so, that gives us really on a daily basis reports, so it's real time.

Obviously, these non-fatal data give us some indicators of the supply of fentanyl, somewhat indirectly, and informs us regarding needs for naloxone distributed.

JOHNSON:
What's the advantage, then, of having all of this information pulled together in one place?

LEVINE:
Well, you know, our goal is to really look at the substance misuse epidemic in a very global way—from prevention to intervention and harm reduction, treatment and recovery—and so, really holding ourselves accountable with a score card that would actually examine all these points is critical.

JOHNSON:
Vermont is not resting on its laurels.

In fact, you have another scorecard, or surveillance data system, in development right now that should come out sometime—maybe this year, whenever it's ready for public consumption.

Tell us about the Health Surveillance Data Publishing system you're working on.

LEVINE:
Yes.

The HSDP is really an attempt to take all kinds of existing data we've been talking about, much of which is on various websites, etc., and combine it all together in a public-facing site so that not only internal state users can use this—whether they be public health analysts, or epidemiologist, or our grant managers, or our legislator, or governor, or other state staff—but also external users—so, that may be a health care professional, that might be a law enforcement agency, that might be a community group, or a grantee, or an advocate—who's working on substance abuse and all of these critical issues, and wants to be able to see data that has been really compiled from a variety of different data sets displayed in aggregate and user-friendly ways.

Ways that would allow one to look at historic trends, to look at geographic data, geospatial mapping, to look at select data elements, perhaps to download a presentation-ready version of materials for an informational purpose.

And then, of course it allows everyone to do what I said earlier: to plan and develop strategies and make informed decisions; and allocate resources where they are needed so that we can address the continuum of substance misuse management across the state.

JOHNSON:
You and your team must be very excited about the potential of this new public-facing database and its ability to impact delivering public health services across the state.

LEVINE:
Oh, we really are.

But it does give me an opportunity to just say that it also illustrates the frustrations and complexity of doing such a system; where, you know, most states have data sets are coming from a variety of different places—some data sets having personal health information, which, of course, one couldn't display publicly until it's been appropriately de-identified—and all of the intense complexity of data-sharing agreements, memorandums of understanding and agreement.

These words translate into months at a time to actually be able to move forward because none of these things just happen with a snap of a finger. And so, it makes it very complex to make this real-time, and it's taken us the better part of a year to get to where we are now. And that's why I'm very optimistic and hopeful that, before the end of this calendar year, we'll certainly be in the place where we need to be.

JOHNSON:
Dealing with this issue of substance misuse—every state is tackling it—how does this tool figure into that fight?

What do you hope it will allow you and the people in Vermont to do as they try to bring this problem under control?

LEVINE:
Yeah.

So, one can never do as well as they want in this arena. We and many other states are beginning to show downturns in the rates of overdose death, which is a wonderful thing; but that doesn't go to a zero, and that would be a really wonderful thing.

Many of us have been able to show significant downturns in the number of prescriptions being written for opioids. Zero would not be the right end result because these drugs do have utility; but at the same time, it may never get to the level where people really believe it should be based on their own preconception of what that number might be.

So, I think it's helpful for people to look at trend data and, over time, come to an understanding of where they think the right sort of normalization point is, what we're capable of doing.

Obviously with deaths, it should be zero, but that's a very challenging number to get to. With the prescription opioids, there will be a sort of settling out point that hopefully most people will become comfortable with, knowing that there is a use for these drugs—just not in chronic pain and other circumstances that they have, perhaps, been overused in.

It's very challenging for people to feel good all the time about management of substance misuse crises, and I think having data portrayed in the ways we're talking can give a little bit more comfort; because every bad event just sort of strikes that the heart of, "We must not be doing enough, we must need to do more," and I think people need to be able to start feeling better about what is being done and how effective it has been.

JOHNSON:
You can find links to the resources mentioned in this episode in the show notes.

Thanks for listening to Public Health Review. If you like the show, please share it with your colleagues.

And if you have comments or questions, we'd like to hear from you. Email us at pr@astho.org—that's PR at ASTHO dot org.

This show is a production of the Association of State and Territorial Health Officials.

For Public Health Review, I'm Robert Johnson. Be well.