CDC held a webinar on June 11, 2026, to introduce the Clinical Drug Test Dashboard. This new data dashboard tracks trends in illegally-made fentanyl, heroin, cocaine, and methamphetamine based on clinical urine drug test data. The webinar includes a demonstration of the dashboard's features, along with a walkthrough of how to explore and interpret emerging patterns in substance use.
AI Generated·May contain errors
Video Summary
AI Generated
Matt Gladden from the CDC introduces a new early warning dashboard that utilizes clinical urine drug test data from Millennium Health to monitor drug overdose trends. The tool tracks the positivity and co-detection of fentanyl, cocaine, methamphetamine, and heroin among adults with substance use disorders to provide more timely insights than traditional overdose death data.
My name is Matt Gladden. I'm a behavioral scientist in the Division of Overdose Prevention
at CDC. Next slide, please. Oh, sorry, that's me. There we go. A little housekeeping. For this
webinar, the chat will be disabled, but if you have questions, please feel free to put them
in the Q&A, which is located on the top of the Teams bar. Also, I just wanted to say the findings
and conclusions in this webinar are those of the authors and do not necessarily reflect those of
CDC. So in this webinar, we really have three major goals. One is why did we build this dashboard?
What was the key function? How would we like you to use it? And what are some of the pitfalls
to avoid? Then we'll go through the three major web pages on the dashboard, the actual dashboard,
its accessibility page, and a key findings page, which just provides simple key findings up front.
And then we'll end by talking about future directions. So we're going to start with why
did we build this dashboard? What purpose does it serve? What gap does it address?
So looking at recent CDC data on drug overdose deaths in 2024, about 96% of fatal overdose deaths
involve an opioid, a stimulant, or both. Indeed, among these 43 jurisdictions in CDC's system,
43% of drug overdose deaths involve both an opioid and a stimulant. If we go down and see, okay,
what are the specific stimulants and opiates involved? We see 65% of deaths involved in
illegally made fentanyls, and most of that is just illegally made fentanyl. And stimulants now,
over 60% of deaths have a stimulant contribute, and the vast majority of those are either cocaine
or methamphetamines. What this means is when we're looking at the drug overdose epidemic,
we really need to quickly identify changes in these drugs and their co-use incidents,
or how much people are being using or being exposed to.
So there's been great strides in providing more timely drug overdose data. There's 12-month
provisional estimates that are published, but they're meant to smooth out rapid changes. So
we also need systems that pick up those rapid changes. And also when we see a change in drug
overdose, we want to try to test our ideas on why it might be going up or down. So that led us
to look at the ability of clinical urine drug tests to act as an early warning, early insight
into drug overdose trends, and also what's going on in the drug overdose epidemic. Specifically,
the data on the dashboard comes from Millennium Health, an accredited laboratory that provides
drug testing services that helps clinicians provide care to patients by monitoring prescription drug
use, illegal drug use, and the effectiveness of their treatment plan. It has very in-depth
testing, and the data is very timely, available usually within less than two weeks of the test.
So for the dashboard, focusing on these drugs I just said, illegally made phenylcocaine,
methamphetamine, we focused on adult patients with a substance use disorder,
excluding alcohol, cannabis, and tobacco. And we really focused on positivity. How often was
phenyl, heroin, methamphetamine, and cocaine detected in this group? And the advantage of
this data is there's a lot of tests, over 650,000 between October 2023 and December 2025.
So as I said, the data is extremely timely, and we're working to update the data one or two
months after the end of the quarter. We probably released quarter one and quarter two together.
There were some unexpected hiccups. But the other real strength of the data is it provides
clinical insight into people with substance use disorder and treatment, and that is a group at
high risk for overdose. And more broadly, accumulating evidence shows it can provide
sentinel surveillance of major changes in the drug overdose, in drug overdose trends. And also
it provides insights into co-detections of substances. And that's important because that
can complicate treatment and recovery. Here I show you two trend lines. The line in orange
is the trend in overdose deaths. With that line in blue is kind of model estimates using Millennium
Health data on phenyl positivity, methamphetamine positivity within phenyl, and then concentration
of both of those positive drugs. And what you can see is it predicts pretty well the line,
but most importantly, it moves up and down when there are major shifts. So this provides evidence.
