Well, thank you so much for the invitation to speak. It's a pleasure to
be here. That was a terrific and really moving patient story, and I think really
fitting as a great introduction to what I'm gonna talk about and show about the
importance of antibiotic stewardship in the outpatient setting. So just quick
background, I am a pediatrician, my clinical focus is in pediatric
infectious diseases. I work at Children's Hospital of Philadelphia and University
of Pennsylvania, and I direct the inpatient antimicrobial stewardship
program there, but I do a lot of epidemiology research and
recently have gotten into the field in the area of outpatient
stewardship, and so now do some QI work as well as research in the
outpatient setting. Oh, I am from Philadelphia I know we don't have a
great reputation in the state of Minnesota, but I've lived in Philly
for a while, but I'm actually from Massachusetts, and I'm a Patriots fan
which was, you know, hard to deal with in Philadelphia, but, you
know, I'm a little friendlier than the folks down there, so anyway --so I have no
conflicts of interest. I think you have your learning objectives printed
somewhere there, and this is a quick outline of what I'm going to talk
about. I'm happy to talk about more.You 36 00:01:28,460 --> 00:01:33,110 can stop me or ask questions after, as
they arise. So, I want to try to make the case for outpatient stewardship -- it
probably won't be a tough sell with this crowd. I want to talk about a specific
study --some work that we've done at Penn and Children's Hospital
Philadelphia to generate what we think is an important target for
outpatient stewardship. Talk about some implementation and interventions that
we've done and then talk about some novel stewardship approaches that other
folks have done that I think are really promising. Okay, so this is Philadelphia
in the background, and I don't know if anyone can see the foreground here this
is a place called Eastern State Penitentiary which was a super famous
prison a long time ago. It's been closed for almost 50 years. It's now a national
historical landmark and it's actually -- you can go and they do
tours, and actually has the best haunted house you've ever seen before in
Philadelphia. Around, you know, October you should go here, but you're
wondering why I'm showing you a penitentiary -- and in addition to being a
great place because they actually were really progressive in reforming the way
people were incarcerated to actually reform and help them instead of just
sticking them in terrible cells --they actually housed some famous criminals so
Al Capone and this man here. Anyone know who this man is? This is Willie
Sutton. Willie Sutton was a famous bank robber and again you're wondering where
I'm going with this, but Willie Sutton helped me think about why you want to
focus on outpatient stewardship. So, when asked why he robbed banks, Willie Sutton
famously said, "Because that's where the money is!" And when you think about
antibiotic use right now, 90% of, more than 90% of antibiotic use is in
the outpatient setting. But most of our efforts and a lot of the regulatory work
has been in the inpatient setting. And it makes sense because it's a little bit
more of a captive, it's much more of a captive audience, and that's where we see
a lot of the multi drug-resistant pathogens and lots of the, quote
unquote, big gun antibiotics, but, but, but, again, more than 90% of human antibiotic
consumption is in the outpatient setting. Prescriptions at places like urgent care
for pinkeye which, you know, even though it was, and I don't know of course,
but even though it was the worst case of pinkeye ever, I still would question
whether that case needed systemic
antibiotics, right? So probably not. But this is where antibiotics are
given most of the time, okay, so just to build on that and give you some scope of
the problem, this is a paper done by Lauri Hicks and colleagues at CDC,
published a few years ago, two or three years ago now, looking at using the
IMS health database is a great database that essentially has a hundred percent
capture of antibiotic prescribing in the outpatient setting and in 2011 --that's
when the most recent data were available -- we saw more than 260 million antibiotic
prescriptions in the United States in one year. So right there,
about 310 million people or 320 million people, so 842 prescriptions per person
about. I'm a pediatrician, so about a third of those, almost a third of those,
were in kids, and in kids it's more like 1.5 to two prescriptions per child per
year, right, so we use a lot and I'll focus again on kids. When you look, just
to give these numbers is hard to think about, you know, where, where
should the number be? We have a lot of people, and maybe those numbers are right,
but we, but using comparative epidemiologic data,
you can help think about it. This is a paper published a few years
back in Pediatrics which shows the top six most commonly prescribed medications
to children. Okay, so the big heavy hitters, asthma, ADHD, cough and cold
medicine, pain medications, the top line is systemic, not topical, antibiotics --so
blows everything else away it's not even close. Seventy to 75 million prescriptions per
year just to children. So, so, we really use 'em a ton, and you might say okay,
again, we have a lot of infections and we need them but if again, looking at the
comparative data, if you look again in the United States and this is one year
later I said 842 the next year was 833 prescriptions per person, if you compare
that to another developed country with a, you know, a first-world healthcare
system, I would argue it that is better than our health care system, that's
Sweden, they prescribed 388 prescriptions per thousand
people. And they think that's too high. Their version of the CDC
has set a goal of 250, okay, and and they're they're gonna meet that goal if
they already haven't, and you know you don't, I don't think people are dropping
dead from colds in Sweden. So, so, we're not doing really well, and I've shown you
it in the zero to two age range, and the three to nine year age range, it's even
higher, again, over one prescription per child, 1300 per thousand, and it's
much lower, much lower in Sweden. It's not just quantity it's also quality, right, so
all antibiotics aren't created equal. We have you know it's it's this is
an oversimplification but we have narrower spectrum agents and broad
spectrum agents, and if you look at the three most
frequently prescribed, broad-spectrum antibiotics, quinolones, you know, your
cipro and your levaquin, macrolides the famous Z-Pak --greatest marketing in
antibiotic history --and then cephalosporins, you know, we're
prescribing quinolones four times more often than Sweden, more than ten times the
macrolides, ten times cephalosporin so so we're not only over prescribing
antibiotics in general, but we're actually prescribing broad-spectrum
drugs much more commonly. And so another way to look at this, and to try to get --
again we don't know the exact number --but comparative data are helpful, and this is,
this is a study looking across our Children's Hospital Philadelphia
Practice Network. So we have, we have 31 now, primary care practices --this shows 25.
These are community-based primary care practices they, luckily for me and other
researchers they all have a common electronic health records, we can pull
their data. And if you look at when somebody has an acute respiratory
infection, a patient comes in for a respiratory infection, what proportion of those
patients are prescribed a broad spectrum, or here you can call it an on a second
line or unnecessary antibiotic, and when you look across practices, it goes from
15 percent to more than 50%. And these are tens of thousands of visits, so these
are not, you know, this isn't somebody who saw, you know, two out of three, or one out
of ten, these are highly statistically significant, and these are standardized.
So these are adjusted for the age, the sex, the race, the insurance type, we've
excluded anybody with a complex chronic and medical condition, anyone who's
received an antibiotic in the prior three months, and anyone with antibiotic
allergy. So these are, essentially, you know, if you, if you do the Epi and you
the analysis see this is taking the same patient, and putting them at different
practices, and they're being treated completely differently, so we don't do
this very well. And I don't want to just pick on kids, so this is, this isn't
this is adult data from Jeff Flender, who's a real leader in the area of
outpatient antibiotic stewardship, and this, this looks at antibiotic
prescribing for acute bronchitis in adults, anyone greater than 18. So what
should the prescribing rate be for acute bronchitis in adults?
Should be zero. Okay, there have been more than a dozen, randomized clinical trials
of antibiotics versus placebo for patients with acute bronchitis, and
there's never been --a difference has never been shown -- so it doesn't help. Yet,
over this 20-year period we still prescribe in either primary care setting or
the emergency department, about 75% of our --70, 75% of our
patients get antibiotic prescriptions for this condition. But, antibiotics are
wonderful, right? And this is, these are quotes that we, we have --I work closely
with Julie Szymczak -- I'll show some work from her -- she's a
PhD sociologist who's a qualitative researcher who is, has an
interest in antibiotic stewardship, and she often goes to the front lines whether
its primary care setting, inpatient waiting rooms, on the wards of in busy
and patient units, and talks to people. And this is one where she's just got
some quotes from parents right so -- "I'm a fan of antibiotics just because the fact
that it does heal them pretty quickly," "I think they're wonderful, they clear
things quickly..." "All I can say is they work...", so we, you know, we have this
perception, a lot of patients have this perception --and I'll talk a little bit
more about --that they, that they are "so great", but there are downsides and again
I'm kind of preaching to the choir but I'll review some of these important
downsides. The first thing we think about around antibiotic stewardship is is
resistance, right? So antibiotic use drives antibiotic resistance. That's it,
that's, that's what happens that's, that's how it works. And bacteria have shown the
ability to become resistant to every antibiotic that's ever been developed,
and in fact there have been some, some really interesting studies where they've
gone to --in the science paper a few years back --where they've gone into caves where
humans had never been, and they've, they've dug up the bacteria and done DNA
sequencing, and found resistance to antibiotics or mechanisms
that haven't even come to market yet. So they're just out there waiting for us
to use them, and they become resistant, you know, once you put a new drug in the
market you see resistance within a year or two there there are a few
exceptions to that. And CDC has appropriately developed a terrific
report, free online, and estimated that we, that more than two million patients
are sickened with antibiotic resistant infections, infections each year leading
to 22,000 deaths. These are, these are vast under-
estimates, these are just what's reported. A quarter million cases of C. difficile, at
least 14,000 deaths from C. difficile. So you, you know people die from C.
difficile, as many in the room know, and this is really a problem. But, I think
it's really important from both a scientific but also a practical
perspective to think about the individual harm, and this why Kaitlyn's
story is so powerful, right? So, when you're in a room with a patient whether
it's in primary care or urgent care or on the wards of an inpatient unit, it's
hard to think about the public health impact of antibiotic resistance when
you're doing and giving that one prescription to that one sick child or
that sick adult, okay? And it's, it's hard to think about that and it's hard for
that, that interaction to really capture that. We should always be thinking about
that as public health experts. But, it's hard to do that, so, so we need to think
about when we need to quantify the individual harm that antibiotics can
cause. So 5 - up to a quarter of patients who receive an antibiotic will have
antibiotic-associated diarrhea. I mean, in the least that's annoying, and it keeps,
you know, kids home from work, and you know, parents home from work sometimes,
but in the worst, it could be C. difficile, a small but significant
percentage is C. difficile, which is, which can be life-threatening.
When you look, one in 1,000 patients who receive an antibiotic will go to the
emergency department for an acute adverse drug effect from that antibiotic.
And, you say well, eh, one in a thousand, that's not a big number but that's the
same rate as drugs like warfarin and digoxin --the drug that we worry about the
most in terms of toxicity. And, also there are some nice public health data out of
the UK where you actually, if you give somebody an antibiotic for an acute
respiratory infection, bacterial or viral -- even things like a otitus media, Strep
throat, peritonsillar abscess, you actually have a one in four thousand
chance you'll actually prevent serious harm in that patient, things like, you
know, that people worry about getting a bad case of strepto, or getting a
peritonsillar abscess, or getting mastoiditis. So, we have some numbers that
actually don't really favor the use of antibiotics in some of these common
conditions, and then, and then there's the microbiome.
