What I Learned at the ATP Reunion Conference Feb 2012

Hi all,

happy valentine’s day!  I hope you’re all doing well.  It was fun to
meet up with some of our class for the reunion conference.  While this
conference was not the life-altering event our class was, I still
found it useful and thought I’d share some of the things I learned at
the conference.  Probably the most important part of the conference
for me was getting refocused and energized.  I also found that I came
up with many ideas for quality improvement hearing about what others had done and
thinking about how I could apply this to my environment.  The content
of the conference for me was also more about seeing where things are
heading and having more of a sense of control of what’s happening than
just feeling like a little boat bobbing along the water waves never
quite seeing the large ships generating the waves.  Please share your
thoughts and perspectives, as things may look different from where
you’re sitting.  Below I submit to you my 13 ruminations on what I
learned at the conference.  I hope to see many of you next year.

What I learned at the ATP reunion conference Feb 2012

1. Capitation – it’s coming.  It’s not completely here yet, so
implement QI rationally so that you don’t go bankrupt decreasing
hospital admissions and procedures, but it’s almost here.   Insurance
companies may well come to an end, and QI will be the way to maintain
healthcare quality while controlling costs.

On the whole the current system “fee for service” the risk is
overtreatment.  In capitation the risk is undertreatment.  The party
line is that QI will protect against this.  I don’t know if that’s
completely true.  On the other hand, we’re currently not just a system
of overtreatment, we’re also a system of undertreatment for the
millions without health insurance or those who cannot afford their
medications.  I don’t know what rationing is if it’s not denying
healthcare to a huge proportion of the population.

Given our current system of overtreatment, the physicians still say
“trust me, I only do this in your interest.”  While there may be
unethical physicians who do things for profit, most of us probably try
to do things that are necessary.  What we perceive to be necessary is
likely shaped by how we were trained, and that in turn was likely
shaped by the system. There’s also “cognitive dissonance” where
instead of changing our behavior, we change our attitudes/perceptions
to match with our behavior.  Undoubtedly there will be undertreatment
of some during capitation. But in the big picture, it seems to me, that
erring on the side of undertreatment is the more reasonable way to go
for several reasons. 1. “First do no harm,” 2. we only help <10% of
patients, the vast majority would do well or poorly regardless of what
we do. Intensive proceduralizing increases the chances of
iatrogenesis, and 3. medicine is so polluted by interests (drug
companies marketing marginally effective drugs, researchers falsifying
data (google Duke oncology for the most recent debacle), etc.) that
many of our treatments are discredited within a few years of
implementation anyway.

2. We are to blame.  We can complain all day long about the ridiculous
patient expectations (can’t walk out of the ER without having at least
some body part scanned) but we are the ones that generated those
expectations.  It’s up to us to adjust them.  Anyone other than
doctors adjusting the expectations will be perceived as rationing and
withholding of care.

3. Primary care and public health will be important.

Once you collect the premiums and are on the hook for the care of the
population, all of a sudden battling childhood obesity in the schools
or homes makes more sense than all the QI metrics about checking
hemoglobin A1c and nutrition counseling for the obese 12-year-old just
diagnosed with type 2 diabetes.  In a world where the government can’t
make the nutritionally recommended “plate” mostly vegetables, without
also declaring pizza to be a vegetable, we have a lot of work to do in
that direction.

4. Salaries will equalize.

No one actually said this, but I’m not the only one who thinks this is
unavoidable and not altogether a bad thing.  When the orthopedic
surgeon says there’re currently 9 docs where they probably just need
3, there’s no way to sell this well.  The current model of “shared
savings” where you can say to radiologists that you’ll pay them the
same not to do a million scans is short term.  You’re not planning on
paying them a huge salary to do nothing forever.  It’s OK if the
dermopathologist doesn’t make a million while the PCP is drowning.
It’s OK if primary care can attract and retain the best and the
brightest once again.

