offene Ablage: nothing to hide

To talk of a "revolution" in health care would be demeaning to the
thousands of people staking their lives on real revolutions right now
in various countries, but there is no doubt that the conflation of
out-of-control health care costs, fancy new technologies, and various
government mandates (not only from the US government, but from many
states including Massachusetts) have forced doctors, vendors, and
other people in the heath care field to scramble and order changes
throughout their organizations. A couple hundred of these people came
to the "Tools
for Meaningful and Accountable Care" conference held yesterday by
the Massachusetts Health Data
Consortium.

I didn't interview many participants (the ones I talked to were very
happy with the presentations) but I wonder whether all of them got
what they came for. They may well be haggling over questions such as
"How many prescriptions do we need to order online in 2011 in order to
qualify for the first stage of Meaningful Use booty?" or "How do I get
an image from the radiologist down the street while satisfying HIPAA
privacy regulations?" What they got, however, was a broad look at the
needs of health care and a set of projections by various speakers that
congealed into what I find to be a coherent vision for health care in
the future.

And I think the communication of this vision is important. Costs will
continue to rise and reform will fail if doctors, vendors, and IT
staffs simply race to meet each stage of regulations and act in an ad
hoc manner without an overall coordination of effort. Just how broad
this coordination of effort must be--we're not talking here just about
gathering an entire hospital around a program, or even a whole
consortium such as Partners HealthCare, the biggest Massachusetts
provider--will come out during this article.

Capitation versus Clinical Effectiveness Research

One of the big changes in the Massachusetts health care scene went
oddly unmentioned during the whole day of talks. I'm referring to the
href="http://www.boston.com/business/healthcare/articles/2011/01/23/blue_cross_ceo_says_providers_must_control_health_care_costs_or_else/">dictat
from Blue Cross/Blue Shield of Massachusetts announcing that they
will change from fee-for-service to a "global payment plan." This
mirrors recent plans from the state government to pressure the health
care insurers and providers to pay for outcome rather than for
individual procedures. But imposed on the current delivery system,
such a "global payment plan" is just a repackaging of old-fashioned
capitation.

Nobody seems to want to admit this, just as people are reluctant to
announce the return of "managed care" and prefer to assign the new,
as-yet untainted term "accountable care organization." It was up to
the CEO of a leading ACO--Dr. Craig Samitt of Dean Healthcare in
Wisconsin--to display a slide in his keynote with an equal sign
connecting "managed care" and "accountable care organization." He said
this moment is our chance to do managed care right.

(It's also sobering that in Massachusetts, world center for health
care, the two organizations singled out at this conference for their
achievements in bringing to life the potential in health care IT both
lay outside the state: Wisconsin's Dean Healthcare and central Texas's
Scott & White Health Plan. Furthermore, the individuals who
traveled here to describe their work had both spent long careers in
Massachusetts health care organizations before traveling elsewhere to
lead these advances.)

Payments for outcome and ACOs can work: they can lower costs while
simultaneously improving health care. But by themselves they're like a
meringue fashioned out of only sugar and cornstarch. The egg that will
make them work is clinical effectiveness research, a topic excellently
covered in talks by two doctors, Harold C. Sox and Michael Fischer.

CER is a medical application of the quality control routinely done in
other industries; it perhaps has its origin in time-and-motion
studies. It asks tough questions such as why one surgeon has far
greater success on the same patient population as another--not in
order to reward or punish, but to convey the best practices from one
clinic and region to another. CER should overcome the enormous
disparities that we all know exist between doctors, between hospitals,
between patient populations (such as differences in outcome by race)
and between different parts of the country.

Dr. Sox pointed out that CER was being tried as early as the 1960s,
but took a great leap in the mid-1990s and continues to make advances
despite such cynical political pot-shots as raising the fear of death
panels. (I highly recommend href="http://www.newyorker.com/reporting/2010/08/02/100802fa_fact_gawande">Atul
Gawande's New Yorker article for a sweeping overview of the real
purpose and effect of end-of-life decisions.) CER is now formalized by
the Federal Government in several initiatives that are not likely to
go away.

Dr. Fischer said that CER required big changes in education and in
how results are delivered. Crude impressions like "death panels" have
to be fought with better outreach to the public. Continuing medical
education (which has impolitely referred to as "broken") needs to be
more hands-on and to demonstrate that doctors understand the material
they've been given. And EMRs have to become much more sophisticated at
delivering information.

Currently, doctors using EMRs are pelted with notorious "alerts" that
resemble the warnings web browsers give all of us when we visit web
sites with outdated security certificates. Most doctors treat the
alerts like web users treat the security dialog boxes--they ignore
them and click through. And that's because they're just too darned
many alerts. Every medication has some potential impact on something
(a rare food allergy, for instance) and the computer systems can't
figure out what doctors really need to know. Furthermore, if a system
displays an alert and the patient experiences a problem later, the
doctor's liability is increased. If a doctor dismisses an alert, he or
she had better type in a reason for doing so.

Making CER work will require vendors to design more flexible systems,
and the IT staff at each institution to choose the alerts that can
actually affect medical decisions. Some of the enforcement can also be
spread around: nurses and other staff can be involved in CER.

All Together Now

The value that comes from aggregating results of procedures and
treatments raises several questions. One is the effects on patient
privacy, because it's well-known that anonymized data can often be
de-anonymized, and we're talking here of widely shared data being
crunched by dozens or hundreds of organizations. (I'm on the planning
committee for an href="http://www.utexas.edu/lbj/healthprivacy">upcoming conference on
health data privacy.) But a deeper question concerns the ability
of many forces to work together to make change.

A small medical practice can't internally collect enough data to
determine statistically what works and what doesn't. Unless someone
coordinates these small practices, they will fall behind and lose both
money and patients. But even a large institution has limited access to
data. Michael Lee, a director at the fairly large Atrius Health
group, said they wished they could see data on their patients after
they move on to other institutions. Better care and lower costs will
require massive coordination across the country.

The Direct
Project at Health and Human Services, which reached a major
milestone last week with the announcement of some successful data
transfers, should become a platform for such exchange and coordination
(and they're taking privacy quite seriously). But it's just a
platform--echoing a point made by Joel Vengco of GE--whose value
emerges only through the proper encoding, exchange, and interpretation
of data, followed ultimately by the dissemination of results and their
use by providers. (Whew.)

This is perhaps why Micky Tripathi, president of the Massachusetts
eHealth Collective, stressed that doctors have to move from just
entering data into their EHRs to entering accurate data, and using
structures that allow it to be parsed and crunched. It was also
pointed out that many of the requirements for receiving meaningful use
payments depend on multiple institutions: specialists, labs,
pharmacies, and other institutions the doctor deals with have to be
set up to send and receive the communications for which the government
rewards the individual doctor.

It used to be that doctors would practice and health care researchers
would do research (with some overlap, of course, at teaching hospitals
and major research centers). Practice and research are now
intertwining and perhaps merging.

All these ways in which health reform becomes a group effort show why
a high-level vision is important. And someone at the top must firmly
hold on to this vision. That's why I had a second surprise yesterday
at a news item that went unmentioned: the upcoming href="http://www.healthcareitnews.com/news/blumenthal-leave-onc">departure
of David Blumenthal, National Coordinator for Health Information
Technology. I trust that the ONC is suffused with enough smart
people holding firm to their goals for it to continue to make change.
If Blumenthal's departure slows down implementation, though, maybe it
will give us a welcome breathing space to re-evaluate our tools and
what we need them to accomplish.

Too expensive, too hard to use, and too disruptive to workflow

That was the three-sided accusation delivered to vendors of EHRs by
Dr. Marylou Buyse, the chief medical director of Scott & White
Health Plan, who spoke at yesterday's conference and won an
achievement award there. Nobody blinked when she delivered the
judgment, and indeed it's one I've heard repeatedly. Dr. Buyse should
have added that their proprietary formats and imprecise
implementations of standards throw up enormous barriers to the data
exchange required for meaningful use, as I discussed in a href="http://radar.oreilly.com/2010/03/report-from-himms-health-it-co-1.html">report
from last year's HIMSS conference.

Few speakers picked up this theme, perhaps because many vendors were
present, and perhaps because the speakers figured we had to soldier on
with whatever we had. My third surprise of the day, regarding
unmentioned news, was the alarming href="http://www.whitehouse.gov/sites/default/files/microsites/ostp/pcast-health-it-report.pdf">
report by the President’s Council of Advisors on Science and
Technology expressing concern about the ability of current EHRs to
carry out the basic data exchanges required for improvements in health
care.

Maybe health care in the US is so far behind the digital age that any
upgrade would produce benefits. Paul Grabscheid of InterSystems
reminded the audience of a recent study showing that two-thirds of
doctors still use fax machines to send medical records out, and the
next biggest medium in use is snail mail. Adoption of EHRs is rising
in this country (it may be up to 20%, depending on how it's counted)
but is still extremely inadequate. Nevertheless, most observers don't
call for moving full-speed ahead with existing computer systems and
workflows. Before making investments, it's important to be smart.

Better standards, as the PCAST report called for, are important, and open source systems would address interoperability. (I feel it justified to insert a plug here for the health care track at O'Reilly's Open Source convention. But most of all, we've all got to work together. Spent forces with nothing new to offer must be pushed out of the way, while the vast majority of people have to set aside maneuvers for short-term advantage and focus on a common goal. To return to the stirring words of keynoter Dr. Samitt, insurers and providers (and he could have added patients and politicians) have to "work together to drive change."

Maybe that's the key to any revolution.