We may be able to use this as an early warning system. But when we move and use this data,
we do have to note its limitations. It's a sentinel system, which means it's not nationally
representative. You can't compare phenyl positivity in the West and the East because different
clinical practices may be involved. You may have patients with more severe opiates disorder in
one region than another. And it only describes a small portion of people using drugs. And while
it's an important population receiving treatment, it doesn't get as broad a view of some of the
recreational drug use. And this is just data from one of multiple companies that provide the service.
The last limitation we should always keep in mind, this is detection. So I can't distinguish between
intentional, the person want to take the drug versus unintentional exposure, and also detection
doesn't mean they used it at the same time. It detects exposure over, you know, depending on the
drug days. So it could be used sequentially. It could be used at the same time. We don't know.
So kind of this key takeaway is we are hoping that the data helps us more rapidly pick up on new
trends and test kind of ideas of what might be driving them. It also lets us look at co-detection
among people receiving treatment. And also because there's enough tests, we can look at the different
regional trends. A big lesson is different regions of the U.S. can face very different problems
because the drug markets are different. And just reiterating the cautions,
convenient sample, really people with substance use disorder, urine that are being tested,
receiving treatment from one company, it's really focused on trend changes more than
absolute values. And like any Sentinel system, we should use other data to confirm its results.
So we provide all this information on the dashboard. When you first open up the dashboard,
there's an about dashboard section we encourage you to read. And then underneath the dashboard
data, there's an important data considerations. So now I'm going to move from kind of why to actually
a quick review of the dashboard, and then we'll end with some key findings, future directions.
When you look at the dashboard, we purposely decide to focus on key drugs that I discussed
that were really involved in a lot of fatal drug overdoses. The other advantage of focusing on
these drugs is they're routinely tested by Millennium Health. So we maximize the number
of drugs. And we want to put this system out and then learn from it and expand based on that
experience. We focus on two-year time intervals because it really is meant to be an early warning
system. And the other thing we heard is some people don't have a lot of time. And when you
hit the dashboard, you might just want a quick summary of results. We provide that. You might
also want a little bit more in-depth information. So we try to provide that with tool. There's three
specific websites. The main one you should go to is the dashboard page on the left. We also have a
key findings page and an accessibility page. I will go and now describe each one of these.
So let's focus on the dashboard page. One of the unique aspects of this dashboard page is right
above the dashboard, you'll see the key findings for fentanyl. Fentanyl has been a major driver in
the drug overdose epidemic. So we wanted to provide key findings up front if you want to
look at them. Findings for other drugs, however, have to be seen on the key findings page. And you
can access key findings either through that key findings section or at the bottom of the page
through related resources. When you go to the key findings page and open it up, you'll see a blue box
that provides the general result and then a very short sentence about maybe some of its implications.
So let's move to the actual dashboard where we turn to the key findings a little bit later and
more in depth. So the dashboard has three key parts shown on this slide. The top part lets you
filter by geographic region, drug, and timeframe. The tables underneath, one provides you kind of
general trends in how often that drug's detected in this population. And then the third part in
second table, it lets you focus in on drugs detected within fentanyl, within cocaine,
within methamphetamine. So let's take a look at the filters. So first you can select a region
and we give you the option of overall any specimen collected in the United States,
and then the four census regions. And firstly, I don't know all the states in the census regions
offhand. So when you select a census region, we provide a map so you can quickly see what states
are included in that region. Moving to the selected drug, just four drugs, fentanyl, heroin,
cocaine, and meth. As I said earlier, fentanyl, cocaine, and methamphetamine are major drivers
and heroin historically was a driver. And so we keep it on the dashboard in case there's a resurgence.
