Subscribe
Share
Share
Embed
Communicable takes on hot topics in infectious diseases and clinical microbiology. Hosted by the editors of CMI Communications, the open-access journal of ESCMID, the European Society of Clinical Microbiology & Infectious Diseases.
[00:00:00]
Angela: Hello, and welcome back to Communicable, the podcast brought to you by CMI Communications ESCIMD's, open Access Journal, covering infectious diseases and clinical microbiology. My name is Angela Huttner. I'm an infectious disease doctor at the Geneva University Hospital in Switzerland, and editor in chief of CMI Comms.
I'm joined by my co-host, Annie Joseph, clinical microbiologist at Nottingham University Hospital's, NHS Trust in the uk, and associate editor at CMI
Annie: Hi, Angela. Hi everyone. Great to be back. So today's topic is Diagnostic Stewardship in Infectious Diseases and Clinical Microbiology, and we are pleased to have two great experts on this with us today.
So firstly, I'm delighted to welcome Daniel Morgan to Communicable. Dan is a professor of epidemiology and medicine at the University of Mary Land in Baltimore, where he directs the Center for Innovation and Diagnosis. At the Veterans Affairs Hospital in Maryland [00:01:00] Healthcare System. He directs epidemiology.
Annie: Dan's research explores probability in medicine, medical overuse, diagnostic stewardship, infection prevention, and ai. Welcome to Communicable Dan.
Dan: great to be here. thanks for the invitation
and I'm really pleased to welcome our second guest, Valerie Vaughn. Valerie is Associate Professor of Internal Medicine at the University of Utah in Salt Lake City, where she's also director of Clinical Research for the Department of Internal Medicine and, Associate Director for Impact, a health services research Center, , a leader in hospital medicine, Valerie's research focuses on reducing diagnostic error and curbing antibiotic overuse, especially during transitions of care.
Her work spans key areas such as physician decision making, cognitive error, organizational culture, and the development of quality measures to drive better patient outcomes. Valerie, welcome to Communicable.
Valerie: Thank you. I'm so excited to be here.
So our listeners know, we usually start these episodes with a get to know you question for guests and also our hosts.
Annie: this time the [00:02:00] question is, if you could only listen to one album for the rest of your life, what would it be? my album I'd listen to for the rest of my life would be Jagged Little Pill by Alanis Morisset. it was the first album I listened to over and over again as a teenager in the teenage angst phase. And I just felt like no one else had songs like her.
No one else sounded like her at that time. Very cathartic to sing along too Good for karaoke as well. So lots of pluses to Alanis Morisette
Angela: yeah. She's a contemporary, I would say. for my answer, . I'm gonna choose classical music because it is the only way to not be uncool.
And I actually think that if I had to live a life and only listen to one album, it would have to be something that's obviously very, multifaceted and,, wouldn't drive me crazy, having to listen to it over and over again.
And I thought about, Beethoven symphonies . it made me go back to when I was a medical resident in Manhattan. I had a. Medical student whose husband worked at Carnegie Hall, and sometimes Carnegie Hall would not sell out, believe it [00:03:00] or not. So at the last minute she would say like, come on, let's go to Carnegie Hall together. So I got to go one night and to hear for the first time in my life ever, the seventh symphony by Beethoven seventh, not the ninth, And the second movement. And , it was amazing. And I recommend everyone listen to it because there's so much to it. You could listen to it so many times and still get something new out of it, I think every time. dan you're up. ,?
I'm embarrassed to say mine still, it's closer probably to Alanis Morissette. you know, probably the combination of like calm, kind of emotionally resonant and not annoying for me would be Sufjan Stevens, like this Michigan album, he's borderline Christian rock.
Dan: which I'm not but for whatever reason, like he's always worked for me he's been the background music for a lot of the work that I've done. and I just listened to it, driving in to say like, really? Is that it? yeah, it still works. and I wouldn't choose to have one album for the rest of my life, but that would be a reasonable one.
Angela: You could live with it.
Dan: Yeah.
Angela: Valerie, you're up next.
Valerie: Yeah. So I had to make [00:04:00] the decision of whether to reveal my deep, dark secret to answer this question.
and I decided I would because, I love you guys, but I am an OG nineties country fan, born and raised in Texas, which I don't know if you guys all knew, but it's like deep down, that was what shaped my teenage years. So I would have to say, it was hard to choose one album, but I'll go with In Pieces by Garth Brooks.
there's like literally not a single song on that album. I don't know the words to, there's super motivational songs on there, like standing Outside the Fire, which is all about taking risks and, knowing that you can't really live unless you, risk, getting hurt. And then, one of the coolest experiences of my life was actually to the last song on that album called Ain't Going Down Till the Sun Comes Up, which is about exactly what it sounds.
It's about. Partying all night. But basically I watched this live at the Houston Astrodome, and during the last chorus, right as he says, ain't going [00:05:00] down till the sun comes up, he rises off the stage attached to a helicopter and flies out of the astro durum like into like wherever.
In addition to being one of the most prolific country music writers, he was such a showman so. Now you guys want my deep park secret?
Speaker 5: and so does, global listenership. Yes. Just so you know. Yes. Everybody now knows.
Dan: But I know Angela, you have some of the redneck American in you, so you can understand that.
But Annie, your mind must just be blown by that. I saw your face, your reacting.
Annie: Yes. Annie, tell us what you think. Well, I, no, I just think that Brits are, we are just too cynical to enjoy things like that. You know, like really wholesome things. We want our music be a bit more.
Whi
Speaker 6: away nineties. Country music is
Valerie: super wholesome. Like, if you wanna just think that the world is a good place to be, just listen to some nineties country music. , If you go too early, everyone's like dying and everything's bad. And if you go too late, it's a little hyper patriotic.
But nineties is the perfect time [00:06:00] where everyone is good, love is good, life is good, and we're gonna be good people together. So it was , very wholesome.
Angela: Yeah.
Annie: So now we know a little bit more about our guests. It's time to dive into the topic of the day, which is diagnostic stewardship. So Dan, we're gonna kick off with you. When did diagnostic stewardship begin as a concept, and how would you define it today?
Dan: the ideas of diagnostic stewardship have been part of clinical microbiology for a long time, you know, 50 years or longer.
but I think that the idea of really translating them into the, clinical area more and having clinicians be more involved with them and maybe leading some of the efforts around them really started probably around 2015 to 2017. There were some different names for it, people called it things like the culture of culturing or testing stewardship, and there's been laboratory stewardship for a while.
I mean, I feel like, at least my kind of intense involvement around it and where the term itself diagnostic stewardship started being used was around 2017. there was an article, about the use of it In molecular tests, around that time as well [00:07:00] as, one from the Netherlands, I think that I didn't see until later.