So, who's heard of the microbiome? Okay good, it's becoming more
and more popular, right? So, so the microbiome is basically the bacteria
that live in and on us --most of it's in our gut, and, and you know your body is
mostly, mostly microbes, you know, this is a fact, and some people get freaked out
by this, but for every human cell there are five to ten bacterial cells in and
on your body, that's it, that's why. So, we, we are mostly microbes. Hundred times
more bacterial genes and more than a thousand different species in and on
your body. And it turns out that these, these are important, and so I just --this
is a nice review article --but we used to think okay your bacteria, the bacteria
just kind of lived there, and they don't cause problems, we don't bother them, they
don't bother us, and maybe they make a little vitamin K, but there's a lot more
that goes on in regulation of metabolism. Well, if anyone's seen some of the
amazing studies they've done in animals to show that when you give mice
antibiotics they get fatter than the mice who get placebo. And actually,
you can actually give mice antibiotics and just take their stool and do a stool
transplant into mice who didn't get antibiotics and they get fatter than the
mice who get stool transplants from non- antibiotic receiving mice. And they've
gone even farther to show that that you if you take humans that are discordant that --
human twins --that one is obese and one is lean, and you take their stool and you
give it to mice, the mice that get --the mice follow that phenotype --so the ones
who get the lean stay lean and the ones who get it from the obese twin, you
know, so there's a lot going on here, and I think development and regulation of the
immune system is a really important one to pay attention to,
because it turns out that the microbes in your gut
educate the immune cells in your gut, for life. And, that's important because as you
know, your immune system is really important but also it can be related
there have been some associations with the changing microbiome and autoimmune
diseases, chronic diseases. A lot of this research is early, and we don't know all
the cause and effect, but, but it's, it's fairly profound, and prevention and
invasion of growth of a pathogen --C. diff is the perfect example there. So, this is --
I am NOT a microbiome scientist, but this is your incredibly basic primer on
the the microbiome. It's pretty complicated, but diversity is good,
right? Diversity is always a good thing in pretty much every walk of life but,
diversity is good, and I've shown a couple slides from Dan Knights, who
actually is a collaborator of mine. He's a computer scientist and microbiome
expert at the University of Minnesota, and, and he has a great website with a
really good primer for those of you that are not microbiome scientists. But, I
borrowed these slides --so he this is your gut, and this is this diverse ecosystem
representing all the bacteria in your gut. And, this is your gut on drugs, right? So
you, you just, you take oral antibiotics, many of them go right through the gut
and you can, you can completely destroy this, this complex, and what is considered
more and more important, healthy ecosystem. And then the problem is okay
you can grow it back as bacteria grow quickly, but you can lead to this, and
this looks like a beautiful photo right? But this lack of diversity has been
associated with harm, both acute and chronic illness, and we want to avoid
that, so these are the individual potential harms from antibiotics that we
have to think about as prescribers as, public health experts, and as patients.
When we, when we think about the decision to make antibiotics --so this was the
cover of a New York Times Sunday magazine article some people may have
read a few years back --this kid is doing the best they can to populate their
microbiome and then we go and give them Augmenin or a Z-pak and ruin everything,
all the work that they've done. Okay so I want to shift gears and talk a little
bit about there are multiple different ways to address antibiotic stewardship,
and I'm going to talk a little bit about implementation approaches that we, that
we've taken and that others have taken. But more foundational than that, is to do
the research and do the studies to find out when you should be using drugs, and,
if so, which drugs you should be using. So I want to talk a little bit about the
concept of broad-spectrum antibiotics that I brought up before. So this --I'm
going to talk about a study that we did. I'm gonna talk about it in five slides --
four years of work and five slides it hurts me to do that. But this is called
the series study compared to effectiveness of antibiotics for
respiratory infections. Most important is the study team and we got this together
this is a PCORI funded study, so that is the Patient-
Centered Outcomes Research Institute, which is a terrific funding organization
that comes off of the Affordable Care Act that hopefully will stick around. And
it was truly a multidisciplinary team, we have people, folks in a family
Advisory Council at our Hospital so these are parents of children with
chronic diseases, primary care doctors, infectious disease epidemiologists,
biostatisticians, and others. And, and so just by way of background, why do you want to
compare --why would we want to compare broad versus narrow spectrum antibiotics? Well,
I think there are --I hopefully have given you some compelling reasons --but in
Pediatrics, and it's fairly similar in adult medicine, the concepts are all the
same. The Pediatrics is pretty straightforward cuz these three
conditions: acute otitis media, acute sinusitis, and strep throat account for
about 70 percent of all antibiotics prescribed to kids. So, you know, this is,
these are the low-hanging fruit here. And, you know, you think it's so bread and
butter, you think that pediatricians did it really well, but it turns out that
they don't and we don't and that some of it's because the guidance isn't very
good. So the American Academy of Pediatrics recommends amoxicillin for
acute otitis media, but a couple years ago, two very high profile, large, New
England Journal published, randomized, control trials compared broad spectrum
Amox-Clav --broad spectrum or augmentin --to placebo. So that sent a
strange message, or a confusing message to practitioners for sinus infections.