And we also provide a timeframe. Most people I think will be interested in the quarterly time
frame. And we did, we avoided monthly to kind of smooth the trends, but six months, you might want
to see how long a trend's going, get a little bit more power in the analysis. And for the Northeast
specifically, we don't have as many specimens. So you might want to use the six months when
looking at that data. So I've already described right under the filters, there's a little summary,
so you know what you filter to. The geographic, we also in the second box show, okay, how many
specimens are in there, and then the timeframe. And finally, because this is a convenient sample,
we want to just remind people all the time of some of its limitations. And so the third box
really focuses on that and provides advice. So if we move to that first table,
we made the table, you know, three major sections, the title, which provides key information,
and you can double check to make sure you have the filter you want. Some ability to select lines
to clean it up or focus on results of interest. And then we can turn labels on and off depending
on detail. And also we provide automatic calculations of some changes. Let's take a
look. Starting with the title, every title will list the drug, the region, and the date. And that's
so you can always confirm that filter is getting what you want. Then say, moving down to selecting
a line, say we were interested in phenyl without cocaine or methamphetamine, that bottom blue line,
it's kind of hard to see on this scale. So, and we might want to have the labels pop up so we can
better see if this looks really close or not. So we just click that line. And what's nice is the
dashboard will zoom in so you get a better look at that trend. And now you can look at the label
so you can evaluate kind of the absolute magnitude of change and what's been going on over the last
two years. Finally, what's this percent change off and how do I use it? If you leave it off,
you can roll over the point and it'll just give you the percentage. And it'll also give you a 95%
confidence interval, which gives you a really nice feeling of how precise this estimate is and
whether we're seeing noise or a real trend. Then if you put change on, what it'll show you is yearly
and quarterly change. And the arrow just shows general direction. It shows the relative change
for the top line, 14.5%. But it also gives the percentages in that yearly change and in the
quarterly change. So you can see, is it a big change or a tiny one from like 1% to 1.5% versus
from 10% to 15%. And I should say, we don't show data that's not on the dashboard. So for the
earliest points, it'll just show quarterly change. So just in summary, this first chart just gives
you a general trend in positivity, which other research has shown does have, has predicted some
major changes. And as a reminder, different factors could change this trend. So it's always
important to confirm these findings with other data sources. The second chart is identically
formatted the same way as the first chart, same functions. Everything's the same except now instead
of the denominator being all specimens tested for those four drugs, it's just one drug. So how often
do you detect cocaine and methamphetamine in fentanyl positive samples? And the same caveats
for this, interpreting this figure as the previous one. So now we're going to just touch real briefly
on the accessibility page. The accessibility page is structured exactly like the dashboard.
It has less descriptive information. So for limitation strengths, go to the dashboard page.
The key difference is it provides the data in a table format, which is much more accessible
and readable to a much broader audience. So we just wanted to make sure you knew that that
function was available. So now I'm going to move to one of the things that we tried to do new,
summarize kind of the key findings for you. You might not want to use the dashboard or just get
a feeling of some key findings you might want to play around with on the dashboard. So let's look
at the key findings page. So as I mentioned earlier from the dashboard, you can get to the key findings
page by opening up that fentanyl box, going to the bottom or going to the bottom of the dashboard.
And every key finding has a blue box. And that just gives a quick headline and the data. And
then there's a short, usually one sentence, max two to describe some of the implications.
So let's actually look at some of the findings. With fentanyl, it becomes really clear the
importance of looking regionally. We saw significant declines in the Midwest and South
from October, December, 2024 to October, December, 2025. And that is consistent with the argument
that there may be less people using and being exposed to fentanyl, which is good. But in the
West, we see a slight increase in a much more flatter trend. And this may be related to fentanyl
being introduced later in the West, differences in the type of fentanyl market. There's a lot more
pills when it was introduced, but it shows we can't just lump all the regions together.
Let's take a closer look. So if we look at the Midwest census region here,
you really see a very steady and large drop over two years, with the one exception of that January,
March, 2026. And I put arrows on the one-year change, which is very promising and consistent
again, with decreased fentanyl detection and exposure. The Southern region, very consistent,
steady drop over two years in fentanyl. But then we move to the Western region and you see
relatively flat with a slight uptick in October, 2025 to October, I mean 2025 to 2026, October,
December. And I should say we focus on the year to take out seasonal effects, but we also provide
that quarterly data. And the advantage of that is with the Western region, while we see this
yearly increase, it does appear to be flattening out and we might even see decreases. We anxiously
await to post the 2026 data. So moving now from fentanyl to cocaine and methamphetamine. I'm not
going to highlight all the findings. I just wanted to highlight that in the Southern region, we saw
a pattern that we wanted to highlight, and that is both cocaine and methamphetamine
without fentanyl and heroin were increasing from October, December 2024 to October, December 2025.