And then actually, when I was at, the only ECCMID I've been to or was ECCMID then was in Vienna. And, I was there with Preeti Malani and Dan Diekema, and, we were talking about it, thought it would be great. And we'd been walking around Vienna 'cause there weren't that many Americans at, ECCMID then.
and Preti said, why don't you write a viewpoint on it? And so I wrote the viewpoint on the airplane. I sent it to them when I got back and it was accepted within a week. And, at least for me, that was one of the, early pieces that came out that really sort of defined what it meant.
And in that piece, we, defined it as modifying the process of ordering, performing and reporting diagnostic tests to improve the treatment of, infections and other conditions. I say over time, the, definition really hasn't changed that much. I mean, it's a pretty broad generic definition. but there's been a lot more interest and I think more specific details around how to do it.
Really culminating, I would say last year with the CDC, putting out a core elements guideline, here in the US about how to develop programs for [00:08:00] diagnostic excellence.
Angela: I do feel like that paper gave it a name. , At least over here, we didn't know necessarily what to call it.
We all had these instincts for diagnostic stewardship, but we didn't have a nice name. And I feel like that was the paper that gave it a really nice name. And I hope I'm not disrespecting other people. but I do think that was a really seminal paper as I remember.
Dan: I'd be flattered to hear that, but it definitely was like a few different people and a number of people working in this area.
But I think just trying to figure out what to call it, and really, you know, Preeti, Malani and Dan Diekema were very helpful for saying like, Hey, these are some good terms. Let's put this out there and get it out before other people come up with a different name that's not as good.
Valerie: Yeah, and I think this is like right during the time when there was this huge movement toward thinking about overuse of things and choosing wisely appropriate use.
Things we do for no reason, like all kinds of different concepts. And the brilliance of the article and writing down diagnostic stewardship was just defining it and really having a model for people to think about these kind of [00:09:00] complex things that we were all trying to conceptualize,
The term diagnostic stewardship. I love it. But some people feel really like strongly about it. I think because it's kind of got that stewardship element in it, in infectious diseases and microbiology, we're kind of so comfortable.
Annie: We're talking about antimicrobial stewardship, antibiotic stewardship. We all have that sense of doing less. It can be better, . It's already tied in. So, although perhaps the term's only been around for 10 years or so, the kind of bedrock of how are we trying to steward something we need to take care of is kind of second nature, in Id world, I suppose already.
Dan: think Id is remarkable for being one of the few fields that really is focused on, sometimes doing less. I don't think there's that many, medical professions that do that. and I do think the term is great. feel like it explains what it is that I think the only negative is that similar to evidence-based medicine or something else that's been kind of popular.
it's overused in a big way and it's used for all different types of things. and I know the lab people are probably the ones you're imagining who don't love the term, cause they've been doing it [00:10:00] already, they're like, that's clin micro, it's not, not this new thing. but I, do think like actually getting it to the clinical level and having clinicians, do more of the efforts around it is the new part using the ideas that laboratory experts have been telling us for decades.
Angela: I were to try to answer your comment, Dan, about why Id people. are championing diagnostic stewardship. I think we're like the only field that has to deal chronically with shortages with non-renewable resources, let's say, I can't think of another field in medicine where people have to do that routinely.
Valerie: Okay. So, so I'll come in as a non-infectious disease doctor here and defend my specialty, Hospital Medicine.
Angela: sorry, Valerie
Valerie: So, this is something Dan and I have had infinite conversations about, so there's no animosity here, but it's just, I think in the United States for many reasons, hospitalists have been put at the forefront of trying to do things to. Restrict overuse of [00:11:00] care. And this is in particular, I don't think these pressures are the same in Europe, but length of stay reduction in the United States is like the number one quality metric that we actually get like incentivized on like bonus structures.
It's the most common, if you survey hospitalists, like your length of stay is what drives your compensation bonus, right? I disagree actually with that as a practice, but I think that idea of, length of stay has moved toward many things in our field being like, well, do we really need to do this?
I would say that while focusing on, diagnostic infectious disease tests has not been what my field has normally done, I'm telling you back in 2017, when Dan was writing this paper, people were doing telemetry projects to try to use telemetry less. They were trying to do CT projects to use chest cts less.
They were trying not to do daily blood draws because people were getting poked every morning, even though their potassium had been the same for the last seven days. So I think some of these concepts really can extend [00:12:00] to, hospital medicine. that's at least what I've been preaching with Dan for the last decade.
Dan: I think Valerie makes good points and I do think that, if you want to actually change the way that care is provided in hospitals too, like hospitalists have to do it, really.
because they're the people providing the majority of care and really organizing a lot of how hospitals work. And I do think a lot of the tests that may be more important than some of the micro ones are general medicine tests. the radiology, the cardiac tests or whatever.
There's certain , projects that have been the focus of hospital medicine, but I, do think, that's where there may be almost, more benefits for patients just 'cause it's a more global, universal type thing.
Valerie: And I think one of the things that we can learn is that we haven't as a specialty conceptualized it and modeled it and thought about it the same way I think infectious disease does. That's part of the gap that I'm trying to bridge is like, can we bring some of the stuff that we've learned about doing this for infectious disease diagnostics and do them for, daily blood draws or other diagnostic tests that we commonly [00:13:00] overuse in medicine.
Valerie, in your opinion, and according to any evidence you might wanna share with us, what are the most effective levers for changing test ordering behavior among clinicians? Is it education? Is it nudges? Is it, electronic health record restrictions?
.
Valerie: The answer is never education.
no, , I have, one simple principle when it comes to any diagnostic stewardship intervention, and that's make my life easier. and I say my speaking again as a non-infectious disease clinician, and if you think about the number of diagnostic tests that I have to understand and know, it's a lot.
So I have to know like a ton about a ton. And the answer at the end of the day is just, I'm not gonna know that much about whatever specific test that the laboratory or micro is gonna be an expert in. So whatever you can do to make my life easier, please do that. And I think this is one of the opportunities that diagnostic stewardship has and can capitalize on and has been capitalizing on that's a little bit [00:14:00] harder than antibiotic stewardship, where antibiotic stewardship maybe feels like you're restricting people.
I feel like you can conceptualize diagnostic stewardship as helping people make the right decision. And so usually the best way to do that is through nudges, right? Where you make doing the wrong thing harder and the right thing easier. so examples of that are prioritizing in a list of tests, which tests you want people to use should be at the top right, or in order sets.
You pre-check the things that you want people to order and you hide the things that you don't want people to order. But I think other good examples are what I call point of care education, which many people do already, and don't call it that, but it's like giving us context for results or for the test themselves as we're ordering them or as we're interpreting them.
so one of the best examples of this is, , we used to always report, sputum culture results and say, GPCs, gprs and whatever is in it. And then finally they changed the interpretation and they said, no MRSA no pseudomonas. [00:15:00] And any microbiologist probably worth their street cred would know that those say the same thing, but , the person interpreting it doesn't necessarily know that.
So it was translating the information from the complex into what people really need at that point of care. And suddenly people deescalated the bank and the Zosyn. so, it's providing information as you need it already. Do this with cascade susceptibilities, that's a big one.