The American Academy of Pediatrics recommends amoxicillin
but the infectious disease Society of America recommends amoxicillin
clavulanic. So that's not great. And then strep throat I don't know if people have seen, but, but strep
throat is, there has never in the history of the world, been a clinical isolate of
group A Strep resistant to penicillin. That's amazing, you can't say that for
any other, yet we, we use all types of things, like azithromycin there's,
there's arguments that cephalosporins are better because of shared
beta-lactamases -- which has actually never been proven, but people are
prescribing different medications, so we thought this would be a good place to
start. Also pneumococcal vaccination, as you know, twenty years ago there was no
pneumococcal vaccination, then we had a seven-valent, and a ten-valent,
now a thirteen-valent vaccine, which is great and, there
there's some evidence that we're getting rid of pneumococcal colonization
and and replacing it with other bugs like H. flu moraxella that are more
likely to produce betalactamases, and so could this resistance lead to the need
for a broader spectrum antibiotics. And then, oh yeah by the way 50% of
antibiotic use for children is broad-spectrum which is probably not
appropriate, and I showed you this slide before, we're not doing it very well. So
hopefully I've made the case that it's good to compare these drugs. So what we
did is, I mentioned we have this this large network of primary care practices
that are all affiliated with Children's Hospital Philadelphia, we did a
prospective cohort study from 2015 to 2016, and 31 of these practices kids aged
6 months to 12 years who are diagnosed with acute respiratory tract infections, are
the ones I talked about otitis media, strep throat, sinusitis, AND prescribed an
antibiotic. We excluded patients who had multiple different conditions, at some
other --you know, if you had a skin soft tissue infection or UTI, you were
out of it, and anyone who got recent antibiotics, we thought that confused the
picture. And then we -- the thing that made this a large and sort of resource
intensive study is that instead of just looking at the health records or
reviewing charts, we actually called patients, called parents of patients,
because we wanted to really find out what the outcomes were so we called them
five days after their diagnosis, enrolled them, and then we did two, structured,
telephone interviews at day 5 and day 14 after their diagnosis. Our exposure you see here,
a pretty simple narrow spectrum penicillin, amoxicillin because they're
first-line recommended agents for the conditions I've talked about and then
broad-spectrum antibiotics augmentin,
cephalosporins --cefdinir or Omnicef is the most common one there, and then
macrolides or Z-Pak the Z-Pak or azithromycin dominates there. And so I
think the key, the key to this study and it's patient centeredness was was our
outcomes. So this is a picture of Julie Symzcek, we mentioned before, she's a
PhD sociologist with interest in antibiotic stewardship, and so first
figure out what outcomes we should look at because kids with --just like
adults with acute respiratory infections-- they're not dying from those, they're not
going to the hospital from those, it's incredibly rare, so we want to look at
the common everyday outcomes that you wouldn't always capture in the chart. So
she first went to busy practices and sat down and talked to parents who had kids
with, with painful ears and fevers in waiting rooms, and said, what do you care
about when your kid comes in, and what are you, what are you worried about, and
what are you trying to fix here. And, so the things that parents identified were
missing school and work, child suffering, so that we figured would be their sleep
quality, we had forgotten about that one right? Katelyn mentioned about sleep, I
mean, I'm a parent of two kids and, and when sleep is really hard to get when
kids are sick, and it's, it's, it's important, and then speed of symptom
resolution were identified as important outcomes. So we work those in,
and, you know, we did a lot of complex analysis, but as you can see from this,
this one summary slide, these are all the outcomes we looked at in the left, and
the one I'm circling which was, which was a peds quality of life inventory that we
did, we assessed every patient for --parent and patient for --the different, there
was no difference between broad-spectrum antibiotics in the first column or
narrow-spectrum antibiotics in the second column in terms of their quality
of life, and then across all these other things --missing school day care, requiring
additional child care, sleep disturbance -- there was no difference, except for
experienced adverse effects, so 35% percent of patients who received a
broad-spectrum antibiotic had some adverse effect, mostly diarrhea, nausea,
vomiting, or rash as opposed to 25% with narrow spectrum agents. So there are
limitations to every study, there were certainly limitations to this one --we
relied on the clinician diagnosis so we weren't prospectively verifying whether
these patients had these infections, but this is the real world, and this is, these
are the patients who are diagnosed with and treated for antibiotics, we felt that
that was legitimate. We only were able to --we were cold calling patients
after they were diagnosed --so we only --only 30% of patients answered
the phone and agreed, which is about where we expected, and we actually
compared those who hung up to those who didn't, and demographically they were
similar. The, the peds --the quality of life inventory
might not be sensitive enough to pick up minor differences, there are always
potential unobserved confounders and and whether this generalizes outside of our
network we're not sure. But, our, our conclusions were that according to these
patient-centered outcomes we didn't find a benefit of broad-spectrum agents over
a narrow spectrum agents for treatment treatment of these conditions, although
broad-spectrum agents were associated with higher rates of adverse drug
effects, and these we thought these confirm and extend the recommendations
of the American Academy of Pediatrics to prescribe narrow spectrum agents for
most conditions. Okay, so I talked about implementation, how do we actually get
this done? We now have some foundational information to say, you know, this is, this
is, these are the right drugs, and we have the agencies behind us to say these are
the first-line recommended drugs, but I showed you that 50 percent of kids are
getting broad-spectrum agents, and it's all over the map, and you can look at
this is not just in our network but all over the place, so how do we actually change
this? And, you know, this conference is about antimicrobial stewardship or
antibiotic stewardship and, you know, most people here, based on the poll,
have programs, are involved in programs that, that do stewardship. But as
I said before most of the guidance around stewardship is in the inpatient
setting, right? So IDSA and, and SHEA and CDC really have great guidance around
the inpatient side, but not a lot in the outpatient setting, and CDC has really
come to the, stepped-up and started to make some recommendations here, but we
wanted to see, you know, is outpatient stewardship achievable? And so these are
some of the core strategies that are recommended in the inpatient setting.
Prior authorization, you can't do, right? You can have 40 people in
the waiting room in Acadia you can't wait to get a call whether you're
allowed to use your augmentin. But, but there are other things, like prospective
audit and feedback, and adaptation of that, education guidelines and optimizing
doses that can I think can be done, and have been done in the outpatient setting.
So as one example, and there are many groups who have done work like this, I'm
just going to talk about some things that we've done
at CHOP. We conducted a cluster randomized study in this network that
I've been talking about, so we enrolled 18 practices which had about 170
clinicians, and as I said they were all in a common electronic health record and
we focused here again on antibiotic choice. We're talking about not whether
you prescribe an antibiotic or not, we have, we have a study in progress looking
at that, but if you prescribed an antibiotic for one of these conditions,
what did you prescribe? And, again, Strep throat, sinusitis, and here we used
pneumonia because there was a separate otitis media or ear infection study
going on, so we wanted to focus on these three. And, again, the nice thing
about these three is the first-line agent of the amox or penicillin and the
others are all second line. So, what we did is, here is our timeline, we pulled
data, we did a, essentially, a cluster randomized trial, but we also used
pre-post study analytics to superimpose. We collected 20 months of baseline data
and then we, myself and a group of other folks on our team, went out to the nine
practices that were the nine clusters or practices that were randomized to the
intervention, and did on-site education to say okay here remind them of the
guidelines, and here's what we're gonna do we're gonna we're gonna study this,
we're gonna look at your data, and we're gonna give you some feedback reports. And
you can see, we gave over 12 months, we gave three feedback reports about how
they prescribe their antibiotics compared to others in their practice or
their entire network and with respect to the the prescribing guidelines by the
AAP. So here was an example of one of those report cards they had, they had one
for each condition and this is, this is a real de-identified example, and so this
is broad spectrum of your proportion of broad-spectrum antibiotic prescribing
for acute sinusitis, and we listed what what the drugs were, and you can see
how you prescribe over time so blue is baseline first quarter, second quarter --we
kept adding bars over time --how your practice was prescribing, and how to
network prescribing. The idea here is to get at peer comparison, right? To leverage
this idea of benchmarking and peer comparison --how are you doing
paired to two others --and that's in the behavioral economics space that's
thought to be a powerful lever. And so we did this, and as you can see again over
time starting on the left, twenty weeks before to zero, which was the start of
the intervention, when you compared the control group is the top --I don't they
can see the colors --green dots and line to the intervention, which are the X's
the red X is --we saw a relative reduction of 50% of broad-spectrum antibiotic
prescribing in those who are who were getting the feedback reports in the
intervention, right? So, so this was great but it was a research study that was
funded for a certain amount of time by AHRQ --should thank them --AHRQ is
a terrific organization, and then we stopped giving feedback reports, but we
still had the data, right? And so we stopped, and what do you think happened?
Yeah, so went right back up to the baseline. And, so we learned, we learned a
lot from this, that we thought this, thought this feedback, leveraging
electronic health record could be effective, but there's something else
behind this, right? It seems like being watched and having an active
intervention was important because I don't think people forgot what the
recommended guidelines were once we stopped sending them feedback reports.
They, I think they knew I think there's something hear about, about a Hawthorne
effect or otherwise of being watched. So, so what do clinicians actually think
about this, about antibiotic prescribing, about this type of intervention? So, we, we again,
Julie Szymczak, our behavioral sociologist, went out to practices and
interviewed clinicians who were part of the, some who were part of the study and
some who weren't part of this study to ask them a few questions about the
intervention about prescribing. And, so it turned out that most did not believe
their prescribing behavior contributed to antibiotic overuse. And this is not
new, this has been shown before in other populations, and we've actually shown --I
don't have a slide, but Julie has actually polled people again, and when
you look, generally speaking, about ninety percent of primary care doctors think
that antibiotic overuse is a problem, but only 10 percent think it's a problem in
their practice. Right, so those numbers don't really make
a whole lot of sense and this is, this is a problem. It's sort of like "not in my
backyard", and that's something, that's something that we really have to think
about when we are designing interventions. And talking to folks, they
reported frequently confronting parental pressure, the first thing --not just in our
network but in other past studies, and
I'll show you some more data --the first --thing people say --well, it's the parents
wanted it or the patients wanted it that's why I get --they pressure me
into it and then sometimes they would acquiesce because they, they worried that
if they didn't give it, they would lose that patient, and this is the reality the
real world, this business model. Kaitlyn mentioned urgent care. Urgent care and
retail clinics have popped up all over the place, and and there's a real fear
that pediatricians will lose their patients to these clinics and
they'll just go and get their antibiotics, and then come back and have to deal with
the complications. So this is, this qualitative work is basically reams
and reams of recorded quotes, and this is one representative quote, "We have lots of
parents who come in, they know what they want, they don't care what we have to say, they
want the antibiotic that they want because they know what's wrong with their child."
And so whether this is really true or not, this is, this is a perception of a
lot of practitioners so we have to pay attention to this. And, and again I
mentioned other folks have done this, this is a picture of Rita Mangione Smith, who is
a health services researcher at University of Washington in Seattle, and
really a pioneer in this field, then and now, is still doing terrific work. And she,
almost 20 years ago now, interviewed ten doctors and 306 parents, and these
doctor-patient diads after they left the room, and had had an encounter or
antibiotic prescription was given. And here the physician perception of parent
expectations for antibiotics, antibiotics, here was the only
predictor of prescribing. So when I, when I, when a doc thought that the
parent wanted the antibiotic they, they prescribed an antibiotic sixty-two
percent of the time versus only seven percent of the time when they didn't
think they wanted antibiotic. So that's powerful, but what, what do the parents
actually think? So when she then went on to talk to these
families, they were direct and actually, they had videos, there were direct parent requests
for antibiotics one percent of the time. And parent expectations for antibiotics
were not associated with the physician perceived expectations. So most of the
time, when Doc's thought that the parent wanted an antibiotic, it wasn't true,
So there's this, there's this disconnect. And parents who expected antibiotics but
didn't receive them were more satisfied if they had a contingency plan. So this idea
of, okay, your child has a viral illness, and this
is what you should expect, and this is what we can do if it gets worse, so that
really was the key. They want an explanation for what's going on, and a
plan --not necessarily an antibiotic. And failure to meet parent expectations
around this communication piece was a key driver of parent satisfaction and
you know satisfaction isn't everything but it's important, right? You want to
have a productive patient encounter so you can develop trust and rapport and
and have things, have things work well. We did we --like I said 20
years ago we did this more recently where we interviewed parents in waiting
rooms, it was part of the same part of the same study that I had showed you
before. But generally speaking, parents did not plan to demand an antibiotic for
their child. These are sick kids with fevers in waiting rooms. They deferred
to the medical expertise about their need for antibiotic therapy, and they seem to
be aware of the downsides of antibiotics, and seem to be generally willing to
partner with, with the prescribers to actually improve appropriate use. So this
is really important to know that this is the case. So more with from Rita Mangione-Smith,
she did later then, and again almost 20 years after the first
work, she did surveys of providers in Seattle practices. Twenty providers, more than
1,200 visits, and show and to try to get at, okay, so there's this there's
some type of communication event that needs to happen for, for appropriate
satisfaction and understanding. And she found -- not to get too much into it --but
I'm collaborating on a study with her, that she's directing right now on how to
frame the encounter. So it has to do with positive treatment recommendations
versus negative treatment recommendations, and, and I won't do it
justice, but it's, it's generally speaking, instead of saying, "Your kid doesn't have
a bacterial infection, they don't need an antibiotic", you say, "Well, you know, they
have a viral infection, it's frustrating because we don't have
it we don't have drugs that can treat that, but here's what we can do, as I said
before, here's the plan, I you, you can expect them to have a fever for a couple
more days, be coughing maybe for 10 or 14 days, but progressively get better, and if
they don't, or if they have these signs or symptoms, you know maybe they start to
have a hard time breathing or they become lethargic, or then I think then
you call me, and we'll see them again." Those are the type of encounters that
really work, and make it so parents are not demanding antibiotics, and Rita's
doing a terrific job of characterizing that. So I want to finish up talking
about these sort of some more non- clinical drivers of antibiotic
prescribing, so what I've been talking about, this parent interaction these
perceptions of wanting an antibiotic is not a clinical driver of, of getting
an antibiotic, right? Just because you think somebody wants one, that has
nothing to do with the patient's clinical condition, so this perceived
parent pressure is something we have to pay attention to. But there are other,
there are other, other things that we have to pay attention to that show you
how powerful these non-clinical drivers are. So the presence of trainees, there are some
really, there are a couple really cool studies done a while back where they
look at doctors on the inpatient setting who are prescribing antibiotics, and they
are much more likely to prescribe an antibiotic when they're by themself
than when they're with a medical student, right? Doesn't, I mean, it's not shocking,
but at the same time, that's crazy, right? That shouldn't be the case, so, so they're
influenced by context. Speaking of that time of day, Jeff Linder who I mentioned
before, who was at Harvard and now is Chief of a General Internal Medicine at
Northwestern, did a really great study, it's a brief report, and I think it's in
JAMA Internal Medicine, or The Research Report where they looked at antibiotic
prescribing in general pediatric practices across the time --over the day
and prescribing is lowest at 8:00 a.m. when people first come in, it goes up too
when you get toward lunch, and then after lunch it drops back down to the 8 a.m.
levels, and they go by 5 o'clock it gets higher, and they've adjusted for the type
of the, patient type and the presentations, as we had adjusted before,
so these are not different patients or more demanding patients coming in, I mean
that's crazy, right, when you think about it, but it's human nature, and it just
reminds you that we are human beings prescribing antibiotics, so you have to
understand human behavior and levers that affect human behavior. Patient race --
so we published a study about six or seven years ago in Pediatrics that
looked at antibiotic prescribing in our, in the same network I talked about, and
black patients receive 25% fewer antibiotic prescriptions and 25% fewer
broad-spectrum antibiotic prescriptions than white patients, and this is not, you
know, there have been a lot of disparities that have been shown in the
literature and it can become complex, right, when you have a big team or you're
talking about different underserved areas that might have different
insurance type, but this isn't this is adjusted for age, for sex, for type of
insurance, and these are, in these are these are, this, this analysis was done
within clinicians so it's looking at the same doctor prescribing to patients of
different race, so it's not like we're, we're looking at one clinic in one
community. So, these, this doesn't make any sense
right? I mean these, these are, these, these, these are levers that we have to be
thinking about. And practice location --the main predictor of whether you get a
macrolide for pneumonia in a pediatric patient in our network is suburban
location. That has nothing to do with the clinical presentation, right? So these are,
these are, you know I just, I, I know I'm kind of beating a dead horse here, and
I want to drive these home, but we are, you know these, these non-clinical
drivers of prescribing are really important we can't just keep thinking
about, okay, make sure you look and see pus behind the eardrum it's more than
that that we have to educate people on. So I just wanna mention two studies,
I'll close with those, that have leveraged this behavior change in behavioral
economics, to think about how to prescribe --change prescribing, and I've
called them novel socio-behavioral strategies. So this one's great another,
and I think it was another short report, this one in, this one also in JAMA
internal medicine, led by Daniela Meeker, Jason Doctor. Jeff Linder also has his
fingerprints on this one as well. But Daniel Meeker and Jason Doctor are
behavioral economists that work at University of Southern California. And what they
did here is an intervention that they said
took advantage of clinician's desire to be consistent with their public
commitments. So they called this, I don't know if anyone's ever read the book
nudge? That book? It's a good book, you should read it. It just talks about its,
sort of it's based in behavioral economics and and the idea here is how
do you kind of nudge people to do the right thing and here they called it a
low-cost behavioral nudge in the form of a public commitment device. So they had a
letter, it's basically a proclamation to prescribe antibiotics judiciously, that
they had, they randomized primary care doctors to looking at this letter,
signing this letter, and then posting the letter in their, in their patient rooms.