And if we take a look, let's look at methamphetamines in the Southern region. You can see that increase
from 8.1% to 10.0%. And it's on the second line because that's methamphetamine without
heroin and fentanyl. And it seems the primary driver of a general increase. You do see that
this increase started around in that April, June quarter of 2025. When we shift now over to cocaine,
we see a pretty steady increase over the last two years in that cocaine without heroin and fentanyl
group. And it started all the way back in April to June 2024. So I hope this has been helpful both
to kind of highlight how this can be used to look at trends by different regions. We encourage you
to actually review the findings and dashboard to get the full group. And it's a reminder that it
really is focusing on people with substance use disorder in treatment. And we are trying to get
available faster to give us an early picture of what might be coming. And then the changes we saw
is we saw drops in fentanyl detection in the Midwest and South over the last year and really
two years. But we did see the West increasing slightly, but hopefully stabilizing. And also
we saw kind of this pattern in the South maybe of people using illegal stimulants without
fentanyl. And there's a concern that as fentanyl use drops, as if treatments with medication for
opiate use disorder aren't available for stimulants, you don't have medications, but you
do have contingency management and other behavioral therapies that are effective.
We're worried there might be a shift in that group or people aren't as successful. So I'll end with
possible future directions, and then I'll open up for your questions to see how we can help.
So when considering future directions, we really want to see what problems and factors are we
worried about or that might drive or change drug overdose trends. So first is the box of just
overall exposure to kind of any of these four drugs. The second would be rapid changes in
exposure due to how people are using drugs, if they're changing methods or how often they use it,
or changes in drug purity. And a lot of concern over the last year or two are kind of drugs that
are mixed in with fentanyl specifically and a lot of the adverse health effects they can cause.
Because if we can't treat withdrawal or wounds of a patient, it might be much more difficult to
maintain their recovery and get them into treatment. And finally, we're always looking for
or concerned about opiates that are even more potent than fentanyl. As it's tended to happen when
we see a shift in the supply to more potent opiates, we also unfortunately see increases
in fatal overdose deaths. Then there might be functions that are helpful, looking at the
patient, maybe a Friday of the year aggregation to get a larger picture. So just a reminder,
the chat is disabled, but you can put questions in the Q&A box if you have any questions for me.
And I know I sometimes I don't have questions. They don't come to me 30, 40 minutes or there's
a problem with the dashboard. You can always send those inquiries to this email at DOPInquiries
at CDC.gov. So again, I really appreciate your time and I hope you find the dashboard useful.
And we will be working to get new data up on 2026 and by the fall. Thank you. I'm going to
read the question and address them. I'm just now opening that up. Thank you for doing that.
Will additional private labs be added to increase national coverage? In the future,
could the data be filtered by state and even county level? Those two questions are related
and good questions. One of the reasons we haven't done state and county filters right now
is because it's a convenient sample. There's really good coverage in some states,
not good coverage in other states. So we were struggling with how to show that data. So that's
something we'll discuss, which is linked to the second point. If we could stack data from other
private labs and combine them all together, that would increase the system. It's something we're
exploring, but require more resources. So I can't make any promises, but it's a great point and
would make the system a lot stronger. Next question. Are there any efforts to build
partnerships beyond to create yet? Right now, there is not. We wanted to release what we had.
And also there's complexities. The private companies have to agree with public data sharing
and providing their information on a timely basis. But I appreciate those comments.
Why are there so few samples in the Northeast Census region? It's just the market that
Millennium Health serves. They have different clients. And again, that's where other complementary
sources would go on. But it's also why we thought it was really important to provide a six-month
view and to provide those 95% confidence intervals so you can know what you're dealing with. Because
the Northwest can look bouncy due to that reason. Is there a place to understand how many tests were
assessed per census region? Right under the filter, we included all tests that had fentanyl,
heroin, methamphetamines, cocaine. And for that, so if you pick fentanyl positive in the South,
it'll show you what the denominator is in that second box for that timeframe. We don't break
it down by might. This data is all, oops, sorry. The HIPAA, this data is all de-identified.
So this is all the tests. We provide a link to the dashboard at the bottom. And from that link,
you can find the accessibility page. You can also find the key findings page. And I really
appreciate these questions. You're kidding. Since there is an FDA-approved medication for
stimulant use disorder, is treatment defined, for example, as those receiving medication for
opiate use? Does the disorder happen to use stimulants or does it include other health services?