That's really helpful. Providing comments , this organism is universally sensitive to penicillins. This is usually a contaminant. Especially I feel like they change names of the bugs all the time, and your non ID people are not gonna remember any of that. I feel like the ID people barely remember that.
So we have no chance of remembering those things. So, Instead of educating me during a conference where I'm gonna immediately forget it. The second you've said it, like the point of care education where you remind me right as I'm interpreting the information is super helpful.
And then I'll give one more example, which is the only time in my life I will say that a BPA or a best [00:16:00] practice advisory is a good thing because usually me and my colleagues hate those, right? It interrupts our workflow. But if you can provide information in that BPA, that's hard for us to get. Otherwise, it's super helpful.
So an example of this is like you go to order a c diff test and it says, Hey, did you know this patient literally just took some MiraLax? Are you sure you wanna order that c diff test? Super helpful, because often the reason that I'm ordering it is because I just left a patient's room and the nurse is like, Hey, do you know your patient's having diarrhea?
And I'm like, oh, okay. I should, order a c diff test. And I didn't go and look in the medical record because that takes like four or five clicks minimum to see that they've gotten, a laxative. So providing that information immediately is actually super helpful. And I'll just cancel that test.
So there are other things like that, do you know this patient just got a MRSA swab? Are you sure you wanna repeat it? , Those kind of things are really helpful to the clinician, but also help, with diagnostic stewardship.
I can't help but notice [00:17:00] that all of these interventions, I think all of them that you named, they actually go above the head of the, treating physician. Right? The lab decides how they're gonna present the results of the. Culture or whatever. I think that's okay with people.
Valerie: Yeah. I mean, I don't think you need to be like, Hey, we're doing diagnostic stewardship on you, right? Like, this is just part of that automatic thing, which is we're doing this to try to make your life easier. And I think the time when it's nice to actually, I think, have frontline clinicians at the table when you're making these decisions because otherwise it's sometimes hard to understand the workflow or why people might overuse.
So for example, with the c diff test, I'm sure people are like, why is someone so dumb to order a c diff test when someone literally just got MiraLax? But if you talk to someone, it's like, well, I didn't know they got MiraLax. 'cause it's part of the standing orders. And the nurses never tell me that they give it.
And this is usually being triggered because the nurse comes to me and says, Hey, this patient is having diarrhea. And I didn't even go look to see if they were on, laxatives. So if you have that conversation and [00:18:00] you realize that It's not people choosing to order c diff even though they're on laxatives, it's choosing to order c diff because they didn't know they were on laxatives.
And that problem you can actually fix and help. So I think that's why it's helpful sometimes, like figure out who are the people overusing the test or misusing the test, and why Is it a knowledge problem? Is it a process problem? And, the solution to answer that is different depending on kind of what the source of the problem is.
And I'd just like to highlight, I, this is part of the reason I love, hearing Valerie talk too, that although she's a great researcher, she always talks like a clinician, with her examples in the way that you wanna go about doing these things. that resonates a lot more with people and is, you know, concrete and clear.
Dan: I, couldn't agree more. I mean, I think, making it simple, making it not annoying another example of how. Sometimes this is what clinicians want, but it's not what they can get easily is, there were people, in a pediatric hospital in Colorado, Kevin and others who were looking at a meningitis encephalitis panel, and they noticed that people were ordering this when people had no white blood cells and their [00:19:00] CSF they were looking for meningitis.
The patient didn't have meningitis, but they were still ordering the, multiplex PCR and they were getting these false positives sometimes with, HH six or other viruses. And they went to ask the emergency room doctors, why are you doing this? Why are you ordering it inappropriately? And they said, we order it all at once.
We poke somebody and we get the specimen and we send it and we can't go back you know, check the result and then order another test. And he said, well how about if we like reflex it, it only will go to PCR if you have white blood cells in the CSF. And doctors were happy with that. It made their life easier to order the test the right way, without any more effort.
So I think that is a key part of like talking to the right people and making certain you implement it that way.
the, HHV six on the C-S-F-P-C-R plagues us all. When I get phoned by the lab with the HHV six at three in the morning, I'm like, no.
Annie: Why have we tested it? But we do have an algorithm in place, but sometimes they slip through. I really like Valerie, your first answer is It's not [00:20:00] education because I think, I've probably lost track of the amount of, conference posters and, things you see where it's like, oh, we need to educate people not to request this.
We need to educate people that, you know, it's a poor pre-test probability of this and no education isn't the answer. It's got to be easy to do the right thing and hard to do the wrong thing, because you just can't hold all that information in the brain. You know, if someone puts you on the spot and asks me, now what's the sensitivity of our h pylori assay, I would have no idea off the top of my head.
The main thing is that you don't order it twice on somebody. that's the main thing that you need to know, isn't it?
Dan: I mean, I think that's the problem also. it's just a modern problem of the explosion of information and knowledge and still this kind of old fashioned belief that doctors or , any type of human can actually keep all of that in mind and make decisions appropriately.
And that kind of cognitive load is, I think just too much to expect. the goal of diagnostic stewardship or what makes it appealing to me is if you can simplify things and make it easier to focus on the big picture and be realistic [00:21:00] about how people actually do order tests, that they don't really say like, let me think carefully about this test and the pretest probability.
And the test performance characteristics, they usually say, I'm thinking about a problem with the lungs, let me order the breathing test. You know, and they're not really thinking that carefully. So if we can make it, easier to do the right thing and harder to do the wrong thing, we can have benefits.
the other, really common diagnostic stewardship intervention that I love is actually providing the price of a test. So the number of times when I'm on rounds and my med student is like, Hey, why don't we order test X and I don't really wanna order test X, but , we're teaching institution and I want them to know that this patient doesn't have whatever rare disease they're testing for, but then you like, go to order it and it's a thousand dollars test.
Valerie: Well, a thousand dollars isn't worth satisfying their curiosity, , so that kind of little piece of information I think can be really helpful. And especially nowadays with these new genomic panels and stuff that are coming out and you just hear people talking about it [00:22:00] bankrupting the lab budget.
But I don't think that frontline clinicians always know how expensive those tests are or that they're not gonna be covered for some of their patients. And , my, Husband actually went through this with a genetic test for a biopsy that he got and the results ended up not being covered by insurance and it was like $5,000.
And so would he have said yes to it? Would the doctor have said yes to it? Had they known like that would be the price tag? Probably not. and it's really hard to get price information, and this may be a totally US phenomenon , but I think a little bit more of the cost information would be helpful too.
And I've seen some successful examples of that, but I think we could do way more.
That's really interesting. You know, there is part of me that was rebelling against this idea of like, okay, diagnostic stewardship goes above, the individual clinician, you know, these decisions are made and that's the way it's gonna be.
Angela: you could argue it's a little bit paternalistic, right? Like, now, we're supposed to have, autonomy as physician prescribers. but when you say, well, actually it turns [00:23:00] out you are prescribing something that actually could just fall on the back of the patient, and that patient then has to pick up the cost, maybe a little bit of paternalism is in order, honestly, like when you have a system where someone's gotta pay for your decisions that you are making as an individual who is not entirely informed on all aspects.