And it's, it's --there was a lot of text, and I just pulled out a couple of
excerpts of it, but it's sort of like this --antibiotics like penicillin fight
infections, but they can be cause side effects. If you have a virus, you know, you
won't get an antibiotic, your health is important to us, we promise to treat your
illness the best way possible. So sort of appealing to that kind of Hippocratic
oath thing and they signed it, and they just stuck them, that was the whole
randomized trial, you sign thing and they put the poster on the wall, so when that,
when the doc was going into the room they they kind of were
reminded of that. And just doing that they showed a 20% reduction in
inappropriate antibiotic prescriptions for viral infections. So bronchitis was
the main driver, but also things like influenza. Pretty powerful stuff. I
mean this is, this is, a 20% reduction is a huge number, and compared to some
things that've been done in the past, is much larger than really expensive public
health outreach and education and all types of things. So pretty powerful stuff!
And the last study I'm going to show you is is work by the same group where they
did, but with a much larger, more research resource intensive intervention,
and this was done in more than 50 practices more than 250 doctors across
two practice settings. One around Boston around the Harvard Pilgrim group where
Jeff Linder was, and one around USC in California where Daniela Meeker and
Jason doctor were. And they, they enrolled practices that all had electronic health
records, it didn't have to be the same platform it just had to be electronic.
And they, they instituted a sort of cluster randomized approach, three
different interventions. So the first one was suggested alternatives. So if you
coded for a condition that we know is, is caused by a virus, so influenza, acute
bronchitis, largely driven by that, and you were gonna, and you try to click for an
antibiotic prescription, the first one you get will pop up and it said, antibiotics
are generally not indicated for this, here's a list of suggested alternatives,
things like cough and cold medications, or humidifiers, whatever it might be. The
second one was accountable justification. So again, viral condition, you prescribed
an antibiotic, and what would pop up is a text box that said, you know, you have to
justify why you're prescribing this, and if you didn't justify why you're
prescribing it, "no justification given," would populate the chart. I mean Jeff
Leonard shared some funny stories about, you know, people would just write like,
"because I want to," and yeah, but most of the time, you know, they would write, you
know, well, it's I think you know it's a sinusitis or something like that. So, and
then the last one which they, they hypothesized would be the most powerful
was peer comparison. And this one did not happen in real time. This was a monthly
report, somewhat similar to the study that we had done, that would show that,
would send you a letter, both I think they did it both by email and regular
old US mail, and they essentially dichotomized. And if you were in the top
decile, so the top 10% of prescribers in terms of how judicious your
prescribing was, you got a letter that said "You are a top performer."
Everybody else got a letter that said, "You are not a top performer," and so these
jokes about, like, how do you, how do you think all the Harvard doctors felt about
getting a letter that said they're not a top performer? And, and lo and behold,
when you look across, and they did them all in isolation, and in
combinations of the three, and in a very complex analysis, which was really
terrific, and you can see, accountable justification, peer comparison, and
suggest alternatives, all the orange bars, if you can see them, are the intervention
groups. All showed a reduction in antibiotic use. Peer comparison
was the most most significant, and most profound. And this was actually really
impressive because the prescribing rates they started with were really low, and
as you can see, and I don't know if you noticed from the study that I showed you in our
network, things were already going down before the intervention,
so once the practitioners knew they were in a study and knew they were being
watched, things were already going down. So, so the bar to move this further was
pretty high, yet they still moved it further.
So to summarize, antibiotic prescribing in the ambulatory setting is common, and can
be harmful both to the patient, this individual harm, as well as society,
this public health risk that we all know about. Broad-spectrum agents are probably
not better than narrow spectrum agents for most conditions, and they cause more
harm. Audit with feedback can be an effective strategy to improve
prescribing --showed you in a couple of different studies where this peer
comparison works. And then other social behavioral approaches such as improving
communication, holding clinicians accountable, and, and applying some of
these behavioral economic nudging principles can be effective. And so I'll
stop there and I'm happy to take any questions. Or I can take them after, I
think we have a panel after. We do. Maybe one question to start it off. I think we
have maybe two --four minutes, maybe for questions? So thinking about the, the
concept of optimizing, not withholding antibiotics, and say the patient's story
situation where perhaps an antibiotic was appropriate, but choosing the right
antibiotic, topical versus systemic, you know, I think some of the feedback we
have heard just through some of the surveys that we've done in general
interaction with health care providers is, there's a role for, say prescribing
guidance, but people don't necessarily
want to be told how to prescribe. So do you think there's a value of talking
about optimization when we talk about stewardship so that people don't think
we're saying you do it or you don't. Yeah, that's a great point, and
so I, yeah, and so this is something that I try to proselytize on the inpatient
side, but also on the outpatient side. It's not about, this is not about
restriction or rationing of antibiotics. And I'm so glad you brought that up, it's about
optimization, right? So it's about doing the right thing, and it's really about
quality improvement. So if you think about, if you think about it from that
lens --I want to do what's best for the patient --it's the way to go because it's
a sell and a win for everybody, and and, and with antibiotic stewardship, nine
times out of ten, doing the right thing for the patient is actually either
not giving an antibiotic, or giving a narrower spectrum agent, or a shorter
duration, or de-escalating. So it just works out that way.
You don't, we're not in the position, I mean you could argue from a public
health standpoint that we're almost in a position where we need to ration
antibiotics, but you don't need to. But we have enough evidence and enough data to
just say what's the right thing to do? And if you do the right thing for the
patient, it almost always works out that way. So I think that framing is so
important, and it can really change it, can be game-changing when you work with --
I mean, like I said, I direct our inpatient stewardship program --and we
work with folks in the neonatal ICU and then in the surgical wards. My, all my
clinical work is on the, in transplant ID. So it's, it's bone marrow transplants
or solid organ transplants. And we work with those --to say let's figure out
what the best regimen is based on the evidence, and let's study it, and follow
it. And it's almost always shorter, when we have, and I think building on even
further, we have interventions where we actually say, well, Oh a wait a second you need
to get antibiotics faster and broader spectrum for, you know, for example of
patients with sepsis. And so--and that helps right? Because it gets you some
street cred where you're saying, listen we ARE really optimizing, and we're
trying to figure out --so start with that principle of what's the best thing to do
for the patient, and like I said, nine times out of ten it'll be fewer days of
antibiotics. Yeah, so we spend a lot of time talking about trying to fix things
upstream, and we spend a lot of time also like dealing with the repercussions of
patient satisfaction as far as reducing outcome or making outcomes worse
especially with opioid prescribing, and so do we think that there's
a role for getting rid of Press Ganey and getting rid of patient satisfaction
scores as it relates to improving antimicrobial stewardship, especially
when you look at, well, parent expectations drive prescribing habits. It's
a great question it's, I don't know enough about that space really, but I'll
make some comments, right. So, so for one, when you look at the Press Ganey scores
in in our practices in our network, it completely disassociates from
appropriate prescribing. So some of the best patient satisfaction are practices
that over-prescribed antibiotics. I would, so, so, so that makes you think
that it's not, at least with respect to antibiotics, it's not helpful and it shouldn't be
there. That said, I think some of the work that I talked about that Rita had done,
shows that there's a way to make patients satisfied, if you explain to
them what you're doing and why you're doing it. So I think those two things
aren't mutually exclusive, and you can, if you approach it in the right way, with
this open-minded attitude, to say that they're not, parents aren't, your patients
aren't just coming demanding antibiotics what they demand is, they want to know what's
wrong with them or what's wrong with their child. And so I think you can have
your cake and eat it too, in a way, and sit down with patients. And I, at least
personally, I totally have had some satisfaction here, what seems like good
patient satisfaction. You sit down you say, okay here's what we need to do, and I
wish we had this but we don't, and if you spend the time and you tell them what
their diagnosis is, and you're transparent about it and you talk about
and quantify the harms of antibiotics, I think you can get satisfaction there. So
you know maybe it's that we need to tweak how those things are rolled out
and implemented and what the questions are, but they probably can both exist. We
had one more question here. I know what that we have an ambulatory focus today,
but you, you had mentioned sepsis, and this is something that I'm struggling
with as a stewardship program manager with our sepsis group. I know that
there's a CHOP protocol. We're still waiting in particular for pediatrics in
2019 for the new sepsis guidelines. How do you align stewardship with sepsis--
meeting core measures -- is the first dose a big deal? You know,
I'm getting a lot of feedback, you know, but, and it's like, how do we treat sepsis
and at the same time how much c-diff can we bring to the world? Yeah, it's a it's a
great question. So I'll sort of address it in two ways. The first one is again, it's
this is about optimizing patient care, and in some situations, you got to get
broad-spectrum agents to patients fast. The problem with that --and that's, that's a
truth. The problem with sepsis --this is my
perception of the problem --with a lot of sepsis measures that are coming out is
that the pendulum is swinging way too far. So all sepsis is not really sepsis,
right? So that the data that, so the data that these measures are based on are
generally retrospective studies of patients with severe sepsis or septic
shock. When you, when you rule a patient out in the hospital, patient has a fever
and you get an antibiotic -- you give, you, you pull a blood culture hopefully --and you
give them antibiotics, nine out of ten of those patients don't have septic shock,
right? So the problem is focusing on the patients that are truly sick or at risk
of really really bad outcomes. I'm gonna tell you right now, we're working on it at our
Hospital. Everybody with a fever and a central line is getting, you know, they're
timing their antibiotics and getting them stat. Those patients, some of them
don't need any antibiotics, you know? Maybe a blood culture, but we, so I think
what we need to do is focus on the patients where, who are really gonna have
a bad outcome, and then get them
antibiotics fast, and then once the appropriate cultures are taken,
deescalate very quickly. If you do that, this is a small proportion of patients, I
don't think you're gonna have a problem. And we can get rid of lots of extra
vancomycin, and Pip-Tazo and cefepime use that we don't really need. So I think
there's actually an opportunity there. It's, it's a little bit of a tough hill
to climb, but there's an opportunity in these sepsis measures to go in and --so
we're shining a spotlight really broadly on patient --shine the spotlight on the
ones who need it --and then actually there's an opportunity to stop giving
broad-spectrum antibiotics to patients who really don't need them. So we could, you
could potentially, either potentially be a win there.
I don't know if I'm saying it clear enough, but I, but I think that we just
have to be careful, you know, there was this measure, I don't know, ten years ago
in adult medicine where all of a sudden everybody who had pneumonia coming to
the ED had to get antibiotics within six hours. It was a complete disaster,
right? Because all of us, everyone just would come in, and they'd just slam
antibiotics in the patients before they actually made a diagnosis.
And so I can see something analogous happening with sepsis, and I
think we're not, you know, we're reacting to data that we're over
extrapolating, and then anecdotes of situations that have been horrifying and
bad, like the incident that happened in New York, New York, but the measures we're
doing in the hospital are not addressing that type of sepsis recognition. We just
have to be careful how we implement those.
For more infomation >> 100 Ideas For Bamboo In The Garden | DIY Garden - Duration: 11:05. 
For more infomation >> Dunkin', Shell team up for Fuel Rewards Program - Duration: 0:54. 
For more infomation >> Beautiful Weather For The Holiday Week - Duration: 3:24. 
For more infomation >> California City Vice Mayor Facing Backlash For Column Calling Gay Men 'Faries' - Duration: 1:03. 
For more infomation >> Getting ready for the beach this summer - Duration: 3:50. 
For more infomation >> "Live Like Jesus" Part 1 - Time for Hope with Dr. Freda Crews - Duration: 28:31. 
For more infomation >> Pet Questions: What training rewards to use for pets? - Duration: 4:41. 
For more infomation >> Learn Colors With Oddbods Toys Paint Candy Cartoon Video For Children - Duration: 4:18. 
Không có nhận xét nào:
Đăng nhận xét