This dashboard, we really only focus on the urine test results. We don't have
patient-level information on the treatment they're receiving. And I think the CDC website
can answer some of the questions on definitions and available treatments for both stimulant use
disorder that are evidence-based and opiate use disorder. I'm trying to see if there's
any other… With some states seeing an increase in orphans, do you all plan to expand testing to
new drugs and populate in drug seizure data or state-level toxicology data? Great point. I think
there's two points here. One is, as we said, in future directions, we are looking at carfentanil
and reporting it, but we have to report it differently because it's almost like number
of samples is the thing. And I think with orphans, it's so early. We need to see whether
it makes it into commercial testing, but the lab does do comprehensive tests, and it's a good idea.
And I also think that your point on data integration is really well taken, like trying
to… It's hard to go to all the data dashboards. How can CDC pull this information together in a
nugget across so you can do this comparison easily instead of having to visit each dashboard?
Oh, this data… I may have missed it, this part, but they're just asking where the data
comes from. This data comes from Millennium Health, which is an accredited laboratory that
serves people with substance use disorder, and so it focuses on patients in treatment,
and the urine tests are part of the care that they're receiving. So, it's from one company
that conducts these, but they conduct a lot of tests, so that's helpful. So, it's not related
to EMS and Biospatial, but CDC does do projects and does collaborate with Biospatial. So, that was…
It's a really good resource to detect trends, outbreaks, et cetera. We added the web address
at the bottom. I'm trying to see if there's any additional questions,
and again, if you have questions that come up later or as you're using it,
please send that to dopinquiriesatcdc.gov. I'll give everybody a minute more. I really
appreciate your time. I know everyone's busy. If I didn't answer your question,
please feel free to put in a follow-up in the Q&A. Daniel, I just wanted to check,
did I miss anything? There was one comment asking if there's any way to access state data.
Yeah. Unfortunately, now we haven't displayed state data, and that's because it's a convenient
sample, and so we have… Some states, the estimates are very precise. Other states,
not. So, on this first launch, we wanted to do region so people can see what's going around,
but it's an excellent question, and it's something we need to grapple with. Okay. I'll
stay on for another minute or two if any other… if any more questions. I think the first phase
is to really focus on these quarterly trends. We will expand, but I should say CDC has other
systems that, such as we're doing biosurveillance, testing of non-fatal samples in hospitals,
and we're also doing really comprehensive testing of drug products and paraphernalia
that we hope to get some of these emerging samples or tests early, and that information
we hope to release next year, and we'll keep you in the loop. Thank you.
What software is used to create the visualizations? Daniel, do you want to talk
about the software used to create the visualizations? I don't know that I can really
speak to that other than this is a built-in CDC template. They custom made a lot of what we see
on the dashboard. Okay. Yeah, and that's when you can send the inquiry. The landscape changes
so quickly. I hope to see quicker data from CDC in the future. I appreciate that comment,
and we're really hoping to get the updates faster. This is a question about cycloraphine,
and please send that one to the DOP inquiry. We can direct you to drug product and paraphernalia
testing that's been detecting it, and I think with cycloraphine, like with nidazines and with
parphenol, it's really important for the testing to look at the testing in your region. Some places
it's not there. Some places it's very rare, and then there's a few places that it has been
detected, so I think it's a real value for more rapid surveillance. This isn't going to serve that
purpose because commercial urine drug only starts testing when it becomes a kind of major factor,
and it gets ordered by clinicians. But yeah, definitely send that to the DOP inquiries.
We can send you resources and more information. Corey Hilton, thanks. Other questions?
Metatomidine, that's great, and I would put xylosine there too, and that is one of the future
directions that I think urine drug testing can help with. It's nested kind of within fentanyl,
and really looking at drugs co-detected with fentanyl and from other sources like drug product
testing, we know it's mixed in together. Metatomidine, Millennium Health has started
testing for, and we are evaluating providing that data more real-time. And I think urine drug test,
well, you know, drug product testing, really comprehensive testing in the EDs might be good
for the rare and emerging substance. I think urine drug testing really helps as a substance
begins to be, a lot of people begin to be exposed to it. So we should get on metatomidine. I really
appreciate that, Tom. And I also, CDC just released a health alert advisory on metatomidine not long
ago, but integrating into our surveillance systems is critical, and that is something
hopefully we'll be able to speak to more in the near future. I'll give everybody just a minute
or two more to ask questions. Well, this, the last thing I just want to say, the webinar will
be, is recorded, will be posted, and please feel free to reach out to the DA, PA inquiries with
any future questions, and the recording will be on the dashboard website page. Thank you for your
time today.