Dan: I mean, one of the, the phrases I like best that talks about nudges and framing is, from one of the founders of behavioral, economics, and I think he called it paternalistic libertarianism. But, the idea that it is kind of paternalistic, it is trying to shape people and push them in a certain direction, but it still allows them to do what they want.
it's gonna shape what the lazy person does. It's gonna shape what the person who's too busy does. But it's not making it impossible to have autonomy and make a decision still, which I feel like that's how you get a lot of reaction from people if you make something impossible to receive.
Dan: But if you just make it a little bit harder, it's more of an annoyance that, they won't do unless they feel [00:24:00] strongly about.
Yeah. And people with strong opinions are always gonna find a workaround, but with most, nudges, it should be relatively simple most people aren't gonna have strong opinions about it.
Annie: I think Valerie, what you're saying there about knowing the cost of the test is really interesting. I mean, in the National Health Service in the uk, because it's free at the point of access and you know it's all taxpayer funded and none of that cost is ever gonna go to the patient individually to pick up either through insurance or out of their own pocket.
I dunno whether we've lost that incentive a little bit. It's almost like well go mad, you know, do the daily CRP because it's almost like that money, it's just not tangible. for the clinician who's ordering or for the patient, the patient certainly would never think about it.
But for the clinician who's ordering, there's no, sort of financial incentive necessarily.
Valerie: Yeah. I don't necessarily think, I'm not advocating that cost should be like our number one determinant of anything that we do.
I just think that, when you meaning the, person ordering the test doesn't know, Hey, this is actually a really expensive test. I think [00:25:00] it. Means that we're ordering it without like full knowledge of things. but there are other outcomes that I think are really important too.
I know eventually we're gonna get to pan culturing and stuff, but describing and showing a clinician what it looks like to get two sets of blood cultures at three o'clock in the morning from a sleeping patient. nobody actually thinks about the patient harm in that you've woken them up, you're poking them.
It's very uncomfortable. these are the things that actually patients hate the most about being in the hospital. And if you do it when it's totally, a low, likelihood situation that their bacteremic that is direct harm to the patient. depending on what you're talking about, I think there are different kind of outcomes to think about.
so cost isn't always the best one, but sometimes, it can be useful.
Annie: So we've touched on the use of order sets and many hospitals use diagnostic panels or protocolized testing like pan cultures. What do you two think about sort of panel-based testing and protocolized testing? Is it a good thing, bad thing
I.
Dan: I guess the first part of the answer should, don't pan culture. I, think that's a pretty clear one. even though I, that's what [00:26:00] grew up as a doctor doing, you get called as an intern and you're like, all right, do the, typical cultures and, , call me back in an hour or something.
there shouldn't be pan culturing or fever bundles as some places call it. There should be a little bit more thinking, order for individual reasons. I ideally have a sense of what you're gonna do with the test result. takes a little bit more time. we were just talking about the difference between the UK and the us but, I was talking to physicians from Japan who I think Japan uses like twice as much radiology as the United States.
So they certainly aren't afraid of technology and using test, but they said they hardly ever use blood cultures and that's 'cause physicians draw their own blood cultures. So, you know, there are different, features that will create a barrier that, make you decide, you don't care that much about that test.
You know, if you had to go, draw it on a patient in the middle of the night as Valerie is saying, like, that'd probably make you very aware that it's not that important and you'll do it if, you think it's really critical, but you know, not, so often. it's funny 'cause we're arguing for protocolized certain tests, but I think we should do it in a thoughtful process. Like what is the output? So, the idea [00:27:00] that tests should maybe have some laboratory reflex testing so that you don't have the test done if it's inappropriate is great.
But I think that some of the other protocols where it's like part of an order set or it's a rote response to a, nurse call. I think those are places where just like stopping those can improve practice a lot.
A hundred percent we should not be pan culturing. That kind of just goes against who we should be as physicians, which is people that are thoughtful and. Think through okay, this patient has a fever, this patient has altered mental status.
Valerie: What's our differential diagnosis? How do we work that up? What's really hard is that pan culturing is often a knee jerk reflex to be like, okay, I don't wanna have to go through that process because I'm busy, so let's just order these, like four or five tests. I really try to talk to my learners about, why are you ordering that?
does the patient actually meet criteria for treatment if that comes back as positive? but pan cultures I think are one of the hardest things to beat in the hospital 'cause it's so easy to [00:28:00] do. and hard to see the consequences, but we'll always remember that one patient who we decided not to get a test in and they ended up having it.
Valerie: Right. So the hard thing about human psychology is we always remember errors of omission more than commission. Meaning if we didn't do something and there was a bad outcome, we're gonna remember that for the rest of our lives. Whereas we forget all of the bad outcomes that patients had because of things that we did do.
and that's human psychology, which means it's clinician psychology too. So I think pan culturing is one of the hardest things to beat. I will say in a positive way, I think I've seen changes over time in this. I think people are doing pan cultures less now than they used to.
At least in medicine sometimes you still see it with some of the, not to throw shade at my surgical colleagues, but some of my surgical colleagues. so I, do think things are changing, but this is gonna be one that you have to change the underlying culture of culturing, and I think that's a hard thing to do.
Dan: I think this gets at problems of overuse too, is that it has actually a lot easier just to order those tests. if you're called, if you're cross [00:29:00] covering, it's just like when you have a patient in your office say, it takes time to talk to them and explain to them, or, you know, assess them and figure out what you think is happening versus it's really fast to order a test.
And maybe that's part of the problem with our system is that we've optimized for efficiency and testing and even giving, drugs or something. whereas what probably is the best value is, spending the time with a clinician who's gonna try to figure out what's going on and what you need as an individual.
Annie: Yeah, I think that's such a good point, Dan. 'cause often, if I get contacted for advice, this patient's got a new fever, what test should we send? This is their background. we're thinking about doing this. What antibiotics should we start? You know, often my answer is, okay, the patient's stable.
We don't know what's going on. why don't you watch and wait, let's monitor, let's see what happens over the next couple of days. And it's almost like that's an impossible piece of advice to follow because it feels like you're doing nothing when actually watchful waiting, is doing something and it's probably avoiding harm, but it's [00:30:00] like, oh, so you want me to do nothing then for the patient, but they've got fever, maybe they're gonna get worse. It's, almost like the temptation to do something with tests, which then ultimately often leads to antibiotics at the end of it.
the temptation's so strong versus, watchful waiting. And you're right, Dan, I think. Some of that is about how our health systems are built in that, have we really got two days to watch and wait a patient to see how they improve or get worse and then decide what to do? Or are we driven towards that?
Reduce length of stay, make a diagnosis, finish and move on rather than watch and wait.
Angela: Yeah, I would say in Switzerland we really have the luxury of waiting , you were saying Valerie, that in the US you get rewarded for short length of stay. We have some pressure here, but nothing like what you're experiencing in the us and we do get that, that luxury of just being able to have somebody be hospitalized and not be, giving them treatments.
the diagnostic maneuver is just watching them and seeing what happens actually.
Valerie: I was just gonna say to put this into context, like the average length of stay for a [00:31:00] patient with pneumonia in the United States is four days. and whenever we see trials published from Europe, it's like people are in the hospital for seven to 10, so at least twice as long as us.
if you just think about that pressure, so telling one of us to watch and wait, whew, that's gonna be hard. the other thing that I hear with, pan culturing is, well, I'm gonna order it in case I need the information, but then I can decide if something comes back, like whether or not to treat it.
And that is a hundred percent false. so we did an antibiotic stewardship study with asymptomatic bacteria where we incentivized hospitals to reduce their treatment of asymptomatic bacteria. So that's an antibiotic stewardship target, right? So reduce antibiotic use for asymptomatic bacteria.
And we found, we were able across the state of Michigan 69 hospitals to reduce antibiotic use, but a hundred percent of that reduction came from, A reduction in inappropriate urine cultures and urine testing. Like once someone ordered a ua, got [00:32:00] that positive urine culture, they were gonna treat it. And the rates of treatment did not decline over time.
They stayed between 80 and 84% over the years that we did this. But the rates of, getting urine culture has decreased over time. And if you don't get a culture, you don't treat it. And so, once you've ordered a test, if there's a false positive result, cat's outta the bag, like that's gonna end up with a diagnostic cascade where you get more testing or a treatment cascade where you do more treatment.
And it's really hard once you have a positive test to convince yourself it's not real to convince the patient it's not real. or anytime care changes hands, now you have to have, like another doctor also believe that it's not real. And that's just really hard to do.
Angela: So along those lines, Dan, I imagine some people grumble that, this diagnostic stewardship, does delay true diagnosis.
how do you reassure them that it's not harming people? That you're not withdrawing? tests that patients should have a sort of a right to [00:33:00] have.
Dan: I think this is a lot like, say sepsis, which is a, condition in the United States has, been focused on a lot for, treating because it's often severe.
but I, think that like you need to treat the patient clinically who's in front of you if they're sick, if they're unstable and they may have an infection, you should give them antibiotics, right? Like, you're not gonna be waiting on a urine culture to figure out, if it's a UTI or a pneumonia or something but if they're not sick. Those are the patients where you have this room. And that's where we see a lot of the overuse is like people who aren't very sick, they're kind of confusing, they're a little weird or odd, they're not quite right, but they're not, somebody with a low blood pressure or tachycardia or something where, you know, you're worried that they actually could die while they're in the hospital.
So I think the biggest part is treat the patient in front of you clinically. And if they are sick, often we give antibiotics, even if they may not have an infection. I think that this tends to be more of a theoretical concern. Also, when we switched to the way that we did urine culturing to only process cultures, if they had pyuria there were people who were [00:34:00] worried.
There were people who made arguments. They were often theoretical, like, you know, what about neutropenic patients? Maybe that doesn't happen, which, who knows? once it was enacted. Everyone stopped. noticing, even like we made small switches to it. And I tried to like, educate people just to get out in front of it so no one would feel like we were hiding information.
And mostly people were like, what? Huh? How does it work? And a lot of test results, people don't read that carefully. And it's a lot like, say the microbiology comment where it'll say more than three organisms isolated in this urine, it's likely contaminated.
And. people just don't think anymore. They're like, all right, it's a negative culture. versus if you , gave all the three of those organisms, people would do a lot with them. But I think that, we don't have the head space to really think carefully through all of it.
And so often people, don't see this as being a problem once it's available. I guess the other key part is, not fully restricting it. If someone wants a test, having an option that they can get it. If they want to put in the effort of calling the lab or clicking a few more, buttons [00:35:00] in the electronic health record.
Annie: I actually really like having those conversations. I mean, we have a few restricted tests in our lab where, , it has to be on consultant microbiologists say so only, and they're the best conversations. You can really get into it. Often they're really complicated patients.
You can really have that discussion about, what is the pretest probability here? What's the, differential diagnosis, what are we trying to exclude, include with this test? Maybe there's a better test, an even more expensive test probably that I might suggest instead of the, the B 2D glucan or something that they wanted, or the procalcitonin or something that I'm trying to restrict.
But they are like such high quality discussions to then have, and often, you can get that little bit of a abrasiveness, like, I wanna order this. Can you just do it? can we just make the lab do it? And then they usually come around really quickly to be like, oh, okay, well I didn't know that actually about the specificity being really poor or Whatever it is about the test. They can be really good for kind of teachable moments. I think restricting, as long as you've [00:36:00] got enough, time and I guess people to handle those queries.
Valerie: I was just gonna add to that, antibiotic stewardship ended up kind of being branded as we're trying to reduce the amount of antibiotics.
And if you hear like Arjun Srinivasan n talk about antibiotic stewardship, he says. that's not what it is. It's trying to get, the right antibiotic to the right patient, for the right indication, so it's about, improving use of antibiotics. It's not about reducing use of antibiotics. And with diagnostic testing, we have focused today on reducing use because there's so much overuse.
I think it's also, helpful to think about diagnostic stewardship as improving use of diagnostic testing, not reducing use. And there was a really good example that we had presented at our pre-course at ID week on diagnostic stewardship, which was about. Increasing HIV testing in the state of Utah, which is worse in the United States for HIV testing.
And basically what they did is if anyone came in and they got gonorrhea, chlamydia, [00:37:00] syphilis testing, for whatever reason, they automatically got tested for HIV as well, to try to increase the use, especially in our super rural communities where it might be hard to get people in to do HIV testing.
and so they had to put in a whole bunch of processes in place. 'cause the, ED and urgent care clinicians were worried about, well, who follows up a positive result. So they, put in processes in place to do that. But that was an example where you're not trying to reduce testing, you're trying to screen a vulnerable population that is getting under tested.
Angela: so Dan wrote a nice piece for CMI comms. He was one of our earliest authors, and we're very grateful and everyone should go read the paper and we'll talk about it, in a minute.
You define diagnostic stewardship, it's the right test at the right time for the right patient.
something along those lines. but Yeah.
Dan: Yeah, , that is another definition that we've used , the right test for the right patient. And then we try leading to the, best outcome, which is the hard part for all testing to know, how it's gonna impact decisions and actual patient outcomes.
Angela: so [00:38:00] Dan, speaking of, that piece that you wrote, which I loved, it's a very short piece. and it says something, I would say perhaps, maybe people might think it's self-evident, but to me it was a really good point and I hadn't really heard it, expressed so, so nicely, so concisely, can you, tell us what you were talking about in that piece and we'll put it in the show notes.
Dan: Great, thank you. And I should say it was great of you, Angela, to invite, a commentary and then you sent it to three reviewers so
Angela: Yeah. Yeah. We're a quality outfit here.