Don’t worry too much about #4, because

5.  Money is not the only or even main motivator.

See Daniel Pink “Drive” (also short video on ted.com).  Pay people
enough so they don’t care about the money, and they will do good work.
Autonomy, desire to do well, etc. are much more important.  People
will still want to do radiology, dermatology, etc. even if the
salaries drop.  (So, btw, will CEO’s of private companies, but that’s
another story).

6. We all need to do a better job explaining QI to all.  Even within
the ATP reunion conference, while the clinicians talked about
improving care, the finance guy said “the orthopedic surgeons wanted
to make margin on their medicare patients.”  Everyone recognizes that
change is coming and although a few people may think they’ll retire
before this affects them, no one can seriously argue to just keep
going the way we’re going.  The challenge here is to inspire people to
see the potential of improving quality and not just depress them that
we’re going bankrupt and no one has money for anything.

7. No change is too small.
I keep having to think through this one to remind myself.  The most
inspiring thing about the conference was that everyone who got up to
speak didn’t whine about the ineffectiveness of the government to make
rational policies or talk about what the government should do but
can’t because it’s paralyzed.  They all got up and said what they want
to do, what they had done, what they suggest that we do that depends
on us.

Perhaps the best way of curing childhood obesity is prevention and
changing a whole toxic lifestyle and culture and limiting junk food
availability, and inactivity.  But there is still room to improve the
care of the obese 12-year-old in your personal clinic and adhere to
the current guidelines.

8. Shared baselines/eProtocols will be very important.

We’ve sort of bought into this already with the ATP course, but I keep
thinking about it, not the least because my research mentor at IMC is
Alan Morris.  Shared baselines/eProtocols are much more effective in
delivering consistent, high quality care and when delivered in an open
loop manner seem to give us the best of both worlds.  The risk, which
we ought to make explicit is that we lose skills.  For example, in
ventilator management where ARDSnet protocols (even if they’re not
computerized) are widely used, there is a “golden generation” where
the experts understand physiology and the rationale for the
recommendations.  In  medicine, however, if you don’t use it you lose
it.  So, future generations understand far fewer details because they
don’t have to manage them.  We might be terrible at physiology, but
we’re just terrific at looking at CTs.  Some of this is unavoidable
and is as old as medicine itself.  The old-time physicians who could
spin their own urine, plate their own cultures, the surgeons who read
their own path, even the pulmonologists who knew how to disassemble
and reassemble a ventilator.  These are skills we no longer have
because the field is too broad and we have all sorts of sub-sub
specialists.  This skill evolution is a key distinction that doesn’t
get articulated in all the lectures we hear about automation and
analogies to factory work.  The airline pilots have the back-up
autopilot and they are expected to be able to take over from the
autopilot.  (So are we – we’re supposed to know when to deviate from
the protocol, which we do on average ~ 5-10% of the time according to
the IHC eProtocols).  But airline pilots have extensive simulation
training.  That is, they recognize that they will lose the skill if
they don’t use it and spend extensive time practicing the skill.  We
do not.  We do not pretend to manage ventilators on simulated
patients.  We have too many other things on our plate.  The protocols
are a black box and we examine them as validated based on clinical
outcomes.  The details of the logic don’t even matter and are well
beyond the understanding of most physicians.  If this black box
protocol has been shown to improve outcomes in multiple different
settings, then that’s more evidence than we have for most of the
individual recommendations.  Who cares what the evidence base for a
particular recommendation is.  This black box eProtocol has a pre-test
probability of being 90-95% correct (remember we validate until
clinicians agree 90-95% of the time).  As we lose skills, we become
ever less confident in knowing when it’s appropriate to deviate from
the protocol.  Perhaps we extend the protocol to other patients.

Even with all this “risk” of using eProtocols, even if we assume 100%
protocol adherence (i.e. physicians adhere, inappropriately, 5-10% of
the time), the risk of using the eProtocol is still lower than the
benefit of using eProtocol. But somehow we are ever more suspicious.
Is it that faced with a fallible human being who makes a wrong
decision we feel that we’re cared for, whereas faced with a cold
recommendation we do not?

