[This Transcript is Unedited]
National Committee on Vital and Health Statistics
National Health Information Infrastructure (NHII) Workgroup
November 12, 2004
Hubert H. Humphrey Building
200 Independence Avenue, S.W.
Washington , D.C. 20201
CASET Associates, Ltd.
10201 Lee Highway, suite 160
Fairfax , Virginia 22030
TABLE OF CONTENTS
- Welcome and Introductions – John Lumpkin, M.D.
- Panel 1 – Overview
- Panel 2 – Organizations and Vendors Offering PHRs that are not Derivatives of EHR Systems
- Panel 3 – Organizations and Vendors Offering PHRs that are Derived from EHRs
- Discussion of Next Steps
P R O C E E D I N G S [9:05 a.m.]
Agenda Item: Welcome and Introductions – Dr. Lumpkin
DR. LUMPKIN: Good morning, my name is John Lumpkin, I am chair of the
Workgroup on the National Health Information Infrastructure and senior vice
president with the Robert Wood Johnston Foundation. We’re trying a little bit
of different technology, for those of you who are listening on the web you’re
not, because we don’t have that quite hooked up, but you wouldn’t know that
because you’re not able to listen yet but it will be going live soon. We do
have people on the phone, I think we have Jeff Blair, we have Richard Harding
and Susan Kanaan, and is that it? And there may be others on the phone.
Just for those of you who may be signing in since we have members of the
committee who are signed in on the same bridge they may have access to speak
which will not be granted to everyone who’s on the phone, we may have time as
the agenda, depending upon the agenda and the discussion for those who are in
the audience here in the room or on the phone to comment and at that point I
will identify and you’ll be notified by the operator how to make that
So why don’t we start off with introductions and we’ll start with John
MR. HOUSTON: Good morning, I’m John Houston, I’m with the University of
Pittsburgh Medical Center, I’m a member of the committee as well as a member of
MR. KAMBIC: I’m Bob Kambic, CMS, and staff to the workgroup.
MS. FYFFE: Kathleen Fyffe, I’m with the U.S. Department of Health and Human
Services working in Dr. David Brailer’s office and I’m also staff to the NHII
DR. LUMPKIN: And what’s the name of that office?
MS. FYFFE: The Office of the National Coordinator for Health Information
Technology also known as ONCHIT.
DR. LUMPKIN: ONCHIT, I like that.
DR. LANSKY: David Lansky from the Markle Foundation.
DR. ROLLOW: Bill Rollow, I’m the director for the quality improvement group
MS. WILLIAMSON: Michelle Williamson, National Center for Health Statistics,
CDC, and we are live on the web with our broadcast.
DR. STEINDEL: Steve Steindel, Centers for Disease Control and Prevention,
staff to the workgroup and liaison to the full committee.
DR. HUFF: Stan Huff with Intermountain Health Care and the University of
Utah in Salt Lake City, member of the committee and subcommittee, or workgroup.
MR. HUNGATE: Bob Hungate, principal of Physician Patient Partnerships for
Health, member of the committee and the workgroup.
DR. DEERING: Mary Jo Deering of the National Cancer Institute and lead
staff to this workgroup.
MS. SQUIRE: My name is Marietta Squire, I’m with CDC, NCHS, and I’m staff
to the workgroup.
MR. HARRINGTON: My name is David Harrington, I’m with MedicAlert
Foundation, and I’m one of the panelists today.
DR. MARTINO: My name is Janet Martino and I’m a physician consultant to
MR. MORGAN: John Morgan, small company called Health Right.
MS. CRONIN: Hi, my name is Carol Cronin and I’m a consultant, I’m here
today working with the AARP.
MR. MARKS(?): Richard Marks with Patient Command, Inc.
MR. KANOUSE(?): Bill Kanouse with Patient Command also.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics, CDC,
staff to the committee.
MR. DEVAULT: I’m Peter DeVault from Epic Systems Corporation in Madison,
DR. WALKER: Jim Walker, chief medical information officer at Geisinger
MR. MCFARLAND: Scott McFarland, chief executive of SimplyWell, also
affiliated with the Nebraska Medical Center.
MR. FALTUNE(?): Robert Faltune, a private citizen not associated with any
MS. KAYE(?): Robin Kaye, also a private citizen not affiliated with any
MR. BURNS: Jay Burns, I’m with the McKesson Corporation.
MS. ROLLINGS: Kristina Rollings, I’m also with McKesson Corporation.
MR. RODY: Dan Rody, American Health Information Management Association.
MS. BLOM(?): Laura Blom from the Joint Commission.
MR. LUDY(?): Ernie Ludy with the Ludy Family Foundation.
DR. LUMPKIN: Great. Can we have the three who are on the telephone?
DR. HARDING: Richard Harding, member of the committee and the workgroup
from the University of South Carolina.
MS. BAUER: This is Cynthia Bauer, I’m also on the phone from Office of
Disease Prevention Health Promotion, U.S. Department of Health and Human
DR. LUMPKIN: Jeff, are you there? Okay, Susan Kanaan are you there?
MS. KANAAN: I am, can you hear me? This is Susan Kanaan, I’m working with
ODPHP and NCVHS.
DR. LUMPKIN: Great. Thank you all for coming, we’re going to spend today in
doing two things, one we have three panels to really talk about sort of where
we are in relationship to personal health records. Its now been almost three
years since we issued our report on information for health and began to push
the agenda for the National Health Information Infrastructure. The early days
after that report was released was a little bit obscure in that it was very
hard to get anyone to pay attention to that, actually now it’s almost four
years, 2001 we did it, right after, so three years.
And at that point about a year later there was a conference that was
sponsored by the Institute of Medicine on rapid advances in health care and it
was a study that was funded by the Department of Health and Human Services. At
that conference one of the four areas that was presented and discussed was the
issue of use of health information technology as being a rapid advance and
transformational change. Secretary Thompson who sponsored that meeting and
participated and all left that meeting with a real, a lot of energy which has
been reflected in many of the changes as we saw it not just more then three or
four weeks ago that energy infused itself all the way up the White House and in
fact the issue of health information technology was raised in the Presidential
debate for the first time, maybe the only time, but it certainly raised the
One of our concerns as a workgroup as we’ve monitored the progression of
activities is to assure that while we see some progress, certainly a
significant amount of progress with electronic health records, that there’s a
lot of discussion, there’s a lot of debate, we have some concerns about
connectivity because these individual electronic health records could be
islands that we created and we don’t just want to achieve individual electronic
health records but we have our vision of interconnectivity across the country,
that the enhanced savings that are projected will only be achieved that way.
The other area that has seen some advance although a little bit slower is
in the public health arena where through funding from the Centers for Disease
Control and others developments of the national electronic disease surveillance
system, NEDSS, as well as the public health information network, PHIN, there’s
again some work in advances. Yet we as a committee have always believed that
when as one of my instructors once taught me when I was studying emergency
medicine, well all else fails do what’s right for the patient. And so part of
these hearings are really to begin to look at again how progress is proceeding
in relationship to the personal health record and the personal health
This is not the last hearing we will hold, in fact we’re planning on
holding an additional set of hearings in January. The goal of this hearing is
to get a look at how things are progressing and then the last part of this
meeting will be the committee meeting, as an open meeting, and so all of you
here are welcome to stay, to begin to discuss how we’re going to frame the
hearings in January and where it is that we want to go with this particular
So for those of you whoa re speakers, who have agreed to come and
participate, we thank you and we’ll be hearing from you throughout the day and
we’re going to now to the first panel. And our first panel have already
introduces themselves, Kathleen Fyffe with ONCHIT, and Bill Rollow from Centers
for Medicare and Medicaid Services, CMS, formerly known as HCVA, and then David
Lansky who is with the Markle Foundation, formerly with the Foundation for
Accountability. So I believe Kathleen will be first.
Agenda Item: Panel 1 – Overview – Ms. Fyffe
MS. FYFFE: Good morning, thank you very much for inviting me to speak this
morning in front of the NHII Workgroup. My remarks this morning will cover four
areas, I’ll talk about President Bush’s vision regarding health information
technology, mention a few points from the strategic framework, those points
regarding consumers, give a brief description of ONCHIT activities related to
personal health records, and lastly I’m going to offer some suggestions for how
the NCVHS NHII Workgroup can help ONCHIT as we look at personal health records.
As you are aware President Bush unveiled his technology agenda for health
care on April 26th during remarks at the American Association of
Community Colleges Annual Convention. The President said that medicine ought to
be using modern technologies to better share information, reduce medical
errors, and reduce costs to our health care system by billions of dollars. He
said that in order to protect patients and improve care and reduce cost we need
a system where everyone has their own personal electronic medical record that
they control and they can give a doctor when they need to. The President set a
goal that within ten years every American must have a personal electronic
medical record and he further noted that the federal government should take the
lead in order to make this happen.
The HHS framework for strategic action that was issued in July of this year
contained a key goal, a goal to personalized care. What does this mean? It
means the ability to assemble and use information that is complete with respect
to a specific person, in other words that is consumer centric. This information
could include consumer specific health findings, health status monitoring
tools, or customized prevention and self care information.
When we think about universe of health information that can be accessed by
most consumers today it is remarkable and frankly quite overwhelming. This
information for consumers needs to be relevant and customized to their needs.
One of the frameworks’ strategies for achieving the goal of personalized care
is to encourage the use of personal health records, or PHRs. PHRs are a rapidly
emerging trend as evidenced by the many organizations who offer PHRs. Generally
the PHR maintains individual personal health information from a variety of
health records, guidelines, and other useful tools for the consumer.
The federal government has and will continue to use information technology
as a central tool and vehicle to disseminate health information and knowledge
to consumers. For example the Center for Disease Control’s website is a rich
resource for consumer health information ranging from birth defects and
diseases such as influenza or the West Nile virus to chronic diseases and
conditions. Another example is the CMS Medicare beneficiary portal which will
provide secure information via the internet. Initially the portal will provide
access to fee for service claims information. The portal will be demonstrated
as a pilot in the state of Indiana over the next few months.
Recently ONCHIT coordinated activities between the CDC and national private
health plans so that the private plans were able to put links on their websites
to the CDC flu website. This mean the critical information about the flu is
more easily and efficiently disseminated to tens of millions of U.S. consumers.
We’ve been working with CMS, specifically Dr. Rollow’s office, on PHR
initiatives and Dr. Rollow will tell you more about this later this morning.
In addition our office has been coordinating activities related to personal
health records that are being pursued by the Markle Foundation’s Connecting for
Health Project and by the American Medical Informatics Association. Lastly we
are examining how PHR usage can be encouraged through consumer enrollment in
health savings accounts.
My last remarks are now going to focus on how NCVHS can help. First let me
say that we’re greatly appreciative of the NCVHS NHII Workgroup’s leadership
and hard work and are especially happy about the current focus on personal
health records. We all have much to learn about personal health records that
exist in today’s marketplace. In addition to the PHR hearings that you’ve
arranged it is my hope that the workgroup will channel its energies toward
questions such as the following. What is the relationship between PHRs and
electronic health records? What is the business case for personal health
records? How can consumers be encouraged to use PHRs? In addition to consumers
what are the roles that health care industry stakeholders should play in PHRs?
Stakeholders such as providers, employers, health plans, PBMs, and researchers?
How can providers be encouraged to use, to urge their patients to use PHRs?
How might employers encourage their employees to use personal health records?
And lastly as you will inevitably raise questions and issues regarding personal
health records and privacy of individual health information how will you
coordinate your activities with the NCVHS Subcommittee on Privacy and
Confidentiality in order to get their examination and input?
DR. LUMPKIN: Thank you very much. Bill?
Agenda Item: Panel 1 – Overview – Dr. Rollow
DR. ROLLOW: Well hello to everyone, this is a nice panel, ONCHIT, CMS and
Markle with Connecting for Health and some degree both I think. Nice in the
sense that I think each of the three entities really has, had a real interest
in this topic and we’ve all been doing some things independently of each other
but also as I think Kathleen’s remarks indicated have been collaborating quite
a bit, so you should see a fair amount about that in terms of the things that
each of us say.
I’m going to talk not very long using these slides from the perspective of
CMS and particularly from the perspective of personal health records and the
relationship between them and improving health care quality. I am the director
for the quality improvement group at CMS that has responsibility for in
particular the work of what used to be the PROs, the peer review organizations
or PROs, now known as Medicare Quality Improvement Organizations, also the ESRD
Networks which have a similar focus in the area of end stage renal disease and
dialysis. And also from the perspective of the person who leads CMS’s
Information Technology Workgroup, CMS has had a real interest increasingly over
the last couple of years in quality as a substantial part of its agenda and an
increasingly significant part of that agenda has really recognized that
information technology adoption and effective use is an important part of what
We’ve been talking increasingly over the last year about needing to change
our quality improvement efforts, particularly those through the QIO
organization program, to move from work that has been largely incremental,
meaning helping make changes to the system that were sort of at the margins of
existing systems and processes, to work that is more transformational, work
that really assumes that in order for health care quality to improve
substantially there will need to be fundamental changes in the systems that
providers use and in some of the processes that relate to those systems for how
they take care of patients.
So we’ve been talking increasingly about promoting transformational change,
the QIO program itself has proposed an eighth scope of work which we’ll start
next summer, that really is about promoting transformational change, QIOs
promoting transformational change among providers. And personal health records
are an aspect of information technology that I think is quite consistent with
So in the eighth scope of work we’re looking at providing assistance to
physician offices, hospitals, nursing homes, home health agencies, to promote
transformation change, and I put this slide up mostly to underscore what we
think that might look at as really having four dimensions or four core
strategies. One is that providers and practitioners to achieve transformation
need to be able to measure performance and report on performance.
A second is to adopt health information technology and use such systems
A third is to undertake significant changes in process, process redesign,
things like care management, patient self management on the ambulatory side,
human factors design principles particularly for systems on the inpatient side
or in larger facilities, those are examples of fundamental process redesign.
And finally organizational culture change.
We think that these four strategies really are key to achieving
transformation change, one of them is certainly health information technology
adoption and use and on this slide what elements of HIT adoption and use have
we been focusing mostly on, it really have included these five things,
electronic prescribing, electronic laboratory results managements and test
ordering, care management or registry type systems, electronic health records
more generally, and then finally the contribution of information to personal
health records or the ability to develop information that’s useful for
information exchange uses. So those five elements are really core elements in
the work that CMS has been doing in promoting transformation using health
Now how about personal health records, what kind of work have we been doing
in this area? One thing which Kathleen briefly mentioned is this portal that
CMS has been developing. The portal really is a significant change in the way
that CMS interacts with beneficiaries and will give beneficiaries to look at
information that CMS has online in a secure way, the information largely coming
out of claims administrative data and also data that is available relating to
the person’s enrollment status within the program.
And that’s a good starting point and as you may be aware that information
should be available shortly in a portal pilot that is expected to start in
Indiana at this point actually in December, it’s bumped back a slight amount as
these things tend to do. But that’s a good starting point but it’s really not a
very substantial move in the direction of personal health records, the data is
limited and it’s something that we expect to build substantially off of. We’re
thinking that in building off of that what we want really is a set of clinical
information and other related services and tools that beneficiaries can use,
our Medicare beneficiaries or more broadly that patients can use, and our
thinking has been that in order to try to stimulate the availability of high
quality personal health records perhaps one of the things that CMS might
usefully do is to promote and perhaps in some way convene a process in which
there would be some description of the kinds of functionalities that a good
personal health record might have.
Now we understand fully that this is, that PHRs are in a developmental
phase, that there is a lot of experimentation going on about what could be in a
PHR, how PHRs might operate, that we’re really at the infancy in knowing about
things that consumers might value, how they would best use them, that sort of
thing. And we’re by no means looking to in any way constrain any of the
innovation that’s going on. We think actually that we could potentially help
stimulate it and particularly stimulate it to develop in directions that would
potentially have a significant impact on quality if we were to come up with or
put in motion a process that comes up with some kind of a descriptive set of
functionalities that PHRs might have.
If we did that what would CMS potentially do with it? Well, we would use it
in ways that might give preference to PHRs that are using, that have developed
such functionalities. For example the portal that I just mentioned potentially
would like to link to PHRs that do more then our portal today or in the near
future will do and so one thing that we might do is to say we will link our
portal to PHRs that meet a set of specified functionalities.
Another thing would be the quality improvement organizations are out
working with physician offices, physician offices contributing to personal
health records is certainly one of the desirable developments in this area and
as QIOs are working with physician offices on adopting and effectively using
health information technology, contributing data to PHRs might be another
place, PHRs that meet these functionality specifications might be another place
where we could add some momentum.
And then thirdly what about the chronic care improvement organizations
which were created in the Medicare Modernization Act? Couldn’t they, aren’t
they like in fact to use PHRs as one of their ways of improving chronic illness
care, could these specifications also in some way be used in the CCIP program?
And finally what about financial incentive arrangements, demonstrations for
example that CMS does, could we also use these requirements in some way to
create specifications for how physicians would be paid in demonstrations or
contributing clinical information to the personal health records. So those are
some of our thoughts about things that we might do that would use a set of
specifications and perhaps give some momentum to the development of PHRs
meeting such specifications.
Now what might the specifications look like? Well, clearly this remains to
be determined and yet having said that what are some of the things that CMS is
most interested in and in thinking about this we’ve certainly been very much
assisted with the conversations that we’ve had with a number of the PHR
developers, so we’ve tried to sort of stay in touch with how the field is
developing. Also with the work that David Lansky, Connecting for Health, and
Markle have done on some of the research in personal health records.
So with that in mind what kinds of things are we thinking that we’re
interested in? Well, first of all, the existence of a kind of core set of
clinical information that is exchangeable and that patients would have access
to and patients would have control of. Perhaps or using some kind of a
presumably specified format, potentially something perhaps resembling or in
fact fitting with what the continuity of care record specifications have
recommended. So that it seems to us at this point is kind of a cornerstone of
what we would want to see in a personal health record.
Secondly and not too far down the list is tools that patients can use in
self management, so particularly most Medicare patients have chronic
conditions, particularly tools that help patients with chronic conditions that
take care of themselves, better manage their own disease.
Thirdly tools that might help with informed decision making, certainly the
Medicare population increasingly is making decisions about health care in
conjunction with providers. Tools that help facilitate the interaction so that
better informed decisions can be made.
Communications that enable physicians, providers, patients, etc., to
interact with each other and arrange services. And additionally information
that helps beneficiaries know better what coverage they have for services, how
best to use that coverage, either in the traditional Medicare fee for service
type program or in the managed care programs available to beneficiaries. And
then of course whatever happens with these functionalities such systems ought
to meet privacy and security specifications.
Now this is my final slide and it says on it in big capital letters with
one exclamation point, I could have put a couple of exclamation points there,
this is sort of the primary reason that I wanted to come to the meeting is
aside from giving this background is to get your thoughts in the discussion
period about first of all whether you think creating such a set of
specifications is a good idea, is this something that would help advance the
field, could it be done in a way that is consistent with promoting innovation,
that sort of thing. But then secondly if this is a good idea how might we do it
and by we here I really mean that CMS is not interested in itself creating a
set of specifications, we’re really interested in seeing a set of
specifications come into being and what we’d like to think with you about is
how could that best occur.
We’ve been talking with a number of entities about this for some time
actually and I’ve indicated that as the first bullet point here we’ve been
talking with David and with Markle about potentially how we might work with
Markle and Connecting for Health on doing such a thing. And have received a
proposal from Markle that indicates that potentially they might be interested
in helping us convene a process that includes input from a number of different
entities in which two things would be looked at, one is what is the business
case, what can help stimulate EHR adoption, how can CMS’s thoughts that were on
the previous slide, how can those thoughts be better fleshed out and made
consistent with thoughts that others might have so that we get a better picture
as to how PHRs might develop and things that CMS might do to help stimulate
And then secondly could we come up with, again using a process of input
from a number of different entities, could we come up a set of, sort of version
one set of technical specifications, functionalities, etc., that PHRs might
have as a kind of core set, a sort of floor set, that would then say we think
that good personal health records would have these things in them, we may even
describe things that may go beyond that that would not be part of a core set of
requirements but that would help describe areas in which PHRs might offer
additional functionalities, could we come up with a set through a process like
that perhaps sometime next year.
And if we did with Markle helping us put that together then what might
happen? Well, potentially a few things might happen, one would be developers or
offerers of PHRs might then indicate through some self certification process or
self reporting process whether they meet these functionalities. Initially that
might be the best way in a sense to get these expectations out there and to get
some sense as to how the market is developing and how PHRs themselves fit with
these set of specifications. So that could be a next step.
Beyond that then we might take more formal steps, it might be useful to
have a standards development organization as we’ve done, many of us, through
HL7 with EHRs put together a more formal set of specifications. And then it
might be useful for there to be a more formal certification process, perhaps
through the certification commission that’s currently beginning its work on
electronic health record systems. We certainly talked with CCHIT about this to
assess their interest, I think they do have interest although I think they also
have an earlier agenda that is more specifically around electronic health
So these are some thoughts about how such a thing might develop, again I
want to emphasize these are proposed for discussion purposes, no commitments
have been made certainly on CMS’s part or on the department’s part about this,
but I’m hopeful that in the discussion period maybe we can talk a little bit
about your thoughts about whether such a set of specifications would be a good
thing, and then secondly whether or not this kind of approach laid out here
might be a good way to go about it, or thoughts that you might have about how
to create an alternative approach for ways that you might be interested in
assisting in developing a process like this.
Thanks a lot.
DR. LUMPKIN: David Lansky.
Agenda Item: Panel 1 – Overview – Dr. Lansky
DR. LANSKY: Good morning, thank you Dr. Lumpkin and committee for giving me
an opportunity to be with you today. I’ve had the chance to talk with you a
couple of times in the last year or two about the work of Connecting for Health
and particularly our work on personal health records, so I thought rather then
spend too much time today reviewing the work we have been doing I would try to
focus on a few of the themes that have emerged from our work and that may speak
to the challenges you have in communicating to the department and the Secretary
what are the pathways moving forward over the next couple of years, much as
Kathleen and Bill have laid.
Just for those of you who aren’t familiar with Connecting for Health it is
a consortium of about 100 or more then 100 organizations and their
representatives representing many of the private sector organizations in terms
of provider systems, vendors, professional societies, consumer organizations,
as well as many of the public sector agencies with a finger hold in the
electronic health arena. Connecting for Health has been active for several
years and over the last two, two and a half years, there have been a series of
committees working on personal health records. We decided not to develop a
definition of a personal health record but instead to describe them, and to
talk with some of the leading thinkers in this arena about what are the
attributes of a desirable personal health record, how far along is the field in
reality in implementing those attributes, and learning what we could about the
public’s appetite for using the features and functions that a personal health
record could offer.
And I have described some of that work to you in the past, we brought today
copies of the last report we issued in July on the personal health record from
our viewpoint and I certainly hope committee members and guests take a copy of
that, I think they’re over there at the table, and that gives you quite a bit
of detail about our past work.
What I want to do for a couple of minutes now is talk about what we see
coming forward that may be of relevance to your work. The first one I think is
self evident by today’s audience and the panel of presentations you have lined
up, there’s an enormous amount of interest in this field right now since the
President’s announcement and Dr. Brailer’s report that we’re pleased to see
this flourishing level of activity. At the same time as we have looked around
at the field we have seen some significant uncertainties, that is it’s not
entirely clear to us what a personal health record is or will be, what it does,
or who it stirs, and how well, and what I want to do today is lay out for you a
few of the things we see as uncertain and
needing additional development over the next few years.
And in some sense the implication of those uncertainties is that we have a
great deal to learn, we collectively in this room and in the country, and I
hope we can do that learning together, hopefully we can do it in an open minded
sharing way so that we will together reduce these uncertainties over time. And
the pathways that both Kathleen and Bill have described I think are ways of
narrowing the uncertainty will continuing to innovate.
And so my final comments will be about what I would call interim policy
development, that is what the opportunity for the federal government and others
interested in broad strategy to do two things at once, one is to continue to
encourage innovation by the developers, the provider systems, the users, and to
continue to encourage responsiveness in the development of these tools, that is
responsiveness to patient needs. And at the same time while we’re supporting
innovation can we also constrain that innovation just enough so we have
interoperable systems emerging from the flowering of good ideas that’s going on
right now. So let me talk about that in a little more detail.
I’ll just pick off some of the things we see as not yet clear to us at
least as we look around the system that need more work. We use this term
tethered meaning, and I know you have two panels today are actually structured
along this borderline, is the personal health record connected to the
electronic health record and other components of the delivery of care, or is
the personal health record a free standing independent piece of software which
the patient can populate with data and control and use as they wish. We’re not
clear yet which of those paths or some other path may emerge for the design of
these systems and their architecture in the larger network.
Secondly we have felt there’s an attribute that’s very important to the
public which is that they control their own health information, who gets to see
it, who gets to use it, how it’s shared. But in reality many of the early
developments in personal health records are under the physical control of the
health enterprise or of some other entity that is sponsoring the health record.
So the reality of patient control is not yet clear to us.
We have felt that in theory, I think Kathleen suggested this, the personal
health record should be comprehensive in its content. But what we observe in
reality is the uptake is greatest for specialized populations who have well
defined needs and for their particular health situation or in their family’s
health situation there’s a set of functions which are very high value which
motivate them to use these products and tools in which in a sense create a
tension with the idea of comprehensiveness, that the value is perceived to be
highly specialized. So whether we will have broad platforms or very specialized
fragmented platforms isn’t yet clear.
We’ve been thinking a little bit about the tension between what I call a
functional approach to the personal health record or a structure approach to
the personal health record, by which I mean do we think of a personal health
record as an array of functions and services and benefits to users, which
therefore might be fragmented out as I just said into specialized products, or
is a personal health record literally a record, is it a structure, a data
structure which contains a body of information upon which can be built a number
of applications that are either specialized or generic. And I think just in our
language we still use the term personal health record which implies a data
record structure but the reality from the point of view of users is what’s
interesting is the set of functions that sits on top of it and we need to
reconcile those two paradigms if you wish.
Next, I think Bill in particular highlighted the importance that they see
in the ultimate transformation of the way the health system is structured and
performs, I think we all hope that the personal health record is a tool toward
a rethinking of the roles of the different players, including the patient and
family, in the delivery of health care. But what is the design considerations
which lend themselves to fundamentally transforming the delivery of health care
services around the patient’s needs and capabilities.
And we certainly understood there are vendors out there who take almost the
opposite approach, that to be highly accepted and convenient the design of
these tools should be almost seamless with the current delivery of care and in
a sense not challenge or not permit fundamental redesign. On the other hand
those products and approaches which imply fundamental redesign of the care
system are going to be difficult for provider organizations and individual
providers to adopt and interoperate with.
Physician reaction to these products turns out to be an important
consideration. So far in our field work we’ve seen that PHRs are most often
accepted by patients and used by patients when their doctor encourages them to
do so or when their doctor seems to be supportive. Originally if doctors say
that’s a waste of time patients are not likely to spend the effort to register
and use the available services. Doctors have expressed some concern to us in a
variety of formats about liability of the personal health record, that is
creation of additional liability for themselves, information that they should
be taking cognizance of, information they should be acting upon, by virtue of
the patients having this new channel of communication and clinical management
with the provider. So we need to anticipate the need to support physician’s
comfort level with this approach if we’re in a tethered environment and we have
to deal potentially with the perception or reality of additional liability.
These products as we have said we hope they will be interoperable, not only
with the health system but with other products and services that are
increasingly available to patients and families in their homes, for example
home glucose monitoring and other telemetry devices, which generate a lot of
data, that data could be acquired by the personal health record, that data
could in turn be shared selectively or in bulk with their physicians,
physicians then have a question of what do I do with all this data that becomes
supported across this new network, what’s my responsibility to that data, when
do I interpret it, how do I interpret it, how do I manage it.
Funding is always a perennial question here, we don’t know what the
business model is for the personal health record. At present and I think I’ve
said to you all in the past for the most part provider organizations are
sponsoring these consumer portals and sometimes a personal health record
extension of the portal as a loyalty and marketing strategy within the health
system. There’s not yet been a strong case made for an independent market, a
consumer marketplace for personal health records although that may emerge. In
the absence of a direct consumer pays marketplace what is the model for
proliferation of these tools.
And I want to talk just more specifically given your charge about the role
of the government in addressing some of these uncertainties. There are several
things, government agencies like those on this panel and others, who had
addressed, and this is a list of possibilities not a set of recommendations. Do
we need a definition of a personal health record? As you’ve heard from my last
slide I think that may be premature but is it up to the government to say this
is what this is and this is what it isn’t, or in the case of the CMS challenge
this is a sufficiently robust definition that we can interact with this
Is it the role of the government to set data standards for what’s inside a
personal health record or is that entirely a marketplace activity? To the
extent that the PHR does interoperate with electronic health records and
transfer data back and forth between them is it appropriate for the government
to define data standards for interoperability of personal health records? Is it
appropriate for government agencies to provide data to patients, like the kind
for example the CMS and other agencies might have, through the vehicle of the
private sector PHR product market? So in this scenario the intended beneficiary
of the data is of course the patient, they have an arrangement with a private
sector PHR, which then in turn with the patient’s permission would acquire data
from the government, is that an appropriate role for government to play?
Should the government establish rules and policies for private sector PHR
products? An example in this category, and I think Bill alluded to this, if a
PHR product wishes to communicate reminders to patients about preventative
interventions that are due for those patients based on guidelines and some data
regarding their past history of getting those services is available from
claims, should the government say if you’re going to use our claims data to
provide a reminder to those patients here are the rules we impose upon the way
you use that data?
Should the government itself offer PHR products and host patient clinical
data? Certainly some that directly provide service like the VA are already in
that arena, those that don’t directly provide clinical services have a lot of
data and maybe should be in that arena, so the question of whether the
government will offer the product and in turn host clinical data in government
repositories is a policy question.
And then similarly as an add on feature to that if it does host clinical
data or has other good clinical markers should the government provide clinical
reminders for example directly to patients regarding their health status and
So as we see it some of the things that are facing all of us in the next
couple years with a policy slant, as I said at the beginning I hope all of us
will continue to support innovation by private sector and public sector
developers but innovation which was responsive to trust patient and community
needs. I hope that this innovative period will lay the groundwork for
interoperability and not be purely a thousand flowers blossoming but hopefully
in a well manicured garden so that we end up with something that is pleasing to
I hope this next phase will lay the groundwork for consolidation of data
across time and space, this speaks to the data standards issue and the
messaging issue, but ultimately patients who move across the country, who
change providers, who transition in their health status over time, want to
bring that data together. We need to have a lot of different data entities
collaborating if that’s going to be possible.
In the short term we need to find out what are the quick payoffs, where are
the first audiences who will receive benefits, whether they are patient
audiences, care giver audiences, professional audiences, who are the early
winners of this enthusiasm for personal health records and how do we build upon
that. At the same time that we follow-up on Bill’s suggestion that all of this
is not only to get short term wins for high need populations but it’s also to
set a foundation upon which we can transform the way health care is delivered.
In terms of Connecting for Health’s activities what we see doing over the
next year or two, we have been developing attributes and descriptions of what a
personal health record is, we intend to continue to refine the work we have
been doing and continue to collaborate with those in the field to do that. We
will try to identify what are the most elemental core data standards and other
standards that are compatible with an interoperable national network, so as all
the folks in the room and elsewhere begin to develop personal health records we
would at least like to try to work with them to articulate what is the
fundamental basis of interoperability in this new arena so we don’t have things
going off in a direction which can’t be brought back together for the patient’s
As all of us have said on this panel there needs to be continuing
evaluation of what are the roles of the purchasers, the public sector, patients
and families, consumer advocacy groups, and so on, so that there’s, I think
we’re all new to this, we don’t quite know what roles we should each play. One
thing we can do at Connecting for Health because we have done some work at
convening a broad variety of parties is talk with each other about the roles we
expect of each other and how we can most be helpful in moving this agenda
We will do some very practical things as well, we have started a group of
user meetings to help the earlier implementers talk to each other, respecting
their competitive and proprietary boundaries, but also seeing where there are
places where they can learn from each other and advance the public interest in
this field. So we’d like them to share their findings with each other where
they’re comfortable doing that, try to adopt some of these common elements I’ve
talked about in their own product so that ultimately data will be interoperable
across those products and recognizing that they do have marketplace goals that
we don’t want to compromise.
A couple of people have mentioned already this morning, my fellow
panelists, that the public is an important audience and player in this process,
we’ve been approached by many organizations that would like to do public
messaging and communications about both electronic health records and personal
health records. And it’s our feeling that that’s great and there are probably
some core ideas that should be spoken to with a uniform voice and uniform
vocabulary, so that while each vendor and sponsor and host might have their own
approach to this hopefully even the terminology is something we can begin to
work in concord about to get a common message out to the public, including the
benefits of IT and personal health records for those patients, what rights they
have to control their own health information, and how all of this can be
helpful to their own family well being, to their health ultimately.
We’d like to continue to work to demonstrate the value in several classes,
the health improvement benefit, the clinical services benefit, the service
quality and efficiency improvements, and hopefully the economic benefits of
implementing these PHR functions for patients.
And finally expect to continue to do work disseminating we’ve learned and
what others are learning so that the health care IT and policy communities can
continue to learn from what’s going on in the field.
Thank you very much.
DR. LUMPKIN: Well, thank you, I think that was a very good start.
Questions? Steve. Welcome, Eduardo.
DR. STEINDEL: I’m not sure if I have questions or comments, I think the
first comment I’d like to make is this was a fascinating presentation. I know
David well and we were commenting earlier are you going to say anything new,
and I can assure you you did. And Bill you presented some very interesting new
thoughts, too, I thank both of you for introducing some new topics into this
area for discussion.
One of my first comments is in regard to something that you said, Bill, and
this had to do with what I got an impression of where you talked about asking
Connecting for Health to continue with what they were doing with the PHR which
I absolutely commend and I think should go forward, but this is an observation
I have from the electronic prescribing work that’s been recently done. What
happened with that was it started with the Connecting for Health report, that
Connecting for Health report actually formed the basis of and fed the agenda
for the NCVHS hearings which were called for in the Medicare Modernization Act.
And the NCVHS report was basically a validation of a lot of what the Connecting
for Health report said.
Then we heard at our Standards and Security Subcommittee meeting last week
that ONCHIT asked for a report on decision support standards, which Jonathan
Teich is working on, and we got a summary of that and it said it relied heavily
on the NCVHS recommendations. And now we heard yesterday from Stanly
Knockensen(?), who was briefing NUCC about the regulation for e-prescribing,
e-prescribing pilots that come out shortly, that draws very heavily on the
I think you’re getting the sense of what I’m saying is that all these
groups are feeding each other and it really was very productive and it happened
very quickly and I think if it works in that environment it should continue.
But what I’m posing is that we ought to look at the various groups that are
feeding each other as we plan these going forward and pay attention to repetity
because that is also what is needed. So that’s my comment in that area.
And then my question to you, Bill, is when you talked about tools for
informed decision making I think while I’m not going to ask you to answer this
but I think we need to pose the question of how far, how deep and how variable
those tools should be, they could be presenting a lot of interesting comments
there. Which gets back to my overall question to the three people, this sounds
like a Jeff introduction a little bit, but one sense that I got from the
presentation is when we’re looking at patient health records how should we be
thinking about these? This was mentioned in all three presentations, should we
look at something that involves a continuous personal health record that
contains some types of information from birth to death, or a life stage
personal health record that’s more keyed to various incidences as people change
and their needs for information change.
And David you used a very interesting word I think for the first time that
I heard this when you talked about consolidating information which implies some
type of organization and which keys into something about the continuous versus
life stage idea that’s also floating around and I’d like some comments on that.
DR. ROLLOW: I’ll say something brief about the earlier part and then David
or Kathleen can pick up on the later part.
There is a lot of circularity about things that happen and oftentimes
things pop up without really, at least from my point of view, it being clear
that people understand exactly what others are doing and how to best coordinate
some of that work. So I do think we want to do as much of that as we can. HL7
is also got some plans in the area of personal health records and I want to
understand better some of that and how that connects to some of the things that
we’re talking about here. So I think the more that we can stay connected the
better off we are and also the less that we can stimulate people to redo work
that others have done also the better off that we are.
DR. LANSKY: I think I don’t know the answer to the challenge you raised, I
hope actually that those who are developing products speak with you later in
the day this may be a theme that they can address as well since they have so
much direct experience. I tend to think of this, I picture a city with a
substructure of pipes and electrical conduits and so on that are moving
resources around and then buildings on top of that that are specialized in
their populations and functions. I think we need both, there needs to be a
substructure, an infrastructure which permits comprehensiveness across life
stages, across settings, across needs, but the experience of the person living
in the house or the building is a much smaller more confined experience, they
aren’t necessarily exposed to the full array of comprehensive flow of
information over time.
And our experience so far in talking with patients, both in small groups
and in surveys, has been that they tend to talk about, this is very early and
that may change of course, but right now they tend to talk about very specific
functions and features that are valuable to them because of a situation in
their life and all the rest of that body of lifelong comprehensive health
information is dimmed in the sideline. So whether from an architecture and
design point of view that has a bearing on how products get built and how the
infrastructure gets built I don’t know but I think it’s good to keep testing
substructure versus building instantiation of that.
MS. FYFFE: Steve, you’ve brought up a very thought provoking suggestion or
question as to whether it should be continuous or life stage. I think it needs
to be both, it could very well be that when you consider the advances in
genetics and other related research that information that’s been in the
background of your medical record for years suddenly comes to the forefront and
would have to be emphasized in a life stage type of application but I think
this is something that I don’t know the answer to other then to say that I
think it’s both.
DR. DEERING: I too wanted to thank and congratulate all the speakers, when
NHII Workgroup had a series of hearings on the personal health record at least
two years ago everybody felt well this was sort of an interesting topic but the
ball wasn’t really in play and I think that you have demonstrated that this
ball is in play.
I wanted to pick up on something that both Bill and David said and be quite
specific in an observation. On one hand it was David’s theme of we want to
1,000 flowers bloom but we want a cultivate garden in the end that at least
allows for interoperability. Bill you mentioned that eventually the CCHIT might
get to this but they have other word to do right now. What I wanted to suggest
is that all of those who are focusing so fiercely on EHRs right now and the
CCHIT in particular start right now to at least provide for the eventuality of
this so that in the need to be very fast and focused to get some specific
deliverables out the door they don’t inadvertently close doors or wall them up
totally, that they will need to open in a few years. So it’s a subtle thing,
it’s hard to give specific operational implications to that, but I think if the
guidance that’s given to them from policy people is that we know we’re going to
have to go this direction eventually so just make sure that the doors are
marked in your architectural drawings.
DR. ROLLOW: I think that’s a very good comment and at least from my point
of view it’s to some degree why we’ve been thinking a bit about it the way that
I suggested, which is that do we need something early that would be not
completely formalized and around which entities in the PHR arena would
essentially kind of self report, that gives some momentum in the direction for
everyone of something that then becomes more solidified when the time is more
right for it. But it does help people I think plan for what they’re going to
need to do as things become somewhat more solidified. That I understood was
really the essence of your comment and I think that’s right, I think that’s how
we should try to do it.
DR. DEERING: There is a model there in the HL7 EHR effort in that as a
matter of fact most of the functionality that would be needed for PHR exists
within the EHR DSTU, it’s just there to be sort of spotlighted and pulled out
with a few gaps filled, etc., but because of diligent work from the get go that
stuff is in there and we know that HL7 is taking great care not to say that
it’s trying, at least in it’s first few stages it went forward to say oh but
we’re not doing a PHR yet, but now in 2005 it begins to turn its attention
there it will find that it has an awful lot of the raw materials that it needed
and doesn’t have to go back and recreate them.
MR. HOUSTON: A couple of thoughts dovetailing on what Steve had said. My
bias I guess or my gut feeling is that these portals will ultimately be used
mostly by people that chronic illnesses, I would suspect that that’s where the
big win or at least the people that are most interested in them, that’s just a
gut feeling and I know there was some discussion about some of this and I was
wondering whether there had been any type of evaluation done, surveys, as to
what these portal, the current portals are focusing on as well as patient
adoption, what are patients really looking to in terms of using these portals,
what are their demographics, are they people that have chronic illnesses. And
then I guess beyond that if that is the case then I think that integration is
very important because obviously if that information doesn’t get back to the
physician I think it’s probably of limited use in terms of trying to manage
that chronic illness.
Which sort of leads me to my next question, has there been any type of
statistics towards physician adoption as well as concerns over liability of
managing or supporting these chronic disease portals and the like?
DR. LANSKY: In the report that I mentioned there is a chapter that
summarizes some of the survey research that others have done and we’ve done to
try to address the first part of your question about uptake. And I think
generally we would support your comment that people who are sick and use the
health system a lot are the most likely early users of these tools and actually
when we quantitated it it was people who see the doctor five times a year or
more are in the most likely to adopt category and naturally they are people
with chronic illness.
But the paradox is of course demographically people with chronic illness
are going to be concentrated in the older population and they in turn are less
likely to use the internet and computers and technology generally. So it turns
out there’s kind of a bell shaped curve and there’s a population of people
between 45 and 65 with chronic illness who are young enough to be comfortable
computer users and internet users and old enough to beginning to acquire
chronic illness and use multiple prescriptions and that sort of thing, that
seems to be from our data both expressing interest and actually adopting the
tools most rapidly.
Care givers of course is another important population but they are also in
that same demographic target area, they’re often people with either young
children or parents who are themselves in need of health services. So that
seems to be the pattern so far and that may change as comfort with the internet
and computers matures demographically and it may change as the products offer
different services that are more attractive to people across a larger spectrum.
So that’s the first thought.
On the question of physician adoption and support the place we’ve been able
to look at that at all has been primarily in provider sponsored consumer
portals rather then pure PHR strategies for example because they’re the ones
who are then marketing to the doctors and some systems have actually said we
will only offer the PHR to those patients whose doctors permit us to, so the
doctor has become a gatekeeper to the personal health record. And then in turn
the question is which doctors choose to allow their patients to use it and when
they do often the doctors are allowed to indicate which patients within their
own panel will be allowed to use the portal.
So there is some data from some systems we can try to help plug you into
and I think it’s a mixed bag, if I were going to ballpark it I would say a
third to a half of the doctors are fairly comfortable making this available in
the systems that have done it, which in turn tend to be fairly sophisticated
systems. So I don’t know if that would apply equally well to the broad
population of small practice providers.
DR. ORTIZ: This is a comment/question but I also, I don’t want to just
direct it at this panel but I want the people who are going to be speaking
later on to be thinking about this as well because I think it’s an important
issue. And also I don’t want to take this as an advocacy position but when you
were talking about there’s a lot of interest in PHRs, and there has been from
the Office of the National Health, Coordinator of Health Information
Technology, from Connecting for Health, etc., and I know that there is a lot of
interest at this point although at this point I’m still not convinced that this
is what we should be doing and I don’t think that that’s been proven yet. And
once again don’t take that as an advocacy, I’m not for or against them I’m just
trying to get to more answers to the questions. And let me talk a little bit
For many years before we had EHRs there was an interest in people having
their own medical information because we had this fragmented system. And I know
a lot of people who have personally thought wow, this is a great idea, they had
lots of energy and enthusiasm, they developed these personal health records
that were paper based, some of them were excellent, but when they tried to put
them out they failed miserably. Why? Because it was a very small percentage of
the population that was really interested in them, maybe ten percent or less,
people who were very compulsive and wanted to collect information. So all of
them basically failed across the board because most consumers, it really wasn’t
high on their priority list so that’s kind of one example
And people do collect certain information, they might have their lab tests
filed, they might have their appointments written down, so they do collect some
information but in terms of really collecting comprehensive information there
really hasn’t been a lot of demand or interest from the public so I think
that’s one thing.
Number two, even electronically, and this goes a little bit to John’s
question, I’ve been working with Kaiser on a really large project looking at
the use of eHealth and once again it depends on what you defines eHealth versus
PHRs, but I think it’s the largest study that’s been done, we have about
240,000 users of eHealth, that will be coming out in January and February in
JAMIA, we’ve got three or four other papers that will be coming out over the
next six months.
But when we looked at this even when, and we looked at it in people who
were members of Kaiser plus members who had access to the internet, computers
and everything, we defined it by people who had logged in, gotten a password,
so we knew that they did not have a barrier, that they didn’t have internet
access, they could use computers. Even despite that the usage over a four year
period never got above like eight percent and in some cases about two percent.
So even when you eliminate all the barriers there just really wasn’t a demand
And interestingly we actually had data on demographics, race, ethnicity,
all these things, chronic diseases and all that stuff will be coming out, even
in those cases transactional use, things like scheduling appointments and
getting refills for meds were used fairly frequently but still much, much less
then ten percent. So we defined care related services, which is things like
asking advice of clinicians, talking to the pharmacist, actually over four
years decreased as a percentage of eligible users, it was really low, it was
less then two percent.
So it’s very interesting that in a way what we’re, we perceive that this is
important, and it might be, I’m not saying it’s not, but I think that we do
need to look, step back, because there may not be a demand and the consumers
may not be embracing this as much as we think.
And then the other follow-up to that is as we move forward towards this
whole electronic health record and all these things and I’m looking at your
book here which is chapter three page 25, when it says here are the attributes
of a personal health record, well, if you look through like two through seven
they should contain information from one’s entire lifetime, should contain
information from all health care providers, they should be accessible from any
place and any time, they should be private and secure, they should be
transparent, they should allow easy exchange of information with other health
care systems and professionals.
Well, a lot of that we needed the PHR because we didn’t do that but as we
move forward and if we do over the next ten years develop these electronic
health records and we give patients access to them I’m wondering whether that
will fulfill what we’re trying to achieve and in that case is it that, should
we be focusing on what we’re trying to achieve and then PHR may not be
necessarily the answer but a lot of it may depend on how you define a PHR.
Could it be that the electronic health record with all these pieces put into
place and patients having access basically takes care of the problem and PHR as
a separate entity doesn’t really need to exist?
So I’m just throwing those out as comments, questions, for you guys and for
anybody else who speaks later because I think that’s something that we need to
explore further. Thank you.
DR. LUMPKIN: Anyone want to bite on that?
DR. LANSKY: Well, I’ll just make a couple comments but I agree, I think
it’s a very correct important challenge for all of us to address and I think we
should be agnostic until proven that there’s something very solid there.
Anecdotally we talked to many of the provider based PHR sponsors last
spring and asked them what proportion of the patients you’ve offered this to
have used it and how much and just as a very crude rule of thumb it was about
15 percent when there’s an aggressive effort made of offering, 15 percent of
the patients seem to take it up. And I agree with you that how much they took
it up and for what purposes and how sustained, I don’t think we know that so
it’s very important to figure that out. I think if it is 15 percent that’s a
pretty good number and probably would justify, if it were a sustainable 15
percent of people became active users that would be a very important 20, 30
million people in this country. Just as a benchmark for example we noted that
Quicken, the financial software, is used by I think 22 million people. So as
relatively predominant and pervasive as a tool as that is at least
metaphorically it’s a very small percentage of people with personal finance use
that particular approach.
And to your point about the EHR in a sense being a superset of the PHR
potentially if it fully is implemented, I think we at Markle and Connecting for
Health would agree that an interoperable system broadly construed that includes
patient access and patient control of sharing of their own information, whether
we think of that as interoperable EHR or PHR, there’s some attributes that we
could assign to a National Health Information Infrastructure that would satisfy
many of the requirements. The concern we have though, and even this is true in
the early PHR days that we’re in now, is that the practice of control, if the
information network is sponsored by, managed by, rules are made by, provider
organizations or other institutional entities and the patient doesn’t have the
opportunity to genuinely input their own data, manage their own data, control
who else accesses parts of their data, then those attributes if they can be
achieved in a non-PHR world that’s fine. But it’s something that makes the
The transformation objectives that some of us imagine being possible in the
next 20 or 30 years would imply a less provider centric or institutionally
centric way of offering health care services to people and of managing one’s
own health. So the tools that would permit me to manage my own health and my
families health and make selective use of institutional and professional
resources, often that are digitally available to me in the new world, that end
game may be harder to achieve if we only think about an EHR paradigm.
DR. LUMPKIN: Thank you. Next, it’s me. Just a comment because I can’t
leave, I don’t want to leave Eduardo’s thing just on the table without putting
some context on it. This is a very interesting device but can you imagine this
device without being able to conceptualize, I’m holding up my Blackberry for
those on the internet, email. And I think that Eduardo’s question is very
important but like many kinds of innovations the use case may not even be
imaginable to the common public at this particular point. And particularly when
we use the terminology personal health record because that doesn’t denote the
functionality that we conceptualize when we talked about the personal health
dimension. And as a committee we struggled a lot with what we were going to
call these things, I don’t know if dimension is the right term, but we didn’t
want to imply that this was a thing. And it was in fact the functionality, it
was the capability of looking at health data from a person’s perspective. And
so I think your question is both valid and, is a very valid question and one
that should ring out there but also one that we have to be careful that we
don’t close the door, that because we can’t conceptualize of the use case at
this point doesn’t mean that it may not be very functional, it may make a big
If you could imagine a future whereby using my personal health information
I feel more comfortable discussing with my physician, I can tell you having
worked in the emergency department, and Kevin can back me up on this one, there
are certain things that you tell a patient and the first thing you say well,
I’m sorry you have very serious heart disease and your mouth keeps moving and
they don’t hear anything after that. And that’s a major portion of a lot of
interactions is how much does the patient here and what is their resources, two
hours later, three hours later when they finally work through the fact that
something serious is wrong and now they want to know what should they do even
though you may have told them, how do we address that and those are some of the
Here’s my question, where we are with this concept of PHR does it make
sense, will it work, can it work, without having some sort of framework for
connectivity? In many ways, and David talked about the Quicken model, I was a
very early adopter and one of the things I always wanted to have from Quicken
was the ability to pay my bills electronically until I got it. And now even
though I still use Quicken I don’t pay my bills electronically because it took
longer and I had less control and then there was like that, anyway, that was
the other thing, then you got locked into the portal. But the question is are
these two as linked as we frequently talk about, connectivity and the personal
health record as we think about it? And you can or can’t, you don’t have to
answer that necessarily today but that’s also something for our other speakers
to address. Anybody want to —
DR. LANSKY: I’ll jus say from our surveys and so on it’s a little bit, what
question you asked is the answer you get so I realize that it’s not a totally
open field. But in general I think from the patients we’ve talked to, people
we’ve talked to who see themselves sometimes as patients, the benefits of
connectivity are important to them and they tend to see some of the things that
Eduardo talked as the benefits of having any electronic management of their
health information is largely fulfilled when they can connect to their health
system. So I think as we talk to others in the room today and beyond the
unconnected free standing PHR has had a difficult time building a critical
audience, a difficult time building a business model, and so I think it remains
to be seen if there is as we say an untethered model that would be viable.
Patient seems to think they want connectivity.
MS. FYFFE: First of all I’d like to thank Eduardo for bringing up his
points and asking those very insightful questions and I’m going to be eager to
see what the data looks like as it’s published in January. And for John’s
question I really think connectivity interoperability is very, very important.
I envision a PHR for myself at a point in the future that is going to need to
have lab results and be able to synthesize information from a variety of
specialists, by the way I’m expecting to live for a very long time, and I don’t
see how all that information could be easily accumulated without connectivity
and interoperability. I really almost take it for granted quite frankly.
DR. VIGILANTE: Two questions, I think the first really for David and maybe
Eduardo. When you were looking at perspective users and folks who might be
interested in using personal health record or dimension the emphasis has been I
suppose on chronic disease expecting that’s where the utilization might be,
just a couple of other populations I was curious about, pregnant women, women
with young children who might have frequent visits as a non-chronic population,
and then in the chronic subset indicators of interest among the behavioral
health, mental health population and just wondering if there’s any sort of
uptick in those populations.
And the other question is maybe more on the CMS side, given the fact that
email may be a part of this, communicating with your clinician with email may
be certainly an important part of this connectivity, as a recent article in the
Health Affairs pointed out that could be, there’s really no limits on how many
emails you can send to a provider in a day or a night or a week and each one
though may be very brief in the aggregate it could be a very significant amount
of time required to respond to these. And if you break each one down the
transactional cost of compensating for any given email communication, the
transactional cost may actually be, of reimbursement may be higher then the
actual cost of the service provided. And it sort of gets into a very obvious
problem in the economics of this and so those two questions if you folks have
given thought to those issues.
DR. LANSKY: I can answer the first one, just from our 2003 survey we did
ask some questions about the respondent in terms of pregnancy and number of
children and so on, and I was surprised to see there was no significant
difference in the rates at which they indicated interest in PHR services. I
don’t know whether those health systems that have offered PHR services to those
groups have had more or less interest but the survey data didn’t show any
DR. ORTIZ: Do you want me to respond? From the study that I was talking
about we didn’t look at that, we did focus on patients with kind of more
chronic illnesses and high chronic illness index and stuff, but we actually can
look at that data, it’s available, so it’s actually an interesting though to go
back and look at other users and see whether or not, what proportion of them
use it, so it’s something I will take back but I don’t have the answer to that.
There is data though, limited data, on places that have implemented eHealth
in terms of the deluge of emails that have come to providers and so far it
hasn’t come to pass as a problem, I mean almost all providers are concerned
that that’s going to be an issue but in the places that have actually
implemented these things it really has not become a big problem, for some
reason its worked out okay and so far at least from every study I’ve looked at
and every site that I’ve talked to they have not had a problem with becoming
overwhelmed with email. But it does become an issue in terms of reimbursement
and how you deal with these things because if you are going to provide a
certain amount of care by email you need to build that into the reimbursement
structure and also into the schedule system because otherwise if you just end
up keeping things as they are and then just adding all these email transactions
on top of it then there can be a problem, especially in terms of reimbursement
because now patients who would have come in and you could have billed for them
are now no longer coming in.
And when you actually talk about, they’ve actually surveyed patients and
said how much are you willing to pay to have this thing done and they kind of
say oh about $10 dollars an email. You ask physicians well how much do you
think you should get paid and of course it’s $50 dollars minimum or higher per
transaction, so there’s a disconnect there. So there are some things that we do
have to iron out.
DR. LUMPKIN: Okay, last question. Stan?
DR. HUFF: Because it’s come up I need to disclose the potential conflict of
interest because I’m on HL7 committee, I co-chair the vocabulary committee.
Having said that, having worked in terminology all my life or most of it
anyway, it’s interesting to think about definitions and I agree in the spirit
that it was said that we don’t want to prematurely define what PHR is. On the
other hand using the same word to mean different things or having many words to
say the same thing almost always leads to confusion and so it seems that we may
not, I think we want to continue to think about the PHR as a dimension or an
area that we’re discussing but I think it clarifies the discussion if we come
up with good terms that are well defined that say when I think about this
particular, when I think about patient held, we come up with what we mean by
patient held, that it’s on my personal computer or a third party, anyway, I
think to whatever extent we can define the terms that we’re using that will
And in that regard I would have the question, you went back to the quote
from the President, what do you think the President had in mind when he said
every American should have a PHR ten years from now?
MS. FYFFE: Actually the context of his remarks was his reference to a man
who had a daughter with a very serious chronic illness and she had to go to see
a lot of physicians, a lot of institutions, and the father was getting tired of
having to repeat the same information to the physicians over and over again and
also became frustrated because her latest test results and very, very long
medical record wasn’t going from one place to another. So he put together two
notebooks filled with her medical information and carried it with him around to
all the doctor visits and institutional visits and the President talked about
that in his presentation and then went on to say that in order to improve care,
reduce costs, everyone should have their own personal electronic medical record
and went on to say that the ten year goal was that every American must have a
personal electronic medical record that is controlled by them, and he also
mentioned privacy, so the context was the paper medical record and the fact
that it’s difficult to coordinate and really carry it around and get it to
where it needs to be.
DR. HUFF: So I guess in a sense then anything that answered that functional
purpose, that is it carries my information and it makes my information
available to those who are caring for me would be a personal health record from
that sort of definition. That’s a very helpful clarification.
A couple of other question, well one question and then just a comment. A
couple of you mentioned whether the government’s role or some other, in one
case it was the government, in another case it was Connecting for Health, said
should we develop standards and I was wondering how you meant that thinking
from the perspective of open health, open consensus standard processes like
DICOM and ASTM and those things versus what you were thinking of maybe as
happening within Connecting for Health or how the government would go about
that in terms of rulemaking or other kinds of activities, what were you
thinking of in terms of standards development through your organizations.
MR. ROLLOW: Well, I guess I think about the work that’s been done in the
EHR area where we both had data standards and also functionality standards that
have been applicable for that area. And I think presumably we need some of that
as well in the area of personal health records. In the early phase we probably
need more then anything to have some descriptive terminology and a point in
Would the government ever have a rule about any of this? I don’t really
have any idea a this point, it’s not generally necessary for that to be
accomplished or for that to accomplish the purposes that we need or at least it
doesn’t seem to have been necessary insofar as the work that’s been done in
relation to EHRs. So I’m not thinking that that’s a likely outcome here but how
things develop down the road I guess really remains to be seen.
I think the mode that we’ve been working in really has been to try to get
consensus and particularly within an industry, consensus about what kinds of
standards, data standards and functionalities are desirable in these areas, and
then to try to use whatever has been agreed in various ways to get people to
create systems that work in those ways. So I would think that the sort of
approach that I laid out was really pretty much in that spirit, let’s create
some descriptor terminology, let’s give people a picture as to where things
ought to go and let’s get some early experience with whether or not that makes
sense to people in whether or not those who are developing such systems feel
like they can work with that and from our part whether or not that advances the
use of such systems, quality of such systems, that kind of thing.
DR. HUFF: My last comment is that, this goes back to my day job at
Intermountain Health Care, and we spent a lot of money trying to create a
comprehensive electronic health record. And one of the things that occurs to us
as we contemplate the record being held potentially by the patient or by a
third party outside of the provider it is that that doesn’t obviate the need
for us to continue to maintain our record. So I see, I don’t think we’d ever
see a situation where they would sort of see their fiduciary responsibility to
some amorphous thing that they didn’t control and couldn’t guarantee access to,
etc., otherwise they wouldn’t be able to perform the business that they’re
committed to do. And so sort of recognizing that IHC is going to have its own
record for a long time to come you anticipate these other records, and at a
macro level what it says is if EHRs are expensive now that they’ll be twice as
expensive if we have a copy that’s held within the corporation and one that’s
held outside the corporation because basically that means there are twice as
many of them and in the end the public pays for all of them whether they’re
held by the provider or whether they’re held by somebody else, the patient, and
so the economics, I think it’s just another question of how the economics work
because it’s not going to decrease the cost of health care to have two copies
of the information, one held by the provider an done held by either the patient
or the third party.
DR. ROLLOW: I think this is a more general issue though, it’s an issue
that’s integral to information exchange and one that I know the technical group
working on information exchange has also talked a fair amount about, is there a
need for a sort of production version of a record that providers have and is
there also a need for another version of it that’s used for information
exchange purposes, whether PHRs or for other purposes. So I think that remains
to be sorted out to some degree, at least as I’ve understood the technical
discussions there is likely to be a need for having both a production version
and an exchange based version or a reporting version, or even internal to a
practice in many cases there are needs to have both versions. So that’s
probably at the limits of my technical understanding but I know that you have a
much greater understanding of such things.
DR. LANSKY: The Connecting for Health approach, the phrase we use is a
network of network, and I think the premise is that Intermountain for example
has very important information about the patient’s experience in life but only
a partial snapshot of the whole that that patient needs access to. So the PHR
in that paradigm is an integrating platform for a variety of information
sources out there that are relevant to me. And again, the Quicken model is not
a bad example in this case where I might have multiple financial institutions,
banks, stockbrokers, etc., and I’m the one who, my authorization retrieves
information and consolidates it. True, some of that’s a copy of what’s resident
at the bank or other institutions but I now have it in a situation where I can
manage and integrate it myself. So it’s not only duplication it’s also
management of information in a way that can’t be done by anyone node of the
DR. LUMPKIN: Well, with that provocative question we’re going to move to a
short break, the issue was raised how do we structure these, we have two
panels, the second panel this morning and then the third panel, which will be
right after lunch, looking at those PHRs that are not part of an EHR and then
those that are derived from one. So we’re going to take a break until 10:45 so
it’s going to be a very short break.
DR. LUMPKIN: For the purposes of those who are here virtually and in person
I’m going to ask the panel to introduce themselves even though they may have
done so earlier. So why don’t we start off with, in the order in which people
are going to speak, so Jim, if you’d go first.
DR. WALKER: Thank you, it’s a pleasure to be here.
DR. LUMPKIN: Just introduction right now.
DR. WALKER: I’m a practicing internist and the chief medical information
officer at Geisinger Health Systems responsible at an executive level for all
of our clinical information systems.
MR. BAUM: My name is Brian Baum, I’m leading the Health Record Network
Program at Duke University.
MR. MCFARLAND: I’m Scott McFarland, chief executive of SimplyWell, which is
jointly owned between the Nebraska Medical Center and about 625 physicians and
private practice in the high plains and central plains.
MR. BLAIR: Could we ask the speakers to speak a little closer to the
microphone because their voice cuts out?
DR. LUMPKIN: Thank you, Jeff, and since you’ve joined us since we’ve
started could you introduce yourself?
MR. BLAIR: Oh, sure, this is Jeff Blair, I’m a member of the NHII
Workgroup. I was able to hear you this morning but I was on a line where you
were not able to hear me.
DR. LUMPKIN: Great. So David?
MR. HARRINGTON: I’m David Harrington, I’m vice president of information
technology and chief technical officer the MedicAlert Foundation.
DR. LUMPKIN: Great. Jim?
Agenda Item: Panel 2 – Organizations and Vendors
Offering PHRs that are not Derivatives of EHR Systems – Dr. Walker
DR. WALKER: Thank you for the invitation to be here today, it’s a pleasure.
The first question is why link and a story that struck me was a Fox news
television producer who got a Hepatitis A vaccine, went to Pakistan, did not
hear the news that the vaccine had been found to be ineffective, got sick, came
back two years later in 2003 and had confirmed Hepatitis A disease, serious
liver damage and is still unable to work. If he had been a Geisinger patient in
2001 he would have gotten a notification in My Geisinger that the vaccine was
ineffective and had a chance to communicate with his practice to find out how
to get revaccinated. So the issue is how to make that a more present reality.
I’m going to talk some about our setting just so you have understanding
what drove us in terms of Geisinger and also in terms of our patients to try to
establish a patient EHR, what the patient experience has been, I’m going to go
very light on the physician experience, I think you have, our experience has
been pretty much consonant with others, and then talk some about the barriers
that we have encountered and are still working on.
To understand us you have to understand that we serve 31 counties in
central Pennsylvania, 23 of which are rural. Our population is older, poor, and
less mobile then United States averages. We have 800 physicians, 650 employed,
42 clinic sites throughout those 31 counties, two hospitals, fully integrated
care but unlike someone like Kaiser we’re 70 percent fee for service.
Our outpatient EHR was completed, the implementation was completed in March
of 2002 so that all of our 650 employed outpatient physicians see all results
at digital radiology images, all records in the system, 100 percent of orders
are entered in the system, and about 70 or 80 percent of visit notes are
documented electronically and then actually transmitted to external physicians
within 24 hours.
So when you ask or when I actually started reflecting on what really
prompted us to build a patient EHR the real kick off was that our outpatient
EHR is designed with making a patient EHR a practical reality in its design so
the patient EHR is simply a view of the same electronic record that I see, the
same electronic record that we share with external physicians. And then the
obvious drivers, Institute of Medicine made a number of recommendations in
Crossing the Quality Chasm and of course HIPAA sets another set of standards
for us to achieve and we intended to meet all of those in the course of this
Our goal is number one are to delight patients. It is clear to us that this
is the killer ap from the patient standpoint. All of the enormous values of the
electronic health record and all the other things we’re doing in clinical
information systems are relatively invisible to patients but in focus groups
and in survey questionnaires we find that this is the thing that really catches
patient’s imaginations and really changes their experience of health care.
The second goal is to empower patients and we intend to do this in a number
of ways. The latest is within a couple of months now patients would be able to
schedule family practice appointments electronically, they will be able to go
into the electronic schedule, pick the time that they want, and then have a
reminder afterwards that they’ve picked it.
To improve practice efficiency, while we think that financial incentives
are important we’re relatively cynical about the speed and uniformity and
appropriateness with which they are likely to be applied and we believe and are
in the process, we have funded research that is in the process of trying to
demonstrate that with an effective patient EHR we can transform health care to
the point that we can decrease our costs of providing care simultaneously with
providing the patient with a far higher touch and far higher quality care then
they have ever received in any system before.
And finally, and this isn’t insignificant, we want to delight providers.
Part of what we are interested in doing is making the clerical work of health
care disappear, both from the patient’s point of view so that you don’t have to
carry two notebooks around with you to get semi adequate health care, and also
from the provider’s view so that all of the things that now take enormous
amounts of time, it’s estimated that someone like me would have to spend seven
hours a day to meet the screening and prevention recommendations of National
Health Task Force, make that seven hours go away so that providers and patients
can spend their time on something a little more valuable.
So patient drivers, when we started thinking about this in 2001, which is
about the time we had the outpatient EHR under our belts, we did this poll with
patients in our waiting rooms, community practice service line or family
practice waiting rooms, I think it was 7,000 patients, and gave them a fairly
wide range of things they wanted to do, you can see they don’t want to purchase
medical supplies on the net, but they did think they wanted to see lab results,
schedule and cancel appointments, request drug renewals, request medical advice
and so forth. That’s fairly consonant with a poll that Harris Interactive
published just about the same time, probably a year later then we did ours so
it may have been done near the same time.
So what is our status now? We have all practices, all specialties, patient
have access to what we call My Geisinger, the patient EHR. There are 17,000
active users and we’re adding 2,000 a month at this point and last month we
were adding about 1200 a month, so it looks for now that the uptake is
accelerating. The denominator for us is something like 500,000 patients. We are
in some way responsible for the care of about half a million patients one way
or another. We have pediatric access, which means that parents, parents of
children under 14 have rights to see the record but between 14 and 18 we have a
set of rules that Pennsylvania and other laws we hope and allow patients to
have access to their record and to share that access with that parents, and
then at 18 to have exclusive access.
And as it turns that’s important, when we asked focus groups, there’s one
group, older people with chronic illnesses who have a lot of traffic with
physicians, another group is young parents, 35 year old parents, who could care
or less about talking to the doctor themselves because they have very little to
say but would kill to have a printable immunization record so they don’t have
to schlep into the doctor’s office every time their kid goes to camp.
And then proxy access is another what we think critical functionality that
lets the patient give other people access to their EHR. The most powerful use
of this, and it’s been noted in other surveys, is 60 year old people helping
their 85 year old parents manage their health care and for the first time being
able to see what medicines their doctor thinks they’re taking, being able to
see when their next appointment is, and by the way being able to send an email,
we had a patient come, not a patient, a patient’s child come to a feedback
group and she said I love this, I was able to send a message to the doctor’s
office and say my father wasn’t eating very well and his temperature was X and
he was having a little trouble breathing, they got in to see him and treated
pneumonia, he didn’t need hospitalization. So the children of patients are also
a very large part of the population for us.
So what can you do with this thing? The first thing is as I said they see
an extract at this point, and it’s a continually growing extract, of the
electronic record that I see, so they have a patient list, they see their
problem list, which of course we have the advantage in our system for the most
part that is all of their care givers and of course one thing that I think
didn’t come up in the discussion earlier is if you have a really high
performing primary care physician part of that, speaking as one, part of my job
is to take all of the patient’s experience 40 years ago and current and at the
specialist and with me and abstract that appropriately so that I have a
complete record of the patient’s health history. And theoretically then can of
course, if the patient has access to that then the patient would be able to
share that with anyone they want to. One of the things that’s probably obvious
to you is that one of our patients can be in Cancun and show the emergency
department doctor their electronic medical record and by the way that means the
ED doc there can see our lab tests, not lab results that the patient wrote in
that the doctor may wonder what lab they’re from, what the standard, what the
normal range is, what it means, so it’s the allergies and medications and the
medication list is linked to explanations for what each medicine is given for,
what the right dosing is, and what the things are that might happen that you
should call your doctor to let them know it’s happening.
Immunizations as I said, young adult standpoint is the killer. Histories,
surgical, medical, family, and social, this is toxins, sexual activity,
smoking, alcohol, drug use. And then lab results with interpretations, as you
will see this ends up being the thing the patients, at least talk back to us
the most about and I’ll talk a little bit about why. Appointment information,
they can see past appointments as well as upcoming appointments, and again in
our system that means primary care and most of their specialists although they
don’t see all Geisinger physicians exclusively obviously.
What else can they do? They can renew medicines, so that medicine list has
a check box beside each medicine so when you need to renew a prescription you
simply click the check box for the appropriate medicine, hit submit, and it
goes electronically to your physician’s practice. You get a message back when
that has been completed and then you can go to the drug store knowing that the
medicines are going to be there and you’re not going to have to go back and
call the practice again and come back two days later to get your medicines,
which at least in my experience is appallingly frequent.
You can request appointments and as I said you can now, well now, within a
couple of months you’ll be able to actually select your own appointment time
and date if for a family practice visit. Incidentally a study was just done,
when you do this with mammography, when you use something like My Geisinger to
send a message to a patient and say it’s time for your annual mammogram, let
the patient schedule it electronically and then provide three automatic
reminders of the date of that test you decrease no shows from nine percent to
one percent, which has obvious huge implications for practice efficiency but
also for American health care quality.
You can request referrals so that we do a great amount of referral
management electronically, and you can ask for medical advice and I’ll show you
something about that.
So this is the use, 17,000 active patients, this is third quarter 2004,
280,000 lab test checks, 280,000 times patients looked at their lab tests. And
what happens with this is as it turns out is that a fair number of patients are
on medicines where their drug levels are significant, anti-epilepsy drugs, lots
of other drugs, and they have a very strong interest in keeping track of those
levels through My Geisinger and then obviously the other things that people
with diabetes, people that you’re managing their cholesterol, find it very
useful. And then as you see also 250,000 messages annualized, that’s based on
It turns out that the patients were about right in that poll but they
didn’t anticipate how strongly they would prefer medical advice over the other
forms of communication. Just a note on that anecdotally, as Eduardo said,
there’s a fair amount of research and so far all of it includes that the
traffic is manageable and physicians find it acceptable, particularly if they
got to paid to find it acceptable, but they find it acceptable. And that’s our
experience, in fact what we are finding anecdotally is that the physicians who
really use this say it’s actually easier, they want more email message because
it’s easier then talking to someone on the phone, it’s faster then sending a
letter, and in our system of course part of what happens is that when the
patient sends a message and I reply all of that is automatically saved into the
record so there’s no question about who said what, when I said it, and whether
I recommended that they come in and be seen because it was too complex a
problem to take care of all the email, or whether I just say why don’t you send
me a couple blood pressures in a week and then I’ll tell you what I recommend
about your blood pressure medicine dose.
So this is the use curves, busy slide, just follow the two arrows. The left
hand arrow is lab tests, lab test views, and as you can see there is
substantial increase month over month in the number of people accessing their
lab results. A quick note on lab tests because I think it’s an important thing
for you to be aware of, all of the third party or sort of national expert
groups that we found to provide lab test interpretations were completely, were
flatly inadequate, things like saying a high calcium can be a result of cancer,
as was it John said earlier, you say something like that to a patient that’s
the only thing they hear for the next month out of your mouth and so we are
actually writing each of the interpretations to be sixth grade level reading
and to say things that are helpful and not frightening or confusing and so that
is slowing us down, we have about 95 percent of tests that are performed we
have an interpretation for.
Another point on this is the way the system is set up the lab results are
not released to this, to the patient view, until I have marked it as done. And
so what our policy obviously is that I’d look at that, I can attach a message
to it that also goes the patient view so I can say everything looks great, LDL
is wonderful, just keep up the prudent diet, and said that. Or if it’s bad news
of any sort, and we recommend all abnormals, I call the patient, do what I
would have done before and let the patient know about it so we don’t want
anybody to get bad news first in My Geisinger. And you can see the messaging
generally, the right hand arrow, is also increasing substantially month over
And within message that left hand white arrow is medical advice, which I’m
not sure, I don’t think we would have expected that that would end up being so
much the major use of messaging but it is. As you see meds renewal has kind of
slowed down in September, I’m not sure what that means.
I want to talk then about some patient experience and some about physician
experience since you probably will not have heard based on an article that we
published in JAMIA this month actually, November/December. As someone else
noted in our population the bulk of users are patients 46 to 64 and our
speculation is the same as that, as David’s I think, that probably what this
represents is the overlap of actually needing to keep track of medical issues
and medicines you’re taking and doctor appointments because of increasing age
and high enough use of internet that you do use it. As you see the 65 year olds
are a slightly smaller portion than their size would predict. As I guess you
all know 65 year old women is the fastest growing group of internet users, a 20
or 30 percent increase use per year, so we’re confident that that number is
going to grow. Interesting male/female ratio stays about the same as the
population and as other people have noted this is clinic visits during the nine
months pre-implementation of My Geisinger but they clearly do have
significantly more visits though again in our feedback group and the
questionnaires both even people who do not have significant chronic diseases
find this very useful.
This is a national study that was done by Connecting for Health that is
consistent with ours, just a note on our population, although it is rural,
poor, and older our internet penetrance is just about the same as national
I want to talk a little bit about how patients want to be communicated with
at this point and of course a lot of caveats, this is an extremely advanced
group of patients obviously, the other thing to say is this is a very advanced
group of physicians, there are not very many physicians in the United States
who do all of their outpatient work on electronic records so there’s a lot of
possible reasons but this is what we find. The blue numbers are measures on
which patient and physician opinion is very different and the ones that I’ve
outlined in white there for routine follow-up doctors still love the phone and
our patients at least wish they would send them an e-message. Same thing with
test results, physicians overwhelmingly prefer letters in this case partly
because our HR makes the letter fairly easy to write but not nearly as easy as
a patient EHR note.
So what do patients like? They want more, they want it faster, age is
pretty much irrelevant in our feedback groups at least. Pediatric access as I
said is a big deal, adult children want to be able to use it to help their
parents, usability is critical. We have gone through three I think iterations
of trying to make the site more and more usable, every time we do that usage
goes up and I suspect that the disparities in usage among systems is partly due
to the fact that being connected to thEHR means a great deal to patients, 87
percent of patients say I trust information that comes from my doc. And the
other thing is making sure the systems are equally usable when we try to
compare their effect.
Other things equal I prefer to go to a doctor who provides it, this is on a
scale zero to hundred. It provides the information I need, one of the things
we’re finding by the way is that for the first time and in significant numbers
patients are calling and saying, or sending an email typically and saying by
the way you’ve got that wrong, I didn’t have that surgery or you missed that
surgery or I’m taking this medicine or I’ve quit that medicine, so one of the
powers of this is that it lets us really have a self healing or self correcting
medical record where we really have a clear and clear sense what the patient is
We were worried about whether patients would be able to cope with having
lab tests visible to them, whether they’d be frightening, see a result six
months ago and not remember what the context was. At first there was fairly
strong response that test results have worried me at times, now that’s gone
down to 21 which is the lowest result we have on any measure we kept track of,
and would I like more lab tests, absolutely.
And I don’t know if you want to stop there, you don’t want that.
Let me talk just about barriers, I think one of the barriers is that this
project would have cost us ten to 20 times as much if it hadn’t been that our
EHR makes it just another module of the EHR. We need to factor in this
evidence, we’re trying to provide some of that, but that clearly is lacking.
Physicians cost of change, I’m talking about an intellectual, energetic cost
that’s going to need to be accounted for. We want to try to do that by making
care processes simpler but clearly financial reimbursement is going to be part
of that. Patience acceptance, right now polls say that about 30 percent of
patients want to do this kind of thing. We believe that that will grow rapidly
and we think our experience is showing that in a fairly disadvantaged
population and one of the things we don’t want to do is create a PHR that only
works if you’re so smart and motivated and fastidious that you can create your
own medical record, I don’t think that will work at all.
A deep problem with providing all of the record that we are wrestling with
is that the record notes radiology results are necessarily written in jargon,
no one would ever get their work done if they weren’t. And translating it,
providing that to patients unvarnished, and this isn’t the doctor saying this
patient’s a crook(?) or something, this is the doctor I can’t read an
ophthalmology note and understand it. And we are going to need to grapple with
what it means to provide the patient all of their record in a way that is
somehow helpful to them and meaningful.
And that’s probably enough.
DR. LUMPKIN: Thank you. Brian?
Agenda Item: Panel 2 – Organizations and Vendors Offering
PHRs that are not Derivatives of EHR Systems – Mr. Baum
MR. BAUM: That’s a very good news story, lots of good information.
Thanks for the opportunity to present to the committee the point of view of
the Health Record Network organization. As I mentioned before this program was
initially launched out of Duke University as a collaborative effort between the
Feuchal(?) School of Business and the Duke Medical Center. Personally my
background in health care is quite limited, I’ve only been in the industry for
about three years or so, I spent the bulk of my career focused in private
industry in business and market development activities for emerging
technologies and general market innovations. I’ve been quite intrigued by the
challenges and opportunities of the health care system.
As I’ve become, traveled the country presenting our point of view and
meeting with various individuals I think it’s clear that the vision is there, I
just chose one example to cite this, this is from the Framework for Strategic
Action, that the vision, the future state of health care would be consumer
centric, information rich, and medical information follows the consumer and
information tools guide medical decisions. So the good news is it seems as
though we can clearly define what there is, what the future state would be, we
know the definition, the challenge basically is just getting from here to
With MyChart there I don’t mean to overly simplify that challenge because
in short we’re basically talking about moving an industry from point A to point
B, and not just any industry but an industry that represents almost a fifth of
the U.S. economy, 13 million employed, thousands of registered hospitals,
processing 36 million admissions a year, half a million physicians, and
potentially, if that’s not enough, almost 300 million customers. And I say
potentially because none of us knows when we may wake up one day and either be
in an accident or be diagnosed with an illness, that moves us from a passive
observer of the health care industry into a front and center very active
consumer of health care services.
I think we also have a strategy to get there and I think the strategy is
based in large part on the reality of the situation and there are some very
tactical issues that need to be addressed to advance the industry. Developing
standards, deploying systems, equipping physician’s offices, raising funding,
providing incentives, I saw the Connecting for Health report in the last two
weeks or so indicating they require $24,000 dollars or so per physician per
year to drive the adoption of electronic health records. We are creating local
networks, we’re calling on communities across the country to form local health
information infrastructures to bring the business community and the health
community together and build these over time. The hope is these networks will
link and we’ll look back in ten or 15 years and we will have a fully integrated
health information network.
So as I say I think all of these steps are very critical, absolutely
necessary to be conducted, however it is a strategy that has inherent risk and
that risk is it’s very difficult to gain momentum in this environment. It’s
very easy to lose focus of the actions over time, just one example having spent
time in the consulting industry, we backed up three years ago every major IT
vendor out there, Microsoft, Cisco, Intel, all were focusing more and more on
the needs of health care industry because the economy was going so south. Now
that the economy is beginning to heat up again their focus and level of
patience to support the health care industry is not what it was three years ago
so that’s an opportunity lost.
This is where we introduce the Health Record Network concept, if there’s
one word that I would continually refer to is I see what we’re proposing is
very complementary with the existing strategy. HRN is essentially grounded in
the laws of physics, it turns out Newton was quite a market developer as well
so his second law of motion, force equals mass times acceleration, if the mass
in this case is the health care industry and we’re attempting to move it to
this future vision of a fully automated information rich environment what are
the options available to try to push that mass forward.
Within Duke and within our HRN program we looked at a number of options,
the physician community or the medical community at large, is it possible to
coalesce them to be an agent of change. Associations, trade associations,
consumer associations, the vendor community which is represented to a degree
here, government regulations or mandates, here again our forward thinking,
Isaac Newton had his third law of motion that for every action there can be an
equal reaction of opposite force and sometimes government mandates actually do
result in more of a push back from industry then open armed acceptance. Then we
looked at hospitals, the business sector I think can play a big role in this,
and finally health consumers.
As we looked at and went through our analysis of where this force may be
evident to drive this massive industry our conclusion was basically the health
consumer held the greatest promise to actually move the industry. This morning
already there have been some discussions about why the consumer may not be a
leverageable asset in this space, over the last year I’ve heard every reason
for why that’s the case. But one of the reasons that we feel that the consumer
is very powerful in this space is that their sheer size and mass, the equation
of ten percent of roughly 300 million population, 30 million people is market
force. What’s almost important in that equation is the inverse, that we’re
saying that 90 percent of the population in the first pass of having access to
electronic health information will not get it, they won’t understand it, they
won’t value it. So nine out of ten people we’re saying out of the gate they
just won’t get it at first pass, not unlike the utilization of credit cards 40
or so years ago, there is an adoption cycle that occurs.
So how do we engage consumers? Well, we start by segmenting the market and
that’s why I mentioned the good news coming out of the Geisinger effort. I
think there’s some commonality in what we’re looking at, who were the early
adopters, we think parents with young children, not trying to solve world
hunger but simply trying to manage inoculation records, highly motivated
audience. Baby boomers beginning to assume responsibility for their parent’s
care, this is a group of the population if nothing else is known for not having
much patience. And as they begin to take control of their parent’s health and
see the complexity of medications, of physicians, of sites, they’re not very
accepting of the inefficiency in which all that is managed. The chronically ill
obviously, good understanding of the need to manage information. And then
anyone who has had a significant encounter with the health system, probably all
of us while we’re very close to this issue could go around and site personal
examples of the breakdown in information management in health care.
So we start by segmenting the market, the next step is really to create
awareness and this is part of the challenge that I think in any of these
audiences they can articulate the problem very clearly, what they can’t explain
is what’s the solution, what’s the so what of it, what can I do instead of
going from doctor to doctor to assemble my child’s inoculation record, what’s
So that brings us to the next element of the Health Record Network model is
providing an easily accessible universal solution. So when someone in any one
of those populations, when we heighten the awareness of their need for managing
information we need to provide a solution where they can take action. It’s our
belief that this can’t be a mix of a thousand different solutions across the
country, that in fact that is the quickest way to lose the momentum of the
market if it is diffused that I am in central Pennsylvania and there’s a very
sound solution for me there but three months later I move to Southern
California and I’m lost. So our belief is that it’s got to be universally
accessible, universally available, it needs to be recognizable.
And here’s an example that I use over and over again to try to provide
context is in the banking industry, we probably all know what FDIC is, we know
that when we walk up to a bank we expect to see that logo on the door and I’d
venture to say probably 99 percent of the American population at least knows
that that logo is a good thing.
What we’re saying is in the health industry a same level of awareness needs
to occur, when you walk up to a physician’s office, when you walk up to a
hospital, you see thEHRN logo and that logo indicates that that facility, that
physician, is utilizing this network, has access to safe, secure information.
Keep it simple. Here our point of view is the market right now has nothing,
the general market has nothing other then those individuals that may be
managing health information on their own. They have nothing so they’re not
ready for full integration and automation. We can start quite simply with the
consumer providing information around family medical history, around medication
list, around allergies, identifying who their clinical care team is.
Then we allow the market to begin to absorb that functionality. What
happens is what has happened in every other industry, once the market forces
are engaged the market is no longer satisfied so the market begins to say why
aren’t my lab results automatically uploaded into that record so that I can
monitor cholesterol levels. Why aren’t my imaging records there so I can go for
a second opinion? So the market forces begin to drive the industry to make
those necessary changes. And finally we need to provide value, even at this
very basic level we need to provide value back to the consumer.
This is a very preliminary view of what we are envisioning for thEHRN
environment and I don’t want to spend a lot of time describing this in the time
I have allotted but one point that I’d like to make, the way we’ve configured
this, the majority of the site is all directed specifically to the consumer and
to managing their care. There is space allocated on the right hand which
identifies who is the payer so in this case United Health Care and at the
bottom of the page GE is the sponsor of benefits. We presented this to Rick
Wagner at General Motors and his immediate reaction to this was this would make
the transition from plan to plan so painless for their employee base because
all of this stays the same, this follows the employee, this follows the
consumer. If I leave GE and go to GM it’s simply an update, all the information
still stays the same. If my employer changes health plans I simply get an
update to that.
So what is HRN? Basically to recap it’s set up as a not for profit
organization. While we launched this out of Duke University we quickly realized
that the scope of what we were proposing needed to be independent, it could not
be affiliated with any one institution. We had extensive debates about a
commercial versus a not for profit venture and our conclusion was that given
the state of the market right now and the sensitivities of issues around health
information that it was necessary to go in a not for profit direction for this.
What we see essentially is creating a national health utility that is there for
the benefit of the nation’s health care system and to benefit all health
consumers. It is not really a competitive space if you will. It can give rise
to many new business models and new businesses but the core value of managing
this we think is best served in a not for profit.
As I mentioned it will be a branded entity, we are assembling a board of
directors right now that will very publicly debate many of the issues and the
rules of operation for how HRN operates. How can information be used, how it
can it not be used, those types of issues.
On the issue of interoperability I typically definEHRN as an overlay
network, our goal is to interconnect to the consumer, provide support to the
consumer, and then linkage back into the EHR systems that are evident across
the country. So we think both elements are very critical to the success of
this. And finally starting simple with consumer provided information.
Our business model, I won’t go through all the detail here but
fundamentally our operating principles are that, our goal is to provide a free
service to consumers where their health information is used in the management
of their care, so whether they access it or their care team accesses it, those
transactions are free. We do believe though that a financially self sustaining
model can be developed to support this network over the long term and that is
by mining the value of the network itself.
Our objectives, business objectives create consumer demand, provide the
nationwide web solution where consumers can voluntarily initiate the creation
of their record, then leverage that market demand to increase the features and
functionality as well as interoperability of the network. We’ll provide
consumers services such as online help, call center help. We’ll offer value
added services to help the consumer utilize the information to improve the
management of their own health care and we’ll also certainly accommodate third
parties for that as well. Create a nationally branded entity and very important
to us our intent is to complement the efforts of NHII and similar organizations
that are focused on building the clinical infrastructure.
The operating model, we’re establishing relationships right now in the
business community, health industry, and consumer groups to advance the agenda.
We are right now planning a series of pilots in different venues to gather
information on who the target audience is, what the most efficient message is,
what the most efficient way to communicate to the consumers is. Our
anticipation also is to fund operations between now and financial self
sufficiency through accommodation of public and private support.
Quick status on where we are, we have legislative approval in the state of
Wyoming to conduct a state wide pilot. At the Duke Heart Center any patient who
is referred to that center will be given an opportunity to establish an HRN
record. We’ll also append to that record the clinical data out of Duke as well
as imaging data. The board of directors right now that we’ve assembled include
Sue Altman, Art Kaplan, Newt Gingrich, others to start defining the roles of
operation and most interestingly we’ve been contacted by representatives of the
Canadian government that are interested in conducting a parallel effort and
they’re actually talking about defining theirs as the Health Record Network,
Canada’s Network for Record Health, so interesting developments there.
Finally again as I say what we see is this as a complementary effort. I
believe if we’re to move this industry into a new era of automation it will
require certainly a push and a pull and I think together we can move what may
DR. LUMPKIN: Thank you. Scott?
Agenda Item: Panel 2 – Organizations and Vendors
Offering PHRs that are not Derivatives of EHR Systems – Mr. McFarland
MR. MCFARLAND: Luckily I’m the grandson of a livestock auctioneer and I can
hopefully get us right back on time.
I’m also going to jump around in my presentation and really focus on the
questions that were raised by the panel in the first session, SimplyWell LLC is
jointly owned by the Nebraska Medical Center, which includes the University of
Nebraska Medical School, the faculty, and 351 physicians and academic in
research practice there, as well as 625 physicians in private practice who
practice throughout the central and high plains at a variety of hospitals. We
believe the success that what we’ve had has been in large part due to the way
that we are owned, we’ve had tremendous penetration with physician’s offices
and tremendous penetration with health systems across the state of Nebraska.
We have been flying a bit under the radar screen and I thank you for
allowing us to come today to discuss the progress we’ve been making, things
that Brian was stating in his comments that they will be doing we are actually
doing in 41 states with tens of thousands of people and we actually have a
population just across the river in Fairfax, Virginia, at a private company.
We basically place our —
DR. LUMPKIN: I’m sorry, did you say 41 or 21 states?
MR. MCFARLAND: 41, yes, we actually, we will be at 49 about the middle of
next year, we’re not going to have exposure in Hawaii unfortunately.
We place our product as part of the employee benefit structure, again we
have mainly large self funded employers make the product parts of its benefit
scheme. I’m going to just quickly go through these slides and touch on the high
points, what’s different about SimplyWell is that it provides yearly work site
health screenings with physical metrics and lab, we will typically go to the
same employer site, say it has 1,000 people, at least three times a year. We do
a fasting glucose, we show up at 4:30 in the morning during shift time changes,
etc., we have a great competence at field operation and gathering real clinical
metrics on working age populations. Also we developed an individual action plan
tied to the history risk, the physical, the biometrics, and the laboratory
findings and again I’ve put in parentheticals who drive why we selected that
I’m going to skip the next slide and also emphasize sort of the public
health database, the dynamic database, and the expansion capabilities were all
driven by our research and academic physicians. Again, they have particular
areas of research that they particularly need good, solid clinical metrics on.
Rather then posting a bunch of newspapers, advertisements, they’re now calling
us to simply query our database and come up with a list of qualified folks who
would be willing to participate in research activities.
Here’s where we’re going in 2005, an online employment management system,
again our patients are telling us, and participants are telling us they want
that. Automatic therapeutic substitution for pharmaceuticals, that’s being
driven by the payers. Interoperable radiology, we’re going to do that across
seven health systems in the state of Nebraska and if you’ve ever driven across
the state of Nebraska that’s like three quarters of the country so it’s going
to be a fun activity but very important for our physicians who practice at
multiple sites and do a lot of rural medicine.
We also have functionality moving forward for the elimination of avoidable
medical claims, particularly those related to lifestyle, that’s being driven by
payers who want to get their populations thinner and also the identification of
individuals in medical treatment for active disease management, and again we’re
utilizing our nurse call center in that fashion to take a lot of the fragmented
disease management population and put them on an integrated model.
Again, we firmly believe and this kind of goes to what Eduardo said
earlier, we knew that if we didn’t incent this thing we would have very low
participation by the patients in this activity. Being part of the group health
plan design they receive points and incentives to view their electronic health
record, to have their lab drawn at the worksite. About 92 percent of our folks
log on at least once, actually 33 percent share that access and information
with their personal physicians. We see about 76 percent of those who sign up
for the lab draw actually show up and are poked with a needle. The online lab
results, typically we don’t give them points, again they’ve already been stuck
with the needle, we don’t give them any points until they view their lab
results and if they have a critical value they have to talk to our nurse, one
of our nurses, before they, we don’t want them to receive bad information, or
bad news online, about 88 percent actually draw or view their results online.
We do the worksite screenings, it’s usually available to 100 percent of the
employees and the spouses, we’re not doing children under 18 currently. We see
participant dependent on incentives and it ranges, we have some groups that
have only about 30 percent show up and many times that’s the working or worried
well, and if it’s particularly incentive, let’s say there’s $1,000 dollars at
stake, you’ll get about 93 percent of the covered lives on the plan to show up.
Again, we believe very, very firmly in access to our nurse call center and
access to materials online, about 23 percent of our alpha site participants
have told us that the resources online and our nurse call center has kept them
out of the emergency room or a physician’s office. A big part of the 2005 plans
that we work with were encouraging behavior change, obviously they’re wanting
to see a lot of benefits tied to the health savings accounts, the flex spend
cards and gift card technologies. Incentives are becoming more financial in
nature due to the low transaction costs associated with fulfillment. We used to
do a lot of windbreakers and tennis shoes and coozies and exercise equipment,
that stuff is expensive to store and it’s expensive to ship. It costs us about
a nickel to do a transaction onto one of these cards.
Thank you to all of you who worked on all of these great things, they’ve
certainly been a guiding aspect of our development functionality.
Why is the SimplyWell model being adopted and I think I’m just going to
mention two of these, I’ll let you review them. We support health care
relationships continuously, not when sickness and illness presents. Again, the
focus of our organization is preventive medicine and risk management and
finally behavior modification or lifestyle change. We also leverage the least
utilized resource in U.S. health care delivery, that being the patient.
Who’s paying for it? Again, self insured employers both in the public and
private sector. They’re financing the SimplyWell investment via stop loss or
reinsurance carrier credits or discounts offered by underwriting carriers and
MGUs in the marketplace. The rewards systems again are going back to cards, and
the annual billing is occurring through the group plan not as a corporate line
item expense. The physician practices who are putting their patient populations
on SimplyWell, what we do for them is basically we do a lab up front, we draw
and process the physician standing lab orders for the patient in advance of the
patient’s appointment, the online lab results and resulting PHR is available to
the patient and the physician in advance and during the patient’s appointment
with the physicians. Physicians love it because they have access to good
clinical information at the point where they interface with the patient. And
again, a big reason of success, we’ve only been placing this with the physician
owners of SimplyWell, we’ve not marketed this externally.
Who is left out? Fully insured employers, small business unfortunately do
not have the resources to invest in this simply because the group plans will
not let them bill this through the group plan. Carriers who are on fully
insured products that serve small business are unwilling to invest in keeping a
population healthy when that carrier will likely lose that population to
another fully insured carrier within three to four years. There’s some great
model legislation coming out in Pennsylvania, it’s House Bill 2501, it’s
basically a disease management insurance tax credit policy that we hope other
states will replicate and the federal government in terms of helping small
businesses invest in preventive medicine and disease management.
We already think we’re an interoperable EHR, we didn’t really get in the
business to do personal health records, again with our interoperability with
our physicians and our health systems across the state we already believe we’re
an interoperable EHR. And again, we think the barriers of interoperability are
being overcome with improved datasets for better health outcomes and lower
delivery costs and I again list those physician references below.
Again no one has talked about consent authorization and controls and we
handle that in a very unique manner. We very early on we embarked on this
effort, our patients, they absolutely never want their employer to find out
anything about their protected health information and they want to be in
complete control as to who that health information is released to including
their doctors. We got that message loud and clear. The doctors didn’t believe
it at first but after listening to the consumer focus groups they’re like wow,
patients are concerned about privacy and want to own that record.
The funny thing is patients actually thought these health records existed
at hospitals and with their insurance carriers and they thought that they owned
it and many times they were told by their insurance carriers that said no,
that’s our information we’re keeping it. So they were very surprised, most
consumers are very surprised to learn that there is no ability for
So what we do in terms of patient consent authorization and control, they
accept the SimplyWell user agreement and disclaimers, we capture their
signature every time we draw blood on them, whether it’s the worksite or at the
clinic, the patient sets up a unique user ID, a password and security
questions. The patient affirmatively directs the application to generate a
unique user ID and password for each member of the participant/patient’s care
The patient/participant affirmatively releases the unique user ID and
password to the individual care team member, the participant affirmatively
selects the length of time each care member has access to the PHR and that
expires every year, so again they will have to renew it if they’ve given access
for a year. An access audit trail is always provided at the patient’s request
and participants/ patients are granting access to family members and also
Again, you all understand the concepts of population health management and
disease management, that’s how we do it, we have not figured out how to do the
Medicare/Medicaid population, we just have not focused on that, we’ve been so
busy as an early stage company growing the market that really doesn’t receive a
lot of episodic care and again resultant, personal health records that result
from an episodic experience are sort of the reverse of what we’re doing, we’re
going to the worksite, getting these folk an electronic health record, and then
when they do, when the time comes when they do receive episodic care it is
there for the clinicians to utilize.
And again, this is my final slide, our alpha site has been in existence
since 1998, it has about 4,800 lives on the plan, the expected cost for this
employer in the Midwest was a trend in the gold and that was about a 15 percent
trend. Here’s what the actual costs have done for the group, if you take a look
at the delta in 2001, 2002, and 2003 the cost avoidance from the actuarial
perspective, that’s about $24 million dollars to this group plan. So obviously
there have been some plan design changes, they’ve made it more focused on self
care and preventive medicine, and it’s saving this particular self insured
employer a great deal of money.
And that’s all I have.
DR. LUMPKIN: Great, thank you very much. David?
Agenda Item: Panel 2 – Organizations and Vendors
Offering PHRs that are not Derivatives of EHR Systems – Mr. Harrington
MR. HARRINGTON: Good morning, I also want to thank the committee for
inviting MedicAlert to participate today, we’re very happy to be here.
I want to tell you a little bit about MedicAlert, we’ve been around for
quite a while and sort of flying under the radar in a lot of cases. We were
founded in 1956, we’re a non-profit foundation, our mission has always been to
protect and save lives of our members, we’ve been focused primarily on
providing information about our members medical conditions and medications,
allergies, and devices, medical devices that they may have in an emergency
But we’ve been keeping electronic health records since 1985 when the first
computer was rolled into MedicAlert I believe around that date and we have
always had what we call our Emergency Response Center, which is a 24 x 7
operation that emergency response technicians, emergency department personnel,
physicians, have contacted by telephone up until the internet age and still
today they contact by telephone, to tell us that one of our members is in an
emergency situation and to get information about their conditions.
We are headquartered in the United States, in California, in Turlock,
California, which is about 90 miles east of San Francisco, and we’re global,
we’re international, we have nine affiliate countries around the world. So we
actually 48 years and quite a lot of exposure, we’ve got four million members
worldwide and all those members have a form of a personal health record.
The benefits that we provide to patients, providers, practitioners, and
payers are that it enhances, we enhance and speed diagnostic assessment, that’s
why people call us 24 x 7 to make sure that they know everything as they are in
the process of diagnosing a situation. Our service and our information improves
the ability to detect hidden medical conditions. There’s a lot of information
that we have in our personal health records, our members have in their personal
health records at MedicAlert that may never appear in any individual medical
record that they may have and I’ll get into that more later, but as a result it
reinforces quality of care standards, it avoids retesting for previously
diagnosed medical conditions. We provide the ability to retrieve digital images
and we provide patient medical history of course, that information we store in
the repository as well but we also have family and physician contact
information. And we serve another vital purpose for many of the health plans in
that we can provide out of network notification if somebody is being
hospitalized in a place where they can moved from that place to ameliorate some
Now I’m taking the questions that Mary Jo asked us to address pretty much
literally one by one so let me talk a little bit about the functionality of our
electronic health record. We offer personal information management, personal
and physician contact information management, the medical information that we
manage includes medications including dosages, we do drug utilization reviews.
This is where the extended information comes in, we keep, capture and keep
information in our electronic health record for our members on all kinds of
medications that are non-prescription medications, vitamins, over the counter,
any kind of herbal medications they may be taking and these are very valuable
in doing drug utilization reviews and often, probably never, will a pharmacy or
a PBM have these medications in their database when somebody goes to fill a
prescription from a physician and they run their normal DR checks as they’re
filling the prescriptions so we feel that’s a vital service that we can offer.
We also store conditions, allergies, and detailed information about any
medical devices that the member may have, manufacturer, model number, serial
number, date of implantation if it’s an implantable device, etc.
The primary function or purpose for our existence has been emergency
response, or responding to emergency responders if you will. We disseminate the
medical information under explicit authorization from our members and we
provide family notification services. There are a number of other follow-up
services that we provide based on emergency episodes that get us into the arena
of outcomes and other statistical information summaries.
The personal health record that the member maintains is accessible via the
web, it’s repository based and we are in the early stages of a pilot program of
using a USB token client application version of that personal health record
that will allow an individual to carry around the application and the
information on a token like the one I’m holding right now, USB token, and it
will also have a provision to keep all the information in sync with the central
Other services that MedicAlert offers are also advanced directives
management, we are the designated DNR issuer and holder of DNRs for I believe
it’s 11 states in the United States, so you can only get those forms from
MedicAlert. We also have clinical trial protocol support and management, we’ve
done a number of clinical trials where the members of the clinical trial, I’m
sorry, the participants in the clinical trial become MedicAlert members. And
the membership and participation in MedicAlert in those clinical trials helps
reduce the time for notification of serve adverse events in clinical trials
from up seven to ten days down to in some cases minutes before the physician
investigator is made aware of any adverse events.
We have a travelers program as well, as service for traveling, it’s a
repatriation service, it’s not really an insurance service, but it’s in our
So we have a very rich offering and functionality that supports that
What basis did we select the functionality? Well, most of these 48 plus
years of service to our members was the basis, constant member interaction and
member feedback. We’ve done a lot of member surveys and focus groups, we’re
just completing a set of surveys and focus groups with Harris Interactive right
now, we’ve engaged them to poll our members and non-members as well as
physicians and other emergency response personnel to assess their views of the
value of the functions and the services that we provide.
We’ve participated in standards organizations, I’m personally active in
HL7, ASTM, the object management group, we are very active in all those
standards efforts, the electronic health record, the CCR, and interoperability
We have close relationships with many organizations in the emergency
response community that feed back into the functionality that we provide
including the Emergency Nurses Association, ACEP, the American College of
Emergency Physicians, as with the American Society of Law Enforcement Trainers,
AMA, the CMA, and the list goes on, I just ran out of slide space there, we are
constantly upgrading and maintaining those relationships. And we’ve published a
teaching manual, we do an awful lot of professional education, seminars and
training sessions for emergency response technicians, and this training manual,
teaching manual that was just published recently called Geriatric Emergencies,
an EMT Teaching Manual, was a compendium of about 15 or 20 professionals in the
emergency health care field and MedicAlert funded through a grant.
The usage, who is actually using the EHR? I’ll give you a couple of
profiles of our members, the age breakdown is 19 percent are age 18 and below,
we’ve got 19 percent from age 19 to 45, 32 percent age 46 to 64, age 65 and
above are 30 percent.
In terms of diseases these statistics are instances of reporting at least
one of these diseases, 31 percent of our members report diabetes, 34 percent
hypertension, 15 percent asthma, 21 percent cardiovascular, I’m sure most of
the health care professionals would expect this kind of a breakdown.
What features are actually being used and how frequently? Well, members use
all of the features and functionality of our electronic health record quite
frequently. The personal information, the personal contact information, and
more importantly the medical information management functions are constantly in
use and I have some statistics on that here. The full functionality for
information update and product orders is available either through our call
center or on the web, about 33 percent today of all the activity in our
repository, whether it’s orders or updates or changes to information comes via
the website and that’s growing quite dramatically actually. We do about a
thousand, this is medical information updates, this isn’t demographic or
contact information, but actual updates to the medical information combined
sources we do about a thousand a day so that’s a fairly large volume for us and
for any facility I think of maintaining health information.
We respond to about a thousand true emergency calls per year and I use the
word true emergency because we get a lot of calls on our emergency line, it’s
the number that’s engraved on the back of the emblems that we issue to our
members, oftentimes we get informational calls, we get lost emblem calls, and
we pretty much reclassify all of those calls as not true emergencies and we
process those as normal business transactions but there’s about a thousand true
emergency calls per year, roughly three plus a day —
DR. HUFF: How many members?
MR. HARRINGTON: I’m sorry, we have four million members worldwide, 2.8
million in the U.S.
So we have a thousand opportunities to protect and save lives directly
because of an incident or an episode each year, which kind of goes back to I
think Dr. Ortiz’s question earlier of is there demand for personal health
record. In our view, and I don’t want to sound sanctimonious about this but if
you save just one life how do you value that. And we believe that the value of
all the information that we keep does save lives, how many of those thousand
emergency calls per year end up actually saving a life because they called,
well there’s really no way of knowing, we think it’s significant and I think
you’ve got to factor that in when you address the issue of demand. All these
surveys that people do and we do them ourselves of the market and how much
people would be willing to use the personal health record to me are backward
looking and they’re not forward looking in terms of the contexts that will be
emerging, the frameworks that will be emerging as we move forward in both the
information age and also the health care information age. So I’d just like to
throw my comments into a very good question by Dr. Ortiz.
Our total call volume at the contact center, we process over 350,000 calls
per year and it’s growing, I get the figures every month and our telephone call
volume is well over 35,000 a month now and our website averages about 18,000
page views per day. So we are here, we have been here for 48 years, and we have
a lot of members that we support.
What standards are we using? Well, actually today we’re as I mentioned
before we are active in the object management group, I’m chairing myself the
joint task force between the object management group and HL7 to develop
specifications for services for interoperability between the health care
systems, particularly those that are going to be based on and will implement
the HL7 electronic health record functional outline. ASTM, we’re active in
working on the CCR specification. The current standards that we do employ,
ICD-9, CPT, we use Sequel(?) of course and TCPIP but the important health care
standards are ICD-9 and CPT coding.
In development, I’ve got a team of developers that are working on the
future of course, we’re using OMG’s model driven architecture and UML. We have
the HL7 electronic health record functional outline, we are implementing, our
future directions will be based on that. Of course the CCR from ASTM E31,
SNOMED plays a big important part of our future in terms of nomenclature,
terminology mapping, particularly since we’re a global organization it’s very
important for us to be able to map terminology. And as an active participant in
HL7 we’re going to be moving along with them with their information model into
the version three.
What is our business model? Well we have a subscription business model
today, we have what’s called advantage membership that we charge $35 dollars
for the enrollment year, which includes emblems and recurring costs.
To what extent do we believe our products should be interoperable? Well,
we’re a strong believer in interoperability, we think they’re very important,
as I mentioned I’m working on the standards for interoperability. We don’t
think the industry will be able to implement this without interoperability.
What are the barriers? Of course evolving standards are barriers,
transition from message based interfaces to services based. The EHR functional
outline is a great step forward and it’s going to have many profiles attached
How is user consent authorization and control handled? Well, our members
give us explicit consent for information to be released on emergencies when
they join, we have a system of authorization that’s more granular then that
that’s in our development organization right now. All the medical information
is controlled by the member or the patient today, as we move forward we’re
going to be tethering ourselves, that seems to be the term, to other parts of
the health care industry and implementing interoperable interfaces with payers,
with providers, with health plans, with physician practice management systems
and other sources of information, so this release of information will continue
to be very important.
And the scope of our coverage, how does our product or approach reach out
to users with varying degrees of access to care, etc.? Well, as a non-profit we
partner with disease related organizations like the American Diabetes
Association, Asthma Allergy Foundation, Peanut Allergy Foundation, etc., again
the list goes on. We share endorsements, we share programs, we point to each
other’s websites, we have even more closer relationships with some of the
organizations. Again our access is via a 24 hour contact center through the
U.S. Postal Service or more importantly via the internet.
We also provide sponsored memberships at the county clinic level, we have a
number of relationships, one of them is at the Tri County Regional Center in
California down by San Luis Abisbo(?), Santa Barbara, Ventura counties where we
provide sponsored memberships for qualified patients of those county clinics
and we do that all over the nation. We have close ties and partnerships as I
mentioned before with professional organizations.
So this is a kind of pictorial of the MedicAlert community pretty much
across the board consumers, government positions, hospitals, public safety,
pharmacies, clinical trials, and finally MedicAlert will be a global repository
of electronic and personal health records focused on quality of care, patient
safety, enhancing the health care experience of our member community. I like
the idea that Brian was talking about, it being a public utility, we sort of
think we are a public utility, another analogy has been similar to a bank where
you can deposit your health care information and control it, the Quicken model
we’ve used, so we feel like all of those are pretty relevant descriptions of
what we do.
And that’s it. Thank you.
DR. LUMPKIN: Great, thank you, thank all of the panelists, this has been a
very interesting morning, very thought provoking. We’re running a little bit
behind schedule so we’ll have time for some questions, we’ll go for a lunch
break, and then we’ll try to get back on time with our afternoon panel. So
DR. ORTIZ: Thank you very much for some very enlightening presentations. My
question is, I want to talk a little bit more about the interoperability issue
and don’t take this as a criticism, it’s because I’m trying to push and figure
out how do we get from here to there, that a lot of the benefits of this comes
if you have interoperability, not interoperability where all the SimplyWell
sites communicate with each other but how does SimplyWell communicate with
Geisinger, how does Geisinger communicate with MedicAlert, because I imagine at
this point that a lot of that’s not occurring.
Now obviously we’re kind of an early stage of that and everybody always
says we’re for interoperability, of course we are, and everybody is saying oh
yeah, we’re using HL7 functional models, but I’m wondering is there anything
actually concrete going forward to actually start putting this into place
because I imagine when SimplyWell, and I don’t know if this is wrong, but your
people go to the employer and they get their blood drawn and stuff, I imagine
when they get that information the interoperability is that they go see their
provider who’s working in a paper based system and they print it out and take
it with them if they remember to do that. But in an ideal situation obviously
that SimplyWell would interact with an electronic system at a provider’s place,
perhaps you’re getting ready to draw a thyroid function test or a glucose and
it says oh, this was already done six weeks ago by the provider, do you really
want to repeat this, or you get the result and it gets transmitted to the
So once again I’m not trying to be critical but I’m wondering are you guys
doing anything or what can you tell us about things to actually move this
forward in the real world?
MR. MCFARLAND: Here’s how we do, granted like with the population over here
in Fairfax obviously very few of those Fairfax employees are served by
physicians in Nebraska, they have not had the long history association of our
CLIA(?) approved lab results, etc. What we do with the patients over here in
Fairfax is we affirmatively, because we capture who’s part of their care team,
all of their OB-GYN, primary care physician, etc., and begin to open a
dialogue, we have two folks in our staff who do nothing but practice management
consults with physicians that are across the country indicating the value of
the SimplyWell medical record, having them put it, giving access to those
physicians over here in Northern Virginia, letting them know who we are. If the
participant gave them access electronically we give them that user ID and
password, and we also fax to them, which is an electronic mechanism, it’s an
automatic fax machine out of the application, the physician summary and the lab
results to these non-network physicians.
But one thing that’s important is even these physicians in Northern
Virginia are part of the network of the self insured plan so again, if our
system will automatically catch, because we also do claims payment analysis if
a physician indeed did order or reorder a passing glucose, hemogram(?),
whatever panel, and if it was not needed because they had already gone to a
SimplyWell worksite experience the third party administrator will not pay that
claim. So that’s how we catch it, we catch it at the claims level, when the
physician that repeats duplicative lab, when they do it, that’s why integration
with the TPA is so important, or I should say the claims payer is so important
or the ASO is so important. We don’t simply pay physicians for duplicative
DR. LUMPKIN: Scott, let me just follow-up on that because of where Eduardo
was going. Back in Nebraska do your physicians who started this up in the
medical center have an electronic health record?
MR. MCFARLAND: It’s interesting, they’ve spent a lot of money as a lot of
health systems across the country on electronic medical records and electronic
health records for provider based, or I should say health system based, and the
funny thing is a greater volume of their patients are on SimplyWell, which we
really didn’t mean to become a personal health record, we just thought it was
important that patients have access to their health information. So the
patients are really sort of driving the physicians and the systems to utilizing
this collection point of information. And we’re built on a Lynex(?) platform
because I’m a huge believer in open source technology so interoperability in
that context with clean datasets that can be verified is actually very easy for
our application to be built upon.
That’s why I mentioned very early about the Medicare and the Medicaid
population, there’s a lot of data out there related to those individuals,
converting that data, utilizing that data, and gaining information and more
importantly gaining clinical knowledge from that data, I don’t see the value
there, I think you need to start with clean datasets, work with working age
populations who grow up and become accustomed with the personal health record
and begin to become a consumerist push if you will towards the systems and
toward the providers for interoperability. And I think we’re, it’s interesting
because we’re really, we’re very, very new to the electronic health record
space and we really have never had a focus become that, we sort of
unconsciously developed it as part of our services for preventive medicine.
MR. HARRINGTON: Could I comment on the issue of interoperability? It’s near
and dear to our heart, I have really a one word response and it’s standards. As
I mentioned briefly I’m deeply involved in the standards for developing, in the
efforts to develop standards for interoperability, it is a critical issue for
interoperating between systems and I would invite everybody, all my panelists,
join us in the HL7 health domain process, that’s working very closely with the
HL7 services project, that’s part of the HL7 electronic health record and I see
Linda Chevy(?) here today, and we’re all working very hard to take this
functional outline, map it with the information model that HL7 has developed
and then come up with standards for interoperability between systems that
implement those models.
It’s absolutely critical important and I’d like to say I think it’s more
important for interoperability standards to be developed then for data
standards to be aligned between disparate system because that isn’t going to
happen totally and so we think that the value is in the interoperability, if we
can agree on what the if you will, the analogy is the RCA jack on your stereo
component looks like, then we’ll pass stuff over that wire and you can just
plug in a new component at the other end.
It’s also critically important, I’d like to invite the afternoon panelists
to join us in these efforts, Cerner, Epic and McKesson because those are the
large vendors, have to be involved —
MR. BLAIR: This is Jeff, can you hear me?
DR. LUMPKIN: Yes, go ahead, Jeff.
MR. BLAIR: By the way the testimony has been just tremendous, very, very
helpful and informative, I’m impressed with all of the systems. This is a
question about whether any of the panelists have had an experience whereby
you’ve run into a conflict between protecting patient privacy and safety, let
me be specific. I recently was at a health care conference where some of the
e-prescribing vendors were indicating that they were receiving medication
history information from PBMs and payers and that was to go into an
e-prescribing system where there would be clinical decision support for drug to
And it turned out that the payers were filtering the information that was
being provided to the prescribes. The filter was to filter out any medications
that could easily be identified for AIDS, for sexually transmitted diseases,
for behavior health/mental health conditions, because they were more concerned
about litigation with respect to violating privacy then the possibility that
they would have incomplete information for drug to drug interactions for
medications. That conflict between protecting patient’s privacy and being able
to provide the information to ensure their safety, have any of you run into
either this or a similar conflict that we should be aware of?
MR. HARRINGTON: I’ll speak for MedicAlert, we have not, we have not run
into that kind of a problem but we haven’t played in the arena that you
described of prescriptions and we are certainly aware and working on the
appropriate release authorization mechanisms as we move forward down that road.
But to date that’s our model, some of the other panelists may have had
MR. MCFARLAND: We draw blood all across the country, we really have not run
into it from a pharmacy or after the fact point of marketplace delivery but
because we gather clinical metrics we find HIV every week, we find Hepatitis
every week, we find women who are pregnant who don’t realize they’re pregnant.
So you can imagine, there are certain public reporting issues associated with
those particular infectious diseases that we need to direct to the appropriate
channels so we have to handle that exactly right, comply with federal law, and
also do everything on our, do everything possible to protect the patient.
DR. WALKER: We have some experience with that in the sense that, Jim
Walker, Geisinger Health System. We provide patient information to external
physicians who care for patients that we care for and when it’s a specific
encounter, a hospitalization, we automatically provide that information because
the physician is involved in the care. When it is providing access to the
entire EHR, which in our case at least often does go back decades or more
because physicians do abstract old information even if it’s not electronic,
sometimes physician access to that record is delayed by the requirement to get
specific authorization under HIPAA. So if that’s, and there probably are times
that’s patient safety issues, certainly efficiency issue. As it turns out we
think we can use the patient EHR to provide the patient the ability to provide
authorization electronically in real time and get authorization in real time to
eliminate at least that form of delay.
DR. DEERING: Well maybe actually I’ll leave this floating in the air
because I know that there are intended to be hearings on consent and
authorization and specifically I think that we’ll get into this. What I wanted
to sort of leave floating in the air is to the extent that consent and
authorization are clearly documented I would wonder to what extent the
providers and their lawyers would view that as covering those concerns that in
Jeff’s scenarios were preventing them from doing, well, posing them with the
dilemma in the first place. Perhaps there’s not an answer to that but that
might be one of many issues to look at.
DR. LUMPKIN: And something that we can do jointly with Privacy and
Confidentiality, there are obviously interplays between that as well as various
generally not federal requirements for reporting but generally state and local
requirements for reporting which may have some impact.
At this time, just one quick question to follow-up on Eduardo’s one for
David, and that was are any of your records updated or populated electronically
other then I mean someone entering data on a website by hand but in an
MR. HARRINGTON: Today no but we have, we’re about to roll that out is all I
can say, we’re definitely working on that, that’s why interoperability
standards are so important to us.
MR. MCFARLAND: All of our blood is bar coded is there’s no patient
identifiers on the vials so it’s computer —
DR. LUMPKIN: But I would assume that those lab results are done
electronically from the lab then updates to your records —
MR. MCFARLAND: Right, automatically.
DR. LUMPKIN: In an interoperable way.
MR. MCFARLAND: Correct.
DR. LUMPKIN: And Brian yours is really just an inside out so that’s not
really an issue.
MR. BAUM: Right.
DR. LUMPKIN: Scott, have you addressed that issue of following up on
interoperability, how you would see that occur? I’m sorry, Brian, yes.
MR. BAUM: Yes, that’s something that we’re building into the model, we are
not live yet at this point so we don’t have actual experience but our building
that into our program.
DR. LUMPKIN: Great. Okay, at this time we’re going to have a break, we’re
going to try to get back at 1:00, and there’s food up one level.
[Whereupon at 12:25 p.m. the meeting was recessed, to reconvene at 1:15
p.m., the same afternoon, November 12, 2004.]
A F T E R N
O O N S E S S I
O N [1:15 p.m.]
DR. LUMPKIN: I’d like to thank our panel for the afternoon, welcome, and
I’m going to ask that you introduce yourselves so the people on the internet
can recognize your voices and associate with the name and we’ll start off with
the panel. So I guess we’ll start off in the order in the books, so Seth?
MR. RUPP: My name is Seth Rupp, I’m a vice president, Cerner Corporation,
in our intellectual property division. I’ve got the strategic and engineering
responsibility for our physician, our nursing and our consumer applications.
MR. DEVAULT: Good afternoon, I’m Peter DeVault from Epic Systems
Corporation in Madison, Wisconsin, I work in enterprise integration which means
that I work on a lot of our standards work as well as our implementations of
our larger customers.
MR. BLAIR: That voice really was cutting out, if they could get closer to
DR. LUMPKIN: I think that’s because he has two of them. Kristina?
MS. ROLLINGS: Hi, my name is Kristina Rollings, I’m a director for patient
strategies at McKesson and responsible for developing the product strategy for
our patent facing solution.
DR. LUMPKIN: Great. Take it away, Seth.
Agenda Item: Panel 3 – Organizations and Vendors Offering
PHRs that are Derived from EHRs – Mr. Rupp
MR. RUPP: It’s a pleasure to be here today and I appreciate being given the
opportunity to present on behalf of Cerner. As indicated in my title slide my
presentation today will address the questions posed by the NHII Workgroup while
accentuating Cerner’s juvenile diabetes personal health record initiative as an
example of our progress in this space. For future reference today Cerner’s PHR
solution set is branded as IQ Health, I’d like to remind everyone today that
November is National Diabetes Month.
Cerner believes that there’s three main drivers for PHR’s existence and
utilization, that is to manage health with clinical motivations, to manage cost
with financial and ROI motivations, and to manage efficiencies within the
general complexities of health care. Furthermore we believe there’s five main
users of PHRs, we believe there’s the healthy but concerned, and I call these
our marathon runners and I do believe that they are the exception. I think
there’s the healthy but at risk so I’m healthy today but I know that dad got
type II diabetes and mom had breast cancer, and I think as our industry evolves
this is going to have a larger play in genomics as this gets pulled in.
I think the chronically ill, so this is our diabetes, our asthmatics, those
with CHF, COPD. I believer there’s the proxy for health care that came up some
this morning, what I’ve coined parents and peds, hospice, general familial
needs, the role of guardians in the health care system. And then broadly those
seeking efficiency so viewing my appointments, HSA again that came up this
morning, clinical systems integration, my ability to facilitate refills.
So addressing one of the questions about the business models in the PHR
space I think they’re as widespread and varying, and having varying degrees of
success. I believe they have spans the models, and these aren’t necessary
Cerner’s models, free to the individual, monthly subscriptions to individual
consumers, monthly subscriptions to the PHR sponsors and then subsequently free
to the consumers. PHR bundling with electronic medical record sales, we’ve seen
per member per month payer type initiatives. We’ve seen charging by individual
e-consultation sessions whereby a consumer may swipe a credit card. And
e-consultation payer reimbursement models, as you may be aware July
1st there was approval of a new CPT code, CPT 0074T I believe, which
facilitates the reimbursement for e-consultations.
Cerner’s vision has been designed around a community EMR and PHR model that
is interoperable and of one common architecture. As of September Cerner has 266
unique clients live and 642 unique sites live on our millennium architecture.
The concept of the community model is depicted in this slide whereby we start
with the automation of the care process and the elimination of paper. This is a
creation of an EMR in our acute care, our ambulatory, our critical care, our
long term care, and our rural hospital spaces, connecting then to the person by
virtual personal health systems. This is a community outreach to the individual
consumers, to health plans, employers, payers, schools and governments, and
PHRs has relevancy in this place. We then structure, store and study the
evidence to create new knowledge through the interoperability and the
aggregation of this data to finally close the loop by implementing knowledge
driven care based on this learning using all things learned to drive evidence
based medicine, health maintenance and clinical rules.
Hence addressing your question against the barriers to interoperability
with the EHRs and across PHRs, vendors and clients, I think the issues are
fundamental architecture and technology based, so not unlike today’s same
issues in the EMR space, lack of integration and common architecture on large
scales relegates vendors and clients to traditional interfaces at best, not
unlike my colleague from MedicAlert this morning talking about standards and
However furthermore in the PHR space this conundrum is exacerbated by
clients in various internet portal strategies and subsequent technologies that
create a new set of integration challenges, questions around data ownership,
branding is something huge that we see in addition to security concerns. And
lastly integration barriers are different in our opinion based on PHR
implementation scale, whether this be individual portals versus communities
versus state levels, like we’re seeing at CMS, versus national initiatives.
Cerner’s implementation has all its data in one common place, the users
simply access the data through different interfaces. In this example we show
that physician X has to do an online consultation and the consumer accessing
data via their PHR portal, other interfaces to the same data may be
vis-à-vis the patient EMR, the ambulatory EMR, the emergency EMR,
pharmacy management, etc., but all interfaces are viewing the same data.
The standards being used for the PHR are hence leveraged from those same
industry standards applicable to the more mature EMR space within the HCIT
industry. Cerner uses standardized medical terminology with vocabularies like
the NCPDP, SNOMED-CT, ICD-9, CPT, DICOM, and HL7.
Regarding your question how is consent authorization control handled,
similar to the stories described this morning also there’s two parties so
there’s two answers. Clinical provided authorization is established by the
relationship to the patient and to an encounter, if the clinician does not have
the appropriate relationship they’re unable to access their information. On the
individual consumer side members have the ability to grant access to those
health care professionals in addition to proxies around health care or
Cerner’s PHR is real, here are four examples of our implementation around
specific conditions at Wynona Health, the Minnesota Spectrum Health, and
Michigan Genesis Health in the quad cities and Heartland Health in Missouri.
Regarding the committee’s question who is actually using them and what
features are actually being used and how frequently are they being used, Cerner
currently has 28 signed clients in advance of the juvenile diabetes initiative
I’m going to tell you about, and 14 of them are currently live. We’re seeing
the highest utilization between people with chronic conditions that have a need
to manage and interact with their health care team besides citing the likes of
diabetes as must CHF and cancer management. The most interaction that we
currently see is between those patients and their doctor’s office, the patients
are able to document their self care and establish a care plan with their
provider. Patients view this interaction between visits as invaluable and
providers are able to provide the education and feedback they may not be able
to handle within every office visit due to time constraints or otherwise.
Providers like the asynchronous communication with their patients.
Another example of iconic use around PHRs in chronic conditions is Cerner’s
October 13, 2004 announcement of its ten year gift of its personal health
record, again branded IQ Health, to every juvenile diabetic in the United
States who’s interested in it. Cerner predicts it will invest $25 million
dollars over the next ten years to provide this personal health record for
children with diabetes. The offering entails a personal health record, a
diabetes management center, and provider messaging capabilities.
Cerner chose to do this project because of the great opportunity to
transform the health care of a very important population around juvenile
diabetes and you can see the motivations here, 13,000 new diagnosed every year,
one in 400 children, it’s just an amazing problem and one that we chose because
it’s very interactive, the children having to one, it’s a family disease, not
just an individual child’s disease, and what they have to do as you guys know
around individual blood glucoses, around exercise, around carbohydrate intake
and insulin, it’s a very interactive and data intensive disease process.
Addressing your question how does a product reach out to users with varying
degrees of access to care, levels of health literacy, etc., this is a very
tough problem to solve and one that gets asked every time I present on the
juvenile diabetes initiative. Access to technology is the starting point, the
ubiquity of the PHR is using internet technology is the hope. In general
Americans are using the internet to find health information, those with lower
incomes may have access via community centers, workplace access, and libraries.
As PHRs are rolled out into the community there needs to be a strategy to give
lower income populations access to the internet.
While the national averages depicted here seem to be around the 70 percent
range our personal observation, unlicensed observation, is that we’ve seen the
lowest in rural Tennessee at around 47 percent, the highest from our client
base that’s been quoted is in Evanston, Illinois, downtown Chicago, at about 95
percent. Our content educational materials are written for common levels of
literacy and education, another key challenge I have not heard talked about
today is non-English language support.
A national website for our Juvenile Diabetes Foundation effort has been
created as an outreach mechanism to individual patients, their families,
clinicians and care providers to sign up for participation in this program.
Interest in the juvenile diabetes initiative has been very high, 37
organizations in a mere month, month tomorrow, have expressed interest in
participating in this since its announcement and inception. Primary interest
and partnership has been from the pediatric hospitals. The network we could
create would provide as we know the numbers today PHR and EMR connectivity to
the patients of approximately 375 endocrinologists geographically dispersed
across the United States. Cerner predicts this initiative will gain further
momentum in enrollment and participation as the program evolves. The
opportunity around this network of care, around the condition of diabetes, is
Regarding your question what functionality is offered, on what basis is it
selected, the juvenile diabetes PHR has modules for both the diabetic consumer
and their family in addition to modules for the clinician and care provider,
and care provider team to interact. Cerner’s IQ Health Solution offers patients
a solid foundation to manage their health, this includes a personal health
record, secure messaging with their care team, a specialized home diary to
track their daily activities, a historical view of those activities, and a
customized plan of care. Basic provider connectivity includes secure messaging
service with connectivity to other providers and their patients.
Providers can offer connectivity with the clinical EMR that is critical to
allow integration into the provider’s workflow and access to the patient’s full
non-PHR medical record. This is big so you’re actually from the consumer view
looking into the acute care or ambulatory care EMR to see the data so questions
earlier about how the data was entered, this is actually viewing the
information that the clinicians themselves put in.
This functionality was selected on expert’s feedback and design from
diabetes, their families, pediatricians, endocrinologists, nurses, diabetes
educators, dieticians, and technology providers like device companies.
In summary Cerner believes that the PHR, that PHRs are a key element in
connecting the consumer to their health record and provider, being able to
afford consumer access to their EMR data and connectivity to their clinicians
is what they desire. It is our experience that with all potential uses of PHRs
that the chronically are the most motivated around the purpose and are more
likely to utilize such technologies as part of their daily health status
management regimen. Cerner’s juvenile diabetes initiative program is a program
designed to give real PHR technology to connect the consumer, their family and
their care provider to better their life and their health.
DR. LUMPKIN: Thank you very much. Peter.
That may be a bad sign, you don’t even have interoperable Power Point
between you two. Just a little bit of a glitch, just while he’s loading up
maybe I can ask a couple of questions. When you talked about the central
repository I assume that’s for an installation as opposed to Cerner having some
big repository some place.
MR. RUPP: Actually all of our personal health record clients today are
running out of one centralized location in a data center in Kansas City. So it
is an ASP model although it is today a national model because of the diverse
geography, but yes, all of our clients are running in one data center.
DR. LUMPKIN: Okay, great, thanks.
Agenda Item: Panel 3 – Organizations and Vendors
Offering PHRs that are Derived from EHRs – Mr. DeVault
MR. DEVAULT: Good afternoon, again I’m Peter DeVault, first of all I’d like
to say thank you for being able to contribute to this very important project,
we’re excited to be able to participate. I’m going to be talking about a couple
of things today, first of all introducing you to Epic’s MyChart application
which is an outgrowth of our electronic medical record system, and also
offering some thoughts on connectivity and interoperability.
A little bit about Epic to begin with, what we offer to our customers is
primarily an enterprise electronic health record system with adjunct services
such as scheduling, billing, production of claims, etc. And our MyChart
application, which is the equivalent of the PHR that we’re talking about today,
is a layer of services offered to consumers on top of that set of applications.
We can for today’s purposes group those into two sets of provisions, one is
access, both to information and to services, so access to electronic medical
records data that the health care provider has as part of Epic care, their
electronic medical record, and also to services, so online scheduling, being
able to track referrals and claims, being able to request services of secondary
care providers, etc. And also very importantly communication with providers,
customer service, and payer. And in fact the way we like to think about MyChart
is really as a set of applications that strengthen relationships between health
care consumers and their health care providers and payers.
The next point I’m going to make I think is important and has been raised a
couple of times in the second panel, which is that because this is, and I
really like the metaphor of a tethered application, I think that conveys
exactly what we’re talking about here, because it’s a tethered PHR it’s
connected integrally to the electronic health record system that the health
care provider has, it’s a filtered view of that patient record. What that means
is that the physician typically chooses which lab results to share with the
patient and which problem list entries on the patient’s list of chronic
problems are divulged to the patient or in what form they’re shown. So for
example the raw ICD-9 codes that describe a patient’s chronic health problems
might not be what’s shown to the patient, it might be a subset of those and
they might have more friendly terms or a narrative description of those
And typically our customers don’t offer copies of their progress notes to
the patients, so it is very much a filtered view and I think that has some
advantages and disadvantages. It really doesn’t solve the problem that we heard
about this morning with the father who is carting around two notebooks worth of
medical records to a variety of providers because in that situation your really
do need I think 100 percent of the patient’s cumulative medical record from all
the different health care providers involved. However, it does allow for a
shared interpretation of that medical record data between the provider and the
consumer of that information, something which I think is very important. As has
been raised earlier today most consumers are not in a position to understand
and correctly interpret the raw health care data.
And that you might have noticed at the beginning of this slide show we
refer to MyChart as a shared medical record or a shared health record, it’s not
purely the provider’s record but it’s also not purely the patient’s record,
it’s a set of data and a set of functionality that allows for shared
communication and interpretation of that information.
I do want to address some of the specific questions that Mary Jo asked us
to, authorization is pretty typical from what I’ve heard other folks talk about
today, one of two models that our customers use, either it’s a shared decision
between a provider and his or her patient and an access code is generated which
then the patient can log into the web service and generate a user ID and
password and from that point on they have access to some set of functionality.
Or it’s offered administratively so it’s not up to the provider, the specific
provider necessarily to give access to those services. Sometimes there’s a
combination and when I talk about business models in just a second I’ll
describe what that means. And in all cases that I’m aware of our customers
require some form of written formal consent that it’s okay to share this
information over the internet in a secure fashion.
Our customers have been pretty innovative and as have Cerner’s, have pretty
much run the gambit on the different kind of business models available for
providing these services to their patients. Subscription is the most common so
a monthly or a yearly fee for having access to a certain set of services. These
are in many cases tiered so for example one of our larger customers gives free
access to a certain set of services, most of those are view only services, so
being able to see your lab results, the orders that have been placed for you,
your medication and prescription activity, your current referrals and claims,
that sort of thing, being able to view that information is provided as a free
service whereas being able to communicate with your providers or to schedule
appointments becomes a subscription service offered on top of that.
Some of our customers have some experience with transactional services
which would be made easier and more standard with the ability to charge for
e-visits in the near future. And then we also have some health plans who offer
these services as part of certain coverages.
Just a few numbers, we have over 150,000 active users as of, I think it was
156,000 as of the end of October, that means they’ve logged in at least once
over the last year, that’s how we count active users. Of those users 35 to 40
percent log in in any given month. The largest demographic segment, we don’t
have real great breakdowns on this because our different customers collect
different kinds of demographic data but one piece of information that kind of
throws out the observations made by others is that the largest demographic
segment is, and the largest growing segment, is women in their 40s.
One question was how do we address specific populations and really the
answer to that is that we provide some tools that allow our customers for the
purchasers of our electronic health record system to reach out to populations
that are important to them. So we offer a variety of health maintenance and
disease management tools that are part of the electronic health record system
and which then allow us to push out reminders, alerts, useful pieces of
information to the patients that are using MyChart. So for example diabetic
patients might be reminded that the protocol with their provider is that you
have a yearly eye exam, foot exam, etc., and that okay, we notice you haven’t
had an HBO1C in the last year you should come in for that, so we can push out a
subset of the same alerts that we make available to physicians and other care
providers and make those available to the patients as well.
On top of that we’re busy at Epic right now developing several sets of
specialty based functionality on top of our base electronic health record
system, specifically right now in oncology. And as we all know cancer patients
are more likely then any other population, or most other populations of
patients, to take a very active interest in their care and to seek out
information about treatment planning. And we’re making a lot of the same
features that are available to our oncology and care providers available to our
patients as well so online treatment planning, being able to schedule upcoming
chemotherapy appointments, etc.
Now at this point I’d like to offer some, some of it’s sort of
autobiographical, how did our MyChart application develop as an outgrowth of
our electronic health record system but I also think that it serves as a model
for what we are likely to see many personal health record systems develop.
Certainly the services that we offer as part of MyChart began as the
services that we offer in our electronic health record system. Some of that was
because it was easy to do, we already know how to schedule patients so lets
just open it up for patients to schedule appointments. We already know how to
display results and problem lists, lets open that up to the patient. And of
course some of it was very much that our customers wanted to provide certain
services to their customers, the patients, as a result of focus groups or
competitive advantages that they sought to gain in the marketplace. So it began
of course as just that, a set of, a subset of the services that were part of
the EHR system itself.
And of course at that point you also want to open up connectivity to other
resources that are available on the net so we allow our customer’s patients to
search outside resources using some of the key words and some of the concepts
that are stored as part of their electronic health record so you kind of get
that for free as a result of some of the coded discreet information that’s part
of the electronic health record system, you can also search some external
Now one possibility for the evolution of this kind of model is that the
tether gets weaker and weaker and the PHR sort of starts to orbit the
electronic health record system or even starts to orbit several electronic
health record systems. And this I think is probably where we will see some PHR
systems evolve so they began as component pieces of electronic health record
systems but then start to develop many of the features of stand alone systems
and the ability to connect to a variety of data sources.
And I think if we follow that analogy it really makes sense to start
thinking about the PHR, and I know most of us are anyway, as a special case in
electronic health record system, that once we, and I wanted to bring this up
because I think there’s a danger that we might start trying to solve the same
set of problems twice, where many of us are already very involved in functional
standards for an electronic health record system, I work on the HL7 committee
with Linda and Mary Jo and Michelle and several others in the room, that’s not
a fundamentally different project from defining a functional set for a personal
health record system. Similarly when we solve the problem of interoperability
of electronic health record systems I think we’ve also solved the problem of
connectivity of PHRs to EHRs. I think we need to think of it as the same set of
problems and not try to solve them twice.
There’s supposed to be some funky animation on here but it must not have
survived the translation so that’s kind of ugly. What I meant to convey here is
that there are really two ways we might see a PHR system, one is just another
node in the network of networks, it’s a system that can talk to other systems,
interoperate with other EHR systems. And another is that it’s really a virtual
record that knows how to go find information from the various care providers
who are using electronic health record systems and accumulate that into a
virtual record at any given time.
So one of the questions was should EHRs interoperate with PHRs and I think
you’d be surprised if at this point in my talk I said no, of course the answer
is yes. And I think there’s a couple of different ways of looking at it, a PHR
may be a special case of an EHR and the work that we have done at HL7 in
defining what is the library of possible functions that an electronic health
record system might subsume, largely or at least in part of the very
appropriate insistence by Mary Jo and some other people those functions do
include a lot of the tools that we require for a generalized PHR solution and
so I really think that we should think about PHR as a special case of an
electronic health record system. And of course it might also be a functional
subset so that’s the path where the EHR spawns off various flavors of PHR.
That’s what I had to share today and of course I’d be happy to answer
questions at the end.
DR. LUMPKIN: Thank you very much. Kristina?
Agenda Item: Panel 3 – Organizations and Vendors
Offering PHRs that are Derived from EHRs – Ms. Rollings
MS. ROLLINGS: Thank you everyone for inviting McKesson to participate in
this hearing today. We believe that we have a significant contribution to these
discussions based on our experience that we’ve had across care settings. For
those of you who aren’t familiar with McKesson provider technologies we provide
information solutions really across the continuum of care, from a patient’s
perspective there are solutions that capture data, everything from wellness and
prevention all the way to maintenance and also in every care setting we provide
ambulatory solutions, inpatient solutions, home care solutions, etc.
So as we started to think about patient centered care strategy it was with
all these back end systems there’s a lot of data there that is very relevant
for the patient to have access to and we looked at how can we leverage what
these existing investments are already in place in the health care
organizations and create an interface that will patient view into that data and
make it portable for that patient as well. So I want to talk a little bit about
how we got to our patient centered care strategy as it is today.
First we looked at, we leveraged the research that was being done by
marketing firms and back in 2002 we learned that health care organizations were
wanting to serve online patient and consumer ranks and their highest e-business
priorities were marketing and branding and patient and consumer interaction.
And so we’ve seen it evolved quite a bit into that patient interaction.
I’m not going to read all of these statistics because most of you have
probably heard these a number of times but the one that was key to us is
certainly that 77 percent of consumers that are visiting a hospital or a health
plan site report that they have a chronic health condition so this is a large
target audience of the solutions that we’re providing today.
Also consumer trends, we went and looked at what consumers are doing in
other industry markets, they’re doing things like scheduling their travel, and
even selecting their seats on those planes online, they’re doing account
management, their banking and their financial transactions, they’re
registering, ordering and paying for events and entertainment and retail and
business, etc. And certainly online education, the biggest use of the internet
today is research and looking for health information, or looking for
information and health should be no different.
And finally we looked at the consumer trends for self service functions and
how consumers were preferring to do a number of different types of functions
like scheduling appointments, prescription renewals, and medication payment,
etc. And so with that we started to develop a portal for patients.
We continued to do ongoing market research, the first thing that we did was
we provided a web survey out to our very large customer base to find out what
is it your patients are asking for, what types of functions are they interested
in. We also have strategic advisory councils that include CIO Council,
marketing and community relations and PR as well as service line leaders. We
also have patient focus groups our customers put together for us as we looked
to develop functionality for patient, for facing patients, we’d like to know
what it is patients are looking to use and how they want to use it so that the
usability is accurate. Market research from reports, we continue to watch what
Gartner, Forrester, and Pew Internet Research.
So that brings us to McKesson’s patient centered care strategy, the
strategy is about enabling health care organizations to deliver solutions
directly to patients and their families that will positively impact their
interactions and provide the kind of consistent high quality experience that
consumers are looking for.
The solutions that are part of this strategy include personalized online
tools and treatment plans that support patients that are managing conditions
and hopefully lead to healthier outcomes. Also state of the art in room
solutions that give patients an access to a wealth of content, education,
hospital and internet services as well as entertainment while creating a strong
competitive differentiation for that hospital. And then finally telehealth
devices that improve patient monitoring, disease management, increase patient
involvement and reduce the strain on clinicians and reduce the cost of care.
The functions of these solutions will span the continuum of care, provide
interactions from wellness and prevention to diagnosis and pre-service,
inpatient and outpatient experience, post discharge, and finally maintenance or
return to wellness.
Key components of our patient centered care strategy, information to
support collaborative care, how can this information become available for care
givers to share and include the patient as they’re making decisions. Certainly
mobile technology, more self service for those consumers and patients, smart
patient rooms so while they’re in the hospital receiving service that they have
access to those services. And finally integrated proactive disease management.
Our strategy as I said earlier leverages the existing technology and
investments that are already in place, there’s a wealth of data available in
hospital information systems, clinical data repository, ambulatory care
systems, departmental systems like lab, radiology and transcription, medical
records, medical imaging, home care, this all provides patient data that’s very
important to the record. What we’re doing is we’re building applications on top
of that with functions that pull data out of those systems.
Those three applications we’re going to talk a little bit about now, the
first one is called Horizon WP Patient Portal. This is an online tool that
health care organizations and physician offices can provide out to their
patients to provide a secure point of access to do things like communicate with
their care team online in a secure environment. Schedule view and receive
appointments. They can view, inquire and pay their bills. Access to their
medical record, if they want to view it, maybe they want to print it or maybe
they even want to save it off to their own device. Receive test results
immediately or get that from the care giver who has filtered and prescribed
content along with that. Monitor conditions and assess access relevant content
and patient education. Monitor medications, side effects. Receive alerts and
reminders, and maintain daily journals.
The next tool is our in room solution for an inpatient experience. This
device will provide personalized content, patient education, access to hospital
and internet services, entertainment, surveys and requests, and also access to
the patient portal that we just described.
And then finally moving to the return to wellness phase our telehealth
advisor. This is a device that empowers patients to actively engage in managing
their own health care in a home health environment with disease management
protocols that assess signs, symptoms, knowledge and behavior. The patient can
collect these vital signs using medical peripherals. The assessment data is
then transmitted to a central clinical repository. Patients are risk stratified
for targeted intervention and adverse event avoidance. Patients are educated on
managing chronic illness and this is provided through a business relation that
we have with Hero Health Network.
Who are using these solutions? Our development partner for the patient
portal is interestingly enough not a U.S. hospital, it’s a Canadian hospital,
Grand River in Kitchener, Ontario. They were opening a new cancer center, state
of the art cancer center, and they wanted to do something completely different
for their patients. They found that when patients were diagnosed with cancer
they, I think it was somebody had said earlier, after they hear that diagnosis
they don’t hear anything for the next 20 minutes, they’re busy processing that
diagnosis that they just heard so they’ve lost everything that they learned in
that interaction, they get home, they’re ready to start their treatment plan
and they don’t know what to expect. So we’ve developed a suite of condition
management modules that allow a patient to log on and follow a personalized
treatment plan against a calendar according to the weeks and cycles of their
therapy. They can record symptoms, monitor side effects, they can schedule and
view their appointments through the care plan, request and order refills, and
securely communicate with their care team.
The portal replicates educational content developed and used by staff
clinicians and provides vital links to resources and support groups. The pilot
included 55 breast cancer and 23 prostate cancer patients who now use the
system daily to manage their illness.
PatientVision, this is the in room solution, our pilot partner was St.
Vincent’s Hospital in Birmingham, Alabama. Their implementation included a
hospital information and content ranging from what is a nurse to when is
breakfast served, entertainment such as movies, music, and games, surveys for
immediate service recovery, patient education, and also access to the patient
Telehealth Advisor, two examples are South Mississippi Home Health in
Hattiesburg, Mississippi, they deployed 12 devices initially for congestive
heart failure. The physicians are excited and supportive of the program and
they focus on improving clinical outcomes and reducing costs.
Neighborhood Home Health in Westchester, Pennsylvania deployed 15 devices,
also congestive heart failure and COPD. This was a grant funded program for
Neighborhood Home Health, they focus on the Medicare population and physicians
and referral sources are excited about the technology and the opportunity.
How do we handle user consent, authorization, and control? Well, there are,
with the portal there are three different user types, we have what’s called a
guest user, a registered user, and a connected user. Guest user is just
somebody who can come to the site, they don’t have to give us any information,
but they have access to any of the content that’s published to the site. A
registered user is somebody who’s going to give us a little bit more
information about themselves and we’ll allow them to do some self service
functions. We won’t provide them back with health information but they can
provide us with information by submitting things like pre-registration forms,
And then finally our connected user, that’s somebody who we have seen in
person, we’ve validated that they’re who they say they are and they’re given an
activation code. And this activation key is used to secure a log in and
password that will give them access to be able to use all of the functions on
the portal as well as access to their medical record, their account summary,
and things of that nature.
Consent at registration, when they do the registration process with a user
name and password they have to electronically accept the terms, when it’s in
person for a connected user certainly written signature. And finally
authorization, as I mentioned they’re given a user name and a password, for a
connected user they have to have an authentication key and some of our
organizations are actually using different biometric technologies like key fobs
and other devices, security devices that are available today.
Standards, all of our solutions are J2EE based, our connections to the back
end source systems are agnostic, we’ve used technologies like JCA, CORBA, JDBC
and HL7, and certainly we’re continuing to monitor emerging portal
technologies, JSR168 and WSRP.
We believe that we’re interoperable within the McKesson source system
background. What we’re looking at now is how can we follow these standards so
that we can become interoperable with other vendors as well as we recognize
that for a patient to have a full record it is going to cross multiple health
systems, it’s going to cross multiple vendors. So we’re standardizing our
solutions to follow JSR168, HL7, all the different requirements that are coming
out to allow source system independence. Standards and modular approach will
allow these solutions to fit within multiple PHR models. Our solutions are very
patient centric versus in the past they’ve been provider or care setting
What are some of the barriers that we see to interoperability? Certainly
provider willingness to share data, a couple of weeks ago I was at a
conference, we were talking to a customer who was interested in implementing
the patient portal and wanting to pull in account data for patients that may be
seen anywhere within the community so there were other hospitals that they
wanted to work with that had different vendor systems. The barrier that they
found is that those hospitals struggle to share that patient data and so that’s
going to have to change I think for us to get to a connected community.
Certainly MPI resolution, as you’re getting this data from all these
different systems how do we know that the data coming in is the patient that
our record is, so we’ve got to figure out how we do that MPI resolution across
Alignment among other solution vendors and getting everyone to agree what
the approach is. And then final a lack of final standards, as these drafts are
coming out, I think until it’s complete it’s going to be difficult to get the
vendors to code to those standards until they’re done.
And that’s it.
DR. LUMPKIN: Well thank you very much. I’ve got a bunch of questions but
I’m going to look to the other committee members to get us started. Okay, in
that case I’m going to ask a real hard one and Peter I’m going to probably put
you on the spot.
MR. DEVAULT: Sure.
DR. LUMPKIN: I think Kristina said that there was interoperability for the
personal health record within the installations that are McKesson
installations, is that true for the Epic installations? So let’s say for
instance I’m at Kaiser and I move to Group Health in Portland, both of them are
Epic systems, will my electronic health record be able to follow me? My
personal health record?
MR. DEVAULT: It will soon. As you’ve probably followed within Kaiser we are
developing our Care Everywhere platform, not specifically for them but that’s
the first set of users, so that at Kaiser Permanente where they actually have
18 different instances of Epic applications your electronic health record will
be able to follow you. Now what’s made that possible is, and I think this is
very important, what’s made that possible is that they have agreed among all
the different Kaiser regions to standardize on some vocabulary and to have a
certain discipline about how they build their systems such that
interoperability becomes possible at a very tightly bound discreet data level.
So they’re actually doing several steps further then I think most of us
actually mean when we talk about interoperability, they’re making it almost
transparent that there are separate instances of the applications.
Now that’s sort of tackling it from the hardest end first and what that
development is going to allow us to do is within the next couple of years allow
exactly the kind of scenario that you’re talking about where we have less
tightly collaborative communities of Epic users, for example there are spots
all over the country where we have several Epic customers who are interested in
sharing data but aren’t interested in sharing procedure master files and other
sets of vocabularies that there aren’t adequate standards for, where they will
be able to in fact share patient data at a less tightly coupled, less discreet
So I would say we’re most of the way there right now but not completely
MR. KAMBIC: Bob Kambic from CMS. I’m a little puzzled, this morning I think
we really focused on a PHR but you folks are the big players and I mean
basically what I’ve heard is an EHR talk called a PHR talk and whoever said, I
mean I just disagree that in fact they are the same thing. Connecting for
Health spent a lot of time and money and so did RWJ and Markle on this report,
I think there’s a lot of issues in how you’re going to translate someone said
this morning all the medical jargon which physicians and nurses and so on need
into the sixth grade or ninth grade understandability for our beneficiary, our
customer, all of us, people who may not be physicians or nurses. How are you
going to do that?
Yours was extraordinarily complex and I don’t understand how, Connecting
for Health tells us that each person controls his or her own PHR, individuals
decide which parts of the PHR can be accessed by whom and for how long. PHRs
contain information from ones entire lifetime, what I thought, here’s a very
interesting, PHRs are transparent, individuals can see who entered each piece
of data, where it was transferred from and who has viewed it, I mean this
I think that there are aspects to the PHR that are very specific to PHRs
and I would encourage you as the big players to think about this a little more
specifically instead of just translating EHR functionalities and specifications
over and saying oh by the way here’s the EHR and we also cut a little piece out
of it and we have a PHR. Maybe I’m wrong, I’m willing to be proven wrong but I
would like to hear more focus on sort of, a lot of work was put into this
report and it seems to me that maybe you guys aren’t familiar with it and you
need to familiarize yourselves with it a little bit.
MR. RUPP: I’d respond to that, I think they are distinct and I think they
have been confused today so I even went to the extent of using EMR for a lot of
my presentation. I would represent that I think there’s a personal health
record, there’s a concept of a stand alone personal health record, there’s a
concept of a stand alone electronic medical record. I think there’s a concept
that we talked about today in part of Peter’s depiction of a patient view into
the electronic medical record. And then there’s this whole community concept
which we haven’t really talked about today which is kind of the RIO and that’s
kind of being coined the EHR. And then to complicate things further Global
represents it as an EHR which is a culmination of the PHR and the EMR data that
sometimes necessarily clinical decisions aren’t made out of.
So I think part of the conundrum is just getting those definitions out
there but I do think there is that continuum of which I just described five or
six and I think there’s distinct offerings within each of those faces tying
back to the business model. So I think by virtue of having an EMR in the HCIT
industry that we can access I would represent that the value is the
connectivity to the physician. It’s a much different space automating HSA
accounts around financial employer models then it is connecting to the
physician around the chronic condition.
DR. DEERING: I have forgotten which of you said it but you talked about the
providers lack of willingness to share data actually as a significant barrier
regardless of the capacity of the system to share that data. And I’d like to
take that remark on one hand and then coupled with this observation that Bob
just made about are you really just putting a PHR face on your EHR, and ask you
whether for all of those of you in your dealing with the provider, the
providers themselves, have you in addressing their reluctance to actually let
their information go, and maybe it was you because I think you’re talking, was
that your comment? It was a little bullet there.
MS. ROLLINGS: Actually it was mine.
DR. DEERING: Do you get a sense that anything approaching the continuity of
care record or a predefined set of data fields could be helpful in overcoming
that kind of resistance and are you thinking of anything like that? Because
right now all of your PHR faces are so imbedded as I can see it, as you say
they all draw from your EHR data, so are you thinking of any of these sort of
partial solutions at all, partial dataset?
MS. ROLLINGS: Absolutely. In the discussion that I was referencing that
happened a couple of weeks ago it was around the patient looking for account
summary data and all of that data having to come from this hospital system as
well as this hospital system down the road which happens to be a competitor of
this hospital. And in order to accomplish that it needs to be in one place. I
do believe that if it was a prescribed set of data that was sent somewhere
neutral that all these different hospital systems could access, yes, that would
probably address the concerns that we’ve come across with other health care
MR. DEVAULT: I think a standardized subset of information that we all agree
that is sharable, it would be a tremendously useful thing, I think a lot of
health care providers would feel more comfortable sharing that kind of data if
it became standard to do that sort of thing.
And just to address Bob’s comments earlier, I think there is a distinct
thing called a PHR however I will stand by my comments that I think it’s
largely a subset or a superset of functionality of an electronic health record
system and that it is electronic health record system technology that will
enable personal health record systems. And I think there are two different
kinds of systems that we might talk about and I really do like the metaphor of
tethered versus non-tethered, and I would suggest that one of the first points
he made which was that we need to be able to translate the raw health record
information into something that a patient can consume, I think that actually
argues for the tethered model where the PHR is an outgrowth of the electronic
health record system because you need that relationship with the provider to
decide in what way that information is going to be presented and to be a
constant resource for that interpretation of the data.
MS. ROLLINGS: Also I’d like to make another comment based on Robert’s
observation and I also believe that the PHR is separate from the EHR and what
we presented today was a toolset that allows for data to come from the EHR and
a patient to have input into that record and then there’s something separate
called a PHR that that data from the EHR, that data that the patient enters,
becomes the patient record. And there will be other sources, whether it’s
employers or all those other pieces that we talked about, and so that’s what
our approach has been to date. So I don’t want to articulate that the solutions
that we presented today are a PHR solution, we believe it’s a component to help
the patient to have involvement or have input into that patient record, patient
MR. HUNGATE: There’s a patient view that says I’d like to be in control,
it’s not uncommon and there isn’t a feeling that there’s a way to do that very
well. What I think I’ve heard here is a discussion of a personal health care
record more then a personal health record, so that I was kind of listening is
there anything here about how the patient’s functioning, is there any
discipline around showing how the patient is really doing, is there something
that the patient, how they’re getting along, what are their activities of daily
living, what’s their health status. And I have the feeling that patients might
be interested in seeing somebody pay attention to that on their part, that
there is a feeling that they need to get what they are in there in a way that
works. Could you comment on that, how that relates to what you’re doing?
MS. ROLLINGS: I’ll start, two of the solutions that I talked about today,
both the patient portal and the telehealth solution, both provide for the
patient to enter data, whether it’s side effect monitoring, whether it’s vitals
and things that they might be tracking at home, and that is passed on to their
care team through these solutions. The condition management portion for example
of the patient portal allows for a treatment plan to be followed by cycle by
week by a patient where they’re recording and monitoring that chronic illness
through these tools. There is trending, this is something that gets reviewed in
the next appointment with their care team and especially in the telehealth
example our customers are telling us that they’re avoiding adverse reactions
and other types of things that they’ve had in the past when they didn’t have
this automatic monitoring information available to them.
MR. HUNGATE: A related question, is that in free text form or is it in some
controlled vocabulary so that it is comparable to other patient’s input?
MS. ROLLINGS: Most of it is in a standard vocabulary through a toolset, the
telehealth example they’re using peripherals at home that upload data. There
are a couple of functions where they can enter free text and they can choose to
make that available to their care team or not, things like diaries and so
MR. DEVAULT: And I think maybe I didn’t successfully highlight similar
aspects of our applications either partly because perhaps Jim Walker who spoke
earlier, their Geisinger system uses our MyChart software and he did a pretty
good job of talking about the self healing record and that sort of thing. We do
certainly allow our patients to, or our customers patients rather, to enter a
variety of data, some of it in free text form, some of it in discreet data, and
the typical model is that that data flows to a provider, and again this is why
I think the relationship aspect between the provider and the patient is
important, that flows to a provider who can then interpret it, decide if it’s
acceptable data to file along side the rest of the collected data for the
patient, so whether it’s lab results, glucose monitoring from home devices, or
reporting that I’m also taking these other meds that you don’t have on file.
That’s a very important piece of the shared, and that’s why we called it a
shared electronic medical record.
DR. LUMPKIN: Steve, did you —
DR. STEINDEL: I don’t know if this is so much of a question or a comment
and it’s in somewhat reaction to what Bob was commenting on earlier. But I
heard a very distinct theme coming through today and this is the continuum of
health records that I don’t think we’ve really talked about, we’ve talked about
the EHR, we’ve talked about the PHR, we’ve talked about this has this
attributes, this has that attributes, we’ve done it in this session over the
years and in general in the community, but when we talked about the tethered
model and your nice solar system type pictures —
MR. DEVAULT: I wasn’t sure if that was a solar system or a nucleus
DR. STEINDEL: I’m not sure which either but this sort of started to drive
things home to me about the continuum of those two sets of records. But
Kristina in your talk it really drove the picture, I don’t know whether you
specifically meant to say it or the picture but the impression that I got was
that the view of the PHR can be looked at as something that starts outside of
when the patient comes in, whether it’s at birth or whenever they pick it up,
but what impressed me was when you were talking about your portal system in the
patient’s room, you’re making the PHR available and an integral part of what’s
going on in the health care in the hospital it sounds like. And then it
continues after they leave and I think this drives home the idea that we’re
really talking about one system that has different types of attributes and Bob
you mentioned a lot of those attributes, they were very well delineated in the
Connecting for Health report. David’s still here, they’re outlined in there,
but I don’t think we need to view them as separate systems, we need to view
them as a continuum of attributes that are applied during the patient’s, either
their care if we’re talking about a disease or their life if we’re talking
about a continual system and maybe some people can react to that and in part
I’m asking this because I have to leave at the break so it’d be interesting to
start the discussion.
DR. LUMPKIN: Further questions?
DR. DEERING: I was expecting some discussion so that I would have more time
to frame my question here, let’s see if I can put my finger on it. It has to do
with, I think it was Peter’s solar system versus fission model and something
that you touched on that goes beyond the tethered or untethered but it gets to
a fundamental architecture. I know that the Markle roadmap architecture is a
locator architecture as opposed to an aggregator architecture and I think you
were the one who said is it actually a subset, was it you who said is it just a
virtual, in other words you said it’s just a finder, does it ever really exist
or does it just find information when it needs it. And I think that got sort of
passed over quickly because as I hear our conversation the continuum that we’re
talking about still thinks about it as a record or a system as opposed to
something that might simply be a finding tool and the tool that exists to
execute a particular aggregation, synthesis, anything other task at a moment of
need. And I guess my question would be is anybody, I mean it’s spelled out what
that would be in a national EHR effort in the roadmap but has anyone begun to
think of what that might look like for PHR?
DR. LUMPKIN: Steve?
DR. STEINDEL: Mary Jo, can I comment on that because this is obviously a
very important part of the discussion that’s going on today nationally about
the architectures of these systems. And what’s concerning me is so far the
successful systems that have aggregated health information have all been the
repository model. And we haven’t had that much experimentation done with the
locator model and I think the only one that I’m aware of that’s trying to use
the locator model is Santa Barbara, no, Indianapolis is an aggregation —
MR. KAMBIC: It’s my understanding it’s a distributed and then they
aggregate it —
DR. LUMPKIN: Bob, can you talk into the microphone?
MR. KAMBIC: My understanding is the Indianapolis model is distributed and
aggregated for the encounter and then it’s destroyed —
DR. STEINDEL: No, I believe it’s maintained in a repository for reference.
DR. HUFF: It’s maintained in a repository.
MR. DEVAULT: I’m glad you highlighted that distinction, Mary Jo, I think
that, it’s my feeling that a personal health record system, somebody brought up
an excellent point earlier that it’s expensive in a number of different ways,
not just in dollars but it’s expensive in a number of different ways to
duplicate data. There’s lots of questions about where would that data reside
for a personal health record system and whose data is it and I think ownership
of data issues are very important and need to be thought through. But it makes
more sense to me to think of a personal health record system as an aggregator
function if we’re talking about a stand alone PHR then it does as, what was the
other term? As a locator, or a locator as opposed to, I don’t remember exactly
the terms, but it makes sense to me that it’s a virtual health record as
opposed to an actual record that you could pick up.
We’re actually using some of that technology, we use MyChart at Kaiser and
at some of the other places that are working on our Care Everywhere model and
that’s actually the way that MyChart works in that situation, we don’t put into
a single repository all the data that MyChart is looking at if a patient for
example in Northern California is looking at, is seeing providers in several
different parts of Northern California, we don’t aggregate all that data for
MyChart purposes, we do for some other purposes, but we actually have developed
some technology where it can go and aggregate that data for the view that the
patient is requiring at the time.
DR. STEINDEL: John, can I comment on that for a second? But you’re working
within Kaiser so it’s essentially the same business entity —
MR. DEVAULT: That’s correct.
DR. STEINDEL: And that removes some of the barriers, not all of them but
some of them.
MR. DEVAULT: It absolutely does and two of them that I mentioned earlier
were just political barriers, the idea that you can share this information,
well, it’s the same business entity as you pointed out, but also the discipline
that they’ve tied themselves to to using a standardized vocabulary among other
MS. ROLLINGS: The McKesson model is very similar today where the patient
portal kind of serves as that locator based on what action the patient is
taking, if they want to request a copy of their medical record, they fill out,
they electronically submit an ROI request form that then goes and locates those
pieces and bring them back. Same thing for their account summaries, their
condition management, their meds, all of that based on the function that
they’re using within the portal it goes and gets that data from those systems
so it is also that same model where it’s a locator. That doesn’t mean to say
that we aren’t watching the trends and couldn’t be that data repository model
as well because we certainly have the technology to do that.
MR. RUPP: And I think the level is important so we’ve talked about the
Kaiser or large IDN, so the scope which is one I tried to make in my
presentation, I think the statewide stuff that we’re seeing some traction on
the CMS, and then the nationwide national initiatives not unlike the juvenile
diabetes effort. And part of the reason we’re doing that is to go learn, I mean
I don’t think anybody really knows.
DR. STEINDEL: I’m encouraged about those comments because it sounds like
we’re going to see some experimental efforts with the locator soon.
DR. LUMPKIN: Is this a question or a comment?
DR. ORTIZ: Question.
DR. LUMPKIN: Okay.
DR. ORTIZ: My question to you as vendors, developers, etc., as we know part
of the PHR, you’ve got data that comes in from the clinical site, the hospital,
the clinics, etc., plus your own data that you can input, and for example Epic
you’re trying to figure out where maybe data from Kaiser and Northwest maybe
could be shared with Group Health or whoever else might have that. But let’s
kind of go to another level and say assuming you got rid of some of the
barriers that aren’t in your control, for example the resistance of providers
to share data, we can get over that hurdle, and you got over some of the
hurdles of interoperability that are being worked on through all the different
standards organizations. Can you tell us how realistic it is from your
perspective as vendors and as competitors in the marketplace for, for example
University of Illinois uses Cerner and let’s say you’ve got another hospital
system in Chicago that buys Epic or McKesson or someone else, how realistic
would it be if you could get over some of the other barriers that are not
within your control for Epic to work out something with Cerner or Siemens or
McKesson or whatever to actually set this up so you could actually share data
and import it in. But be realistic about this, how likely is that to occur and
what are some of the things that you would need to do to move that forward that
are within your control?
MR. DEVAULT: Sure. Well, it’s happening right now, in fact at HIMSS coming
up through IHE we’ll be demonstrating with I think McKesson is involved to but
GE and Siemens and some other organizations demonstrating interoperability
using a document repository model and using CCR and some other standards to
actually exchange data across system boundaries —
DR. LUMPKIN: Peter, could you talk into either one of those microphones?
MR. DEVAULT: Do I need to repeat anything I said?
MR. BLAIR: Yes, could you please describe again the HIMSS demo that you’ll
be participating in?
MR. DEVAULT: At HIMSS IHE, Integrating the Healthcare Enterprise, we’re
participating in a demonstration of a project using existing technology,
existing standards, to share documents across system boundaries, so between GE,
Siemens, ourselves, and a couple of others, I can’t remember who —
MS. ROLLINGS: McKesson is part of that as well.
MR. DEVAULT: And I think that’s very exciting because it’s not requiring
new standards at this point, there’s limits to what we can because there aren’t
as many standards as we would like. But I think it’s very encouraging, it
proves that it can be done and once we agree on some of those other vocabulary
type standards the fact that we have CDA and we have CCR and we have some
technologies that develop from electronic business outside of the health care
realm, it can be done so as you said there is some cultural, legal, political,
whatever other barriers that need to be worked on but assuming that those can
be overcome I think the technical ones can be overcome as well.
MR. BLAIR: What documents will be shared? Is it going to be radiological
images or documentation for radiological reports or is it going to be CCR, is
it going to be what?
MR. DEVAULT: There are, and I don’t want to be a spokesperson for this
because I’m not by any means in charge of it and there might be other people in
the room who know more of the details then I do but there are a few different
scenarios in the demonstration, some of it is radiological, one of the pieces
that Epic is working on is to share data between a primary care provider, a
hospital, and a gastroenterologist through the course of caring for a patient
with gastro-reflux, so it’s a variety of different scenarios, it’s not all just
DR. ORTIZ: Does anyone else have comments on that as well? Because the
thing I was getting to, not only the technological perspective but kind of from
a corporate philosophical perspective how you guys felt about that.
MR. RUPP: We do do it today, in fact I think many of the clients that have
been cited today, that there’s a footprint of all three of our solutions where
we do have that interoperability. The common denominator today is HL7, I don’t
think any of us would represent that we necessarily seek it out to do of our
own volition and I think the broader initiatives, whether it be some of the
LHIIs or the RIO type concepts is what’s going to force the interoperability
but I think it’s going to be at a level broader then where we traditionally
either find ourselves playing today or come across one another which I think is
the traditional health care setting.
MR. DEVAULT: I think you’re right, Seth, and like anything else the market
is going to drive it so when people start making decisions about electronic
health record systems based on whether they’re interoperable with other local
or regional systems then the vendors will develop those systems.
MS. ROLLINGS: I think the technical piece is the easiest piece to solve, as
Seth said we’ve been doing it for years and years and years, McKesson admission
systems interfacing to Cerner lab systems, etc., with interface engines in HL7,
that’s the easy piece to solve, it’s just a matter of standards and everybody
following those standards. I think the harder piece, even when you get past the
whole corporate issue, is the workflow in the health care industry, physicians
and how they practice in their daily workflow, nurses and all of those who are
actually giving care and accessing those systems and changing their practice
and how they use that information to collaborate because they haven’t had it in
the past and I think it’s a huge paradigm shift for those that are practicing
medicine to be able to share that information. And patients as well, they’ve
never really had this type of access. So I think that’s the harder problem to
DR. LUMPKIN: I’d like to thank the panel, at this time we’re going to have
some conversation, I did get one request to make a comment, you have a couple
minutes. Go to the microphone and introduce yourself.
MS. KAYE: My name is Robin Kaye and I’m an attorney whose manner of
personal interest has been medical privacy issues since HIPAA was passed in
1996 and I’ve co-written some papers with the Institute for Health Freedom
about medical privacy issues and as a private individual not associated with a
corporation and unfortunately not having the best of health myself and going
through quite a bit of health problems with my dad who died at age 56 who was
also a physician, some things are of interest to me, just to make sure that I
understand what provision do you envision for patients who do not want their
medical information to be formatted in a certain way or put in electronic
format or access interoperatively without their consent and perhaps want to
remain with the paper record, I have just a few questions but that’s one of
DR. LUMPKIN: Well, actually that’s, I’m going to ask you to just make some
comments because this is a hearing of the committee, it’s not a session where
we’re going to respond to questions, because we haven’t formulated any policy
and we will be having hearings, some of them jointly with the Privacy
Subcommittee to get at in fact many of the same questions that you’re asking.
So if you do have a series of questions you can put those into the record and
then we’ll move on from there.
MS. KAYE: I understand. One of the concerns I had was back, there was a
National Committee on Vital and Health Statistics full meeting around 1997 that
I got up as an audience participant and I said, I think Ms. Fyffe was there,
are you planning what would be tantamount to a national patient information
database and at that time the answer was absolutely not. But if we’re going to
have a system of networks that are communicating with the same codes and the
same electronic tons(?) and transferring information essentially what we will
have is a national database which will consist of interfacing networks and that
puzzles me and alarms me because the answer was a definitive no at that time
from the full committee.
And basically I want to also just state that there was a survey done by the
Institute for Health Freedom, it was a Gallup survey, and it cited that
patients really don’t want the information that they tell their doctors in
private to be shared with third parties even other physicians and that
information was sort of mirrored in the comments, the public comments to the
medical privacy rule, HIPAA medical privacy rule, and basically these
committees have overridden, even the Subcommittee on Privacy and
Confidentiality overrode what was it, 50,000 comments by the public saying they
wanted that privacy, they go to doctors to be healed and not revealed and not
have that information further disseminated.
So that’s also a concern to me, and the request that since HIPAA is so far
down the road perhaps the way to resolve this would be to have an opt in or an
opt out available to patients who really don’t want to participate as you’ve
been talking about today.
The other thing is that under HIPAA the patients are supposed to be able to
ask for restrictions so I’m wondering how you’re going to interface with HIPAA
if you’re going to require as President Bush has said electronic records, how
we’re going to interface that with the HIPAA allowing the patients to ask for
And I also just want to say that there’s a presumption it’s always
wonderful to share patient information and I have to tell you that from my own
personal experience when my father who was a physician was diagnosed with
cancer it was very late, they gave him three months to live, he was offered
only radiation therapy, we saw several doctors, he had the radiation therapy
and what we decided to do, and this was in the mid-80s, was to cover up the
information of the prior doctors that he saw, I had worked for 13 years for my
father before becoming an attorney, we covered up the information and we went
to other doctors.
And he was able to get immunotherapy and many, many courses of chemotherapy
and other experimental treatments, we crisscrossed the country and he lived for
four years. And I really can’t say that if the poor prognosis and diagnosis was
shared with every subsequent doctor that he would have been able to get that
kind of treatment that he received and perhaps not have lived for the four
And so you can’t always presume it’s a wonderful thing to share information
and prior to HIPAA being implemented if you had a personality conflict with a
doctor, which does happen, or a misdiagnosis or a malpractice situation you
could start over fresh with a new doctor and not transfer that other prior
doctors’ negative opinion about you. And under a comprehensive record whatever
happens with each prior doctor could be duly reported and go with you for the
rest of your life and that is something that we should all be concerned about.
So thank you for letting me speak here today, I appreciate it.
DR. LUMPKIN: And thank you for your comments.
Agenda Item: Discussion of Next Steps
DR. LUMPKIN: At this point we’re going to move to committee discussion, we
have before us a, well, actually Stan you wanted to make a comment.
DR. HUFF: They would be just as appropriate as part of the discussions.
DR. LUMPKIN: Okay. What we have before us as a workgroup is having
identified the personal health record as being something we’re going to work
on, we have scheduled a set of hearings, two days in January, January
5th and 6th, which is like very quickly so we need to
start identifying what it is that we want to discuss based upon today’s
hearing, what are the issues we want to raise and also begin to work on who
we’re going to invite since the time is very close, even though it’s next year
it’s like next year.
I kind of pulled out a couple of questions from the discussion as I was
going through and maybe we can pull a few others. The first one that came to me
is, and I think it was reflected somewhat in the presentations we had this
afternoon because there seemed to be a fair bit of market study, it was also
addressed by David but I think we may want to hear from some patient advocacy
organizations and how do people, patient’s view of the PHR and what is it that
they would want to have out of that.
And the second one that seemed to come up and it was raised also in sort of
a comment that was made and that is what are the antitrust issues of sharing
patient information, so if you create a patient record which may include
financial, i.e. billing data, are there antitrust issues that I as physician A
sees what physician B is billing, therefore I may adjust my billing rates and I
think that that’s an issue that we haven’t really addressed.
Are there other issues that come up that we may want to talk about?
MR. BLAIR: John, this is Jeff, can you hear me?
DR. LUMPKIN: I can hear you but I cannot see you.
MR. BLAIR: I very much agree with those two groups of testifiers with
respect to the patient advocates and I’m assuming that you mean like privacy
advocates, or maybe that’s separate. But with all of the testifiers that we had
today they have folks that are voluntarily using personal health records of
some form and maybe they can give us some leads to a number of those individual
patients that have been using, that have some experience using their systems. I
would be very interested in not just folks that are going to testify that all
these systems are great but that could wind up saying what things they do use
and what things they’d like to have that they don’t have and what areas are
frustrating or capabilities that they’d like to have that they don’t have.
DR. LUMPKIN: I think what your last comment was was perhaps more to my
point and I think the second issue that you raised, what you said, you
commented on which is the privacy components of this I think are separate and I
think we need to keep them different. So I would take from you did in addition
to talking about maybe getting some people that have actually used them with
experience but the second issue is looking at the privacy components of that.
Other areas that we —
MR. KAMBIC: John, Bob Kambic, picking up on what I said and what Steve
said, tethered or untethered, Peter Waegeman(?) wrote an interesting article
that’s online and he talks about, devise it into about four or five separate
categories and he talks about the silos, for example the end stage renal
disease might have, various chronic diseases might have their own PHRs.
Institutions like Geisinger have their own PHRs but Geisinger has obviously
bought in, I guess it’s what, to Epic or one of the others, it would be useful
I think to help clarify our thought if we could begin to somehow categorize or
segment this continuum.
DR. LUMPKIN: The other area that I think I heard mentioned and actually
it’s interesting because it really came up for the first time in Kristina’s
presentation was the whole interface of the personal health record and
telehealth. Geisinger’s presentation was very interesting because it talked
about aspects of rural communities and rural communities and rural health
organizations have been very active in the telehealth field. And there’s some
interfaces here that may be significant to look at. Stan?
DR. HUFF: Well I think it turns out my comments would have been better
before so let me just say them anyway.
A couple of things, my experience would, well, the discussion about whether
PHRs are a subset or a superset of EHRs and that sort of thing, I think maybe a
better way of characterizing it is saying that much of the PHR builds on the
same technology. And what I mean by that is that in an electronic health record
system in fact you have to worry about role based authorization and
relationship based authorization for access. You have to have write audit
trails and read audit trails. You have to have a database that contains the
data and you have to use standards for the terminology if that’s going to be
And all of those things I think are a basis from which, that is a common
base for both the EHR and the PHR and then your point is well taken that there
are then other things though, for instance the typical EHRs right now are
focused on medical clinicians consuming the data, not on the patient consuming
the data, and there’s some huge things that need to happen there in order to
make it consumable by the average thing. So that’s a thought.
The other one, I don’t know if there’s a conflict of interest but Utah has
been funded to do a locator model of this kind of EHR and some of the
interesting questions there, it’s not as clean maybe as it would appear between
the locator model and the aggregation model. And what I mean by that is our
model would probably be one where we would have the locator and we would
request information from the system and when we received though we would
incorporate that into our own record. That is we wouldn’t request it each time
we needed it, we would request it once and we would incorporate it, and so all
kinds of interesting questions come up then, how do I prevent creating
duplicates of that, how do I recognize that I’ve already received this
information from somebody else once.
And once you get many parties in the system you can imagine basically
echoes in getting, you could send it to somebody and your own data could get
echoed back to you and you don’t recognize it, I mean there are lots of
interesting kinds of things that go on there. So that was another thought.
DR. LUMPKIN: If I could just sort of break in on that one, I think I’d be
more then happy to have Standards and Security look at that because I think
that really is going to be a pure EHR issue.
DR. HUFF: And the other one was a question which if you have the record
held and is secured by an independent third party, sort of the banking model
that some people mentioned where I basically make withdraws and deposits in
this database and I control who has access, in that model how does, how do
population based decision support things get implemented? In other words if a
new rule comes out that says this is, it’s recommended now that mammography be
done every six months instead of every year or any of those kinds of things
change and you want to apply that to the whole population of the United States,
what’s the model for dissemination of that kind of knowledge and how would that
happen? It’s not clear to me that we don’t, if we’re not careful that we don’t
lose a bunch of the population based decision support capabilities because this
record is now either fragmented in lots of places and we’re locating it or it’s
now in a sense inaccessible in a way that we can implement general health
policies through the infrastructure.
MR. BLAIR: John?
DR. LUMPKIN: Go ahead, Jeff.
MR. BLAIR: This is just an administrative thing, it’s been a little bit of
a challenge for us to identify all of the folks that could articulate privacy
issues related to the National Health Information Infrastructure and personal
health record and e-prescribing and all, and I believe Robin Kaye, maybe if
she’s still there, if she could give her business card to someone in the room
that could pass it on to Amy that’s pulling together the testifiers for the
privacy hearings on these topics.
DR. LUMPKIN: Thank you, good suggestion. Mary Jo.
DR. DEERING: I have some specific things and then something a little bit
more amorphous. The first and most precise is that as we talk about the
continuum of PHRs and EHRs and tethered and untethered there is a model that
some of us might, I’ll try to be very neutral in how I describe it but it’s
almost a pre-PHR in the sense that there are at least two models that I know of
where there is a third party receiving paper text, gathering it themselves, and
not digitizing it in any standard interoperable format but either in one way or
another becoming just the holder of the file literally as opposed to
electronically although it’s electronically in the sense that then they
re-electronic(?) but it’s not interoperable in the sense of most of the
discussions that we’ve heard today. There’s People Chart and Patient Command
who are here today who are at least two that I know of who are doing that and I
think one issue for the committee is does it wish to see and learn about those
models in real time or not and let me pass on to my other two issues and then
the committee can perhaps, so I don’t put people on the spot right now so you
can choose what you want to say.
The other specific thing to suggest is that my former office did pay an
excruciatingly modest amount of money for the importance of the topic to some
very talented people including our bellowed Susan Kanaan and David Lansky and
his group to do a small mini white paper on what are the policy issues of the
federal government getting involved in personal health records. And I happen to
know that there should at least be, the final report will not be ready by then.
Cynthia, are you still on the line, or Susan?
MS. KANAAN: Yes, we’re here, thank you very much.
DR. DEERING: So it’s a plug for saying that I think, and they’ve
interviewed virtually everybody in this room whose, or a lot of people in this
room including the policymakers who are working on it so if the committee would
like to get some preliminary synthesized feedback on that paper I believe in an
informal way it would be ready. Is that a correct statement, Susan?
MS. KANAAN: Yes, definitely, Cynthia is here as well and yes we’d be very
interested in that, we’ll have some preliminary findings.
DR. DEERING: And then the third thing I wanted to put on which is a little
bit more amorphous that I’ve been listening to and it does get at this issue
that Stan was the most recent to comment on, which is the reality of the
digestibility of health information and the communication of medical
information. And it occurs to me that what this has already spawned and it
already exists out there is a whole ancillary industry of content, of medical
content translation both for general information and education but also in the
area of quite precise medical decision support for lay people.
And it might be, I know that Nextcura(?) for example is a company that has
a phenomenally deep array of clinical information and the medical literature
and then a whole stable of people who digest that and so if you plug in your
symptoms online it says, it gives you the evidence based treatment options, or
the array, it says based on the available literature it appears that a person
with your diagnosis and in your condition might be interested in, please pursue
this. And I sort of suspect that Cerner and Epic an McKesson, not to mention of
course all of the untethered PHR vendors, already have quite a large stable of
those kinds of people, it’s a whole sector within a sector perhaps, sort of the
knowledge development synthesis management presentation part of it.
And maybe if we could characterize that segment and find out what’s going
on there we’d have a better feel for how quickly these kinds of things might be
available to help both providers who would just assume, somebody said it had to
be their own docs who looked at all the lab reports and then who had to
personally sort of restate what they wanted to share, etc., etc. Well, if there
was a trusted licensed source for interpreting lab results, I know that
individual said this because there are no national sources. But again, to get
the individual doc or the individual provider out of having to do that
DR. LUMPKIN: And I think that’s something that’s interesting is an area
that we should look at, I’m just going to take issue with getting the doctor
out of interpreting laboratory results —
MR. BLAIR: Don’t trust them, huh?
DR. LUMPKIN: No, I don’t think that that’s something that we want to do
because a laboratory test in and of itself doesn’t stand, it’s usually done
within the context of the patient’s history, their current condition, a whole
host of factors. And trying to offload that interpretation I think would be a
misservice to patients. That’s why they have a clinician, that’s why they began
to develop an ongoing relationship with that clinician. So I think we do have
to look at, and I forget who said this, but one of my favorite quotes is that
the danger of the future lies not in machines thinking like men, rather it lies
in men thinking like machines. And I think that there is a component of health
care that is related to people that we don’t want to lose.
MS. FISCHETTI(?): I want to go ahead and echo because I think it’s the most
important issue, the need to hear directly from consumers as well as consumer
groups. I also want to echo the importance of learning more about these lay
terminologies, I think that’s something that is an industry just starting and
we need to know what that’s about. And then my third issue is we heard a little
bit this morning about the fear that clinicians have and I want to know more
about that, not specifically from just one disciplines, one of the slides this
morning said that there were 13 million people employed in health care and .5
million of them were physicians. Who are the other 12.5 million? What are their
fears? Do they have fears that are the same or distinctly different from the
physician fears? And it’s that group that I’d like to hear from as well.
DR. LUMPKIN: Let me drill down on that a little bit. Of that 13.5 million
or 13 million or whatever the rough number is there’s only a certain number who
are engaged in contributing to the medical record or would see that as being
where they communicate, so I think, is that what you mean, those who currently,
so we would include nurses perhaps, physical therapists, case managers —
MS. FISCHETTI: Educators, the people, health educators, exactly.
The other group that I’d like to hear from are the people who are in charge
of the business of health care, what are their fears and are the people in
charge of keeping a hospital in the black instead of the red, do they have
issues that are different then the provider issues?
DR. HUFF: I think one of the issues that I would like to hear more about,
it’s really kind of an extension of the question you raised about antitrust
things but I think a broadening of that would be what are the real business
case and business issues for this, I mean it’d be interesting to get Kaiser in
and say how would you feel about sharing clinical data with your toughest
competitors within your community and I mean why would you do that, why would
you not do that. I’ve heard Clem talk about in Indianapolis that in fact if
there wasn’t government funding there isn’t anybody right now who would step up
and fund his network. So what is it, is that going to be true forever or
there’s something that’s going to drive, is it going to be something that the
government has to fund forever, I mean we ought to find out from Clem whether
that’s really true or not, I don’t know, but what is the sustaining business
model that in fact would cause people to exchange this data and share the data.
DR. ORTIZ: I know I’m just repeating but I want to emphasize as well that I
think, I can’t overstate how important I think it is to get more consumer
input, I mean obviously this is something geared at the consumer/patient and I
think it would be interesting to have a whole day just of focusing on the
consumer. But as Linda was talking about one of the things I was going to say
is sometimes I get concerned because we get consumer advocacy groups
representing a consumer and oftentimes they’re not really representing the
consumer so we get a lot of the same thing. I’d like to as much as possible get
that perspective but also as much as we can within reason drill down to why
consumers do what they do, why they use these, why they don’t use them, what
they value and what they don’t value, and we don’t really know very well. So to
the extent we can get real data and real consumers and real people in here and
spend some time really digging in I think that would be very helpful.
So that was the first thing, and the second thing, I was just going to
support, when you said the thing about telemedicine I thought that was
interesting because I think that’s an issue that now we’re struggling with,
like at the VA we’ve got this care management program which is telehealth, we
also have My Healthy Vet which is our e-health portal, and I think in a lot of
segments they’ve both kind of come up separately but yet there’s a lot of
opportunity for integration but I don’t think we really know how those
integrate and should they integrate and should they be one entity, should they
be separate entities, so I think that’s an interesting thing that we might want
to explore further as well.
DR. LUMPKIN: Well, let me sort of try to pull this together what I’m
hearing and not surprising I don’t think we’re going to be able to do
everything we talked about in what I would call a day and a half of hearings.
We have two days schedule, I think I’d still like to preserve the afternoon
of the second day for workgroup discussion because we too often schedule
hearings and we just don’t have a chance to process it as a group. I see about
a half day in relationship to consumers, that we would invite those who are
experts on consumers, which would be advocates, those who may have done focus
group and other kinds of research to see where the trends are, but also that we
would gear as part of that to bring in individuals who can share anecdotes, of
those who have used them and those who if we can find them anyone who has used
it and has had a bad experience, I think that would be helpful for us to
understand it in an individual’s voice.
The second segment, and I see these as half day segments, I think would be
to look at the provider based barriers and that would be as Linda mentioned not
only the clinicians and I think we would want to see a group of clinicians, we
may do a panel of physicians and then do a panel of other care givers, who
would be able to speak to the issue of what are they looking for in this and
what are their fears and what are their concerns and obstacles.
The third segment, half day segment, I think is related to the business and
business case, and so we could do two panels in that segment, one on the
business case, and then on the business aspects which would include perhaps a
discussion on issues of antitrust but what are the issues that would concern an
administrator of a hospital or an administrator of a health plan about getting
involved in a personal health record, are there aspects where they feel they’re
losing control their ability to protect themselves or vulnerabilities.
So that’s kind of my boiling down, I feel no major sense of authorship in
having boiled it down so I think if there are other suggestions or swap outs
for that structure —
MR. BLAIR: John, can you hear me?
DR. LUMPKIN: Yes we can, Jeff.
MR. BLAIR: I think it makes sense when we separate out, I’m getting back to
the privacy issues, obviously we’re going to have more then a day of hearings
on privacy issues related to all of these information technology subjects but
there may be, it may be a good idea to have a little bit of cross fertilization
so that we’re not completely unaware of all of the issues that come out of that
and maybe if someone that’s attending the privacy hearings could give a ten or
15 minute summary of the major things that were heard there so that the NHII
Workgroup is at least aware of what those issues are.
DR. LUMPKIN: Yeah, Jeff, I think what I’m hearing is is that we’re actually
going to have our hearing before Privacy has their hearing.
MR. BLAIR: I thought they were on the 19th the Privacy
Subcommittee was going to be having some hearings —
DR. LUMPKIN: I think that’s on e-prescribing.
MR. BLAIR: But you’re right, they’re going to have theirs the second week
as well of January, they’ve got two sets of hearings.
DR. LUMPKIN: But Jeff I think that’s a good idea and one of the things that
I don’t really think at this point we can discuss privacy if we don’t know
exactly what it is that we’re talking about and let me discuss for example the
Quicken model. If you have a Quicken model where the data only goes one way and
so I download from my bank on a monthly basis all the transactions that my bank
believes has happened and then I reconcile my account and then I print out
checks there really is no privacy issue because I control everything that’s in
my personal financial record. If that’s the model we end up with then that’s
one kind of privacy discussion, how do individuals protect that.
If we’re talking about the portal model that we’ve heard about where it’s
just one organization then it’s I’m working with my provider, that’s a
different one. If we talk about a connectivity model that’s a different issue
of privacy so I think we need to get closer to what it is that we think we
might be talking about and at that point we can ask very specific questions and
get specific answers from those who are involved in advocacy for patient
privacy. I’m a little bit concerned that if we just ask the broad question
we’re not going to get answers that are going to be specific enough that we can
focus in on particularly the personal health record.
MR. BLAIR: I completely agree with you and I can think of two or three
other examples that would support your thesis as well.
DR. LUMPKIN: So I think what we’re seeing as a committee, that’s one of the
things we want to do but let’s sequence this in the right way.
MR. BLAIR: I agree.
DR. LUMPKIN: Kathleen.
MS. FYFFE: Mary Jo had mentioned a couple of other types of models of
personal health records or personal health type recordkeeping and are we not
going to hear from those others in the one and a half days that John has
DR. LUMPKIN: My proposal would be no, at this time, and I don’t think we’re
going to be done with hearings, I think what we’ll discuss on the second day of
that hearing is where else we want to go. We are not describing the whole
continuum which I think was raised, there were a number of other issues which I
think were very important issues, the interrelationship of telehealth, so I
think there’s a fair number of things on our plate and I think my suggestion is
just a methodological approach to eating it.
MS. FYFFE: Okay, so we have dates for hearings, these next two days, what
are those dates?
DR. LUMPKIN: January 5th and 6th.
MS. FYFFE: Okay, and then do we have dates beyond that?
DR. LUMPKIN: I think we’re going to have to find another two day segment
sooner rather then later.
DR. DEERING: Well, just to remind us that at some point in February we will
be virtually co-managing a hearing on privacy and my understanding is is that
it could all be shaped by our agenda at this point or at least the great
proportion of it depending on what they have left over. And so that could be
another two days on privacy alone which would mean that one of the things we
needed to do in our discussion in January or by conference call would be to
formulate those models of PHRs around which we wanted to ask our privacy
questions. So that would be a homework item.
I wanted to lay out a couple of other introductory things here, I wanted to
mention again the ODPHP paper that would be available and that could be
presented but the other thing is that we’d talked about is we’ve heard of at
least a couple of related, and they are in fact very closely related, standards
activities here and while I think it’s premature in that neither of them will
have delivered much I’ll just put out on the table do we want to, first of all
one is that HL7 is in fact moving forward with its PHR functional descriptor
and requirements and minimum set, etc., and the uses cases, on the other hand
we’ve heard a little mention of the OMG HL7 around interoperability that we
know just given the interest of some of the parties on it is going to have a
strong impact on PHRs.
Again, both of those are still standing, or just starting up. I only put
them on the table because I think one of the things that we’ve seen in these
hearings is that often the fact that people are there spawn some thinking or
puts a placeholder, sort of plants a flag and can help generate more interest
in those efforts as appropriate. So just wanted to, they would not either of
them be anywhere near reporting on much more then process at that point.
DR. LUMPKIN: And I think it may be good to hear from them. I think that
there are a number of developments that are going to be happening that will add
to it, the whole issue of the background and I can’t remember which one of our
speakers said it today, that if we can solve a lot of the issues of
interoperability of an electronic health record it will point the way to solve
those same solutions, I’m kind of paraphrasing, with the personal health
record. Given that and some of the progress that’s going on in this environment
I think that we will probably be in a better situation to look at where we’re
at in January in how to address some of those issues. I don’t think we’re quite
ready yet because all the players aren’t clear, there’s HL7, hopefully we’ll
begin to see some of the results of the work from the funding just went out
from AHRQ, we get a better idea of what those are, and to look at how those
interface with personal health record where there’s a major connectivity
activity. So I think that there’s some development we may want to look at as
part of our discussion. And I would be remiss in saying that we probably could
start off the first day with the ODPHP paper and then move into the panels.
MR. KAMBIC: A question, John, being relatively new to the committee, what
will be the end result of the committee’s deliberations?
DR. LUMPKIN: That’s a very good question and all I can do is give you my
opinion. One of our challenges as a subcommittee is that we have, in many ways
our success makes it very difficult for us to function. Having now pushed the
issue of the NHII, now with the development of the Office of the National
Coordinator, things are moving very rapidly in the department and I think we
want to enhance that but not to delay it. So where there are issues that we
think we can contribute we want to. What we’ve seen over the last year or so is
that we’ve identified an issue and because we don’t move as quickly, as people
who like Kathleen spend every day of the week working on this particular issue,
we can identify issues, we begin to work on them, and they’re taken up and
people run with them, that’s great.
My guess will be is we will identify early next year some barriers and we
may either address those barriers or it could be that ONCHIT will identify
based upon our hearings and say okay, we’re going to address these barriers and
that work will be done before we have a chance to make recommendations. So I’m
not really focused on being able to say at the end of X time we’re going to
have a report, I think what we want to do is contribute to the process.
MR. KAMBIC: That’s good because I’ve been looking into this a little bit
for CMS and Bill Rollow has the extraordinarily complex, and I would hate to
see us end our hearings just in January because I think that there’s a lot of,
I could come up with a laundry list, I won’t do it, we’re going to ask David
Lansky to do that.
DR. LUMPKIN: And that’s good and I think part of our relationship with the
department, because we’re not a policy making group, we’re advisory, and if
there are issues that are coming up from CMS and ONCHIT and so forth that they
would like us to do a hearing to sort of address the issue I think we’re more
then happy to shed some light on an issue and then work towards a solution
which may or may not result in us making a recommendation.
DR. DEERING: I was just going to actually say then would CMS be in any
position to want to be saying anything further two months from now? I mean is
it helpful to you?
MR. KAMBIC: One of the things that I was going to actually ask is that this
morning Dr. Rollow actually, we wanted to get, CMS wanted to get some sense of
the workgroup as to whether or not the outline of progress that Dr. Rollow
suggested was appropriate and Steve said he thought it was a good idea but I
didn’t actually hear anybody else sort of chime in. Getting a business case, I
think the idea is to ask Connecting for Health to gather some experts together
to outline two things, a business case and the second would be the technical
specifications and then once we have those I think we’re working into next
summer, perhaps next fall, going to a standards development organization and
HL7 is going to be working right along with this but saying okay, here’s what
Connecting for Health thinks, here’s what the real experts in PHR think, and
outlining some actual technical standards that then could be translated into
interoperable, into the interoperable PHRs.
MR. BLAIR: May I chime in on this one a little bit? Can you hear me?
DR. LUMPKIN: Yes, go ahead, Jeff.
MR. BLAIR: I have the sense that the traditional SDOs that are working
together, like for example this panel where we had Cerner and Epic and McKesson
and they’re all able to work together in a demonstration project at HIMSS, the
interoperability is moving forward at that level. The areas where I think we
may need to try to encourage interoperability is where we have these different
models of PHRs like with SimplyWell and MedicAlert and Geisinger maybe not as
different but what are the challenges to try to pull those models in and the
reason that I’m bringing that up is that we don’t consider those as outliers
and we don’t treat them as if well they need to conform with what’s already
been done in HL7 or ASTM or whatever because they’re kind of the pioneers that
are meeting the needs that others haven’t met and I just want to make sure that
when we talk about interoperability we’re inclusive and broad, broad enough so
that new ideas and new models can be accommodated by the interoperability
models we create.
DR. LUMPKIN: And I think that’s what made me feel comfortable with the
timeline that Bill was presenting because it wasn’t talking about trying to go
immediately to a set of guidelines, we’re talking about ’05, I mean ’06, that
gives enough time for the market to mature and I think that what we need to do
is to do this in a sort of dynamic way, what are the things that need to be
done to assure interoperability are different then what are the things that
need to be done to establish standards for personal health records.
MR. BLAIR: Correct.
DR. LUMPKIN: So I think there’s time to get a little bit of clarification
and having CMS move forward keeps the momentum.
MR. KAMBIC: So I can use this as general agreement that this is a process
that the workgroup agrees should go forward, it would be helpful if it went
MR. BLAIR: Whose voice is this?
MR. KAMBIC: It’s Bob Kambic.
DR. LUMPKIN: Mary Jo, do you have enough to, enough of a framework —
DR. DEERING: Any names suggestions for any of these things? I mean could I
just repeat for example —
DR. LUMPKIN: Let’s work through with the, I think that we have a number of
people who presented today who could help us with individual patients. My guess
would be is that we may need to do a little bit of exploration, I would be hard
pressed to ask somebody to could you give us the name of somebody who doesn’t
like your system, so we may need to do a little bit of exploration to find sort
of a balance but I think we should be able to get a couple of names of
individuals who would be willing to talk about their experience with the
personal health records that were presented today.
The consumer groups I think, we tried to get AARP, they’re working on their
own EHR model rumor has it, they weren’t willing to testify, I don’t know where
they will be in January but certainly I think that working with the department
there are a number of consumer advocacy groups we’ve worked with that we can
begin to reach out to them. Kathleen?
MS. FYFFE: Linda? Are there any folks who have used the electronic health
records or personal health type record applications within the VA/VHA system
that might be wiling to speak?
MS. FISCHETTI: We have two types of applications within VHA, one is just a
journaling where you go in and you keep your own information, the other one is
a tethered model where we have it at nine solid sites and actually here in D.C.
we have a very strong users group that would be interested in I’m sure
participating but I would have to check that out.
DR. LUMPKIN: But that might be a good example of instead of going to the
users group, I mean there are a lot of veterans organizations that would
represent, the veterans would maybe be able to give us a sort of different
viewpoint of a system if they would be willing to sort of query their members
on who’s using the system.
MS. FISCHETTI: I will follow up on the request to get a spokesperson from
the veterans service organizations who have specifically thought about use of
personal health records.
DR. LUMPKIN: Okay, great, thank you, good suggestion. To the providers
there are always the usual suspects, certainly with the physicians, I think we
would want to focus in on primary care, AMA, maybe Family Physicians —
MS. FYFFE: What about the endocrinologists? Every hearing I’m at we talk
about the diabetics.
DR. LUMPKIN: Sounds great. My original thought was eh, but maybe we can do
the AMA, either family physicians or interests for the generalists, and then a
specialist such as the endocrinologists because they’ve been so tied up with
diabetes related things. For the other providers I would think the American
Nurses Association may be one and this is kind of, our foundation has 13
hospitals we’re working with in an initiative called Transforming Care at the
Bedside and maybe selecting one of those hospitals, one of those systems to
talk about this, if any of them are looking at it but they’re really looking at
some sort of innovation, that may be one way of addressing it. And then we
probably need to just think if we can get suggestions about which other
organization we may want to reach out to beyond nurses.
And for the business case perhaps CMS and the folks from Markle can help us
select for the folks who are beginning to look at the business case.
MR. KAMBIC: We’d be glad to try to find people that actually have some data
on cost, effectiveness and so on.
DR. LUMPKIN: And then I would think that we could perhaps get hospital
association, AHIP, to talk about the potential business obstacles.
DR. DEERING: I’m thinking also that Dr. Brailer himself and people from
Santa Barbara have spoken eloquently on the extent to which the business issues
were barriers so it’s also possible to go to some one of those organizations
that has tried to put the RIOs together. Now those don’t get directly to PHR
but it does get to the information exchange and so I guess the issue is do we
want to focus on concerns that they might have in looking at PHRs specifically
or at information exchange broadly.
DR. LUMPKIN: Well, I think we’d want to look at the issue of PHRs, which is
getting information to patients about their health and what are the business
issues related to that. And as you relate to that that means that we could look
at a panel, trying to keep about three of them, so we have hospital association
as one, the second being American Health Insurance Plan, AHIP, being the second
from the health plan perspective, and then the third interesting approach that
may be from an organization like Pacific Business Group on Health, the Midwest
Business Group on Health from the other aspect of business, employers.
So I think we’ve filled out the panels. It’s done. Do we have anything else
to discuss? But I think we’re done, thank you all, I think this may be our last
meeting at least for this group until January so I hope everyone has a great
set of holidays, don’t get too much triptofan(?) when you eat your turkey in a
couple of weeks and have the best of holidays and we’ll see you next year.
[Whereupon at 3:20 p.m. the meeting was adjourned.]