Dan: You guys are, are rigorous there. you
Angela: got a lot of reviewers. I, I did not have a hard time finding reviewers for that paper, Dan. So this stewardship is a very sexy topic, you see? Yeah. Yeah, I remember that. Yeah, it's a sign of your good work
Dan: mean, Essentially the point of, that article was, this idea that diagnostic tests are often thought of as like kind of an easy part of medicine, like a part where we don't, need very much regulation or assessing of them beyond that analytically within the lab.
They [00:39:00] measure what they say they measure. whereas as a clinician, I really wanna know like, what does it mean for caring for somebody? And you can get at that with clinical studies that aren't randomized, where you can get a sense of how good are they at diagnosing a disease versus a gold standard or something.
But really what would be ideal is randomized controlled trials, which of course this is more of a theoretical in the United States 'cause we don't do randomized controlled trials for , big interventions, much less diagnostic tests. I do think that diagnostic tests.
Especially some of the more expensive or difficult ones or ones with harms to patients to be looked at this way where you can see like what's the real impact downstream after they get the test or don't get the test. And, there is an interesting review by John Ioannidis I think he reviewed this around 2015, looking at diagnostic tests, how often are they tested in randomized trials and he found a number of examples, but they're really mostly cancer screening.
So things like mammography or PSA testing and notably in those they often found it's a very small effect. there's harms and [00:40:00] benefits to, doing tests. So, think that if, there's a common usage where you can assess it in an RCT it'd be really great to see what's the real downstream impact on patients, both, harms and benefits for some of the, tests we do.
Angela: Yeah. So the title of that piece was very simply, diagnostic tests should be Assessed for Clinical Impact. It's so obvious because it should be being done, but of course nobody's doing it the people the companies that make diagnostic tests, they're not gonna have an interest in doing that. Right. you, could draw a parallel with, pharma companies that make antibiotics, right? They're gonna do the, phase one, two, and three studies that they need to do, and then they're gonna stop there.
And then it's up to us, the regular clinicians, the patients, to do the post-market work. I unfortunately think that your dream, Dan, of having RCTs, for clinical impact of a diagnostic test that's gonna be on us, that's not gonna be, implemented in the workflow of developing a new diagnostic test.
Sadly.
Dan: [00:41:00] I mean, maybe there's people on CMI Josh Davis and others down in, Australia, where they seem to do like clinical trials for everything throw this in for, some things. But
Speaker 8: yeah,
Dan: here clinical trials are just really, difficult and kind of burdensome to do so.
although I talk about it, I'm not proposing that, that I do one notably.
Angela: Well, it's good that you put it out there. somebody else will have to take the torch and do it.
Annie: I think it's so tricky, isn't it? Because I, I completely agree with everything you argue for in that paper, Dan, how could you not, you can't not agree that this is, , clinically important to evaluate diagnostic tests in this way.
But in terms of the cost of performing those. Versus doing a simple, A versus B in the lab. Analytical comparison is astronomical. And I just wonder, without kind of a change to the regulatory framework for diagnostics, in vitro diagnostics, then I just, I can't see who would be funding those sorts of massive studies to show probably when they are done very small impacts and [00:42:00] overall length of stay, or antibiotic use.
I'm kind of thinking maybe about the adapt sepsis trial and procalcitonin reducing antibiotic use by, on average one day, was it Angela? Yeah. Less than a day. Less than a day, yeah.
Valerie: if you think about the millions we've spent on procalcitonin. You know, the cost of a clinical trial to show its value, is tiny compared to the cost that, procalcitonin has ended up costing us.
this is a place where clearly now is not the time to be saying the government needs to do more trials, but the government needs to pay for more trials. And, I think you could do some of these things relatively cheaply with like pragmatic clinical trials, EHR embedded trials where you randomize people, with a waiver of consent to get test A versus test B.
Those are things that we've started to see with the ACORN study coming out of Vanderbilt. Like I think we could start doing that with diagnostic testing too. But you have to have a lot of things in place. You have to have the informatics capabilities to do it. You have to have, an IRB that is forward thinking [00:43:00] and realizes that people are getting randomized based on what doctor they see.
So why not instead randomize them in a way where we can study it and we have to have, a health system. That is willing to do that. And then people that have the interest in, doing the research and kind of formulating the question. So I, don't think those things are fully impossible. I think there are barriers, but that would be my dream is we should be doing a lot more of these kind of, pragmatic clinical trials, for these diagnostic tests.
Dan: I think, Valerie, you described it wonderfully, like how this could be done simply and maybe not in a way that is cost prohibitive. and I, do think, America is in a time of like a kind of academic medicine carnage, where we're gonna be shrinking a lot for a few years.
But I do think how we rebuild is important and hopefully, building back , the academic medical research world to be better and I think be more realistic as far as what is the appropriate regulation to protect patients while still actually answering more important questions more easily.
Angela: fully agree, Valerie. I love the Acorn study just, [00:44:00] because it exists. , We have tried, in Switzerland, there's just no way, no way that any ethics committee would allow, any randomization to occur, without informed consent, a written informed consent, from the patient.
Even if, as you say, we're pseudo randomizing patients all day long, to diagnostic tests, to different treatments, it's happening all the time, but when it's. Just in the clinic, you know, it's not their problem. yeah.
Valerie: How many times do we do shared decision making about the blood cultures we order at three o'clock in the morning?
Right. we don't consent patients to a lot of the diagnostic tests that we order in hospitals. And so to think that you need to go above and beyond that for these pragmatic trials, I think is just so hard. you end up just providing worse care to everybody because you don't have the evidence to figure out what the best care is.
Yeah, I fully agree. And maybe Dan's right, maybe when the US picks itself up and rebuilds, maybe it could be done in a better way to imagine [00:45:00] because like you say, Valerie, those trials wouldn't be that hard to do. many hospitals now have electronic prescribing systems and data collection systems.
Angela: it wouldn't be actually that hard to implement if you do have the infrastructure. there are a few major hurdles to get over First.
Dan: Yeah. Have to change the belief structure, I think just like the hardest one.
Angela: Yeah. Yeah. So, Dan, switching gears a little bit, tell us how will AI play a role in guiding appropriate test utilization? I know you have another hat you wear and you're doing a lot of work in AI in medicine. So can you tell us what is already happening in that space?
Dan: the AI space, I'm mostly talking about generative AI as being the, different part from the, predictive AI that we've had for a long time, like the sepsis prediction rule type, examples.
I think people are really. Talking about it a lot and there's a lot of interest, but outside of you know, the AI scribes that can help people write notes faster, it's, probably not actually implemented that [00:46:00] much in hospitals right now. But I, think that the key issue will be how we implement and people aren't talking that much about it.
I mean, there's mostly talking about capacity or capability that AI can do a great job making a differential diagnosis. or, you know, it can, uh. do a great job talking with patients. But I think we haven't thought about , well, what information do we provide to clinicians from ai?
How do we implement it? How do we give the right amount of information and not guide people to do too much the way that often we do too much in our current system anyway? And I think that all these subjective elements are being decided currently by companies who are building this AI that then is plugged into the EHR.
I mean if you go to a chat. Bot and you put a case in and you ask it like, Hey, what's the best high value care I could do for this patient? You'll get a different answer than I'm worried there could be cancer. Please tell me what to do. 'cause it has this problem with sycophancy, of trying to answer what people want.
And so I think that both the financial and, [00:47:00] regulatory type barriers that are out there for clinicians right now will drive the same processes in ai that if we're not careful, it may tell you to do more or it may give you a differential diagnosis of 10 items. And then, you know, for, poor Valerie who's admitting someone as a hospitalist will say, oh well we didn't really answer that ninth item.
Let's order a test because it's there now. So I think we need to be really careful about how, humans make these decisions around AI being built into the clinical world. 'cause they are subjective and they can lead us down different pathways.
Angela: Okay, so what I'm hearing is it doesn't sound like it's implementable at this point or that you would trust it.
Dan: I think it had to be implemented carefully. I think it's really good at making things more efficient and like reading lots of notes and maybe summarizing them or helping people find information.
So, Valerie's example of like not knowing if someone's on a laxative, that's one, at Stanford. the generative AI has access to the entire medical record. So, it could be great for that. 'cause then you could just ask it like, Hey, has this patient received a laxative?
And it'd [00:48:00] say, , yes, two days ago, or, earlier this morning they got it. So I think it can be helpful for that, but I think that it could really go in different directions, and especially if we're just asking it to make care decisions, it will have a philosophy that is either unstated, but we'll be driving towards like, action bias and doing more.
Or it could be, , driven towards high value care. But it's, how we decide to do it or how the companies decide to do it. And right now, I don't think there's a lot of thoughtfulness around what, those decisions will be.
Valerie: I'll also say that, if we think AI is a thing happening in the future, like it is already here, I don't know if you guys have seen this, but every time I walk into a workroom, it's like every computer is open to, new England Journal of Medicine open Evidence.
and that has replaced UpToDate for our learners. Like they're all using AI for everything nowadays. And I think that's probably not the right answer right now, but nobody cares what I think and nobody cares what we think. Like that is [00:49:00] what the young people are doing. And I will say that, all of the ais that I use for, writing and things like that have gotten infinitely better in the last six months.
Like I haven't seen a hallucinated reference in the last like two months that I've been working with the LLMs. So the problems that they had six months ago are better now, and I suspect that come a year from now, it's gonna be even better. , The speed of improvement is just so. vast .
And I think what would be smart to do is kind of try to see how can infectious disease, how can, people who are thoughtful about overuse, how can we be helping the AI's bias, , can we be the ones that are actually, influencing the underlying motivations and framing that the AI has, rather than letting that be externally controlled.
Annie: Now we have a sort of lightning round of questions, so here we go. What is the most overused test in hospitals today, Valerie?
Valerie: So I'm actually gonna say chest cts. [00:50:00] We just got some really interesting data from the pneumonia collaborative I work with in Michigan. in, patients who have a diagnosis of pneumonia, have a pneumonia on chest x-ray, 50% of them still get a chest ct. And we saw this jump after COVID. So I think some of it is probably people were concerned about concurrent pulmonary embolism, but 50% of people do not have a PE that have pneumonia.
I mean, that is a huge overuse. and it hasn't gone back down after COVID. It's like people started doing it during COVID and now we're just doing chest cts on everybody who walks in to the ER with pneumonia.
Angela: is that also just availability, like physical availability?
Those, hospitals bought more machines during COVID because there was demand and now the machines are there and, it's just easy. Better use them.
I also think,
Angela: oh, sorry. Go ahead, Dan.
Dan: the test I was gonna throw out there is, one that also is like, I think I would blame COVID for too, and that's CRP, and I know Angela loves CRP and thinks it's a great test, but, [00:51:00] CRP and even Lactates, I feel like those were tests I always thought like they're just not very good tests.
Like we don't use them much. And then like, they both popped up at different times. I mean, lactate, it was the sepsis people who got us testing that all the time and crps, it felt like it was COVID. And people were like, yeah, we're looking for inflammation. what does that mean? I don't know.
We have a white count. You know, like, what, what is it gonna be done for? And this isn't like that. We should never do a CRP. I'm not trying to hurt anyone's feelings. Like in a bacteremia, it's a great test. You can shorten antibiotics, I hear. But, daily CRPS is probably not what every patient needs in the hospital.
Angela: Okay. You're killing me. All right. First of all, no one should have a CRP every day. I'm fully on board with you there. crps, it's a tool that
you need to know. I don't wanna waste everyone's time on this, but, yeah, no, it's an very imperfect tool. A CRP, and we all know this, it's, very sensitive, but not at all specific.
You can't use it willy-nilly. You need to know how to use it. but guess what, it costs like [00:52:00] $5. It costs almost nothing compared to a chest ct or, a PET scan, which is what we are doing
here like every other patient's getting a PET scan now.
Anyway, we can have that discussion offline. or you can read
the letter that, I wrote with Daphna Yahav and, Pranita Tamma to defend.
appropriate use of CRP. Let's put it that way. In CMI.
Valerie: I see a pro con debate coming for our next, diagnostic steward pre-course.
Yes. Where we, we wanted to do biomarkers next year. We should have you guys debate. CRP.
Angela: . So, next question in the lightning round. Valerie, what is the most underused test today?
Valerie: Does the physical exam count? Or does that just make me sound like an old person? Okay. I feel like, feel
Angela: like it's a test.
Valerie: I feel like I have a story to tell. which I think you wanted quick answers for this, so I'm sorry, but story to tell.
So you guys all know my, husband. two years ago now, he was having daily high fevers and like, fevers so bad at night. [00:53:00] He was like hallucinating. course we all thought it was COVID. And finally with like the third negative at-home COVID test, we took him to the er. the ER of course redid a COVID test, which was still negative.
And then did our favorite thing of all the pan culture. so blood cultures were negative, urine culture was negative. Chest x-ray was negative and they sent him home. And I remember, 'cause I was there with him that there's no physical exam. The ed doctor did not do a physical exam. So when he got home and he again had a fever to like 102, I was like, okay, I'm gonna do a physical exam on you.
and I remember I got to the abdominal exam and I was like, I can feel your liver if I can feel your liver. Something is wrong with your liver. Like my exam is not that good. And so I wanted him to immediately go back to the ER to get a CT of the abdomen. But he waited until five o'clock in the morning when he knew that census would be low and he could get in quicker.
very doctor of him. And of course he had this massive liver [00:54:00] abscess. it's not the first thing that you would think of with high fevers, but if you'd actually done a physical exam on him, you could have saved yourself all of the other tests and gone straight to a CT of the abdomen.
And so I just think we need to do a better job of thinking about physical exam. And a good history, obviously, but his history was not helpful. So physical exam, the most underused test.
Angela: Wow. Okay. . And you, Dan,
Dan: I would fully endorse Valerie's answers better than mine. I mean, I was thinking maybe rapid id, testing of blood cultures when they're positive.
I feel like that's a test that, , isn't universally done. That's, quite useful. And it's, testing more, but it, gets to better outcomes. especially when it's been implemented in the right, way with having stewardship interpret the results.
This is like a PCR, a molecular test for the, organism identification, like when you have GPCs coming out of a blood culture. So like you know the culture is turning positive, but it's gonna take a day or two before you get a growth to tell you that like, this is staph aureus, or this is coag negative staph. This will tell you which one it [00:55:00] is.
And so instead of getting gram-positive cocci and then you need to treat them with vancomycin and wait until the results come back, this will tell you right away what it is. And so you can say, oh, this is a real blood culture that's positive, versus this is a contaminant most likely.
Annie: There's a few comm, there's a few commercial ones that have like, sort of 18 pathogens that cover 99% of your blood cultures. But we, have a four arm maldi-TOF on our blood cultures, and then you can set it up for eight or 12 hour susceptibility tests.
So, yeah, I'm with Dan Rapid id, but Rapid A STI would put in there as well. We've not got there quite yet.
Dan: Now, even better. And I think there's lots of technologies that, work okay for this. So it's whatever the lab can do. Easiest.
If you could remove one protocolized order from electronic ordering systems tomorrow, which would it be? Valerie?
Valerie: urine cultures.
Annie: Hey, I think this is one of those
Valerie: tests that they are hidden everywhere. They're in, like pre-op order sets, ed order sets, like, daily neuro ICU order sets.
Like they're in such weird places. And, the first step to [00:56:00] reducing asymptomatic bacteria treatment is finding all those places that you're like, why is it there? And removing them
Annie: completely with you on that, Dan.
Dan: Yeah, I mean, there's one where I was like, oh, Valerie's gonna go first and, take my answer.
'cause uh, urine culture is completely it. I don't know if like sputum culture, wound culture cultures appear around some, but they're definitely just not as common as, urine cultures for being everywhere for unclear reasons.
Annie: Yeah, hard agree.
Angela, is it gonna be a full sweep on the urine culture from the order sets?
Angela: Yeah, I would have to agree with urine cultures because of the, the scope of the damage that is done, by the ordering. But I have to be honest that I think what I truly dislike more is the sputum culture because it's just, you know, honestly, in a urine culture, I think it's easier to interpret. Of course, if you have the, what do you call it in English? the sediment, the smo, the
Dan: urinalysis speak
Angela: English. The urinalysis. Thank you. The microscopic exam. Yeah. like if you have that, a urine culture is quite easy to interpret, whereas a sputum culture, like [00:57:00] for non ID people, I mean if it's just, it's so misleading and, it just leads to such abuse,
but I think it's done less often. I think Valerie's right, that a urine culture finds its way into everything. So you'd have to attack there first, if you wanted to improve things. yes. Next question is Valerie, what's one diagnostic habit you've changed personally?
Valerie: Yeah, so I think one, and this goes back to maybe after, again, having Matt in the hospital and seeing him woken up at five o'clock in the morning for his morning blood draws. But also just like the experience of, seeing how others order tests is I really try to bunch my testing. and that's probably not the answer, that is super infectious disease related.
But instead of ordering like, oh, let's think about this and I'll get it now and then, you know, I try to make sure someone only gets poked once they've started doing a thing here, which is just brilliant. And this goes back to make my life easier. Diagnostic stewardship practices is if you can [00:58:00] actually add on what you want, it'll say, do you wanna add this on to something we already have?
When you go to order it, it'll say, do you wanna add this onto this morning's labs? And oh my gosh, I remember as an intern and resident at Michigan having to call down to the lab to add on a test. And now the fact that we can just do that, makes my life and, the outcomes for our patients so much better.
Annie: And finally, if you woke up and found out that our current public funding context was really just a bad dream and you actually had a billion pounds or a billion dollars to do the study of your choice, which one would you do?
It has to be diagnostic stewardship, obviously.
Well, I'd hearken back to the conversation we were having earlier, and I'd create the Ultimate Learning health system where every diagnostic decision that we made was part of a pragmatic clinical trial. So it would be constantly having evidence to support the use or non-use of all of the diagnostic tests that we're doing.
before we retire. We will see this done. We will
Valerie: love it somewhere
Angela: besides just [00:59:00] Acorn, but anyway,
Valerie: the Acorn people actually just did this with MRSA NAS swab test in the ICU. So the Vanderbilt people are making this a habit, We just hired them for a clinical trial. We're putting it in January. So, hopefully it's gonna be routine in the United States in the next five years, especially if you can't pay millions of dollars to do a real clinical trial. This is the only way to move, knowledge forward in the interim.
Dan: Great. And if I had that much money, I'd probably give it to Valerie for this idea., If we could really change the way the system, generates information, it would be key. And I do think you could do these practical trials where, you really get the diagnostic now, or you can order it again in a day if you want it or something like, have things that don't really change the care that's provided, outside of that one test.
and especially with say CT scans, I think you would, find pretty remarkable results how much it doesn't actually change practice or may even lead to harm in ways that, it's hard to assess without a randomized study. and a true learning health network, system would, be well worth [01:00:00] it.
we talk about that a lot in America, but it, it rarely happens.
Angela: still things we can aim for before we retire. so for the two of you, are there any last messages that you want our audience to hear before we close Valerie?
Valerie: Yeah, I'll just say again what I kind of said in the beginning that my favorite thing about diagnostic stewardship is you can both improve patient care but also make the lives of the clinicians that are doing the care easier.
And so, please, please make my life easier.
Dan: similarly, I'd say just think systematically about tests. when they're helpful, when they're harmful, and then try to build that into a system if you can. And that's, what, I think it's really the basis of diagnostic stewardship is changing the system so it's easier to do the right test.
And, it requires a lot of time to think about when tests are useful or not. But then, trying to make that, more hardwired so it's not dependent on each doctor to learn the right way to do each of the decisions we,, make a hundred times a day.
Annie: thank you so much to our guests today, professor Dan Morgan [01:01:00] from the University of Maryland and Professor Valerie Vaughn from the University of Utah in the USA.
And thank you for listening to communicable CMI Comms podcast. This podcast was hosted by Angela Huttner in Geneva, Switzerland, and me Annie Joseph in Nottingham uk, editors at CMI Communications, ESCIMD's Open Access Journal. It was edited by Dr. Katie Hostetler, oy and peer, reviewed by Özlem Türkmen Recen of Çınarcık State Hospital, Yalova, Türkiye.
Theme music was composed and conducted by Joseph McDade. This episode will be citable with a written summary referenced by A DOI in the next eight weeks, and any literature we've discussed today will be found in the show notes.
You can subscribe to Communicable on Spotify, apple, wherever you get your podcasts, and also find it on ESCMID's website for the CMI COMMS Journal. Thank you for listening and helping CMI, comms and ESCMID move the conversation in ID and clinical microbiology further along.
[01:02:00]