Do we trust these recommendations too much? When we do a physical exam
we appreciate the subjectivity of hearing a murmur and exactly where
it’s loudest and what it might signify.  If we read a transthoracic
echocardiography report (TTE) we assume a certain kind of certainty
regarding the valvular and heart function that may, in fact, not be
there.  And if the uncertainty is made explicit, we complain that the
radiologist or cardiologist or whoever interprets the test is hedging
and somehow isn’t “brave” enough to go out on a limb.

9. Patient choice and shared decision-making.

This concept is quite important, but will likely be focused on a few
key areas where there is controversy.  The decision aids are $500K to
produce and $100K to update.  Interestingly patients shown decision
aids re: PSA screening were 14% more likely to decline it.  That’s
interesting, because with NO NEW DATA the USPSTF recommendation was
changed AGAINST PSA screening.  This means that if patients were truly
well-informed, the vast majority ought to have declined screening, but
because there are all sorts of additional pressures and interests to
accommodate we have to appear balanced.  A journalist interviewed on
frontline was talking about this problem of having to appear
“balanced.”  The example she gave was that sometimes you see stories
akin to “some people say 2+2 is 5, but some mathematicians disagree.”
As if both sides of the argument are equidistant from the truth, which
is to be found in some compromise.

10. Patients don’t know anything about their procedures and why they
get them, but they think they’re very well-informed.

This, based on surveys of patients who’d had cardiac catheterizations
and thought the procedures would make them live longer (they don’t),
didn’t know basic complication risks about the procedure, yet
uniformly said they were well-informed, shows once again, it’s all
about the doctor-patient relationship and all about trust.  Even in
this age of internet where you can read 100 differing recommendations,
or pubmed 100 different 10 person non-randomized trials, we still need
someone to sit down with us and tell us what part of the literature
they think applies to us, how this 10 person study fits in with the
overall field and why this or that action is reasonable.  We want to
share in the decision, we want to understand the rationale, the risks,
the benefits, what to expect, but we still want someone to help us
sort through the overwhelming amount of data.  That’s why the
healthcare provider is still the most important source of information
for patients, rated way above the internet, family/friends and media.

11. Guidelines should be prioritized. (David Eddy)
There should be some global outcome score that we’re all trying to
maximize.  Increasing depression screening (which we can’t effectively
treat ) is not the same as increasing aspirin use for CAD [coronary artery disease].  (After the
presenter went on about the increased use of depression screening in a
clinic, a very poignant question from the audience was “how many
people did you help?” Point being, probably not many.)

David Eddy
makes the point that all these guidelines are issued with the same
volume and we’re supposed to follow them equally.  But really, some
are more important than others and rather than having to hit a certain
percentage for each guideline, could we optimize our efforts going for
a “global outcome” score.  To that effect, rather than complaining
about the terrible quality measures currently implemented, we ought to
be dictating what those measures should be.  The power of Archimedes
and IndiGO (available below cost for nonprofits) is that you can more
easily see where to target your efforts.  (Both as a system but also
as a patient – is it more important to lose weight or to afford the

12.  There’s lots of low hanging fruit.  Start there.
I took this lesson from our classmate – Ashley.  Talking to him about
what he’s been able to accomplish in the past year in his hospital was
quite inspiring.  What was refreshing about Ashley’s accomplishments
was that he hadn’t spent too much time developing data systems, which
is out of reach for many of us.  Because we are collectively so far
from where we could be, many of the improvements can be seen by large
scale measurements (length of stay, readmission rate, patient
perception of quality, ED throughput time).  We talked about the
complexity of limiting the number of CT angiograms in the ED, then he
noted that he focused on standardizing the protocol of abdominal CTs
and improved the ED throughput time, patients were happier, and
everyone knew what to do because there was one protocol, not several.
So maybe picking PE evaluation and CT angiograms in the ED is not the
best choice, but there is no shortage of things that need to be done.

13.  Do not let the perfect be the enemy of the good.
(Sort of said by Voltaire as: “Le mieux est l’ennemi du bien. The
best is the enemy of the good.   La Bégueule (1772))

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: