[This Transcript is Unedited]

Department of Health and Human Services

NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS (NCVHS)

Subcommittee on Standards

in Collaboration with the

Subcommittee on Population Health and the
Subcommittee on Privacy, Confidentiality, and Security

Hearing on
Public Health Data Standards

November 12, 2013

National Center for Health Statistics
Auditorium
3311 Toledo Road
Hyattsville, MD 20782

Proceedings by:
CASET Associates, Ltd.
Fairfax, Virginia 22030
caset@caset.net

TABLE OF CONTENTS

P R O C E E D I N G S (8:47 a.m.)

Agenda Item: Welcome

DR. SUAREZ: Well, good morning everyone. I am Walter Suarez. I am with
Kaiser Permanente. I am one of the members of the National Committee of Vital
Health Statistics and a co-chair of the standards subcommittee. It is a really
great pleasure to be able to welcome you all here today to sunny, but cold,
Washington.

It is very exciting really to be able to call this meeting to order because
I think it is truly one of the first, if not really the first, national hearing
on the theme of public health data standards. We have had a number of other
meetings about data standards and standards in public health. Really, this is I
think, the first time from the National Committee’s perspective, and probably
from other perspectives, as well, it is the first time that we are able to
gather so many people to talk about the state of standards in public health.

This also follows very closely and is very much in line with one of our new
national committee themes, if you will, that have come up as one of our new
directions. That is the theme of convergence. I think in many respects, the
standards that we use in health care for communicating electronic messages used
to be, and continues to be in some respects, separate. If we were talking about
administrative transactions, if we were talking about public health, if we were
talking about clinical data changes. What we are moving towards really is
finding a lot more convergence across all those standards so that the same data
that is being collected and captured in electronic health records can be used
for different purposes, whether it is administrative purposes, clinical care or
public health and population health purposes.

We are pursuing that convergent theme very closely. That is why this hearing
is a joint hearing of the National Committee’s subcommittees on public health
or population health data standards, and certainly privacy and security, the
other major pillar of our structure in the National Committee.

I think there are so many people to acknowledge and to thank. I wanted to
say just as part of helping put this together, I have been involved in a number
of listservs and communications messages. I have seen the amount of work that
everybody has put in helping prepare your testimonies, and the excitement and
the interest that you all have expressed in bringing this together. I want to
first of all thank, on behalf of Ob and the subcommittee, and the National
Committee, everyone for the immense amount of work that you put in helping
bring this testimony together.

Certainly, I think from the National Committee’s perspective, people that
help put together this are the heart and soul of this. I want to recognize
Michelle Williamson and Kamahanahokulani for their role in helping us bring you
all together and keeping all this in line. Thank you so much on behalf of the
committee.

I think one of the things I wanted to say also at the outset is this is
really the beginning of a new journey for the National Committee in terms of
the standards subcommittee moving and beginning to work more in the space of
standards in public health. We have been working a lot, as you know, as the
standards subcommittee on the space of HIPAA standards and administrative
transactions and standards related to administrative simplification.

We are going to be moving forward into the next year and into the future
much more in the realm of public health data standards. I hope and I expect
that this is really the beginning of a journey this first hearing on public
health data standards. I know there is a lot of interest on our other
subcommittees on this topic, and so we are going to continue to work
collaboratively and together on this.

The other important thing, I think, is that the scope of what we wanted to
try to do today was a lot larger than what we could accommodate. We do want to
cast a white net and think holistically about public health and population
health. As much as we know how much advances we have made in the standards in
certain areas of public health, we want to also look at the wide variety of
areas in the public health domain, including things like environmental health
and maternal child health, and all these domains within public health that
require and use and exchange a lot of information. While we have a lot more
experience in things like public health laboratories and immunization registry
data exchanges, and even vital health statistics, we also want to look at these
other areas.

Lastly, I want to highlight one of the most important intents today or goals
today was not just to start the debate and the discussion about where we are
with standards in public health, and how they need to begin to or continue to
converge with all the other standards that we are using in the health care
industry, but really to identify opportunities and areas where we can work more
to advance the use of this standards.

I think out of this hearing, I think we are going to begin to identify
certain areas that I expect will be the focus of future endeavors of the
National Committee. That is one of the roles today is to identify some of the
challenges and issues that you have seen, and then the opportunities where we
can move forward.

At the end of this, I think the expectation is going to be, as we regularly
do, that we will put out and put together probably what would be a letter
summarizing the hearing, and identifying a series of observations about what we
heard through the testimony. Then, also identifying potential recommendations
for the secretary and for HHS to consider, and much more widely, for the
industry to consider with respect to standards.

I think that is where we want to probably move forward with this. Again, my
thanks to everyone for coming today, for joining us for what would be a very
exciting day. It is going to be a long day. We are going to try to make it very
fun and engaging. I am going to let Ob say some opening words before we turn to
Martin.

MR. SOONTHORNSIMA: Good morning. My name is Ob Soonthornsima. I am a
co-chair of the Subcommittee on Standards. Very briefly, I just wanted to echo
one of the things that Walter talked about and that is the concept of
convergence. You may or may not know this, but the National Committee is also
working on a roadmap, if you will, of all the standards and mandates and things
that we are asking the industry to embrace over the next four or five years.
This is a huge opportunity for us to learn and be informed about public health
standards and how that might fit into the overall roadmap.

It is not just administrative simplification; it is public health standards
and other HIT standards, as well. Those are different dimensions, if you will,
of the various standards that we think, if they come together, sort of
harmonize the timeframe, decide some scope and scale of these initiatives, then
we might have a better chance as a country to embrace them, to implement them
and begin to get good adoption.

That is another aspect of what we are hoping to come out of this, again, to
inform ourselves as to how this would impact that longer journey roadmap for
the industry. I am looking forward to a very engaging and informative day.
Thank you.

DR. SUAREZ: We are going to do introductions in just one second. I wanted to
mention, it couldn’t be more appropriate to be holding this really first
hearing on public health data standards here at the National Center for Health
Statistics. It couldn’t be a more special occasion for us to also hold it
during the, I guess it would be the last month that we have Marjorie with us
here working on the National Committee. I wanted Marjorie to perhaps say a few
opening words because we go around and do introductions.

MS. GREENBERG: Thank you, Walter. I am feeling a little weepy actually. It
really is a thrill to be able to host this meeting. I thank the Standards
Subcommittee for organizing this before I retire, which will be the end of this
month. I look around the room and I see colleagues, and Walter is definitely
among them, who started work with me right after HIPAA on the implications for
public health and health services research of HIPAA and the whole standards
movement.

I am just delighted to have you here, even though I am retiring, and
although there are many bets going on in my family since I have changed the
date any number of times. The papers are submitted; I have put it on Facebook.
What can I say? I will certainly follow this work with interest. You might even
see me showing up if you have another hearing, and I am not traveling
somewhere.

I want to thank again Michelle and Kamahanahokulani for all the work in
setting this up. I want to mention that we started talking about this, I don’t
know maybe in the summer. Then, we had something that you may remember called
the government furlough. The two of them got out the invitations to all of you
right as we were having to shut down our laptops, as we were being told on
October 1st that we could not use our Blackberrys, et cetera.
Somehow they pulled this off so that when we didn’t come back until October
17th, the responses were starting to come in. Here we are today.

We did have to delay certain things obviously, for the committee and all of
our agencies. You can’t just forget about 17 days. This wasn’t, and I am
especially appreciative to them for that because they really pulled this off
under extremely difficult circumstances.

That is all I am going to say. I am looking forward to everybody’s
testimonies. Some I have had an opportunity to look at, and others I am really
looking forward to receiving. I am just delighted that this is taking place. I
do feel that although a lot of the conversation in other circles of
convergence, and it is in other circles, is about administrative data and
clinical data. That is very important.

The public health uses clinical data and administrative data. I take every
opportunity I can, and this is my one plea to all of you, to follow that after
I have retired, I am quite sure you will, to not forget about public health and
population health. As I always say, vital statistics are the bookends of the
electronic health record. You are born; eventually we all die. Somewhere in
between is everything else, it goes on.

Yet, public health has not gotten the resources certainly that are needed in
order to be a full player, but that doesn’t stop us. If you all, and I know you
will, continue this passion that I have and share with so many of you, then I
will be a happy retiree.

DR. SUAREZ: Thank you, Marjorie. I think we are going to go around the table
and start with Bruce, introducing ourselves.

DR. COHEN: Good morning. I am Bruce Cohen from the Massachusetts Department
of Public Health. I am a member of the Full Committee and co-chair of the
Population Health Subcommittee. I have no conflicts.

DR. FRANCIS: I am Leslie Francis from the University of Utah. I am a member
of the Full Committee and co-chair of Privacy, Confidentiality and Security.
And no conflicts.

DR. CHANDERRAJ: I am Raj Chanderraj, practicing physician and member of the
National Committee and member of the subcommittee. No conflicts.

MS. BEBEE: Good morning, I am Suzie Bebee and I am staff to the
subcommittee. I am with the Assistant Secretary for Planning and Evaluation.

MS. GOSS: Good morning, I am Alix Goss. I am the Pennsylvania Health IT
coordinator and the program director for the Pennsylvania eHealth Partnership
Authority. I am a member of the Full Committee and a member of the Standards
Subcommittee and I have no conflicts.

MS. FARRAR: Good morning, I am Kamahanahokulani Farrar. I am with CMS in the
Office of E-Health Standards and Services. I am the lead staff to the
Subcommittee.

MS. WILLIAMSON: Michelle Williamson with CDC’s National Center for Health
Statistics and staff to the Standards Subcommittee.

MS. GREENBERG: I will do this even though I don’t have to because I am not
going to get to do it too many more times.

Good morning. I am Marjorie Greenberg. I am from the National Center for
Health Statistics, CDC, and Executive Secretary to the Committee.

MS. KLOSS: Good morning. I am Linda Kloss, a health information management
consultant. I am a member of the Full Committee, co-chair of the Privacy,
Confidentiality and Security Subcommittee, and member of the Standards
Subcommittee. No conflicts.

DR. FITZMAURICE: Michael Fitzmaurice, Agency for Healthcare Research and
Quality. Liaison to the Full Committee, staff to the Standards Subcommittee and
the Quality Subcommittee.

DR. SORACE: Hi, I am Jim Sorace. I am ASPE staff to the Subcommittee.

(Introductions around the room)

DR. SUAREZ: Thank you so much again. We are going to go ahead and start with
our first panel. I apologize, we do have people on the phone. Sorry for that.
Could you introduce yourselves on the phone, please?

DR. CHEATHAM: This is Marcus Cheatham. I am the co-chair of the joint Public
Health Informatics.

DR. BRAND: This is Bill Brand, Public Health Informatics Institute in
Atlanta.

Agenda Item: Panel 1: National/Federal PH
Perspective

DR. SUAREZ: Thank you. We are going to go ahead and start with the first
panel. A couple of quick logistics. We have tried to have about 10 minutes or
so for testimony, and then have enough time for the subcommittee and the
committee members to do Q&A. You will be seeing some messages about your
time coming up in a few slides here that we have.

We are going to skip our public health data standards landscape. Michelle
had prepared a beautiful presentation about some of the basic concepts around
public health data standards. We are going to post that on the website for
everybody to see and review. We are also distributing, by the way, through the
National Committee meeting which starts tomorrow and on Thursday. In the
interest of time, we are going to skip and go into the first panel.

I believe we are going to start with John Loonks first with his testimony,
and then go into the order that we had originally.

DR. LOONSK: Thank you very much, Walter, and good morning everyone. I am
John Loonsk, and I am here representing the American Medical Informatics
Association, the professional home for over 4000 informaticians, where I am the
chair of the public health informatics working group.

We would very much like to thank NCVHS for the opportunity to testify today.
I would commend to you also the written testimony that we submitted, as well.
There are some complex concepts here, and it is hard to express them all
quickly.

As I indicated in the written testimony, I also testified to NCVHS on
September 11th of 2002, over 10 years ago. It was the one year
anniversary of 9/11 and after the anthrax attacks. In that testimony and in
other settings, when I was then at the CDC, we were trying to promote
electronic case reporting from clinical care to public health. Unfortunately, I
am still here today, pushing for the same basic capability.

Basic case reporting is the place where the public health data standards, in
terms of implementation, are the weakest. This is particularly true in the
context of how important case reporting is to public health surveillance and
outcomes. Back in 2002, we were trying to get differential reimbursement from
CMS to incent providers to report. Now, we are facing the risk that despite
over $15 billion in high-tech funds, we may not get basic case reporting into
the last incentive phase of Meaningful Use Stage Three.

There is great concern if all of meaningful use incentive stages are missed
for this capability. As I can personally attest, incentives do not come around
frequently, and there is great uncertainty about whether there will be
subsequent penalty phases at all.

Electronic case reporting can do a lot for public health. It can improve the
timeliness of reporting. It can address the very low current yield of
legally-mandated reports. It can improve the completeness of reports that are
received. It can also help support further investigation, tracking and other
operations. To do so, it must include the needs of both management and
reporting, as both are parts of surveillance.

It must also help in emergencies with flexibility that can address the
dynamic nature of each emergency event. Indeed, many of the most important
surveillance needs, emergency or not, are not about the standard norm, but are
about variations from that norm. Variations from the norm may be seem to
conflict with standardization, but case reporting must support them, too.

Electronic case reporting will also take great coordination, particularly in
the face of two long-standing public health issues. Number one, the many
different conditions and diseases, and the categorical programs that align with
each of them. Number two, the jurisdictional variations and their implications
for the nationwide presence of clinical care EHR vendors.

Together these issues present a classic integration challenge. This
challenge is more than just additive; it is actually exponential. The
integration challenge is represented by the number of EHRs times the number of
reporting jurisdictions times the number of conditions.

This integration complexity translates into a very significant public
health standards issue, how to implement public health case reports in
standards. Essentially, there are two divergent approaches to this problem. One
is to code each individual disease or condition into an individual message and
ask the EHRs to implement them all. What we face now truthfully, though, is
that the EHR vendors are already up in arms about public health variations in a
single immunization message.

The second approach is to try to harmonize all the different diseases and
conditions into one electronic message. This is in many ways definitionally
challenging. We believe there is also a third approach that should be
endeavored for Meaningful Use Stage Three. That is to implement a core
non-inclusive, possible case message that is part of a larger process.

When you think about it, creating an inclusive message, whether through
approach one or two, as I expressed, is really a manifestation of the paper
reporting process. In the paper setting, the provider submits a report when
they have concluded there is a reportable condition, and they have all the data
together.

Unlike paper-based reporting, however, electronic case reporting inherently
involves several transactions. Importantly, the first step of the reporting
process should be automatic and not depend on manual initiation by the provider
of care. As an example, in electronic laboratory reporting, or ELR, which is
operational in many jurisdictions, the first step is trigger codes are
distributed that the reporting system uses to automatically determine what to
report. These reports are automated and are attractive because the automated
nature provides much better yield and timeliness of reporting.

Clinical case reporting needs an automated initiation process, too. In case
reporting, the trigger code should help automate a possible case report to
public health. The automation is central to addressing the reporting yield
problem, and can provide the kind of immediacy that other countries are now
showing in their electronic case reporting systems.

The automated process should be sending a core possible case message to
public health. Why core? All previous nationwide efforts of general case
reporting, whether they be NETs or NEDs or several other acronyms have utilized
a core message. A core message can easily be specified for stage three. It
focuses on data that are in EHRs and are invariant.

Why possible? The message is also acting to ask public health, is this truly
a reportable condition, and if so, what other data must be reported and to
whom? In this picture, public health should and would maintain the rules that
show what is reportable where.

The core message leads to a subsequent edition of more variable data by
providers or their staff via the web or an approach like the ONC S&I
framework structure data capture project. The data beyond the core that are
input manually are data that, one, are less likely to be recorded in any HR to
begin with. Two, vary by disease or condition, three, vary by jurisdiction,
four, can be dynamically configured in the case of an emergency.

By moving these data to supplemental data input, we will have addressed the
exponential integration problem by fixing two of the variables. Over time, as
the EHRs mature, more and more data can then be mapped, automatically populate
data needs and maximize automated data input.

It is a critical time for public health data standards. The approach to case
report standardization that I have described here is not really new. It builds
on and aligns with much of the good work that is already going on. The
initiatives in the S&I framework, the existing standardization efforts,
work at the CDC, basic case reporting is also importantly in the concept of
convergence the critical public health standards issue for another reason.

A basic case report where case data are moved from an EHR to a registry or
another non-EHR management system is the critical element for an architecture
for chronic disease management and population health. These systems are not
usually in health departments, but are in clinical care or elsewhere. To
achieve chronic case, chronic disease and population health management outcomes
however, public health should be helping to lead the architecture of these
systems, as well. It is another effort where public health standards efforts
are desperately needed. The architecture should not be left to being addressed
by purely commercial goals, as well.

There are a great many standard needs in public health. Each need is
important to people. Few are as central or connected to so many other needs as
basic case reporting. Hopefully we can leverage the incentive phase of Stage
III to put a basic case-reporting platform into EHRs to support many public
health needs for years to come. Thank you for the opportunity to testify today.

DR. SUAREZ: Thank you very much. We are going to go to Nikolai.

DR. LIPSKIY: Good morning. Again my name is Nikolai Lipskiy. I am health
scientist, lead of standards in intereligibility program for CDC, Centers for
Disease, Office of Surveillance of Epidemiology in Laboratory Services. I want
to thank the leadership of the National Committee on Vital Health Statistics
and the subcommittee for the opportunity to present today testimony public
health data standards, federal laws and regulations.

All of my colleagues in this room are well aware about importance of
standards for this country, as well as for the whole world. As it was in this
court in the United States standard strategy of 2010 by Patrick Gallagher,
today more than ever, standards are an imperative undertaking. Standards
provide the common language that keeps domestic and international trade
flowing. It is difficult to overestimate their critical value to both the U.S.
and global economy.

The United States federal government operates under the legal
standardization framework and policies that are a series of statutes,
regulations and administrative orders. I just want to name four critical
statutes and regulatory documents that are important for building framework
that we just described we need for standards.

This is National Technology Transfer and Advancement Act of 1995, Office of
Management and Budget Circular A-119 of 1998, Federal Engagement in Standards
Activities to address National Priorities, Background and Proposed Policy
Recommendations that were developed by Subcommittee on Standards, National
Science and Technology Council in 2011. Finally, Memorandum from Executive
Office of the President for Heads of Executive Departments and Agencies of
2012, Principles of Federal Engagement in Standards Activities to Address
National Priorities.

The first document of these four is National Technology Transfer and
Advancement Act of 1995 or NTTAA. This act is a public law that was developed
with the purpose to prescribe to federal agencies the adoption of voluntary
consensus standards where possible in lieu of creating non-consensus standards
by federal agencies. NTTAA direct federal agencies with respect to their use of
and participation in development of standards. Also, this act direct the
National Institute of Standards and Technology, NIST, to coordinate the
standards and conformity assessment activities of federal agencies, as well as
state and local governments with the private sector in order to reduce
unnecessary duplication and complexity of conformity assessment schemes.

Finally, this act prescribed to a report on progress accomplished in NTTAA
in collaboration with the Interagency Subcommittee on Standards Policy.

The NTTAA codified existing standardization policy in the White House Office
of Management and Budget, OMB Circular A-119 guiding federal agencies on use
and development of voluntarily a consensus standards and conformity assessment
activities. This circular establishes reporting requirements and authorized the
NIST to coordinate conformity assessment activities of agencies. Finally, OMB
Circular A-119 issues guidance on the responsibilities of agency heads and
agency standards executives in development and implementation of standards.

Federal executive departments and administrative offices operates in the
court of authorities that is assigned to them by law, fulfilling the national
roles in the scope of this authority. They write standardization policies and
regulations. The Office of the President, as authorized by Congress, guides the
United States government to coordinate agency standards activities and
engagement through establishing positions of agency standards, executive and
supporting agency structures.

Recognizing the integrity of the existing United States standards process,
the Office of the President guides agencies on two standardization roles. The
first role is to act as an active participant in standards development. The
second role is to act as user of the standards for regulations and procurement.

Even though participation of federal agencies is a rather supportive than
leading, there are a limited number of areas where federal agencies shall lead
standardization. The White House Executive Office provides regulatory guidance.
Health IT is one of four priority areas where federal agencies lead and
coordinate standardization activities.

CDC takes a leadership in health IT standardization in a scope of its
authority to protect America from health, safety and security threats, both
foreign and in the United States.

Conformity assessment; A conformity assessment is one regional component of
the national standardization framework. The full definition of the conformity
assessment was developed by ISO, International Organization for
Standardization, and International Electrotechnical Commission. It was included
international standardization regulatory documents. Based on this definition of
conformity assessment, an activity concerned with determining directly or
indirectly, that relevant requirements are fulfilled.

Our national regulatory guidance described that conformity assessment
include such tasks as sampling and testing, inspection, supplier’s declaration
of conformity certification and management system assessment in the
registration. It also includes accreditation of the competence of those
activities by a third party and recognition, usually by a government agency, of
an accreditation program’s capability.

In response to NTTAA and Circular A-119, in 2000, NIST published the
following regulation, Guidance on Federal Conformity Assessment Activities.
This document helps federal agencies improve the management and coordination of
their own conformity assessment activities in support of their regulatory
procurement and other missions.

CDC has a well-established conformity assessment system for the personal
protective equipment that is conducted by the National Personal Protective
Technology Laboratory within NIOSH. This is one example of CDC’s conformity
assessment activities.

CDC published a report on standardization in accordance to the NTTAA/OMB
Circular A-119. Each November, we send this report to HHS and NIST. It contains
information about success stories, and standards implementation, and specific
standards that were adopted by agencies from implementation that we are
reporting here. Thank you for this time.

DR. SUAREZ: Thanks so much. I think we are going to go to Marcus on the
phone. Marcus, are you on the phone?

DR. CHEATHAM: Thank you very much. Thank you for making it possible for me
to call in. I found that I needed to be in three places at once this morning.
When I saw this stellar panels that you had lined up, I knew I was not going to
make or break the testimony by my presence.

All I really want to do is briefly characterize some of the relationships
that exist between the panelists who will be speaking this morning, basically
to make the point that there really is a national standards development effort
in public health that is characterized by a high level of expertise and
commitment. Our poverty bedevils us, but our past success shows what we can
accomplish. We welcome your partnership and critical attention.

The Joint Public Health Informatics Taskforce, JPHIT, is a collaboration of
public health associations committed to improving population health through
informatics. JPHIT represents governmental public health, and its board of
directors is comprised of the executive level leaders of public health
professional associations at the national, state and local levels. Here comes
the alphabet soup; I won’t read all the names out. ASTHO, NACCHO, AIRA, APHL,
CSTE, ISDS, NAACR, NAHDO, NAPHSIS and PHDSC. Virtually all of the panelists
participating in the morning session today are either JPHIT board members or
participate regulatory in JPHIT activities.

As a consortium of member associations, JPHIT plays a unique role in
standards development. JPHIT provides a forum that enables coordinated and
collaborative developments to occur. It helps present a unified voice on
national informatics policy issues. When we say unified voice, we don’t mean an
enforced consensus. We welcome healthy disagreement. We learn from each other
by sharing points of view around the JPHIT table.

JPHIT itself is not funded or staffed to be a standards development
organization. Instead, it helps coordinate standards development and also
brings together federal, academic, non-profit and vendor organizations. Since
its inception, JPHIT has clearly influenced important standards policy issues
such as meaningful use. It has helped to create a vision for the future of
standards and interoperability. It has placed public health representatives on
national health IT advisory committees. I want to emphasize again, the
brainpower is in the member associations. JPHIT strives to unleash that
brainpower.

I myself am a full-time local health officer. I was trained as an
informatician, but I haven’t done that work for six years. Now I employ two
trained public health informaticians. Together with my JPHIT co-chair, Dr.
William Hacker, retired state health officer from Kentucky, we chair JPHIT to
ensure that it focuses on projects of practical, operational importance to
public health. We see in our work every day the critical role that standards
play in enabling public health to be successful.

For example, in my state, Michigan, we have an outstanding immunization
registry and a great disease surveillance system. Many local health departments
in Michigan have EHRs that are partly interoperable with these systems. The
systems link my jurisdiction to neighboring ones and to the state. At this very
moment as we are speaking, staff are using these systems to partner with the
state to drive up immunization rates and to investigate disease outbreaks. We
just learned of a new meningitis case, in fact. The follow-up to that is
happening in real-time. It is entirely paperless, and it creates great
collaboration.

It took many years to build these systems, but the payoff has been enormous.
Many of the organizations you will hear from today played vital roles in
building these systems. The budget and staffing of my department has been cut
steadily for many years, but our productivity has not been reduced because we
have been able to leverage technology, the most important pieces of which are
standards-based.

As I read through the draft testimonies of the associations, one thing
jumped out at me. Most of them were emphasizing that the greatest barrier to
successful development and implementation of standards is the human factor. I
am sure you will hear this theme, too. Where do we find the time and talent to
do the work of developing standards? Public health is not staffed or funded to
do standards development work. In fact, the work force continues to shrink.
There are other resources threats, too. In fact, the system that I just spoke
about always states the danger of being defunded because of the solemonic
choices that state governments have to make in this era of austerity.

Efforts have been made to train the public health workforce on informatics,
and to increase the supply of people with required skill. Under the Affordable
Care Act, some community colleges developed informatics programs, and
universities and non-profits have created good informatics curriculum. As we
have seen, public health has not been able to hire many of those who were
trained. Clearly, we cannot expect staffing levels in public health to change
any time soon. We continue to search for innovative ways to improve the skill
of the workforce, and alternative ways of collaborating on standards
development that enable a constrained public health system to participate
fully.

I just really found Chase has partnered with the Public Health Accreditation
Board to weave informatics competencies into the standards and measures for
national accreditation. We continue to try to find ways to improve the
workforce. There are now efforts to enable public health staff to train in
place the CDC and the Public Health Informatics Institute, among others, are
exploring this. The real accessibility of these efforts is not yet proven.

As the association has testified this morning, you will hear about
experiments in doing standards development work under conditions of scarcity
such as the Meaningful Use Task Force, the ISDS-led effort on syndromic
surveillance, the Public Health Reporting Initiative and many others. These
examples need to be studied for what they can teach us about the possibility of
making progress even with limited resources. I thank you very much again for
the opportunity to testify.

DR. SUAREZ: Thank you so much, Marcus. We are going to go to Anna from the
Public Health Data Standards Consortium.

DR. ORLOVA: Thank you very much for the opportunity to present at this
committee. We structured our very detailed response to the questions that are
being posed to us in the same way I structured our presentation following the
questions that were posed by the committee to us.

We believe that public health, as well as the health care industry at large,
is at the very early stages of health IT standardization.

Though the individual standards have been under development by health care
and even public health community over the last 20 years, the so-called
interoperability standards are becoming a true challenge right now. We are in
the very, very early stages of developing those standards.

First it was introduced by the Health Information Technologies Standards
Panel, and right now, this work is carried out by the Integrated Healthcare
Enterprise where the Public Health Data Standards Consortium was one of the
two, being invited to create the Quality Research and Public Health Committee.
You will hear a lot of connections between what I am saying to the work that
has been done through IHE through these years.

According to the community-led Public Health Reporting Initiative of the ONC
S&I framework, we identified the following seven domains in which the
working on interoperability standards had started.

Last year, the administrative data and also this data, also occupational
health data were brought up to IHE development process.

Answering the question about the coordination and representation, PHDSC has
developed a business case on the role of public health and HIT standardization.
Most methodologists for a presentation and participation of public health
interest in the standardization efforts is defined in this document. I am proud
to see a lot of colleagues here in the room who helped us to shape this agenda
in 2009 with the help of the CDC.

Public health has been participating in various standardization phases. To
which extent participation of public health and standardization efforts have
been very low. It didn’t change since 2009. The only positive dynamic that we
see is the establishment of IHE Quality Research and Public Health Committee,
where some of the work has been shifted after HITSP was dismantled.

On this graph, it shows you the different phases of standardization, I just
tried to visualize what are those domains that were selected by the PHRI
initiative, and where they are in terms of covering different stages of health
IT standardization.

What are the strongest and the weakest positions of public health being at
the table? Whenever public health comes to the standardization table, it
demonstrates high professionalism, devotion, and interest to collaborate. The
weakest issues that we have found is the very low participation, absence of
support for coordination and facilitation of public health standardization,
absence of certification program for public health information systems, and
continued siloed approach of health IT standardization in some of the domains,
and continued funding of non-standardized public health infrastructure.

What are the incentives and drivers for adopting and using public health
standards? They are the same as the health care. Public health practitioners
would like to have big data to do their job. They would like to share data
electronically with their stakeholders, and they would like to use the robust
infrastructure to support this data exchange. They believe that standards will
do that.

What are the challenges? The business case in the role of public health
standardization highlights those challenges. I will read for you from this list
only one at the bottom, the lack of standardization leaves the general
population needlessly at-risk when the lack of standardization means that
public health response to disasters, emerging threats, is invisibly
handicapped.

There are seven barriers for public health participation in standardization.
Here are the challenges that you can see. Lack of awareness about standards,
absence of coordination and facilitation, lack of federal, state and local
participation in the standardization process. Difficulties in the identifying
entities where you could participate, lack of technical knowledge and skills to
participate, and lack of funding to support participation.

In the return, we are getting insufficient number of standards to support
public health, low level of maturity of existing standards, and low penetration
of standards-based solutions in health data exchanges with public health and
clinical care.

What is the level of public health readiness in terms of exchanges with the
electronic health record system? We believe that we are currently in the
piloting stages of understanding how connectivity between EHR systems and
public health information systems can be established. PHDSC, with the support
from CDC, conducted five trial projects in five jurisdictions on communicable
diseases in New York State, Delaware, San Diego County, and on early hearing
detection and intervention in the state of Oregon and North Dakota.

The types of standards, we have been asked this question, what has been used
there, is listed on the right of this table. An interesting point, which was we
were expected that infrastructure that has been used for both communicable
diseases, and early hearing detection and intervention reporting, was very
similar across the jurisdictions.

What are the main drivers for these states to embark on this endeavor of
building standardized data exchanges? The interest and need to try the
standardized exchanges so they will get data for those who need to know. The
main challenge, accept those vendors who participated in the IHE Connectation
HIMSS Interoperability Showcase. We see overwhelming and alarming lack of
awareness about standards at all.

What are the challenges for implementation for EHR-based solutions for
public health reporting? I would like to spend some time on those slides. In
terms of standards, a lack of standards to support public-health needs and low
maturity of standards. In terms of standards-based products, absence of public
health information systems certification program that assures the penetration
of standards-based systems in the public health space. Lack of incentives for
both health IT and public health vendors to transition to standardized IT
solution.

Personnel and workforce, lack of IT personnel with an overall awareness and
required technical knowledge to operate with standards-based solutions. Lack of
program personnel capable to express the business needs and development
process. Almost absolute absence of personnel designated to represent the
agencies in the standardization process.

In terms of the infrastructure, today we can say that we do not have
infrastructure in this country to support health IT standardization in public
health. There is no process for identifying and prioritizing needs for HIT
standards. Today’s standards are developed by those bright individuals, usually
in the federal government, who would like just to standardize a particular area
without aligning these efforts with the Healthy People 2020 goals and the
Learning Health System goals. There is no infrastructure for coordination
collaboration. There is no infrastructure for continuing maintenance of
existing standards. There is no infrastructure for coordinating of standards
deployment.

Funding, I just would like to split it in two categories. Today, there is no
funding for the positions of health IT standards personnel at the agencies, no
support for their participation in the standardization process, no support for
primary and continuing education on HIT standards. In terms of the
infrastructure, there is no infrastructure today to support public health
standardization, as I said before, which should include the regulatory
framework which Dr. Lipskiy pointed out in his presentation, as well as true
infrastructure which would make all public health reporting knowledge-based.
There is no support for coordination again, and there is no support for
coordination of standardization efforts with the other stakeholders besides the
public health.

Answering the question on the privacy, we believe that privacy should be
embedded in the workflow for the ongoing public health reporting. The least of
privacy and security standards that we have been asked to produce could be
obtained from the PHDSC white paper written together with the Association of
Public Health Laboratory on the standards for public health domain. Appendix
two contains this list of privacy and security standards. We believe it is
reusable for other domains.

With regards to interoperability questions, that was added two days ago. The
AHIC Biosurveillance Use Case Charge called to transmit essential data, about
40 data elements, from electronically-enabled health care to public health and
agencies in real-time. The HITSP interoperability specification identified the
follow categories of standards. There are 107 types of standards to be
reflected to support the transition of 40 data elements. Naming one standard in
meaningful use regulation for laboratory reporting, immunization and syndromic
surveillance could not enable interoperability when other standards were not
specified.

Public health data standards calls this committee’s attention to the
proposed call to action. We believe we need to establish the National Public
Health Standardization Framework for the 21st centry to strengthen
state, federal and local capabilities for exchanging data. This framework
should have the proper resources to be quantified and allocated to support
these efforts so we can move from connectivity of today to the interoperability
of tomorrow. Thank you very much.

DR. SUAREZ: Thank you very much, Anna. We are going to go to Denise.

DR. LOVE: I could make it short and just say ditto to everything that has
been said. I am going to talk about a little different kind of data,
administrative data or the data collected by my members, the hospital discharge
data. Why I think it is maybe significant when we talk about health information
technology in the future is that perhaps there are some lessons learned that
could be applied. We have been at this kind of administrative data aggregation
for many years, I would say decades.

I am honored to be here. I am representing the National Association of
Health Data Organizations or NAHDO. NAHDO was established in 1986, and it was a
spinoff. Actually, the Washington Business Group on Health, who was worried
about health care costs and wanted data, and I feel like we are back to the
future all the time here.

The fact that 25 states had health data reporting programs, either for
certificate of need or rate review or other purposes, came to their awareness,
I guess, the business community. They found that there was variability on the
ground. The 25 states came together under NAHDO, and have been working ever
since then on promoting the uniformity and the use of such data.

We worked again for decades to make the data sets look like each other on
the ground and improve them. The greatest data gap was that we lacked at
patient care and cost. Much of our time these days is directed towards
establishing all payer claims databases. That sort of is hospital discharge
data on steroids, as far as challenges for standards and comparability. We have
quite a bit of work to do, but it is an exciting data set.

I just wanted to follow a few themes based on the questions, but also kind
of connect some dots. I was here last spring; it seems like yesterday. I think
I will hit on some of those recommendations and themes.

One of the things from NAHDO’s perspective is a lot of our standards
development are ground up. The standards are applied at the state level. The
states vary. We work to align and harmonize, and you will hear some of this
today with Bob David, align those variable standards at the state level with
national standards. It is not a one-off deal. It is a constant feedback loop.

As those standards are developed that reflect state practice, in the
national level, they have to be applied back at the state level. Sometimes they
don’t work as planned, and so it is a feedback loop. I will talk about that a
little bit.

What have we done? To align those states and harmonize those states, we have
the health data services reporting guide where we took the state practices for
hospital discharge data and developed a reporting guide that was similar to the
claim, but to reflect state reporting practices. With all payer claims
databases, I will just give a shout-out to the work that has been done with
X12N, the National Council for Prescription Drug Programs, to harmonize the
pharmacy claim. The Medical Professional Institutional Claim Reporting Standard
called Post-Adjudicated Claim Reporting guide workgroup, PACDR.

We are engaging with HIX transactions, a special appointed committee
addressing payer enrollment transactions, state reporting needs. We continue
our state representation, thanks to many folks in this room and currently with
the Agency for Health Care Research and Quality to help continue our state
representation in the National Uniform Billing Committee and the National
Uniform Claims Committee. Again, these are the content groups that decide
content for much of the data that my members collect and really set the state
level.

We feel by tapping into the workflow process of these payer and provider
data systems, despite the limitations, it is proven to be cost effective for
aggregating statewide core data sets. Again, the standard is a starting place.
We work with providers and payers to design reporting requirements that match
their capabilities today.

We create a business case when those data elements are insufficient for
cost, quality, access or population health reporting. We create a business case
to add incremental fields to those reporting requirements, including making a
case for the utility of those data fields for the providers and payers
themselves.

We have found that when a data element has no relevance to the payer and
provider, is not part of the workflow process, those elements are likely to be
poorly reported, missing, and just no one pays attention to them. We really try
to, in this incremental improvement, work with our payer and provider data
suppliers.

Again, educating training providers to collect and improve these new fields
by public health is a huge task, which we have worked on with many of the
groups in this room. Race ethnicity is one. We heard that pushback wasn’t
needed to pay a claim. Then, we have providers and payers admit they don’t just
pay claims with their administrative data. They do a lot more, including
community assessment, et cetera. President on admission, ICD10 are other
examples of these expansions.

What are incentives and drivers for adopting standards? Again, standards
reduce a provider reporting burden and improve data comparability, resulting in
better information. Collective action, though, is imperative. No organization,
no state who maintains these data systems can fix these problems alone. Even if
they had the money, they couldn’t do it alone. We need collective action.

We need to band together because again we don’t have the budgets or the
workforce to fix the data problems that we have. Our reliance on volunteers is
admirable, but not sustainable. I want to give a shout-out to many in the room
who have spent their time, hours of their time, sitting at these standards
meeting. I agree so much with you; there is only so much of us to go around.

What are the state of information exchanges for public health EHR systems?
For population health, and for my members’ purposes, we feel that EMRs and HIEs
are not shovel-ready for population measurement today. In theory, we are
excited about the potential for the needed health care infrastructure. We feel
like there are laboratories for innovation, but again, it is mostly
aspirational.

While clinical data exchanges is an exciting frontier, NAHDO believes that
equally exciting innovations are within reach today. One of these, I wanted to
go back to Marjorie’s comment about administrative and clinical, I think
incrementally bridging those two systems together is an exciting opportunity or
low-hanging fruit. We are seeing states linking their hospital discharge data
bases with laboratory systems to get better information for risk adjustment and
information that is in play today. We call it the hybrid approach. It is
enhancing existing data with clinical data.

An example of this, we feel that the laboratory systems are most scalable
among all the MR domains. They are not pre-text or expensive abstracted
reports. When merged with hospital discharge data that has been classified with
the clinical condition classification system, such as CCS or other, the risk
adjustment models are proven more robust. In the near term public health
information, the future may come from both of these systems put together.
Implementation challenges, I think you have heard plenty about the workforce
issue. We have incremental improvements that we have worked on with NAACCR
harmonizing the demographics between hospital discharge data and cancer
registries.

Repurposing data poses a challenge. We are struggling right now with the
national provider identifier. Everybody wants cost and quality information at
the physician and clinical level, but nobody has a good MPI. The MPI is not
feasible for that. We have states building expensive provider directories and
trying to resolve this. We need some help there.

Again, privacy and security is embedded in everything we do and should be.
We can do better with data exchange and data sharing. I tell people all the
time in public health, we are in the information business, not the data
business. I think sometimes the culture supports the data business more than
information exchange business.

What are some suggestions here? I go back to incremental improvements such
as linkage and harmonization of demographics across all the public health
databases. I will put my neck out a little bit and be a little provocative
because I don’t want to be here. I may not be around in five years; I may
retire if I am lucky enough. I just feel like we need better national
investment.

I will go back to I would love to see a health statistics modernization act
to fix all these problems, or at least create new ones, to have sufficient
funding to build the infrastructure that we so desperately need, the legal
framework, the workforce framework and the health information framework. Maybe
it is a public health trust; I don’t know. We need some sort of ongoing
investment.

We need to fund, either through that mechanism or another, the organizations
and associations to share cost for a full-time standards core. This would cut
across data systems and be collaborative. We really need a dedicated and not
volunteer workforce. The current model is not working.

Again, the human factor, trained workforce, analytic workforce. I just come
up with automation is great, technology is great, but I heard this at a NAHDO
meeting once about it cannot alone solve our problems. When the stores were
starting to automate the scanning codes, I guess there was an unhappy
situation. Maybe it is an urban myth where Target sent a coupon for baby
coupons to a young girl who bought a pregnancy kit, and she hadn’t told her
parents. The coupon went to the home. Without that human intervention,
technology can only take us so far there.

I want to give a shout-out to the JPHIT folks, the Joint Public Health
Informatics Taskforce in the room. Marcus, I appreciated your comments. I think
that is a nice starting place to resolve some of these cross-cutting issues. I
will have my testimony posted on the web. Thank you.

DR. SUAREZ: Thank you very much, Denise. I think we are going to go Bill and
Dave.

DR. BRAND: Good morning. Thank you for the opportunity to comment on this
important topic. I am Bill Brand, the director of Public Health Informatics
Science at the Public Health Informatics Institute. My oral testimony today
will be accepted from the written testimony that I submitted to the
subcommittee staff.

Since the committee will be hearing from many areas of public health today,
we at the Informatics Institute really wanted to share our observations and
recommendations from both a historical and a future perspective across the
public health enterprise. To begin with some general comments, experience tells
us that standards have emerged and evolved, driven by real needs for health
information, at times by economic needs, and at times driven by technological
innovations that compel their use. Standards, like IT itself, are a means to an
end.

As Dr. Loonks referred to earlier, the legal federation of public health
jurisdictions in the country complicates and challenges the formation and
adoption of standards. Whether we are talking about the development of
standards as a normalized process of work, or the standard information system
tool, standards means of transporting data, or even a standard data element,
all have been made problematic historically because adoption of a standard in
public health in the U.S. is influenced by politics, funding, very importantly
and often by idiosyncratic jurisdictional decision-making, and very
importantly, by the absence of an overall strategy and funding to guide and
support modernization of the public health information infrastructure as you
have heard from other commenters.

We will respond to the questions you ask of us within the context of our
federated public health system here in the U.S. You asked us to first comment
on the current state. In our view, standards have emerged and have been widely
adopted in several key program areas of public health. Earlier, you saw a list
of those from Dr. Orlova. For example, the immunization registry community has,
over the past 20 plus years, collaboratively developed a whole suite of
standards that have been adopted nationally. Similarly, cancer registries and
vital records have developed standards and have succeeded in moving those
standards from concept into widespread adoption. Now, they are increasingly
focusing on reducing jurisdictional variability in how those standards are
implemented. Electronic lab reporting and syndromic surveillance are more
recent examples driven in part by the common need to respond to the meaningful
use program.

In terms of the lessons and the drivers, I think the lesson we learned from
these experiences is that standards will emerge and be adopted when there is a
clear need and business case driving their creation and adoption, and when the
community of stakeholders, most affected by the adoption of the standard, drive
the process. I think the examples of effective standards adoption in the U.S.
show that no single standards or other organizations were needed to create or
force the adoption.

Rather each of those enterprise systems develop standards because of the
sheer willpower of the community of users and developers who agreed on the need
with the help of knowledgeable informatics experts working in cooperation with
relevant standards development organizations. In other words, they had a
compelling information problem to solve that links directly to an understanding
of a health problem at issue.

In terms of the incentives and drivers, I think when you look at those
examples, it is clear that the incentives and the drivers was in the compelling
programmatic need, and in the understanding that all parties either are or
should be doing the same work in much the same way. The recognition that
agencies have more in common than different undergirds effective diffusion of
standards than our federated public health system.

The implementation challenges we see include how long it takes for new ideas
to diffuse the nearly haphazard way in which public health information systems
are funded in the categorical funding streams, and how jurisdictions choose to
support adoption of new systems. Because each jurisdiction dictates its own
legal and financial future, we depend on standards adoption to be driven by
compelling business case rather than on edict from a funder.

Adopting new standards within hundreds, not thousands, of autonomous
jurisdictions depends on their ability and willingness to fund a transition to
new or enhanced systems, that they have the technical capacity to adopt the new
system, and they have the understanding of the business purpose driving their
use of a standard.

As Dr. Cheatham and Orlova mentioned earlier, clearly the biggest
implementation challenge comes down to the capacity of public health
practitioners and their respective associations to engage in standards
development work, starting with building the value proposition and going
through widespread adoption and effective use. Too many current
standards-related activities suffer from minimal practitioner involvement.

Any solution to this problem must help ensure sufficient numbers of people
available for the length of time required to do this work. That they have
sufficient knowledge in both the program need and the overall standards
development process. That there is an availability of standards development
expertise who can help navigate these projects through the rather arcane
process of standards approval. Very importantly and largely missing today is a
clear overall strategy that is shared by CDC, and local and state practice
communities, on ensuring the widespread adoption and effective use of a
standard.

As we look forward to an era of e-health and e-public health, you ask what
interoperability standards are needed to assure public health being able to
work in tandem with health care. We believe the answer to this question doesn’t
lie in specifying specific standards, but rather in public health specifying
value propositions that have meaning to their data exchange partners.

Our point is simple, but really very important in that creating the
standards means nothing if it is not used. In fact, we think the standards
community at large needs to assess its impact based on actual implementation in
use, not on existence of a standard no matter how elegant it is. To achieve
use, one needs to reach the hearts and minds of the people who much use it and
pay for its implementation.

To reach them, they themselves need to see and have a clear and compelling
reason that adoption of a standard brings tangible benefits to both them and
their community partners. It is really upon this foundation of rational joint
economic interests that we see the right standards being developed and broadly
adopted.

In conclusion, we want to reiterate that it is programmatic needs that drive
creation of the standards. It is a joint business case that drives the adoption
of new standards. Standards would be meaningless absent the commitment by
public health agencies to their health care partners and to each other, for
that matter, that they can hold up their technology side of the exchange.

We urge the committee to make an urgent and emphatic recommendation to the
secretary that this sort of omni-directional data partnership among public
health and its health care and other community partners is a business
relationship with will require new financial commitment, one needed to support
and modernize the information infrastructure of public health. We ask the
committee to recommend the creation of a public health information trust fund.

Such a public health information trust fund can support major activities,
some of which earlier testifiers alluded to. One is to support the public
health informatics, and public health services and systems research communities
in assessing current adoption levels of standards, and particularly identifying
where the barriers and gaps are.

A second activity could be to conduct analyses of public health program
areas within which a clear value proposition for a standard may exist, but has
not yet been acted upon by the community, including what support they need. For
instance, some areas of chronic disease and environmental health may be areas
to investigate. A third possible activity that could be supported by this trust
fund is to engage standards experts and make them available to the public
health program community, not to drive the development of the standard, but to
support the community in the process of establishing its standards based on
defined business needs.

Another area is to provide cost-effective training both in informatics and
in standards development to the public health practice community. Lastly, and
very importantly, to support the collaborative development of an overall
roadmap that lays out a path for how public health can transition away from
today’s monolithic and rather inflexible information systems toward a more
modular, reusable, sharable services-based and standards-based design of
information systems.

As Denise just alluded to a minute ago, public health is an information
enterprise and standards clearly play a role in public health being effective
as an information enterprise. Absent a broader engagement strategy, clearly
articulated and compelling business needs, and adequate funding and expertise,
little of real and actionable value can be achieved in a timely way and in an
ongoing way in our federated public health system.

Thank you for the opportunity to testify. We hope the committee will use its
credible voice to advocate for real change in how public health invests in its
information infrastructure in an e-health era. Thank you.

DR. SUAREZ: Thank you so much, Bill. It was really an excellent way to
finish this first panel. We are going to go into our discussion and questions
from our committee members. If anybody wants to jump in and ask some questions,
we will move along.

DR. FRANCIS: This is a question that Denise’s comment about making the
business case leads me to ask. Also, some of what Bill was saying a minute ago.
I wonder whether the development of huge private sector databases of the sort
that, for example, the Cleveland Clinic or the blues, I mean, I am looking at
both administrative and clinical that are being developed to do within their
systems or marketed for others is another barrier to seeing this happen on the
public health side.

I just wonder if you have any thoughts about that, and if you have any
suggestions for the committee about whether another issue here isn’t just
starve the government, but do it in the private sector in a way that serves
private interests, and sometimes public interests, too.

DR. LOVE: That is a loaded question. The position of NAHDO has been that the
private sector has excellent data resources. In fact, many of them are NAHDO
members who have private data systems, university health systems consortium and
others who do similar work. They aggregate across enterprises. I think that is
probably the unique thing about NAHDO and some of its members.

The role of government is a very interesting NAHDO session we have talked
about many times. I do think it is not at odds with what is going on in the
private sector. In fact, and I welcome debate on this, but I feel like if we
are doing our job in the government, in the private sector, with
multi-enterprise data aggregation, we should be pretty much in sync with what
the Kaisers are doing and others because the data that I am talking about are
similar data. The capacity for aggregating is only so much at a given snapshot
in time.

The government has a unique role in that they can convene competitors and
compel them to share the data that everybody wants. I have been in multiple
debates with legislators and others across the country. It really is a powerful
sort of agreement that there needs to be a convening body and an overarching
independent and neutral source.

Now, we are seeing more foundations another step up to that. Government has
to be at the table to level the playing field so that what everyone decides on,
the stakeholders, is collected in the same way that there is no gaming, and
that the benchmarks that come out of that benefit the data suppliers. It is an
intricate dance, and I say it is more political right now than technical. I do
think that government just has to stake out its role and own that role.

DR. SUAREZ: Anna is going to make a statement and then you can go next.

DR. ORLOVA: The legal documents that Dr. Lipskiy just presented to us define
the role of government in threefold categories. The government as a participant
in standardization process, as a leader and as a user. Today I believe we see
government playing the role as participant, which is very telling what you see
from those slides and also these groups that you assembled today.

In our recommendations from PHDSC to this committee, we are actually calling
for the alignment of incentives for collaboration between public and private
sector. When we are going to health information management and system society
interoperability showcase, we see a very strong interest from the private
sector in public health standardization. We did not see this interest in the
regulation for public health standardization that government will produce.

We would like to call that to the literal and remind government about the
literal that it should play by mandate in this process, so that maybe one day,
government will also become a user of standards.

DR. BRAND: I think it is a longstanding principle of governmental public
health, that public health does what others cannot or will not do. The growing
interest in the private sector in population health activity, I think, is an
important development, and one which public health needs to support. It is very
important that, as Denise was saying, public health be at that table at the
community and regional level in having access to and analyzing these large
clinical repositories in population health and management databases.

In part because as you know, that term population health is rather
problematic because it suffers from many different definitions, depending on
the eye of the beholder. Clearly, governmental public health is perhaps the
most inclusive definition of population health and the most experienced today
with decades of experience in analyzing population health metrics. The key, as
we have alluded to before, is having sufficient people with the skills and the
savvy to participate with the private sector in a way that helps both their
needs as well as the public needs.

DR. SUAREZ: We have Bruce and Jim and then Alex.

DR. COHEN: Thank you all for a very provocative testimony. I want to pick up
on a theme that Bill just began to discuss. First of all, it is heartening to
see that federal government does have the regulatory framework that allows us
to become more leaders in facilitating discussions and leadership around
standards development.

My concern is the focus on clinical standards rather than on population
health standards. My underlying fear is that public health is being high jacked
by medical care payment reform, and we haven’t figured out in particular how to
integrate true community needs into the development of standards. When we talk
about population health, the estimates range from 10 to 30 percent are affected
by clinical interventions. If the federal government really wants to focus on
its role in supporting community health and quality of community life, how do
we integrate community needs into the discussion around the content of
population health standards? Anyone feel free to respond or comment.

DR. LOONSK: That is the 64 thousand dollar question. I think that one of the
things that we have been doing is looking at the convergence of population
health and public health informatics particularly in the context of the
Affordable Care Act. Where the Affordable Care Act endows clinical care with
certain functions that historically have been associated with public health, to
quote someone, not my origination, that in many ways population health is
clinical care trying to reinvent the public health wheel.

While we may have the regulatory capabilities in some respects to achieve
standardization, the incentives that have been so hard-fought to try to get
EHRs to be standardized, and to be clinical care to pay attention to health
information exchange for continuity of care purposes have not really been
oriented to population health needs. This is clearly a critical time for that
because it is going to get harder and harder, not easier, in terms of the way
in which population health systems are architected inside of clinical care, and
how they meet broader public health functions. One of the reasons that I was
stressing so strongly getting into meaningful use in a more significant way to
try to get population public health functions into the developing population
health infrastructure and clinical care.

DR. CHEATHAM: This is Marcus. That is a fantastic question, and I don’t
think it has answered. I think it underscores the importance of public health
partnering with the private health care system. NACCHO wrote a really
interesting paper on health disparities and informatics, calling out some of
the critical challenges that exist in implementing informatics and not losing
their focus on community need. Again, I don’t think the question has been
answered, but I think public health needs to be at the table to help find those
answers.

DR. LOVE: One of the concerns that we have is that the databases that are
serving some of those community needs, and I will go to the hospital discharge
and the all-payer claims, there is no federal funding. They are state-funded
and under-funded. I would say probably vital records are the workhorse data set
that is starved, as well.

Again, the infrastructure is a little rickety, but I do think that with
payment reform and ACOs, I have been impressed that the business case for
all-payer claims databases are just coming out without us pushing too hard. An
ACO only has a single payer glimpse of their population. When you get down to
the clinic and the physician level, you run out of observations at a CPT line
item. You need to aggregate across the system. It really is sort of
developmental. I am hopeful that they will evolve and provide some of that
information that you want.

DR. SUAREZ: Thank you, Denise.

DR. ORLOVA: When I hear the term, population health, and thank you very much
for bringing it up, I always think about non-clinical data systems that we are
not even touching base on right now. I would like to see that this National
Public Health Standardization framework concept, the roadmap, whatever this
infrastructure that we are building together would also include the
environmental data systems that right now are under the jurisdictions of EPA,
HUD housing systems, Department of Energy, Department of Commerce and so on.

Without that approach, which is focused not only on the clinical outcomes,
but on the full-scale of the public health landscape, I do not believe we will
be able to reach to the Healthy People walls as well as Learning Health System
walls. To my knowledge, this committee, in fact, already discussed this
particular issue that public health is beyond this particular healthcare thing.
We should be definitely thinking boarder. Thank you for this question.

DR. LOVE: We need identifiers on all of those databases.

DR. SUAREZ: Thank you. We have four questions from Jim and Alex, and then
Mike and Linda. We have about 10 more minutes. I am going to give five more
minutes for this. If you can just target your questions, I would appreciate
that.

DR. SORACE: I am just interested because we hear about EHR data interchange
with public health. A lot of the public health data actually begins in
laboratory information systems that are somewhat distinct from EHR systems. How
do you envision lab to public health data flows, and are they necessarily
dependent through an EHR intermediary? Are they sometimes just going directly?

DR. ORLOVA: In our response, we called them clinical information systems.
That includes EHR, laboratory information management system, pharmacy systems,
payer systems and other systems. We believe that all these systems with the
current technology can utilize the same infrastructure. We may call it cloud;
we may call it health information exchange. The full type of infrastructure
technical implementations that could carry CDA document, or message for that
matter, from point A to point B through this health information exchange is
highly reusable.

We see it happening at IHE connected on every year. I believe that this
country could also implement laboratory health information exchanges the same
way as the rest of the world is doing already for five years.

DR. SUAREZ: Thanks, anyone else?

MS. GOSS: It appears to me that I have consistently heard this morning a
call for orchestration of resources and funding a focused set of resources to
really play facilitator and ombudsman in bringing all of the various pieces
together in the public health with the clinical administrative financial
efforts. If I extrapolate this, I am hearing that the public health arena as an
industry is asking for its own consolidation with some overarching federal
leadership. Have I misinterpreted that, and have you brokered that agreement
among your brethren? I would love to hear from all of you.

DR. LIPSKIY: This is even more than a 100,000 dollar question. Now, I am on
the spot. To make it short, I think the critical issue for us, for all people
who, in this room and on the phone, is understanding how to build regulatory
agenda for public health informatics. When we cry, and we don’t have budget, we
don’t have this convergence, et cetera.

One of the primary questions is how to do it through legal framework. When
people say yes, we have everything, we have some gaps that may be improved,
that may be added to the regulatory agenda that will help us to achieve this
huge agenda that we are just thinking about.

MR. ROSS: In response to this question, I just offered my own opinion. My
opinion, and it was reflected in our testimony, is that we will not see a
massive coming together, or consolidation or federalization of the public
health systems. It is what it is. It is a decentralized by constitutional
design public health system. There isn’t the political will in this country to
change that.

Consequently, where there is a strong value proposition for working
together, our experience, and what was reflected in what Bill said with major
domains of public health, they can see they have a common need, and they can
come together. I think it is driven by a highly refined, clearly articulated
value proposition. It must begin there.

Denise made similar remarks much to the success I think that NAHDO has seen
is their willingness to work with the various health care partners and state
partners to build a value proposition around each standards piece that they are
working on. To think that we can put through legal edict to force it, I just
don’t think that is in the American political spirit. It hasn’t happened in 250
years; I don’t see it happening any time soon. I think this is going to be
incremental.

It is going to have to be driven by localized value proposition. Therefore,
public health has to do, as Marcus Cheatham said, be out working with your
local partners. I think that is where value gets realized. Where we need in
public health is enough resources to get technical talent so that when
standards are needed, it becomes obvious that you can build that political will
across public health and with its data-sharing partners.

DR. SUAREZ: Thank you, Dave.

DR. ORLOVA: Last year, in this room, the Public Health Data Standards
Consortium approved its strategic plan. The first objective of the plan, as you
correctly pointed out, unified public health. The second, align public health
vision objectives and activities with another partner. The third objectives is
to unify public health information systems.

The topic that we are discussing today is that foundation for what we did.
It is actually our ground we stand. Our hats, our hands, our body all moved
because of the activities that involve those standards. I think unification of
public health, presenting public health as a sector, as a unified sector, that
understand that each particular programmatic activity is so important to
deliver public health services and public health mission.

I think that is something that we really have to work on. I hope very much
that in this public health national standardization framework, the priority and
needs for standards would be based on that particular notion of unifying public
health as a sector.

DR. LOONSK: I would agree with Dave that consolidation of the public health
sector does not seem to be in the cards. The way I would articulate it because
IT and the standards part of it, the technology part of this, pressures public
health to be more coordinated than it has had to be in the past, to the
infinite detail. What is really happening is that clinical care and meaningful
use are putting to the question to public health, can you coordinate in a way
that you can meet this side of the bargain.

That is part of what we are seeing in action right now, in terms of
meaningful use, in terms of the public health participation numbers or not, in
terms of meaningful use participation for public health measures. What we will
see for some years to come is whether indeed public health can coordinate to
the degree that it can hold its part of that bargain.

DR. LOVE: I agree that we are in silos, and we will continue to be. That is
the way that our system and the infrastructure is built. From my perspective,
the states will continue to promulgate some unique some state-specific fields.
We will have to percolate them up and align them. That is iterative. I think
that loop will continue.

What I think we all can come around the table with, regardless of the silos,
is consolidate and unify how we collect our identifiers. If I just take two or
three public health data sets, and Bob knows this, and harmonize across those
data sets, the same fields of patient demographics, the programs are collecting
them, sometimes in different ways, even though the standard may be the same. I
think we can consolidate around certain aspects of our data enterprise, but not
one voice answering everything.

DR. SUAREZ: Mike?

DR. FITZMAURICE: First of all, I want to thank the panel for such thoughtful
presentations and some good statements of needs. You really serve us well as we
help to give guidance to the secretary in the public health area. I see the
public health and private health have a lot of the same issues and some of the
same solutions. You have to get the data right. That is a good part of why we
are here.

You then have to get the analysis right of that data, whether it is public
health or whether it is other. You have to get the implementation right once
you decide what the truth is and what action you should take. Now, we are
seeing a large issue with a national website. Behind that, a lot of decisions
have to be made. I am going to ask Denise a rather specific question she has
already half-answered already.

The accountable care organizations, the states, will need data for risk
analysis, and for setting payment rates and for doing a lot of other things. Do
you see a stronger value, a stronger push, toward all payer claims data, for
more states to adopt them, or them to come together in some kind of governance
to make the data better? Not to make health decisions, but to make the data
better.

DR. LOVE: Absolutely, they are coming together. I would say the greatest
barrier has been heretofore lack of funding. I have talked to more states, red
and blue, that want all payer claims databases, but the business case, how to
pay for them, how to sustain them, has been difficult.

We are seeing some states taking advantage of cycle three funding from CMS
to build that infrastructure. Absolutely if everything they want to do,
transform Medicaid, payment reform, consumer engagement, all maps back to you
need information. It is going somewhere; I don’t know where.

MS. KLOSS: I would like to add my thanks, both to the panelists and to this
excellent discussion. I have had a follow-up question to Bill. Your seminal
recommendation that the committee recommend the creation of a public health
information trust fund with the functions that you so clearly laid out. In
formulating that recommendation, did you give any thought to the model of
governance for that fund? How did you envision it working? If the committee
went forward with such a recommendation, how did you envision it? I think it
would be helpful to us if you shared some of the additional thinking you had
behind that excellent recommendation.

DR. BRAND: Thank you for that question. I think I would say that a key
principle would be that it needs to be jointly governed by CDC and a group that
represents the state and local practice community, such as the joint public
health informatics taskforce which you heard earlier from Dr. Cheatham. I think
that is really key that while the federal government, as Dr. Nikolay said, has
some regulatory powers and duties around standardization, they generally have
their federal perspective which is a different perspective than at the state
and local levels where public health hits the ground much more.

The needs for data are much different at a federal level than at a state and
local level where it is largely about using that data with their community
partners to really understand population health need. It is not about the
surveillance, the large-scale surveillance at the federal level.

I think a joint collaborative approach to that trust fund between a group
like JPHIT and the CDC. The key is that it can’t just be within one center at
CDC, as well. It really needs to be as much as possible, socialized across the
CDC because of the fact that all these silo fundings come from the various
programs within CDC.

I think the only similar initiative like this was back in the ‘90s with
the information network for public health officials, the Info project which
didn’t last long enough to really be socialized across CDC, but had that
purpose of really sort of helping to update and modernize the public health
information infrastructure. We could also begin by looking at what was learned
from that experience.

DR. SUAREZ: Suzie, last question.

MS. BURKE-BEBEE: Extending that a little bit further, Bill, how would that
governance incorporate, include what is already in place clinically. How would
you see the stakeholders being involved from that perspective with high tech?

DR. BRAND: I am not sure I caught the beginning of it. Could you say it
again?

MS. BURKE-BEBEE: I am wondering how this trust fund and the governance that
you just described, the CDC and others, would incorporate the meaningful use
component, high tech, the clinical component and all of those stakeholders, the
crossover.

DR. BRAND: That is a very important principle, as well, about engaging the
stakeholders. It is a key principle in any informatics initiative, I think. In
any such initiative, really chartering this process and getting the governance
right is critical. I think representative from those various associations,
whether that is AHIMA, HIMSS, HIEs, the Office of the National Coordinator,
many of those as well because this is not just about meeting the needs of the
public health agency, but meeting the needs of the public health agency and all
of their community partners.

It is about really creating sort of the win-win win-win kind of strategies.
As we alluded to earlier, the value propositions here really have to be able to
meet everybody’s business needs in order to be sustainable and to have real
lasting value.

DR. SUAREZ: Thanks so much, Bill, and thank you everyone on this panel for
an excellent way to get us started. We are going to be really enforcing our
time now. I have 10:41, and we are going to give you a 14-minute break, and we
will come back five minutes before the top of the hour.

(Brief recess)

Agenda Item: Panel 2: State/Local Public Health
Perspective

DR. SUAREZ: We are going to get started. I think we are going to follow the
same order that is listed in the agenda, so we are going to start with ASTHO. I
know that we all probably know what each of these acronyms mean, but it will
always be helpful to decrypt those acronyms.

MS. WILLAMSON: Speaking of the acronyms, I want to bring all of the members’
attention to a document that you should find in your packet. It is called
Health Information Standards and Related Organizations, the Players. This
document was something that started within CHS providing support for the Public
Health Data Standards Consortium, so it was one of the organizations documents
in collaboration with NCHS. We decided to update that document and all of the
acronyms that Water just references including a host of others are included in
this. I have to give kudos to my team members, Dr. Tamara John Paul and Heddy
Kahn who spent a lot of hours updating this document. We hope this will be a
useful resource to the committee members. We plan to put it on the share point
side as well. It can be an evolving document. If ASTHO is not there then we
need to change.

DR. SUAREZ: I would like the name the players.

MS. WILLAMSON: We will take note. That was definitely a slip back to have
ASTHO in the group and I am getting other recommendations. This will be
evolving. Committee members can keep it as a rich document.

MS. SOPER: Actually I am going to defer for the moment to my colleague,
Roland Gamache, because my slides aren’t loaded yet.

MR. GAMACHE: Thank you very much for the opportunity to present today for
this committee. What I tried to do is find examples of a lot of what we talked
about this morning and a lot of the areas that were discussed this morning and
find real concrete examples of some of the skin on them so that we had some
examples of public health applications of this. I am looking for the new
emerging integration of public health with electronic record or with the health
information exchange.

Again, these are the topics, and these are the questions that were asked. I
was thinking about addressing each one of these individually, but so much of
any answer overlaps in each one of these different topic areas I thought what I
would do is present our use-cases and where we are trying to use this
information and how we are trying to use these standards that are already out
there and where they are not out there and where the issues are. Then we can
try and fill in the gaps as we go forward.

The use-cases I am looking at are flu and chronic disease monitoring from
syndromic surveillance systems as well, looking at these both for outbreak
management and mitigation issues. Maintenance and value sets. We heard several
people talk about this. I will define that term as well. Then the
interruptability mapping of local code sets to thinks LOINC or other standards.

The first one is what is a value set? This is just a collection of concepts
drawn from one or more terminologies or systems grouped together. The example
that I am going to use here is the reportable condition mapping table that is a
joint product of CDC and CTSE is also involved with that as well. Before I get
to more of that what I wanted to do is show the values set and show the
definition of some of these. This is an area where we don’t have a standard set
per se that is accepted by public health. What we have is a group that has
gotten together and formulated a bunch of different conditions. I wanted to
show this one because sometimes picking what that values set should be isn’t
very easy.

This one shows several different examples. These are all different examples
of the definition of influenza-like illness with a different standard
definition. One of them was laboratory tests that were ordered. One of them is
an ICD-9 code confirmation. Some of these are from syndromic surveillance of
conditions as well. Notice that one by itself doesn’t tell you what was
happening in the community.

In the middle that shows people showing up in the emergency department with
influenza-like illness. If you looked at some of the other ones, you would miss
some of that blip and would miss what was happening in the community at the
time. This was again the worried well coming in. This was H1-N1 outbreak. You
could miss worried well people when you are trying to do the outreach and
mitigation efforts. That is what I wanted to show in some of these standard
development processes that it can get complicated as well when we move forward.

This is one page of the different conditions that can be used for that
particular definition of flu. There are six more pages that go along with this,
identifying different indicators. The indicators are set mostly from the
literature, but from different people who put the different definitions
together. There is no set standard yet for public health to use in this one.

I think this is also part of the development. We heard about this iterative
process of standards development. For someone moving into trying to do flu
surveillance in a local health department, trying to pick from six pages of
indicators gets challenging, so some guidance here is also what we are talking
about. It is the standard development process. What are some of the things that
we could use to help public health work better with the clinical community as
well because they are defining this also? So this is joint clinical and public
health development of an indicator list.

I also wanted to show this other example because although I am talking about
outbreak management and mitigation efforts that is something that goes across
the different sides of the health department. Large and small almost everyone
does some type of mitigation or outbreak management in their community at some
point in time. These syndromic surveillance systems can also be used to detect
conditions. In 20 years in public health this is the first time that myself, or
any of my colleagues, had a public health condition identified by the public
health department before we got a call from clinical medicine. So this was
really exciting in the sense that it helped us identify something, not that we
actually had an outbreak.

This is another process to maintain, notify condition mapping table. We
heard several people talk about this with the funding and so forth. This is my
plea here where I really feel we need to show that we can be stewards of this
data. When we can’t maintain our existing value sets the way they should be
maintained, then we are not showing good stewardship in the process. For me,
that is the underlying number one thing. We can’t go and ask for more money
when we can’t be good stewards of things that we are supposed to be dealing
with.

CDC and CTSE do an outstanding job trying to keep this up to date, but it is
really challenging to keep up with all of the different definitions. LOINC
codes change twice a year. Keeping this set up to date is an important issue as
well. Then we are also moving to ICD-10 so we have that challenge as well of
trying to integrate those changes into this table. What are we going to do in
the interim until we get those definitions in place?

Again, here are some of the different ways to try to do that and the
different conditions that are out there from people mapping these conditions
differently. These are both recognized institutions in developing these mapping
tables. You can see the number of conditions, 930 weren’t in one of the tables
and 730 were not in one of the other tables as well. How do we integrate this
coordination and make this happen in a more timely manner? It is critically
important for public health.

There is another area to look at, and this is our partnership with clinical
medicine. I think it is really important as well. A lot of clinical groups when
they start moving to LOINC codes or standardized datasets, they look to code
these top 2000 codes first as a first step in getting all of the LOINC codes
coded. Meaningful use requires them to have all of it, but the first step is to
get the majority of them.

The top 2000 represent 98 percent of the clinical code sets. It only shows
65 percent of the public health codes that are needed. Here is another are
where I think we need some work working with our clinical partners as well in
defining what those conditions are for public health that should be included in
this set right away. Some of these are critically important conditions that
need to be included in there.

Another example, and I think this is really important, is this is a public
health alert being sent out. I thank partners from Marion County Public Health
Department as well for sharing this with me. A public health alert is not the
same as a clinical alert. When we talk to our clinical partners, a clinical
alert means something different. A public health alert carries more weight, but
if we put the word alert on it, it becomes a clinical alert to them. There is a
big difference here, and we need to help identify what that difference is and
what that terminology would be.

Again, part of the standards terminology that we are talking about, but I
think this partnership with clinical medicine in defining some of these
differences in terminology are and how we need to define them for use both for
public health and for clinical medicine is important. This is a big part of the
bi-directional communication that we need to really further develop with our
clinical partners as well.

We heard a little bit about CDA. This is part of the automated filling in of
the form for a case report. Again, this is a lot more challenging than you
think it would be. Again, the CDA is something that needs to be developed and
we have been talking about it for a long time. Whatever that tool is this is
another big area where public health is ready to start moving forward, but we
don’t know quite what to do yet. This is another big area for standards
development as we move forward.

Finally, I want to talk about a few of the areas that I didn’t talk about,
newborn screening. We heard about this birth and death as well that people are
going to discuss here. Immunization and cancer registry. There has been some
outstanding work done. The Data Consortium and other groups have developed
use-cases for this. I think a lot of these are very well-developed at this
point in time. We have good things in place. What I have tried to present here
were use-cases of emerging conditions as well. CDA is one I mentioned,
bidirectional.

Chronic disease is another area where people are trying to move in. What are
those definitions that we can use to help? Communicable disease is something
that we have established now. I think the next step for public health is trying
to help impact this chronic disease development area. The standards for what we
need to have in place for identifying what the chronic conditions from both a
syndromic idea and also from a laboratory confirmation are going to be really
important as we move forward.

We would like to think about birth and death in a number of ways. We talked
a little bit about security. This is part of the national security issue as
well and not having that information timely hinders our ability for a number of
different reasons. There two components here. There is a statistical dataset
which I think we can get going pretty quickly, but that legal requirement as
well can be also what is challenging.

Again, although it is something we have very well-defined, I think there are
still some places for some major improvements in this spot, and I think one of
our other partners will discuss that. Really quick, this is summary standard
definitions for syndromic surveillance conditions and its potential use to
monitor and aid and resource deployment for chronic diseases. Work with our
clinical partners I think is going to be important for the future in our
standards development, particularly as we start in the initial implementation
of EHRS systems with them.

Bidirectional communication with primary care and standardization of what
that public health alert and the operational maintenance of value set of public
health importance. I think of all of the things is probably the most critical
to move forward. We have this good definition and these values sets that we are
currently using already and have them up to date and maintained. Thank you.

DR. SUAREZ: Thank you, Roland. Maybe we can go to APHL.

DR. KENNEMORE: Good morning, I am Dr. Wes Kennemore. I am here representing
the Association of Public Health Laboratories. My slides are up there if we can
figure out how to get them to show. The Association of Public Health
Laboratories just a general comment for the sake of brevity I won’t read my
entire written remarks. I would recommend that the committee can go and review
those at their leisure. I will make a few brief verbal remarks and then we will
pass on to the next speaker.

So the Association of Public Health Laboratories is the national nonprofit
and we represent the governmental laboratories that monitor and detect public
health threats, ranging from lead contamination in drinking water to metabolic
conditions in new newborns as well as genetic conditions in newborns. We have
been working extensively in the development and implementation of both data
standards and messaging standards for over eight years and obviously in the
case of things like newborn screening a lot longer than that.

Just to talk a bit about the complexity of the landscape, we are going to
address our remarks at APHL strictly to laboratory reporting or reporting of
laboratory data as opposed to other forms of clinical data I should say. When
you look at this chart, and it is by no means all inclusive, it is just a
representation of the number of players that are involved, the number of
organizations with whom we would have to communicate and the number of
organizations and types of players that we send data to, as well as a brief
listing of some of the standards; actually not the standards, but the actual
type of data that is being communicated.

If we look at a more comprehensive layer in a simpler way this light
actually looks at the individual stakeholders and some of the individual
standards that are involved in the way we exchange laboratory data today.
Again, I want to emphasize that this is by no means comprehensive. It is just a
snapshot of some of the complexities involved as we try to exchange data and
standardized data in the way we communicate across all of the multiple players
involved in this field.

This slide is addressing the comments about how we get public health at the
local level, involved in standards development activities. I am showing in this
slide APHL kind of at the center of this HUB, but I want to emphasize that APHL
is only one such hub. We collect data and information from our member
laboratories. They can be environmental labs, territorial public health labs,
any one of a number of different labs that perform any kind of testing of
public health significance.

We bring input from them through our various communication methodologies and
then we meet and disseminate information from our members out to all of these
other organizations on the bottom half of the slides from CMS to JPHIT to the
SNI framework. Likewise, many of those organizations we communicate to operate
as hubs in their own right. They have additional members that collect
information from just like we do and then they communicate that information
back into the national forums.

One of the weaknesses of this is that we can only go so far. There is a
limited number of us, and in my case, participation in some of the national
organizations, JPHIT for instance is one very, very small segment of what I do.
I am able to spend literally a few hours a month focused on JPHIT activities
because a lot of my time has to be focused on helping to implement particularly
the ELR standards in assisting state and local labs with implementing ELR
standards.

Again, there is kind of a great vacuum of resources and focus on how we
develop the ability to work together. We are certainly putting a lot of effort
behind it and each of these individual organizations would tell you how much
time and effort they are spending working with these types of networks to try
to enhance the development and deployment of standards across the public health
community. We are very limited and we are stretched very thin in those efforts.

There are also a number of different systems that we have to collect data
from. To address one of the questions that was asked of the previous panel,
does laboratory data typically get communicated back to an EHR before it is
reported to some public health entity? To be quite honest with you, our
experience is quite often not. It is obviously critical that that laboratory
information be integrated into the EHR system. To initiate the public health
reporting activities that information typically comes out of the laboratory
information management system.

Now, one of the weaknesses there is that in the early stages as we looked at
stage I in the 2011 standard much of the information that is required wasn’t
necessarily contained in those lab systems, so we wound up having to use a lot
of integration engines as you see in the middle here whether that be Rhapsody,
Mirth or Cloverleaf and the list goes on and on.

We have actually had to use a number of integration engines both to acquire
supplemental data that we might not have had readily available to us in that
initial extract. Also, in some cases just to format the message into HL7-2.5
message or in some cases 2.3. Add to that the complexity that as we report that
into the public health environment there are a variety of different siloed
systems that we are reporting to.

Again, that has been alluded to already many times this morning whether that
be the HIV group, whether that be some other surveillance group or some other
disease group. There are a number of different systems. So these integration
engines wind up having to work on both sides both from the integration of data
before it is transmitted and then from the standpoint of parsing that data back
out. Again, there are just a number of different standards that we have had to
use and that we have had to address as we look at that type of information. As
a quick summary and again that is addressing the integration standard and the
integration engine.

As a quick summary, the state of some of our standards is very well-adapted,
very well-developed and utilized. I take for instance the HL7-2.31 message that
has been used for Influenza reporting for a number of years. It is very
well-developed. We are reporting data from about 48 states live right now to
the CDC on that information. Some of the others like the 2.51 standard, while
it is well-developed, the utilization is certainly emerging as we approach the
later levels or the new levels of meaningful use.

As well, the CDA data and I think Dr. Orlova referred to this earlier. Some
of that is very well-developed for some applications, but not so well-developed
in other areas. Obviously there are barriers. This is a consistent theme,
inconsistent funding, lack of resources, software limitations, training
requirements. These are going to be universal across all of the different
entities that are trying to move and implement these standards. Obviously the
incentives and the drivers are the efficiency and utilization of personnel.

The data centers are driven by government financial incentives. Again, I
point back to the early days of meaningful use when public health actually
wasn’t funded as part of the meaningful use stage I incentives. That was kind
of an afterthought where some of the organizations were funded to help do some
technical assistance. It is important as we move forward that public health be
included in all of these efforts for standardization, but also that we are kept
in mind as funding initiatives are put together to deploy these initiatives
into the public. That concludes my remarks.

MS. BURKE-BEBEE: Your slide has an acronym I am not familiar, FHIR, your
slide.

DR. KENNEMORE: It is actually called FHIR. The concept is basically
replacing some of the HL7 standards 2.3 and 2.5, et cetera, with some more
common transmission messaging methodologies.

MS. BURKE-BEBEE: Is it specific to public health?

DR. KENNEMORE: It is specific to sending information to public health.

DR. SUAREZ: Fast Healthcare Interpretability Resource. It is a complimentary
resource in the HL7-2 set. It is really becoming a major force in accelerating
the way in which HL7 standards can be implemented. It is really a developer’s
field. It applies to all standards and no just public health. Certainly it
applies to all of the body of standards in HL7. It is evolving. It is
developing. It hasn’t been adopted in the meaningful use realm yet or anything
like that. I don’t whether it will be expected to be, but it is a major force
now moving forward.

MS. WILLIAMSON: In public health I know the primary focus has been with PHIR
and immunization.

DR. SUAREZ: Thank you so much. I think we have now ASTHO’s slides ready.

DR. CHANDERRAJ: Actually ASTHO is listed as an external partner. I am sorry.
PHIR might be one.

MS. SOPER: In the interest of time, I will go ahead and get started and just
skip forward on my slides once they are up and running. My name is Paula Soper.
I am the senior director for E-Health at the Association of State and
Territorial Health Officials or ASTHO as we now readily know. I want to thank
you all for the opportunity to speak today on this important topic. I am first
going to talk a bit about the role of ASTHO in informatics and standards. Then
I will highlight some of the work of our member states.

ASTHO represents governmental public health agencies in the 50 states as
well as the U.S. territories and freely associated nations and the District of
Columbia. Our member is the chief health official in each one of those
jurisdictions. We also represent and support the work of over 100,000 public
health professionals nationwide. I also want to mention that we work very
closely with many partners through JPHIT as well as through the ASTHO
affiliates program, many of whom are represented today.

The current activities of ASTHO in public health standards and informatics
really focus around meaningful use, the transition from ICD-9 to ICD-10 which I
am sure will make a few in this room pleased. Structured data capture through
the SNI framework and participation in those activities as well as some work
around interstate data exchange that has become a very hot topic for our
members in the last year or so as well as data privacy and security which has
also become quite a hot topic in the last year or so.

We are actively engaged in supporting our members through policy analysis,
advocacy and technical assistance. Some of the areas that are working on for
technical assistance really are primarily ICD-10 and meaningful use at this
point to assist with implementation. We also work quite a great deal in the
area of member engagement, specifically through the E-Health Policy Committee
which advices and direct our work in the policy arena for public health
informatics as well as the informatics directors’ peer network which includes
the informatics directors or some facsimile thereof in jurisdictions that do
not have that. These individuals have the role of leading the informatics
activities in state health departments. We use this peer network both to help
guide our work as well as to provide technical assistance in peer mentoring
across jurisdictions.

Moving on to some examples from some of our members that we wanted to
highlight, as most public health data is generated in the healthcare sector as
we have talked about today and current initiatives are pointing more toward
integration of primary care and public health through many activities that
ASTHO and NHL and many of our partner associations as well as a partnership
that is run by ASTHO includes more than 80 organizations. We recognize that it
is critical for public health to work closely with EHR vendors and HIEs and in
standards development to facilitate the exchange of information from clinical
care to public health and ideally, bi-directionally back to clinical care.

In New York State, the New York State Department of Health in collaboration
with the New York E-Health Collaborative has initiated a multistate
interoperability working group, which consists of 19 states representing more
than 52 percent of the U.S. population, 22 EHR vendors and 23 HIE vendors. The
workgroup was launched in February 2011 to leverage existing standards in
developing and develop consistent implementation guides for interoperability
between HIE software platforms and the applications that interface with them.
The participants in the IWG commit to comply with the standards that the IWG
members jointly develop.

Thus far, the IWG has focused on clinical summary document, also known as
consolidative CDA or CCDA, which is the standardized summary of medical and
clinical data about any patient. The IWG has ratified interoperability
standards related to statewide send and receive, patient record exchange and
statewide patient data inquiry service. Based on these standards, the IWG has
designed a certification program for EHR, HIE and HISP vendors. The
certification program is currently in development as being managed by the
certification commission for health information technology or CCHIT.

To ensure maximum participation in these activities, the IWG functions
primarily through conference calls but has funding through cooperative
agreement from the ONC for staff at the New York E-Health collaborative to
provide standards development services. We believe that this approach to
standards development should be explored for public health. Having HIEs and EHR
vendors at the table will ensure public health standards can be realistically
implemented by these critical partners. The eagerness of vendors and HIEs to
collaborate on initiatives around biosense lends credence to the feasibility of
this model.

New York State wishes to include the State Department of Health, which is to
bring in state and local public health agency representatives which are
necessary to play a leadership role in public health standards development on
the IWG. However, funding would be needed to support staff participation and
public health participation. New York State feels that this is important. ASTHO
support feels that this is an important mechanism to continue to more deeply
engage public health in the standards development.

I will try to direct you to the appropriate slide. I believe we are on page
4. I will speak quickly. I know I am running out of time. I will now turn to
data privacy and security. Again, as public health becomes more involved as
exchange partners in HIEs, clinical partners and HIEs need assurance that
public health agencies are following strict security standards.

By way of example, Utah recently experienced a significant breech that
required a thorough review of privacy and security policies and practices. As a
HIPAA hybrid entity, only the covered portion of their data was protected by
HIPAA security standards. However, the Utah Department of Health made a
strategic decision to protect all public health data with HIPAA level privacy
and security standards as the lowest level of protection. They implemented
HIPAA-level security measures and adopted more than 100 new privacy policies.

ASTHO believes that all public health agencies should consider HIPAA
security protections to be the lowest level of protection for all public health
data to protect against data breeches and ensure that exchange partners,
including HIEs that their data will be rigorously protected downstream.
However, the cost of system improvement and staff time to develop these
policies and modify systems will likely be significant. Given the current
budgetary pressures at the state level, funding sources will likely be needed
from the federal level to support policy development and system security
enhancement.

As you can see from these examples, state health departments are aware of
and desire to participate in the development and implementation of standards.
However, additional funding is required to ensure that they will be able to
fully participate in these activities as the new health data infrastructure is
built in the United States. Thank you again for the opportunity to present.

DR. SUAREZ: Thanks so much, Paul. We are going to go to AIRA.

MS. COYLE: Thank you. Good morning. I really appreciate the opportunity to
share our comments with you. I primarily want to focus on what we have done and
the strategies for moving forward within our organization and also some
suggestions. My name is Rebecca Coyle and I am the executive director of AIRA.
AIRA is a membership organization that promotes the development and
implementation of immunization information systems or IISs as an important tool
in preventing and controlling vaccine preventable diseases.

As a leader in immunization informatics, AIRA is committed to sharing
collective knowledge and successes representing IIS programs on emerging public
health issues that may impact IIS operations and identifying and representing
IIS in involving health informatics trends. AIRA is the organization that
represents the immunization information system community. AIRA grew out of the
desire of the IIS community to really collaborate as one entity and to respond
to the various challenges of managing and implementing a public health
information system.

Since AIRA was founded in 1999, we have formed many committees within the
organization to assist with the standards development, education,
interoperability and resource development. Our strength truly is within our
community. We recognize the importance of standards and also the long-term
collaborative efforts that need to keep pace with both the external and the
internal development.

We focused on message standardization, standardization of terminology and
semantics used within the community and have a strong foundation in these
areas. We also understand that standards are a living and changing entity. The
communities experience pressure to change and to develop some new requirements
necessitated by new use cases such as the recent request for the collection of
the vaccines for children or VFC eligibility code at the visit and then also
moving towards the bidirectional that would be query response interoperability
between IIS and electronic health record systems.

As an organization, we are actively involved in other organizations — I
think that is very common to the other organizations at the table today -–
in our standard development and activities. AIRA in coordination with the CDC
originally created the HL7 implementation guide for IIS and continued to
jointly maintain this standard. Code sets imbedded within the HL7
implementation guide have strong acceptance and use as well as inclusion in the
ONC meaningful use final rules, which institutionalized certain code sets for
IIS date of submission.

Functional standards were also developed by the community with the
immunization information systems support branch, CDC, NCIRD through a consensus
process involving input from a variety of managers and technical experts from
across the U.S. During the process of updating the standards it was recognized
that there was a growing importance of the IIS to really broaden the health
information technology landscape and that the revised standards are intended to
lay a framework for the development of IIS through 2017.

Prior to that, we didn’t know if it was a floor or a ceiling. Now we have
some goals to move towards. They supersede the minimum functional standards for
registries which were adopted by the National Vaccine Advisory Committee in
2001. The bidirectional data exchange standards have been developed in response
to an increasing movement towards exchanging data between the EHR and IIS.

AIRA has committees comprised of community members that are devoted to
developing and maintaining standards. Again, this is a volunteer community. It
is not a paid community. AIRA has a modeling of immunization registry
operations workgroup that really looks at and reviews the trends that are going
on in the community and then helps develop the best practices for these issues.
These best practice guides have become the true standard for IIS.

Seven best practice guides for functional and technical IIS-related topics
have been developed. Each of these guides includes an overview of the topic,
some principles, the business rules and general recommendations associated with
the topic. It also includes recommendations and challenges and solutions for
implementing some of these practices. Each of these guides is available on our
website and is free. They can be accessed both at the true big guide and also
in a miniature guide form.

In 2011, CDC began the task of developing the clinical decision support and
logic specification for the immunization schedule. This was finalized in 2012.
I want to stress that this is really a huge undertaking and a major effort
towards standardizing the forecast that providers use to evaluate the needs of
their patients. However, we still have a lot of work to do related to the
implementations, but at least the foundation exists. There have been many
standards that have been developed within our community. However, we have also
been slower at developing or even agreeing on the need to create a
certification system for IIS. We are currently in the process of evaluating
what that looks like for our community.

One of the major needs that we have identified is the need for increased
communication with the EHR vendors. These are the contacts that are not
necessarily in prominent positions within the EHR community or the vendor
organization and thus are not necessarily connected to some of the other
national efforts.

Ten years ago, there were a few IISs that were exchanging data with EHRs
because it made sense. I think that is truly what we have to go back. It makes
sense to exchange the data. Data exchange can reduce the errors, increase
productivity and create a consolidated immunization record to prevent missed
opportunities and over-immunization. Meaningful use has absolutely been a
driver to help guide or move some of those providers that were reluctant 10
years ago to really exchange data with the IIS. It has also created some pain
points for us as well.

Some of the incentives for adopting and use public health data standards to
include establishing more accurate and complete immunization records for
individuals on populations. Incentives for the public health data standards as
drivers for interoperability between EHRs and IIS also include improved
workflow for clinic staff making it easier for medical providers to assist the
public in creating more complete data sets and improving public health. Some of
the barriers and challenges to adopting data standards include legislative or
policy restrictions. As ASTHO just mention and just being able to exchange that
data across jurisdictions has been a tremendous issue.

Additionally, funding is disproportionate across the states and it is based
upon the priorities of the state. There is not currently a line item for
funding for IIS. Short-term funding for specific projects has surfaced in
recent years, but truly IISs are unable to channel such funds for maintenance
or long-term stabilization efforts.

We also have seen an increasing trend to centralize IT services which moves
some of the technical staff further away from the program staff or even out of
the public health agency, which creates a general lack of understanding of the
public health need. Additionally, the public health cannot often compete with
the private workforce, both at salaries and also with the staff turnover it
creates a less experienced workforce. There is a lack of informatics trained or
even informatics aware workforce. Hiring freezes have also contributed to some
of these issues.

As far as the current status, I think I have mentioned primarily where we
are at and some of the drivers and the incentives. One of the issues that we
have seen is a need for standardization of the EHR functionality when
connecting to an IIS which does not currently exist. IIS standards are
established in the HL7 implementation guide for immunization which drives what
information is shared and how it is shared from an IIS to an EHR.

However, there is currently no standard for what EHRs are expected to do
with immunization histories and forecasts that are received from the IIS. In
this vacuum some of the EHR systems are creating functionality to allow medical
providers to update the EHR with shots that were sent in a query response by
the IIS and were not already in the EHR.

Some EHR systems are also creating a functionality that allows a medical
provider to quickly see if the EHR has record of shots given that are not in
the IIS and then report them to the IIS in seconds. Others are simply
displaying the information received from the IIS and then others are not
storing it at all. IISs can, making suggestions for EHR implementation but they
cannot require or expect specific implementations due to the lack of standards
or mandates in this area.

It is crucial that standards and best practices be developed as divergent
approaches in logic intensive development needs. It means that no two EHR
systems are displaying immunization histories received from IIS in the same
way. This limits the opportunity for providers to have access to a patient’s
complete history when needed and undermines the public health potential of EHR
and IIS interoperability.

Additionally, some of the implementation challenges from a public health
agency perspective are varied depending on the maturity of the IIS. Challenges
include state laws or policies that prevent the exchange of IIS data across
state lines or the re-disclosure of data that prevented IIS from storing
immunization information obtained by IIS. For jurisdictions that are supported
by an in-house IT department, the ability to implement or enhance systems is
dependent upon the priorities of the department and the IIS often is prevented
from responding as quickly as the changing environment calls for.

Health information exchanges or HIEs have increased the presence in the
public health world. In some states this has created additional hurdles for the
IIS and redirects the resources to create new, but often duplicate connections
within EHR and IIS.

In some states there is a mandate that all IIS data must be exchanged by the
HIE. This creates a vulnerability for the IIS should the HIE fail, or if it
doesn’t support the technologies that are currently being utilized. Often
decision makers are not familiar with the capabilities of the IIS and assume
that an HIE can replicate an IIS. Vaccine ordering accountability and vaccine
forecasting are very specialized and intricate systems that HIEs are often ill
prepared to replicate.

I know my time is limited. What I think I will do is I will kind of
summarize for you the following areas that are in most need of additional work
or development to support the advancement of IIS standards. First is the clear
standard around how EHRs will interoperate with the IIS regarding data
elements. That includes vaccine eligibility codes, active and inactive status
codes, et cetera, standards for the receipt and consumption of IIS data into
EHRs and receipt and processing of error messages and responses messages. We
would also like to see a meeting of EHR vendors, EHR users, IIS representatives
and national policy makers to develop a common vision for how bidirectional
interfaces should be used.

Interstate data exchange. This one is probably one of the major topics. I
think it would be a good topic for a second workshop hosted by this
subcommittee. Our federated system sets a floor for disclosure for covered
entities and also public health exceptions. Much of the confidentiality
protections are established at the state and through laws and policies.
However, variations in those laws and policies may be real or perceived
barriers to interstate data exchange especially of identified data that can be
re-disclosed by IIS. Thank you for the opportunity to present here today.

DR. SUAREZ: Thank you very much. I think we have NAPHSIS next.

MR. NIELSEN: Good morning. It is a pleasure to be here. My name is Larry
Nielsen, and I am here representing NAPHSIS, which is the National Association
for Public Health Statistics and Information Systems. We are an association of
the 57 vital records jurisdictions across the United States. These
jurisdictions are made up of the 50 states, fiver territories, New York City
and Washington D.C. We know that vital records are like the premier public
health dataset. It comes about a close as you can get to being
population-based. It has all kinds of rich information that a lot of people are
interested in. As an example some of the systems that we feed are immunization
registries, cancer registries, newborn screenings, et cetera, et cetera, et
cetera.

To give you a quick overview of what vital records data flow is all about,
the green arrow shows the flow of birth data information that comes from
hospitals and birthing centers. There are a few births that out of institution
areas as well, but primarily from hospitals. Death information is represented
by the yellow arrows. The demographic information comes from funeral homes and
the cause of death and medical information comes from a medical certifier or
from the medical examiner.

Vital records have this huge public facing component, which is somewhat
unique among public health agencies. That is that we issue to public birth
certificates and death certificates. These are among the most important
documents that are out there. Then the blue arrows represent the data
interchanges that we have. We do a lot of data interchange with CDC,
specifically NCHS, with SSA, with various public health agencies both at the
state, local and federal levels. We also have a data interchange with other
states that I will talk about just briefly in a moment, or an
inter-jurisdictional exchange.

Most of our jurisdictions, not all, but most have incorporated electronic
birth registration systems and electronic death registration systems. These are
primarily web-based, and they collect data directly from the users of the
providers of the vital records information, hospitals in the case of birth
data, funeral homes and medical certifiers in the case of death data. This has
been a huge project getting these things into place, but it has made a huge
improvement in most jurisdictions in the timeliness and accuracy of vital
records data.

The electronic transmission of our information to NCHS and between states
uses a standard data format and standard transmission protocol that was
established by NCHS and NAPHSIS members over a period of time. The standards
for vital records start with the certificates themselves. For example, this is
just part of the U.S. standard certificate of live birth which shows some of
the legal information and a little bit of the statistical demographic
information. Standards for vital records have been evolving over many decades.

The most recent standard is the standard that I show before you which is the
2003 revision in the U.S. standard certificates of birth and death and the
report of fetal death. Most of our jurisdictions are using the 2003 standard
and those that aren’t will be on the 2003 standard within the next year or two.
We are very hopeful. It is a big re-engineering process for our jurisdictions.

This is a little fuzzy, but in addition to the demographic and
administrative information that we collect with our standard certificates we
also collect quite a bit of medical information on the birth certificate. As on
the death certificate, cause of death, manner of death and a fair amount of
medical information as it relates to contributing to the death. For those of
you that are interested all of these standards are available on the NCH
website.

I will just talk a little bit about the internal standards. Vital records
standards have been developed for use to transmit data between the
jurisdictions and between the jurisdictions and NCHS. That is also using
standard data formats. We use standard data transmission protocol. It is a
secure protocol. Security is a big deal to vital records, first of all to
protect the privacy of the people that are named on the certificates and also
especially birth certificates are documents for all kinds of other documents
such as driver’s licenses and passports.

It is very important that we keep this stuff private and secure. That is a
little bit about the internal within the vital records world transmission of
data and standards related to that. We are also interested very much in moving
forward and getting information related to electronic health records to help
make our data even more timely and more complete and accurate as we implement
these standards.

So NCHS along with input and feedback from a lot of our very dedicated
members have done a tremendous amount of work in creating a set of standards
for the interchange of information between electronic health records and vital
records. This is a new area to vital records professionals because a lot of
vital records professionals aren’t really accustomed to talking about clinical
codes and stuff like that.

We did develop ICD-10 over a decade ago for causes of death, but currently
we are also involved in how do you go about translating SNOMED codes to the
codes that are on the standard birth certificate, et cetera.

So we are very much interested in getting electronic health records involved
in this process.

We are currently doing two pilots. These pilots have been demonstrated at
this year’s Connect-a-thon at the HIMSS conference and their interoperability
demonstration and in our annual joint meeting between NAPHSIS and NCHS. There
was a lot of interest among our people and among our membership in these
pilots.

The first pilot is happening in Minnesota. It uses forms, pre-populating
some data from the EHR and presenting the form to enter data that is not part
of the standard electronic health record. For these next two slides I want to
thank Michelle Williamson. This information is presented to a form receiver at
the Minnesota Department of Health where it is used to populate the record in
the EBRS. In the Utah Department of Health partner in the Intermountain
Healthcare gives the EHR user access to a standard medical certification screen
to enter and certify cause and manner of death and other medical information
used on the standard death certification.

So there is a lot of interest. There are some challenges and a lot of these
are going to sound very familiar from a lot of the other presenters that have
gone before me. First of all our members have done a huge amount of investment
in adopting the 2003 U.S. standard certificates in their current EBRS and EDRS
systems. That has taken a lot of resources, a lot of expense, and a lot of
time.

Frankly, there is not a huge amount of incentive to move forward into the
EHR world. Vital records are not part of meaningful use. There is no strong
push from either the hospital side for birth or from physician side for death.
There is not a lot of push from the state side, from the public health side to
get this stuff going. Of course, there is a huge lack of resources.

Most jurisdictions received a lot of help, financial and other types of
support from NAPHSIS and NCHS in getting their EBRS and EDRS systems up to
place. That kind of help just isn’t really on the horizon except for the pilot
stage which is being funded by NCHS for going forward with EHR information. In
a lot of jurisdictions there is also a tremendous lack of IT support. Some of
them have to compete for IT resources with other state agencies or just within
their own departments.

Lastly, some people are a little bit concerned because there are specialized
requirements for this civil registry. As I mentioned before, birth certificates
especially are breeder documents for other very important things. It is very
important when names and places and dates come down from an EHR into a vital
records system that they be absolutely correct. Those are legal records. They
are not just clinical records now. Jurisdictions also have laws in place and
regulations that stipulate what it takes to make a change in a vital record. It
is not just something that comes down from an EHR that says this was incorrect,
make this change.

Those are some of the challenges that we have to meet in terms of furthering
our standards and efforts in vital records. Thank you very much for the
opportunity to testify today.

DR. SUAREZ: Thank you very much. I think we have Janet, CSTE.

MS. HAMILTON: Thank you. My comments will be provided through the committee
following the presentation. Good morning. I am Janet Hamilton and I am
representing the Council of State and Territorial Epidemiologists. I am looking
forward to this discussion and thank you very much for taking comments on this
important subject and for inviting CSTE.

CSTE works to advance public health policy and epidemiologic capacity. We
are specifically a member-driven organization. In my day job while I do
represent the CSTE executive board here, I work for the Florida Department of
Health managing surveillance systems. CSTE is an organization of member states
and territories representing applied public health epidemiologists at the
state, local, tribal and territorial level. We are charged with the
responsibility for defining and recommending which diseases and conditions are
recordable and nationally notifiable and which of these diseases and conditions
will be voluntarily reported to CDC.

Over the years, CSTE has grown to provide information, education and
developmental support of practicing epidemiologists in a wide range of areas as
well as expertise for program and surveillance efforts. I was asked to comment
on the areas of bio-surveillance. I think you will hear a lot of comments from
very specific groups. My comments are trying to be broad and over-arching.

Public health officials need timely access to valid and reliable data
necessary to monitor the health status and health-related events and factors.
We rely on data to develop and implement our programs that protect health and
ensure safety. Like counterparts in other domains, public health professionals
increasingly turn to advances in IT to support there need to capture, exchange,
manage and analyze data. These data include diagnoses and treatment information
for infectious and environmental chronic diseases and conditions, patterns in
patient complaints or symptoms that may be suggestive of an outbreak or other
health event, vaccination records, birth and death records, resulting screening
tests from newborns and other priority areas.

Public health departments have adopted a variety of information systems to
help capture data from healthcare providers and other sources. This theme is
common, but regardless of the funding source there are serious challenges with
maintaining steady funding to state and local health departments especially to
support surveillance efforts. Public health currently accounts for less than
0.1 percent of the GDP and less than 3 percent of federal health-related
expenditures. This information comes from the 2009 report, Expenditures for
Public Health Assessing Historical Perspectives and Trends which was published
in the American Journal of Public Health and the 2010 IOM report for the
public’s health, the Role of Measurement and Action and Accountability.

Public Health has different data needs from clinical care for different
surveillance activities. We have heard about vitals such as death certificates
or the granddaddy of all surveillance, registries, cancer, birth defects,
population surveys such as the behavioral risk factors surveillance system,
reportable disease and condition surveillance, outbreak management activities,
periodic active surveillance activities, where there is intense effort in a
short amount of time often done in the context of outbreaks and also supporting
surveillance efforts in certain emergency situations such as post-hurricane,
large national events, such as the Republican or Democratic National
Conventions, ice storms, the Gulf oil spill or post-Haiti earthquake.

Some of these data flows are more predictable than others. The traditional
core of public health surveillance is really to learn about every person with a
reportable disease to identify promptly all cases of diseases or conditions
that require public health intervention to detect outbreaks, changing trends or
patterns in disease occurrence, plan and assess or evaluate control and
prevention interventions. A lot of how we have accomplished our work is to
actually make something reportable through state laws, infectious diseases,
birth defects, cancer, occupational diseases, environmental diseases, child
abuse, et cetera. Some of these areas have more fully developed standards.

In the area of infectious reportable disease surveillance, there has been
much successful project and standards development for example around electronic
laboratory reporting. In order to more effectively support case reporting, we
still need effective triggers and standards around them so that the report will
actually reach public health. There has been some activity in this area. For
example, CSTE is working in conjunction with CDC on development of the
reportable conditions knowledge management system, which would expand on the
existing reportable conditions mapping tables.

There have been two states, Wisconsin and New York that have actually done
some very initial testing of receiving a case report for an infectious disease.
While they have done a lot of work in this area, they have mentioned since
individual case reporting is not part of stage III, there has been a lack of
interest on the part of providers to actually participate as they are so busy
with other meaningful use activities.

We need effective affordable disease case reporting. How does public health
get some of their needed information? The initial report alone, either from a
clinician or a laboratory often does not have all of the information that
public health needs for completing follow-up and documentation of individuals
with reportable diseases. Public health needs, person, place, time, the who,
the when, the where, clinical information to confirm the diagnosis, treatment
or medications given to the patient where the exposure or event occurred, for
example travel history, history of pre-travel immunizations and during travel
prophylaxis or history of arthropod exposures.

The denominator present, for example, the total exposed or the total
impacted. The environmental setting is the workplace, lifetime occupational
exposure histories, how the patient may have become ill, insect bites, foods
consumed and other travel locations and information about future prevention
actions that may have taken place or are needed, for example, exposed family
members that need treatment or vaccine. Many of these items are not in the
healthcare record or in today’s world are only inconsistently recorded.

Not all data needed by public health is always reported in an EHR. Not all
data in an EHR is reported to public health. Standards are needed to further
this and to support bidirectional communication. Some of the pieces of
information that I mentioned would be useful for public health query after an
initial case report and standards would be helpful for furthering that process.
In conjunction with this, standards must be developed to support two
conflicting processes, the timeliness of receiving data quickly and rapidly and
completeness.

We are often lacking in public health drivers and incentives to support the
development of standards for public health surveillance activities. We need
better incentives to further support the standards development. Public health
tends to be an afterthought or a downstream data recipient. Decisions are made
for billing purposes or care of the individual without actually looking at how
that record will be systematically received and processed and not incorporating
the often broader public health and public health functions and perspectives.

I want to focus in the couple of minutes left on outbreaks and how things
change during outbreaks. What does public health need for an outbreak
investigation and management? We have a couple of examples. For example during
H1-N1, we have very complex contact tracing in the home, work, healthcare
contacts, information that we needed for case isolation and quarantine of
contacts, laboratory diagnosis was challenge, especially early on when CDC was
the only location where testing could occur and public health was not prepared
to be the only clinical diagnosis location.

Public health had limited capacity and capability to quickly return the
results back to providers. Essentially our public health surveillance needs are
to establish data flow connections, to leverage electronic feeds prior to
events, strike the balance between the speed of information versus completeness
of data. Our public health goal is to reduce the time accessing and gathering
information. The patients are contacted sooner. The source of illness is
identified more quickly leading to improved disease prevention.

In sum, we need effective processes for defining standards, refining
standards, accessing them and establishing an enterprise continuous process.
More common national standards once developed need to be applied with equal
vigor by all parties who need to use them. Building standards that aren’t
adopted or used is frustrating and we need good measures for adoption and
incentives for adoption.

Public health roles are changing and providers do not understand these
evolving roles where public health is spending less time actually digitizing
the information and more time analyzing it. We need to identify and expand on
demonstrated opportunities of retrieving and returning information from public
health to clinical care good messaging standards. Interoperability and exchange
are important, but at this point in time, messaging is probably the single most
important piece to public health. We need a flexible means to access data
during outbreaks which doesn’t necessarily mean sending data. While have
assumed that health information exchanges have built query repositories of
extracts of health records that is not always the case.

As we work towards developing recommendations for data sharing processes we
need to include the role of content and data sharing agreements, include the
ability to rapidly institute standards-based data collection systems and
response to national outbreaks, more concisely we need to consider maintaining
security and privacy of individual records when they are appropriately shared
within and between agencies.

Finally, if we only remember one thing from presentation today it would be
the need for ongoing funding support, sustained funding to support continuous
quality improvement, a manifesto or a call to action to secure funding to
support public health information systems, resource capacity training and
standards development and adoption with broad engagement and a broad engagement
strategy which supports collaborative development that includes state and local
public health.

DR. SUAREZ: Thank you very much, Janet. We are going to take a few minutes.
We are going to move a little bit ahead of schedule. I know we are almost at
12:05 now, and we have a lunch break, but we want to have a chance to have some
questions and some discussion with this panel, so we are going to spend about
20 minutes and then move the schedule. I see some people already jumping up the
questions, so we will start with Leslie and then go to Raj.

DR. FRANCIS: Several of you have mentioned the importance of privacy and the
absence of HIPAA standards applying to public health and the variance in what
different states do. I am asking this in my capacity as the co-chair of
privacy. Do you have any specific suggestions that you think would be helpful
for this committee to pursue on the privacy front?

MR. NIELSEN: On the part of vital records, there is a fair amount of
variance among the states. The majority of states now are what we call closed
records states, and in this states, there are only a very select set of
individuals who are entitled to get a birth record or a death record. There are
a few states remaining that are open record states. Some of them are very open
where anybody can come in and request and get a certified copy of a birth
certificate for example, which opens changes or fraud.

Some state attorney generals have said that HIPAA applies to vital records
because there is health data in there. The closed records states, however, have
laws in place that protect vital records in really a more stringent manner than
what HIPAA would do. From the NAPHSIS side and from the NCHS side there is a
lot of encouragement in our jurisdictions who are open record states to adopt
two model laws, one from 1992 and one from 2011 that closed those records and
made them private.

MS. SOPER: I believe that there is a tremendous amount of variation not just
across jurisdictions but also within jurisdictions depending upon whether the
data under consideration falls under HIPAA or does not and falls under the
public health exception. At ASTHO through discussions with our member over the
year or so are recognizing that similar to HIPAA states need to have policies
in place or a law in place that requires a certain base level of protection of
all public health data.

In large part not because public health thus far has not taken great care to
protect the privacy and security of data, but to create a baseline for all of
public health especially as we are beginning to in greater numbers partner with
and exchange data through HIEs and potentially with ACOs and certainly with
clinical care. I think guidelines moving in that direction and recommendations
to that effect would be beneficial and certainly funding recommendations. We
always need funding to implement new privacy and security.

MS. HAMILTON: I might just expand on that. I think even within states while
privacy and security and handling of that data is paramount, you will find that
different practices are even in place for different diseases and that makes it
challenging when we are working even within our own agency or across agencies.
Having clear guidelines for how this should be handled and not looking for the
exceptions but to try and have it be as systematic across them as possible
would be very helpful.

DR. SUAREZ: Thank you. Raj.

DR. CHANDERRAJ: My question is to Mr. Kennemore. When you are transmitting
the lab information to the public health authorities are you taking informed
consents for the patients? If so are you stated this is independent of the
physicians consent to draw the lab. There is a liability issue for example in
an HIV patient. If you transmit that information to the public health record
who is liable in that case? The patient may claim that he told the doctor don’t
release this information and then you released that information to public
health authorities.

DR. KENNEMORE: That is actually a very good question and it is crosses the
line in many cases between what is and is not HIPAA protected data and what
does and does not fall under the public health exclusion. We at APHL obviously
don’t operate those laboratories, but what we do find is in the public health
lab, each of those jurisdictions basically has their own form of consent that
that they have to obtain in order to draw the lab and do the test. Once the
result is obtained, in most cases, if it is a reportable condition it has to be
report irrespective of consent of the patient. The consent must be obtained
obviously before performing the invasive procedure of drawing that original
lab.

Now, we get into a little bit of a further complication as we now look at
the data networks that now transmit that data. We at APHL have actually just
completed a fairly extensive legal investigation into when that data would or
would not fall under the public health exclusion. The response to that were
about 12 pages of fairly extensive legalese back from the lawyer that did the
research for us.

There are a few fundamental principles that apply. It basically looks at who
are you acting on behalf of? So if you are acting as an agent of the public
health entity, you have certain protections and certain exclusions that you are
allowed to utilize. If you are doing some form of transformation on that data,
for example, a meaningful use compliance activity, and you are acting on behalf
of the provider or the eligible hospital for instance, at that point you fall
under the HIPAA regulations and other similar rules. That was a fairly wordy
answer. Did I address your question?

DR. CHANDERRAJ: Fundamental question is who is liable?

DR. KENNEMORE: Well, there are a number of players along the line who are
liable. In the case of the initial performance of the test, my understanding
-– keep in mind I am not a lawyer -– but my understanding of that is
that the laboratory who draws and performs the test is liable for not only
obtaining the consent, but also for how they handle that data once the results
are obtained. Then it becomes a matter of how that data is handled from that
point on. When that data is now transmitted outside of that laboratory
environment it depends on who that transmitter is. In other words if you are
using a third party transmitter that party may or may not be liable depending
upon who they are considered to be operating of an agent of.

Again, wordy answer, but if you are operating as an agent of public health
you have some exemptions from liability. If you are operating as an agent of
thee provider, for instance an eligible hospital would draw a test. If you are
transforming that rest result into a meaningful use compliant message and you
are a third party health information service provider and you are acting in
that capacity, then you have a liability as well, not from the informed consent
of drawing and performing the test, but you have a liability in terms of how
you manage their data. I believe this a fairly universal principle. If you have
a reportable result, you must report that result to public health and that
there is no consent required and consent cannot be denied.

DR. CHANDERRAJ: Do you notify the patient that you are releasing this data?

DR. KENNEMORE: Once again there is a very significant discussion on that and
it is jurisdictional specific to be quit frank. The laboratory is required in
all cases to report that data irrespective of any rule which may or may not
require them to notify the patient. In some jurisdictions the laboratories are
actually not permitted to communicate that lab result directly to the patient.
It must go through a licensed physician or medical provider. In other
jurisdictions, that result may be provide to the patient by the laboratory and
in other jurisdictions that information is mandatorily provided to the patient
by the laboratory. It really is jurisdictional specific.

There were a number of years ago a national proposed rule that would require
laboratories. There were two variants of that as I understand it. One of them
would allow laboratories to communicate that information to the patient without
the provider being involved. Another recommendation was that the laboratory
would be required to provide that information to the patient. Those proposed
rules have not been acted on in about three years if my recollection is
correct. Please do not hold me to that data, but it has been sometime since
there has been any action on the proposed rule. I am unaware of any deadline
upcoming to make that a final rule.

DR. SUAREZ: So when you say there are proposed rules that have been
published as a proposed rule by the federal agency, these are clear
modifications or are they something different?

DR. KENNEMORE: Walter, I have read the entire publication, but it has been
sometime since I have read it, so the specifics escape me, so I am sorry.

DR. SUAREZ: From your perspective, is that an area where it would be
beneficial to have some resolution with respect to the ability for labs to
disclose results directly to patients?

DR. KENNEMORE: It affects our meaningful use implementation because one of
the big drives under stage II is patient engagement. As of now in many
jurisdictions the patient has no ability to get access to that laboratory data.
There is the argument that many MDs make that the patient may not be able to
interpret that data and therefore they should have access to it without some
sort of counsel involved. I can see arguments.

I am a MD myself. I can see arguments on both sides of that coin. One of the
interesting point is I am a MD and in my jurisdiction, I can’t get access to my
laboratory data unless the physician that ordered it gives it to me. There are
some interesting points to go along there.

MS. SOPER: The question you are asking has much broader public health
implications for both providers and for public health. It is not just lab
reporting that would fall under that. Certainly any disease or condition that
is legally required to be reported to public health would fall under the public
health exception whether it is reported from lab or from a provider or from a
business associate from one of those.

Public health does not have the same downstream liabilities and the same
upstream liabilities to the reporter. It would be public health that would have
the liability if we broke our own rules and a breech occurred, but it would not
fall under the HIPAA liabilities and also would not have the same upstream
effect on the provider who reported it or their business associate. The
question that you had about notifying the patient if something was reported to
public health, providers who are covered entities or their business associates
are required to maintain the fact that the disclosure was made to public
health. They are not required to notify the individual that a disclosure was
made. If a patient came back if a disclosure was made they need to be able to
answer that question yes or no based on their records.

DR. CHANDERRAJ: The lab doesn’t communicate to the provider that the
disclosure was made. How would the provider know that the disclosure was made
by what entity?

MS. SOPER: I would defer to my APHL friends on that given their vast legal
review of the issue.

DR. KENNEMORE: The answer is that is a very good question which is not
capable of being answered clearly at this time. You actually landed exactly on
a gap because the laboratory is actually not required to provide the physician
with the notification that they have made that notification in any case. It
depends upon a lot of the local jurisdictional rules.

MS. HAMILTON: I just want to expand on that. Most states and territories and
locals have reporting regulations that govern the lab and the physician. The
physician is required to report and the lab is required to report. They don’t
need to have knowledge that one reported in order for them to act upon their
required duty under the state regulation.

DR. SUAREZ: That is a very interesting about this issue. It certainly is an
important issue to highlight.

MR. SOONTHORNSIMA: This is a much higher level of question. Based on my
observation this morning, there are some common themes. One thing that really
stood out was the need for additional resources, financially, human resources
capacity and even brain power.

One question, as I listen to different domains and experts, how do you think
at a national level one can try to ally these priorities across various domains
at the state level, at the informatics lab, vital records, bio-surveillance as
all of the analysts provide their insight. There seems to be lots of different
challenges and issues. When it comes to resources, if we have limited
resources, how might we apply the resources at the national or even state level
given a common set of priorities? Can we get there?

MR. NIELSEN: The experience of vital records has been that a lot of the
standards have been arrived at through common consent, but a lot have been
arrived at because the jurisdictions have entered into contracts, specifically
with NCHS and also for some data with the Social Security Administration. Those
contracts provide funding to the jurisdictions to satisfy the standards that
are promulgated under those contracts and the reporting requirements for those
contracts.

In this case it is the federal government that has provided to the
jurisdictions a carrot and then behind that a little bit of a stick to meet the
standards and to satisfy the reporting requirements behind them. In the
experience of vital records a lot of the standard reporting has been
established by contracts with agencies at the federal government.

DR. SUAREZ: Paula.

MS. SOPER: I think at the national level versus the state level it is
probably a bit of a different animal. At the national level certainly
organizations like the Joint Public Health Informatics Taskforce where most of
the organizations that are presenting today are very active members. We have
already had a history of identifying priorities and advancing those priorities
forward.

So national collaboratives such as JPHIT, I think are a very powerful way to
move standards forward and apply resources at a national level, both at the
national level and at the state and I dare say even at the local level. Some of
the larger jurisdictions we need to be sure that we are part of the broader
prioritization of both resources as well as priorities and that public health
needs to have a leadership role in groups. We certainly have leadership from
Dr. Art Davidson through the policy committee to ONC. I think greater level of
participation of leadership at the national level from state and local public
health jurisdictions.

Then activities such as the Interoperability Working Group that New York
State has spear-headed, are important opportunities to be able to bring states
together who have a desire to work collaboratively with HIEs and EHR vendors. I
dare say we would need to bring in public health vendors as well. That would be
a powerful way to apply resources to get the work done with the partners that
need to be at the table.

MS. HAMILTON: I would like to comment on that also. I think that is a great
question, and if there were an easy answer, we probably would have already gone
in that direction. One of the challenges that we face is that there is a lot of
categorical funding. Many times that categorical funding is in support of very
specific kinds of activities. With individual states as you have heard this
morning, different things are prioritized in different ways. You have seen
advancement in almost a mosaic. One state is maybe very advanced in their
immunization registry. Another may be very advanced in another area.

I think trying to have influence into those categorical funding dollars to
support standards development and support groups that are collaborative in the
way that say JPHIT has been collaborative across organizations. There has been
discussion this morning of almost a public health core. I think moving in those
directions and trying to leverage that kind of activity would be helpful. That
goes whether we have the same amount of money or not, I think that would still
be very helpful.

MR. GAMACHE: So most of my comments have been spoken already. I think we
talked about this as a framework of how to set the priorities for public health
and what the standards JPHIT along with some of the partner members and other
government organizations is a great place for some of that foundation work to
be done. I think that is a great place to look for this in part.

The other avenue and I think Janet mentioned it briefly as well, is
sometimes we have collaborations that happen among leading states that develop
the initial set of standards. I think it is critically important for us moving
forward. When we get collaboratives of three or four states and they agree on a
set of standards that sets a precedent for us to be able to move forward with.
I don’t want to lose that initiative at the same time. I think that is one of
the places where JPHIT would look and say here is an initiative states that
stared and somehow we need to collect that as well. That is one of the other
avenues I think in the standardization process that we don’t really have an
avenue to collection information or meeting that looks at these standards from
the field per se. We have the ability to comment on what is out there or what
is being proposed.

A lot of organizations such Standards Works let their members of some type
initiate when they need a new standard and set it in a queue of some sort. We
don’t have a process for that. I would like to see it established also. I think
some of these state collaboratives do that as well, but they do it internally
within their group and move forward. I think that is a great ability for public
health to move forward. A bunch of states have specialization in certain areas
for great reason. They have diseases that they are dealing with that other
states aren’t. Or they have a specific condition that they have to deal with.
To allow them to move forward and develop those I think adds a lot to our
field.

DR. SUAREZ: Thank you so much. Thank you again for this great panel. It is
12:30, so we are going to give you all 50 minutes for lunch. We will be back at
1:20 and we will begin to try to catch up some of the time. At 1:20, we will
restart.

(Luncheon recess was taken at 12:30 p.m.)

A F T E R N O O N S E S S I O N

Agenda Item: Panel 3: PH Data Standards
Developers

DR. SUAREZ: We are going to invite our first panel of the afternoon, which
is the panel on Public Health Data Standards Developers. I think we’re going to
follow the same order that is listed in the agenda, unless there are any issues
among the presenters. But if not we can just follow that order.

Since we don’t have a break between panel three and four, if we can have the
members of panel four come to the table as well that will help speed up the
process here. We still have the phone live? Anybody on the phone? We have the
two presenters on the call this morning. So let’s start with the HL7 Public
Health and Emergency Response Work Group.

MR. POOL: Thank you. I would like to make on behalf of myself and my
colleague John Roberts, who is also co-chair of Public Health Emergency
Response for HL7, on behalf of HL7, I would like to express our appreciation,
to say thank you.

HL7 is committed to an open, collegial, and constructive vehicle to advance
interoperability in healthcare, and particularly has long recognized the
special needs of Public Health, and the Public Health Emergency Response
Workgroup is the home to serve those needs at HL7. Within HL7 PHER has been one
of the most prolific workgroups within the last few years, and also frequently
acts as a voice for Public Health domain perspective on standards that are
being developed, even by many other workgroups within HL7, either those that
are impacted or are impacted by Public Health.

You’ve heard already that there are certain areas of strengths we’ve had
historically related to Public Health standards, particularly those of vital
records, immunization, lab reporting, case reporting – although I know
that’s somewhat depending upon the view of the beholder – and newborn
screening.

I’d like to point out that this fairly impressive record of a number of
health related standards at HL7 has been achieved however with a very small
number of contributors. There is a young lady you may have heard of, her name
is Michelle Williamson. Michelle is the role model for my perspective for
Public Health contributors to standards development. She as not only shepherded
several individual standards, she also has done a spectacular job of engaging
and representing a broad consortium of Public Health stakeholders who are not
at the table personably, but no matter how much trouble I give her has
staunchly defended her stakeholders and their perspectives. And perhaps, I wish
she wasn’t here, would like to say that she also has been extraordinarily
strategic in the selection and coordination of the individual standards she has
developed to give her constituency a tremendous amount of power.

Our challenge is we only have a fraction of important data exchanges between
providers and Public Health in place today. We’ve only done a small sliver of
what we need to do. To do more we need two things. We need more hands. It takes
time sitting at a keyboard typing, sitting in conferences hashing out the
details, it takes physical bodies. Writing standards is not something you do
during your lunch break two times a week and have a successful outcome. We also
need more minds. We need more views, more perspectives, more experience, more
needs, and more challenges. We need people who bring pieces to the puzzle.

I’d also like to raise a topic about threats to standards. First off,
standards will be deficient if they are developed without the insights and
input of those who will ultimately use them. If the people who are going to be
using the standards do not participate in the development of those standards
they’re going to be wrong somehow. They may be good approximations, but they
won’t be right.

And the problem is deficient standards aren’t just a waste of time, they’re
a liability. Because once you have a standard, even if it’s a deficient one,
they can be disruptive, and they can prevent you from getting the initiative to
go ahead and get a proper standard. Well, you’ve already got a standard. If you
aren’t going to adopt it why should we go and prove the standard for you?
Deficient standards are not just a waste, they can be a liability.

I wouldn’t be from a standards development organization if I didn’t put out
the phrase Standards with a capital S. I think I’ve heard already this morning
some people echoing things that went close to my HL7 heart, and that is
standards are not just documents and specifications, they are the product of
open, transparent, consensus based pieces. They are living and breathing an
evolving. They need longitudinal publication and curation and care. They need
an SDO. You guys know the reasons why we want standards – it is god,
mother, and apple pie – it’s cost effective, quality, quantity, and timeliness.

For Public Health the big issue I think you’ve heard already is awareness.
We’ve heard issues of needing knowledgeable staff. We’ve heard the idea about a
cultural shift from controlled silos to shared. And also I would point out that
if we imitate changes in adoption of standards in Public Health, if you start
at the program level and try to develop standards from the program level up the
existing silos make that impractical. For providers, they’re waiting on Public
Health. They can’t do it first. And there is some cost, but it’s mostly who am
I going to talk to.

As an SDO there were several things we don’t think we’re prepared to
address: the state of information exchanges, implementation challenges, and
security standards. I would point out that we do believe in privacy and
security. We’ve specifically and intentionally developed what we call
separation of concerns, that is having the content standards, the security
standards, and the privacy standards be separate standards that can be brought
together and used together in modular fashions without limiting. So you can
have the content moved in a secure way with different content. We tried to make
them modular, create some flexibility.

I’m going to touch quickly on a couple of more points. Vocabulary standards,
you’ve heard that a little bit this morning. I cannot emphasize it enough.
There are few areas where the Public Health silos are more problematic for us.
We need contribution, we need to develop them, we need to adopt them, and if
they aren’t right fix them, don’t just avoid them. We need content standards,
particularly CDA. You’ll hear that through the day, we need more CDA adoption.

Privacy standards. The data query, you’ve already heard questions of, on
dynamic, can I ask of somebody data. You’ve heard clinical decision support
protocols, triggers, logic, those pieces. And case level reporting from Public
Health back out, we’ve heard that intimated a couple of times today.

So, in summary, from an HL7 perspective we believe standards are an
investment and their potential is limited only by the level of involvement in
their development. If you want to reap the benefits of an investment in
standards you have to invest in them by actually contributing. Standards that
aren’t used are worse than none at all, and Public Health has to adopt them
first. Public health is a big universe. You have a few stars out here, guys,
but it’s a long way to creating a universe. And lastly, as your parents always
told you: SDOs are your friends. Use them.

DR. SUAREZ: Thank you. Bob?

MR. DAVIS, R: First of all, I would like to go on record of saying how
closely aligned HL7 and X12 are, in terms of what Ken’s comments are. Secondly,
I would like to reiterate about Michelle. If Michelle wasn’t there I actually
would have not failed retirement, but with Michelle here, here I am. X12
appreciates being invited to present today. I would like to start by saying X12
is very proud of the relationship that already exists with its Public Health
partners. When my slides come up, and I’m going to have you just keep with that
slide for now, if it fits use it, I just want you to think about that in the
background as I go through my presentation. And I’m going to kind of keep it
there for a little bit.

A little-appreciated fact: Public Health Discharge Data systems have been in
existence since the late 1970’s. We think of these systems as being new,
they’re not. One thing that has stayed constant in old and new systems is that
these data systems are driven by two questions. I remember when I first became
this health standards guy, my mother never quite knew what I did, because
what’s a data collector? The two questions are what’s wrong with you, and how
much does it cost.

These Public Health Discharge Data systems have for a long time been the
canary in the mineshaft for what the data needs are. They have anticipated
needs before the rest of the industry realizes it. I’m sure people in this room
know, but a lot of the state data systems develop their standards from the
NCVHS core elements that were originally created by these committees. So a few
examples of the canary in the mineshaft for data is the present on admission
indicator, even diagnosis on a bill, because it used to be there was just these
UV billing codes, and all of a sudden diagnosis became part of the scheme.

These systems have long been the basis for quality measurement by states and
the publication of reports intended to improve the quality of care in health
facilities regulated by each state. I think a really important thing as we
start moving forward is to remember that the role of these systems is to create
data-driven health policy. It’s not just to administrate health care. Let me
say that one more time: The role of these systems is to create data-driven
health policy. And I think we need to keep that in mind when we start thinking
of these systems.

My title was if it fits use it, and I want to talk about why it fits with
the X12 standards. Before I go into much detail with this particular point, I
think it’s important to note that the relationship that X12 has with the Public
Health community was driven by the needs of the healthcare community, not by
the standards organization. So the needs came first, and then came the
standards that seem to fit.

Why did the X12 become a good fit for a lot of these discharge systems?
Well, these systems use the same, and I really emphasize this, the same
provider processes. It’s not the data so much, it’s the processes that created
the data that were the same in the X12 world as opposed to what the Public
Health community wanted, at least for their discharge systems, and soon these
All Payer Claims Databases, which I’ll talk about in a second.

These systems use the same payer processes to collect what we’re going to
call Post Adjudicated Claim Data, this APCD data. Using compatible X12
transactions reduces the burden on providers and payers with better quality
data reported, by the fact that we’re not trying to change the process. We’re
trying to leverage those processes to do what the data needs are for the Public
Health requirements.

Question number one – and there are actually two sets of questions –
one is the question for creating these systems, the what’s wrong with you
question. The data elements needed by states to answer this question are
already or easily made part of the X12 standards. When you started looking data
element by data element, when you looked at what was on a UB data content,
demographic data, data about the diagnosis codes.

The Healthcare Service Data Reporting Guide, which I was probably the
instigator of that, is an approved X12 standard created to provide a standard
messaging vehicle for Public Health reporting. It is very closely aligned with
the HIPAA Institutional Claim Guide. X12 provides a cross-industry form for
having meaningful discussion about Meaningful Use. Fostering important
relationships, X12 has a long-standing relationship with some of the data
conduct committees in the Public Health community.

Question two: How much does it cost? The fit with the what’s wrong with you
question was very good. The how much does it cost question, ever since these
systems were created we always wanted to know about cost. Up until the time we
started getting data from payers the data we had was always a surrogate for the
cost. Well, with this relationship between the payers and the Public Health
community for these All Payer Claims Databases, all of a sudden we had a fit,
and as it turns out the fit again was with an X12 standard.

I’m going to quickly move on to addressing some of the questions that the
committee had. The question about current state of Public Health related
standards, we got the good news, which is that we have relationships. With X12
we have relationships with HL7, we have relationships with the data content
committees. That’s the good news. The bad news is, just as Ken said, there is a
small number of individuals who are providing those contacts, and we need to
have that enlarged. We need more minds involved. And I 100 percent agree with
Ken. Our water is being carried by a few individuals, sometimes they’re not
even being paid to do this.

What are some of the incentives and drivers for adopting the data standards?
When we have data standards we all know we have more compatible data. Certainly
when we have 50 states doing these discharge systems we know that we can now
compare data across the state lines. Less of a burden to collect, if we use the
processes that exist.

The barriers: Resources to maintain existing systems. A lot of times we get
resources to create the systems, a lot of times we don’t get resources to
maintain the systems. We all know data systems are just like children, they
take a long time to mature.

Implementation challenges: The short answer is we’re resource starved. We’re
resource starved to develop and maintain suitable standards meeting the Public
Health needs. We’re resource starved to maintain standards that meet the needs
of Public Health systems. We’re resource starved to educate about what even
exists today. In spite of these challenges states still have been creating
these systems. States have still been taking and comparing data across state
lines. States have been creating new systems, these all payer claims databases.
So states have moved ahead along the way.

Privacy and security. Just like Ken said, certainly the X12 organization
understands that privacy and security are both very important concepts. From
our perspective, if you look at the state model, if we’re talking about if you
use it, it fits. States for a long time have had data that is privileged data.
States for a long time have figured out how not to have abuses of that data.

So clearly there are a lot of lessons to be learned by states. We’ve heard
earlier that we have states doing things in different ways, which is kind of
the nature of the states. So yes, we need to do some standardization there. I
will maintain that the bigger privacy question for Public Health is more of a
question not about security, not about privacy, but the idea that we need to
share the data, and share it appropriately. We’ll put that one down on the
table.

Interoperability, and I’m going to finish on one of my high notes here, we
need to work harder developing definitions across the spectrum of Public
Health. I believe that this is the low hanging fruit, in terms of when we start
wanting to coordinate standards across the spectrum of Public Health. We need
to have definitions that are constant across the spectrum so that in a clinical
world if you have a definition for provider that definition is consistent with
the same provider that we’re using to set policy over here. That’s a place
where we can go, that’s a place where you can do resources, and that’s a place,
just like Ken talked about, the vocabulary I think is a very important
recommendation that we’re going to have.

ASC X12 stands ready to create implementable standards for Public Health
reporting, as evidence by the body of work represented in a balloted approval
of the Health Care Data Service Reporting Guide, which is the discharge data
and also the post-adjudicated claims reporting guide as approved ASC X12
standards. X12 stands ready to develop new yet to be determined data standards
for the future, with confidence based on successful pasts. ASC X12 standards
fit the needs of Public Health reporting systems long used to establish health
policy in this country. With that, I know my time is short, thank you very much
for letting us participate, and I hope the dialogue begins.

DR. SUAREZ: Thank you.

MS. DAVIS, D: Thank you for again inviting for inviting IHE, Integrating the
Healthcare Enterprise. I am the co-chair of the Planning Committee for
Integrating the Healthcare Enterprise Quality Research & Public Health
Domain. I have the pleasure of also serving on the testing and tools committee,
and I’m on the IHE international board. So on behalf of IHE thanks again for
having us.

IHE really covers multiple domains. This QRPH, which is the acronym for
that, actually is one of 13 domains that IHE covers. These clinical domains
include radiology, laboratory, pharmacy, and so forth. Now, a little bit about
IHE to set the primer. The domains themselves have the development
responsibility to create what we call profiles, or the standard implementation
guides.

We are a harmonization process recognized by ISO as a Standard Development
Organization. So we are an SDO, but we’re also recognized by ISO as a Proven
Standards Adoption Process. Our process includes not only the development of
the standards, but also the testing tools that are required to test for
conformance, or conformity assessment like we’re getting ready to branch into.

Now, IHE in general. IHE started in 1997 by radiologists. They were tired of
getting an ADT message into a system and having to reenter it into PAC system,
reenter it into the Radiology Information System, and move that data. So that’s
what IHE was born for, and that’s what they have leveraged along the way. There
is an infrastructure underpinning that includes a lot of the ubiquitous ways of
moving that data around between the systems, but Public Health reporting is
just one of the places that this touches. And Michelle again I want to thank,
she is my newest co-chair coming into the domain, so with her help we are
actually trying to do some evangelization of the works that we’re doing within
IHE.

The barriers that we have within the IHE domain itself really is the
involvement of the stakeholders. I know a lot of people have testified to date.
We are very proud that we have over 30 people coming to our weekly calls to
help us evolve these standards, but we need more and more people to understand
what these are and how they can leverage them. We want to take the data that’s
collected within the electronic health records at the point of care, be able to
move that for that population health management, being able to provide the
improvements that are necessary.

The barriers that we really have are the HER vendors aren’t incentivized
right now. They’re really up to their eyeballs in trying to develop everything
for the Meaningful Use requirements. And so I work for a health information
exchange in the state of Tennessee, I’m the Chief Technology Officer, and the
hardest thing is trying to make sure that these stakeholders, the hospitals,
the eligible providers know what their vendors are even capable of doing. So a
lot of this stud is education, and that’s what we really need to improve upon,
but the funding for this is an important piece.

I think some of the people testified earlier today that you have to be able
to send the data to something, and that’s what we find as the biggest barrier
in our ecosystem. We have all of these wonderful standards that are ready for
the industry to adopt, but we don’t have enough of that adoption really to get
that momentum moving. We do have some very good pilots, thanks to the outreach,
within our process.

Real quickly, if you think about the process, I’m going to take you through
IHE’s process so you understand it. It starts and it evolves over an 18 month
overlapping cycle. The process starts with users – those hospitals, those
health systems, different organizations like CDC, CMS, HHS – come to the table
and provide something that they’re having issues with, and they provide a use
case.

That planning committee evolves those use cases into something that we then
scope for work that’s handed over to the technical committee, and the profiles
then are adopted through a six month period into written specifications.
They’re put out for a public comment period for 30 days, and that’s where I
encouraged everybody, all of you are subject matter experts in Public Health.
We need more of you to be reviewing these profiles and helping leverage them as
you do your work, as you continue to roll out to your stakeholders.

The actual profiles themselves, once the comments are pulled back, they are
republished for what we call trial implementation. The trial implementation
phase within IHE really is a period of time where it allows the testing of
those specifications to make sure that they were written appropriately and that
everyone is clearly understanding the points and the requirements of that
specification.

Once an IHE profile has been tested at two different Connectathon events
-– IHE does host the largest IT interoperability events in the world, and
they are the Connectathons. We average over 100 different vendors that come to
the table that test multiple domains at each one of these, both in the United
States, in Europe, Asian continents, as well as Australia and Canada. So what
we have is this framework within QRPH that allows for the development of early
hearing and detection type of profiles, profiles that handle healthy weight,
cancer registry reporting.

We have 14 different profiles all in trial implementation. That’s a barrier
of ours as well. We need to move those to final text, but we’re not getting
enough of those testers, people coming to the table to really test those
profiles. Once we do then we can start evangelizing and doing that marketing
for the ongoing hopefully deployment.

IHE development, the international realm does that development of the
profiles. The regional deployment arms – IHE USA, IHE Europe, IHE Korea
– they’re the ones that test at Connectathons around the world, and
showcase them at place like the HIMSS Interoperability Showcase that’s been
mentioned before. The Public Health folks have had PHIN conferences that we’ve
done demonstrations in as well. So the IHE profiles continue to be evangelized
in these various venues. Hopefully that allows us to educate the physicians and
providers of what is possible.

QRPH was founded in 2007, it is in year eight of its development processing.
And we do have a lot of good pilots that are starting, but we need more and
more of those. The testimony that I provided to you includes two additional
documents, including the submission letter. There is a PowerPoint which you’re
seeing here, as well as a written 16 page document that goes into a lot more of
this detail, because ten minutes isn’t enough.

So on behalf of IHE international I want to thank you again for the
opportunity, and I stand ready to try to help, and I look forward to ongoing
discussions and next steps to evolve our roadmap and hopefully get it deployed
more and more, so thank you.

DR. SUAREZ: We are going to go next to CDC. We have two topics on CDC
actually, one is on the vocabulary standards.

MR. GANESAN: My name is Sundak Ganesan, thanks for giving an opportunity to
present about the vocabulary work we have done within CDC. Again, this is not
about a tool. PHIN VADS is a tool that came as a result of work done by STO’s
and public health and various other organizations like APHO and CST and others.
So I’m just going to touch on the vocabulary standards, development, and
implementation.

I think this is trying to illustrate what is a value set, that’s what the
PHIN VADS is. You can see here on the right side, the standards like SNOMED and
ICD-9 and all of those are vocabulary standards. If you try to take a smaller
subset of the concepts, in this case say influenza related microorganism as a
smaller subset, that’s what we call it as a value set. I’m going to use the
terminology of value set, so that’s the main focus of PHIN VADS.

Again, just to give the history, the first Public Health values, I think it
was done in collaboration with HL7 for the immunization guide back in 1995. So
at that time all the value sets were embedded in the HL7 implementations guide.
So we got all the vocabulary recommendations and everything from HL7 Vocab
Technical Committee.

As a result, in 2004, VC we founded as a need for distributing the HL7
related messaging guide vocabulary related things. So PHIN VADS was created
mainly to distribute the Public Health value sets, focusing on Public Health.
So folks behind PHIN VADS, several people from CDC programs and other folks
also who are vocabulary and messaging experts, they actively participate in all
the organization during IHE and HL7 and APHL activities, as well as the CASTE
informatics activities, including the ONC Meaningful Use and various other
organizations.

So the main purpose as I mentioned was PHIN VADS to distribute HL7 2X, it
also distributes 3X and CDA related vocabularies. And also a CDC vocabulary
team acts in a way like SDO, they maintain the standards like race and
ethnicity and vaccines and healthcare service locations in collaboration with
HL7, so we do those things too. And then the main purpose is really to share
the value sets across all the messaging guides. We don’t recreate different
value sets for different guides.

And then again, one of the things we worked with HL7 and IHE was there are
several systems they wanted tracked through the API or Application Programming
Interface or web services to extract the vocabulary. So CDC was kind of in the
middle actually. On one side HL7 common terminology services was going on, on
the other side IHE was doing a sharing value set. So we did support both of
them, we did go to the showcase, as well as I think some harmonization talks
are going on, too.

And again, this is a history actually. All the standards we have used
basically originated from the E-Gov Consolidated Health Informatics. Obviously
NCVHS hearings, and the main thing is HL7 Vocabulary Technical Committee, that
is the main driving force for all the vocabulary related standards.

This is just a list of activities supported by PHIN VADS. There are a lot of
different things, including as I think Ken mentioned we do support case
reporting, case notification, electronic lab reporting, and then healthcare
associated infections, vital statistics, and various different things.

This is just to highlight that we did actually in 2010, there were two
testimonies given to the ONC Vocabulary Task Force. The main task for the
discussion was about how do we federate the distribution of the vocabulary
standards. So I just kind of provided the link in those things.

And recently in the last two years we have been collaborating with the
National Library of Medicine. They worked with us and then we share our
experience regarding the distribution of the value sets and implementations. I
think further that led to the development of the value set authority center,
which is mainly meant for distributing the quality measures of relate value
sets.

And then this is one of the organizations that use PHIN VADS actively for
certification, they use the CTS web services to do the Meaningful Use relate
certification.

This one, I think several times we heard about this one from CST and others.
This is one example of Public Health activity data. CDC, CSTE, APHL, and Public
Health Labs, and epidemiologists, everybody did together actually, including
HL7 standards too. So it is a collaborative effort to basically identify for
any reportable conditions a laboratory test and results.

So we did work with LOINC an SNOMED and everybody else. So that’s mainly
being used by the electronic laboratory, reporting mainly for filtering the
incoming laboratory test to identify which one is reportable, which one is not
reportable, in addition to even within the state as well as in CDC those tables
are portable condition mapping tables, provided more of a routing guidance
actually to various different programs.

And again, this is a list of things we have worked with the different users
within CDC, several web applications within CDC or other Public Health
applications like EpiInfo, PHIN MSS, Messaging and Quality Framework, and
various applications. They did use the web services, and we also worked across
all the programs, not only in Atlanta. Even here in Hyattsville, too, actually
we doo closely work with them on ICD-10 transition and vital statistics relate
activities.

This is again to illustrate how we are collaborating with others, all the
federal agencies on the top, and then also you have standard development
organizations. We work with the vendors, again we participate in the PHIN
conference, we want to make sure the standards we have created and the
messaging guides we have created are implementable, so we work with them
closely and make sure whatever they need for implementation, especially with
the vendors because they are the main implementers for the case reporting an
other stuff.

I think there are a lot of questions people have, there are so many
registries within the US itself. Each one has a different scope. As I
mentioned, CDC’s focus is on the Public Health vocabulary value sets, mainly
focusing on messaging. That’s the main thing actually, for messaging we do all
the specific metadata. Again, NLM, National Library of Medicine, their focus
was primarily the UMLS. That focus is really for SDO vocabulary distribution as
well as a mapping between the vocabulary standards, and now lately they’ve been
tasked to do the Meaningful Use quality measure relate value sets.

And again, other agencies like National Cancer Institute, they actually do
the cancer related stuff, plus they also distribute FDA related vocabulary
value sets too. And then AHRQ is also the same thing. They distribute the HITSP
information metadata models, as well as some of the information models from IHE
and other stuff. So again, I’m trying to illustrate here there’s not one system
that can actually handle distribution for all the Public Health. Even for PHIN
VADs, for NRAM and relate things we do work with EPA and get the vocabularies
from them, and for cancer-related things we do collaborate with the National
Cancer Institute.

So even though it seems like there are a lot of different registries and
seems like duplicated work, I think everybody has a focus. So I think one of
the things that we can make sure is not duplicating and giving clear
instructions to the implementers and saying that these are the focuses for
different registries. Again, I just want to briefly highlight NIOSH is again
primarily working on trying to incorporate occupational and industry related
data elements into the EHR.

So the things I want to make at the last point was some of the questions you
have addressed. We still have challenges in mapping the Public Health data
elements to standards, especially we need to have the data elements modeled
correctly in the LOINC, that is still a challenge. It’s not just HL7 alone, the
distribution of information between LOINC and SNOMED and HL7 that needs to be
done, and we are actively working on implementation of the various case
notification messaging guides. Right now we have five or six messaging guides,
so now we came up with a generic version two and stage mapping guide for case
notification, we are hoping the states could adopt and implement them for all
other conditions very quickly.

And I think the training has been emphasized by others too. I think we do
want to mention those things. And the last point I wanted to make sure is I
think it’s essential between the federal agencies and other terminology
distributors, including the vendors, to make sure they all collaboratively work
and give a clear instruction for the implementers where to get things related
to the vocabulary. I’ll stop here.

DR. SUAREZ: Thank you. So we have Nikolay next.

MR. LIPSKIY: My name is Nikolay Lipskiy, and I will present right now a
topic on BioSense and implementation of data standards. My colleagues from CDC
BioSense Program ask me to provide this testimony today. And I would like to
start from background on BioSense. First, what is the BioSense program? The
BioSense program is a Public Health surveillance system that was developed to
increase the ability of health officials at different jurisdictional levels for
efficient, rapid, and collaborative monitoring and responding to harmful health
effects. BioSense is a common electronic health information system that
provides to Public Health officials standardized tools and procedures to
rapidly collect, share, and evaluate information. This system is a CDC
collaborated effort with CSTE, ASTHO, NACCHO, and other partners to create a
user-driven nationwide syndromic surveillance system. We believe that
implementation of BioSense promotes utilization of standardized syndromic
surveillance.

Now people are talking about BioSense 2.0. What this new version is about is
BioSense 2.0, the latest version, which is streamlined collaborative
data-exchange system that enables its users to track health issues as they
develop and to share this information quickly with other Public Health
jurisdictions in the system. And these users have agreed to share health
related data.

BioSense 2.0 provides a mechanism for collection and sharing information on
emergency department visits, hospitalizations, and other hospital related data
from multiple sources such as VA, DOD, and civilian hospitals from around the
country. BioSense 2.0 helps participants in exchanging data in a more secure
way.

BioSense 2.0 provides data in a distributed cloud environment. This is the
first department of health and Human Services system that attributes completely
in a distributed cloud environment. Implementation of a cloud computing
approach gives participants easily managed on-demand access to a shared pool of
computing resources such as networks, servers, software, tool storage, and
services, with limited need for additional IT support. Use of these common
resources provides to users significant efficiency, cost reduction, and
information sharing capabilities. BioSense 2.0 provides to participants a
secure environment for exchanging data, free data storage, and an easy to use
data display dashboard.

As you can see on the presented BioSense map, at this time 51 jurisdictions
signed Data Use Agreement, eight DUAs are under review, and 35 are recipients
of BioSense 2.0 cooperative Agreement.

BioSense 2.0 and Meaningful Use requirements. Implementation of BioSense 2.0
expands the capacity of state and local health departments to support
Meaningful Use programs intended to expand the use of electronic health
records. Also, it provides the capability to submit electronic syndromic
surveillance data to Public Health agencies and actual submission according to
applicable law and practice that was added as the objective into the CMS Final
Rules HER Incentive Program.

BioSense 2.0 brings to participants additional capabilities to share
analysis and data. The Society for Disease Surveillance – ISDS – and
CDC, working with partners, developed the PHIN Messaging Guide for Syndromic
Surveillance: Emergency Department and Urgent Care Data Release Version 1.0,
which is based on HL7 Version 2.5.1, and compatible to Version 2.3.1. This
document provides a minimal data set and definition that supports syndromic
surveillance practice. It was implemented by redesigned BioSense 2.0 program.

BioSense implementation of ISDS Final Recommendation: Core Processes and HER
Requirements for Public Health Syndromic Surveillance of 2011 is one of the
approaches for security, privacy, and standardization. The ISDS document guides
on establishing secure and standardized data exchange environment for
participants through described business process for data sharing partnership,
conducting quality assurance, also by recommending minimum data set and
recommending standards for data elements.

Another important technical document that BioSense 2.0 uses for
standardization is the Public Health Information Network Messaging System,
PHINMS, Guide for Public Health Surveillance for Emergency Rooms, urgent care
facilities, and ambulatory clinics. PHINMS allows users to effect the effective
exchange of secure bidirectional messages. Also, it provides a common approach
to security requirements such as encryption and authentication, as well as a
standard method for addressing and routing content.

Analyzing the BioSense 2.0 data exchange we can specify existing national
security regulatory requirements and standards that should apply to Public
Health data exchange, which is one of the questions for this meeting. The first
of these documents is the Federal Information Security Management Act of 2002,
FISMA. This act requires that each federal agency shall develop, document, and
implement an agency-wide program to provide information security for the
information and information systems. Also, it authorizes the NIST to provide
detailed information towards compliance with FISMA.

The second document is the NIST Special Publication 800-53, Recommended
Security Controls for Federal Information Systems and Organizations, 2013. This
document lists the control enhancements that must be put in place and
documented to maintain the security of the data.

And finally, the third document is the Federal information Processing
Standards publication – FIPS – 199, Standards for Security
Categorization of Federal Information and Information Systems. These standards
provide a method to assess the impact level of the information and the damage
that may be done if the confidentiality, integrity, or availability will become
compromised.

BioSense 2.0 data elements. Recommended data elements for BioSense are 32
data elements from the ISDs Final Recommendation: Core Processes and EHR
Requirements for Public health Syndromic Surveillance, 2011. These elements are
categorized in three groups, such as treatment facility identifiers, patient
demographics, and patient health indicators. The treatment facility identifiers
comprises of the facility name, ideal location, facility visit type, and report
data/time. The presented elements of patient demographics include a patient ID,
medical record number, and information about age, gender, race, ethnicity, and
residence, and Sundak, just a few minutes ago described the elements in that.
And finally, patient health indicators is the group that contains unique
visiting ID, visit day/time, data of onset, et cetera, that is included in
VATS.

Recommended by BioSense 2.0 data standards. Implementing by BioSense 2.0
data elements aims participants in standardization of data exchange. The ISDS
recommendation, which allows us national data standards such as for example for
gender HL7 v 2.5.1 Administrative Sex, for race CDC Race Categories, for
ethnicity CDC Ethnicity Group Value Set, for chief complaint/reason for visit
LOINC Codes and ICD-9-Clinicla Modification as well as ICD-10CM or free text,
and so on, all this information in my testimony.

CDC participation in standardization of syndromic surveillance. As we’ve
previously discussed, development and implementation of BioSense 2.0 is a
collaborative effort. So, CDC Office of Public Health Scientific Services,
OPHSS, leads and coordinates CDC participation in BioSense 2.0 development as
well as in development and implementation of syndromic surveillance
standardization, gathering privacy and security recommendations. The OPHSS
Standards and Interoperability Program works with internal and external to CDC
partners on assessment and sharing privacy and security rules as it regulated
by HIPAA, HHS, ONC, et cetera.

Also, this program works with internal and external to CDC partners on
assessment and sharing Syndromic Surveillance Data Interchange Standards
through a publicly available website. I made a screenshot of this website on my
slide. And finally, CDC Vocabulary and Distribution System, VADS, maintains a
repository of standardized syndromic surveillance value sets that are utilized
by BioSense 2.0 which gathers them through a publically available website.

DR. SUAREZ: Thank you. We have Jim now.

MR. DANIEL: My name is Jim Daniel, I’m the Public Health coordinator with
the Office of the National Coordinator for Health IT. Thank you for inviting me
to provide this testimony this afternoon, and also thank you for inviting me to
speak about the S&I framework as opposed to the Public Health Meaningful
Use measures, which is what I always have to talk about, so I’m glad to talk
about something a little bit different. Although I’m still going to say
something about the Public Health Meaningful Use measures because I have to at
the end.

I’m going to challenge people to think a little bit differently about how we
approach Public Health and our interactions with EHRs. I heard several people
today say it’s not a part of Meaningful Use, just sort of like we’re giving up.
Meaningful Use is driving everything, if it’s not part of MU we can’t do it.
But I think there are some different ways to think about how we approach the
problems of Public Health and interoperability with EHRs

And instead of just focusing on specific measures like case reporting, or
specific measure like birth reporting, there is a stack of technologies that I
think we could focus on that could cover multiple Public Health use cases. And
I want to bring to light three S&I frameworks that actually I think could
help us get to that point, where there are multiple Public Health use cases
that we could address if we look at three S&I frameworks.

And one of the great things about these S&I frameworks is they’re moving
forward with or without Public Health. There are multiple other stakeholders
that are involved in these S&I frameworks, and so they’re moving forward
anyway. They’re going to probably be part of certification at some point, and
it’s up to Public Health to decide if we’re going to fit in and play or not.
But these are moving forward with or without us. I think as a community we’re
doing a good job getting involved in some of them. Some of them I think we
still need some work on. Some of that is because I don’t have time to
coordinate all three, but we’re trying to work on that problem too.

The three frameworks that I want to talk about are structured data capture,
data access frameworks, and healthy decisions, and a little bit about how each
of these could actually help us with Public Health. My good from Didi from IHE
did a really nice job of sort of describing the whole process that they go
through, and it’s exactly the same for S&I framework, so I will not repeat
that. They do it so well, we just copied them. And then I’ll end a little bit
with some of the work that I do think we need to do on some of the current
Public Health transactional standards and some of the things that we still need
there.

Let’s start with structured data capture. Structured data capture is really
this concept that came about because everyone in the world and their brother
came to CMS and ONC and said hey, if you made EHRs just include these five data
elements we could do our job so much better. And when we started adding all of
those up together we were like wow, the vendors are going to kill us, we cannot
do this.

So instead of taking the approach of adding all these new variables for
things that would help, even something simple like current pregnancy status,
which would be so important for so many Public Health conditions, or some other
things that would be really important for clinical resources, instead of
putting all of these into certified EHRs we came up with this concept of
structured data capture. And it’s really based on an IHE profile RFD where it’s
a forms manager approach. And instead of putting all that content burden on the
vendors, instead the concept is we’re going to be certifying them for an
interaction with the forms manager.

We’re going to be following standards that are already out there, we’re not
going to completely be reinventing our own standards, but there is some work
that we need to do to make it meet the multiple use cases that have come up for
structured data capture. And like I mentioned, it’s not just Public Health. FDA
is out there for adverse event reporting, clinical research is out there. So
there are several domains that are interested in structured data capture. Like
I said, it’s moving forward with or without Public Health.

The good news is we have a really great Public Health tiger team that has
been focused on structured data capture, and we have some really amazing work
that was done under the auspices of the Public Health Reporting Initiative as
well that we were able to build on. They did some great work defining common
data elements for Public Health, hats off to Nikolai and Anna who were
instrumental in leading that. And they had some really great pilots as well,
and we have some new pilots going on with CSTE around reporting of some
communicable diseases using the RFT profile as well. And that’s really what
we’re going to be building upon in the structured data capture.

So again, it’s this idea that instead of putting everything in HER it’s sort
of this forms manager approach, we’re going to certify that interaction between
them. And then it’s great because I heard Janet talk about outbreak management.
All of a sudden you need to know where someone has traveled to, or if they
traveled to a Middle East country because they have flu-like symptom and all of
a sudden we’re worried about MERS.

Now, instead of asking every HER vendor to add that field to their product,
we can just put it in the few forms managers that we’re hosting as Public
Health, and it’s automatically going to appear in those EHRs as they try to
report a flu-like illness, because we’re going to ask for any flu-like illness
that’s reported to include this new data element.

So it’s a really great way I think of thinking about multiple things that
Public Health wants to do. Case reporting is obviously one of them, we had
other use cases that came up in our Public Health tiger team around doing
community referrals for something like smoking cessation. If you want to do a
community referral you might want to get some additional information from
people about language preferences or other things that you might not
necessarily know that the structure data capture could help with.

Here are some of the standards that are being developed as part of
structured data capture. I will mention that auto population that I think we
heard about and RFD is a critical part of structured data capture as well. Any
data that’s part of a form that needs to be delivered to Public Health that
could be auto-populated will be, and then it says extra data elements that we
all want in the EHRs that aren’t there yet, those will be the ones that are
blank that will be left for the physician to fill out.

We’ve identified three pilots as part of our tiger team that we think will
be great to move forward with: case reporting around infectious disease, early
hearing and detection, and cancer reporting from physicians, which is currently
part of stage two Meaningful Use. And you’ll see a lot of those aligned with
the work that PHRI has already done. We’re really building upon their
successes.

The second one that I want to talk about is data access framework. Again,
this is moving forward with or without Public Health, but data access framework
is all about making the data that’s within an HER available for consumption by
people who need to have that data. There are lots of stakeholders there. Just a
physician who needs to move their data from one system to another, they decide
that they don’t like their vendor anymore, they need a framework and standards
to move their data instead of being charged $300,000 just to get rid of a
vendor.

So there need to be some open standards around that data access framework
that allows for that accessibility. Consumers may need some access to their
data as well. So these same standards could be applied for consumer access. And
then there are issues like ACOs, who might need to look at more population
health measures across a group of practices. These are the standards that would
be developed for that.

And then we have Public Health, who often doesn’t care so much about
individual case reporting for some conditions, but they want to know some
aggregate measures about the health of their community: who that have diabetes
also has their A1C under control, or people that have high blood pressure, how
many are on appropriate medication, how many are following the ABCDs. So these
are the types of questions that Public Health could start thinking about asking
with the data access framework.

This is a group that we’re still trying to get active engagement in. You’ll
see they have sort of different levels of the data access framework, and right
now Public Health fits in with the last use case, which is aggregating data
across multiple sites for a single query. We’re trying to push them into saying
that’s really critical, we need to think about that one sooner. So as the
Public Health community we’re trying to work our way into the data access
framework and make sure they move that third use case up, because that’s going
to be the one that really benefits Public Health.

There are a lot of standards that go along with this. I’m getting signals
that I’m running out of time, so I won’t go into all of these. There are
transport and security and content standards. Just like any standard we need
all of those to make something like this work. They’re being developed as part
of the data access framework. And we do actually have some pilots that are very
similar to the data access framework, which were query health pilots.

And query health, for those of you who have heard about it before, is kind
of being consumed underneath the data access framework umbrella. And we had
some really great population Public Health pilots in New York City and
Massachusetts, where they did some of this using query health, pulling together
information across multiple providers.

And finally, the last one is healthy decisions. I know a lot of times we’ve
talked about bi-directionality, where Public Health has information they want
to get to physicians as well. Sometimes that’s clinical decision support, you
want to make sure that they’re following recommendations around screenings or
other Public Health related activities. There is this concept of putting
clinical decision out there as a service, and Public Health is very involved in
this.

One great example that’s currently being investigated is just when is a
condition reportable to what Public Health jurisdiction. That is a huge
problem. And so we’re actually taking some of the work that Janet talked about
in the Reportable Condition Knowledge Management System, and figuring out how
that can be put into clinical decision support language, just like we heard
from Rebecca that immunization did, so that that could be put out as a service
and consumed by individual EHRs but maintained by one person, or one group, or
multiple Public Health groups.

Finally, my quick spiel on the Public Health measures. I think everyone
knows what they are so I won’t go over them. But what we’ve learned I think
with the Public Health measures that are currently out there is where we need
to be, and I’m quoting by data idol here, Doug Fridsma, we need to be sending
conservatively and receiving liberally, and we’re not doing that in Public
Health. We’re asking people to send really conservatively, and we’re accepting
even more conservatively.

And we need to think about how we can actually move to being more
interoperable, and that’s really getting to the send conservatively receive
liberal. And I think there are some things we need to do with our
implementation guides in our next round of implementation guides, making sure
that ONC can certify all of the critical data elements.

And Charlie has been through this process with me many times. And it’s just
a regulatory issue that we cannot certify optional fields. So we have to make
those required but empty, and then when we make those required but empty
everybody has to agree that if they have data and you don’t want it you’re just
going to ignore it, you’re not going to reject that message.

So there are some things that we really need to do on the Public Health side
to help us get to this point. And then we have great certification for our
senders, I think we also need to start thinking about some optional
certification for our receiving systems as well.

DR. SUAREZ: Thank you very much for that testimony. I think we have our last
testifier on this panel, Charlie.

MR. ISHIKAWA: Good afternoon, I am Charlie Ishikawa with the International
Society for Disease Surveillance, and I have the unenviable position to be
between you and a break. But thank you so much for inviting us to come here and
speak with you about the state of health data standards for syndromic
surveillance among US governmental Public Health agencies.

ISDS is an international 501C(3) non-profit organization based in Boston,
Massachusetts. We are a society of innovators and early adopters of information
technology for better disease surveillance practice and more effective Public
Health action. Our community spans disciplines, market sectors, governmental
agencies and geopolitical boundaries.

In recent years, we have become the de facto organization for developing
syndromic surveillance data standards in the United States. And you’ve heard
about some of this work from Nikolai and others before me. The Markle
Foundation, CDC, ONC, NIST, our JPHIT partner associations and many others
institutions and companies are our valued partners in this work — work that we
all do to help local and state Public Health agencies meet contemporary demands
for relevant and timely health information.

To speak about the drivers and facilitators of health data standards for
syndromic surveillance, I will share with you ISDS’s experience in developing
national standards in two areas: aggregate health data exchange, and then
secondly HER messaging for Meaningful Use.

In 2009, we leveraged an aggregate health data standard to establish a
national, near-real time influenza-like-illness, or ILI, surveillance system
for the H1N1 pandemic response. The system we created was called Distribute.
Distribute was a system of 43 local and state jurisdictions sharing their daily
counts of ILI emergency department visits with one another and the general
public. We worked with the CDC and PHII to build-up this system of systems
within three months, and at its peak, Distribute data covered more than 40% of
all emergency departments in the United States.

There are three things that we learned about adopting a Public Health
aggregate data standard from Distribute that I’d like to share with you.

First, adoption of an aggregate health data standard requires surge capacity
and strong level local and state surveillance system capabilities. Driven by an
emergency need for national situational awareness, federal resources enabled us
to stand-up a national system a remarkably small amount of time.

Although that infusion of surge capacity was critical, it was really just a
small part of the Distribute infrastructure, local and state syndromic
surveillance systems were its foundation. Notably, many of those systems were
built and maintained with Public Health Emergency Preparedness and Response
grants.

Second, aggregate health data standards must be simple and flexible for
successful adoption. Distribute data were messaged using a comma separated
value file sent over the internet via sFTP.

Only four data elements were required of the health departments to send to
Distribute: The encounter date, the date of the visit. A three digit zip code,
to give a geological sense of where the observations were being made. An age
group. And for each of those age groups a percentage or count of ILI visits
using the agency’s preferred ILI syndrome definition.

This simplicity of Distribute’s technical requirements provided a low bar of
entry for participation. The allowance for site specific ILI case definitions
provided the flexibility necessary for Public Health agencies to send data that
met their quality standards, and their quality standards for sharing with
external stakeholders for their use.

The third thing that I’d like to share with you about Distribute that is
relevant to this hearing is that philanthropists, non-governmental Public
Health associations, and academic institutions are critical for to the success
of data standards. Initial work on the Distribute model was supported by a
grant from the Markle Foundation. PHII, ASTHO, NACCHO and CSTE were our
partners with the CDC in making it a nationwide system. And that effort was led
by researchers from McGill University, Harvard University and the University of
Washington. Notably, our project leaders for ISDS were volunteers.

Supporting a Public Health response to influenza is just one way that
syndromic surveillance systems are used. They are also used to help monitor,
gauge and assess the population health impacts of extreme cold weather events,
hurricanes, terrorist attacks and other natural and man-made disasters. In all
these situations, patient encounter data from healthcare settings are critical
inputs.

Mainly because these systems are used for similar purposes, there is some
degree of standardization among different syndromic surveillance systems that’s
possible. Generally, health data for syndromic surveillance represent all the
patient encounters of a particular type within a treatment facility, not a
subset based upon specific patient health criteria.

Most systems collect these data at least once every 24 hours, often more
frequently, to maximize signal sensitivity with acceptable losses in
specificity. And since syndromic analyses, by-in-large, do not require the
identification of individuals, many of these systems do not collect limited
personal identifiable information.

Currently, Meaningful Use and federal assists from the CDC are driving
greater standardization in how hospitals provide these data to Public Health
agencies. The PHIN Messaging Guide for Syndromic Surveillance Emergency
Department and Urgent Care data is now the standard to which EHR technology are
certified, and hospitals attest for the syndromic surveillance Meaningful Use
measure.

It is a relatively simple standard. Ubiquitous admission, discharge and
transfer, or ADT HL7 messages are used for packaging data that support common
or more advanced syndromic applications. It is also a flexible data standard
that health agencies can tailor to suit their jurisdictional and system needs.
This standard was developed in 2009 with support from the CDC to ISDS, and it
was done to ensure that Meaningful Use will maintain the integrity of
established syndromic surveillance systems.

Now, as Meaningful Use Stage 2 begins and on-going syndromic surveillance
data transfer is core for eligible hospitals, BioSense 2.0 is providing added
capacity for some Public Health agencies to receive these data and develop core
syndromic surveillance capabilities for the very first time.

We are, however, at an early stage of adoption and have yet to see whether
or not the certified EHRs will actually send the right data in implementation.
Patient chief complaints, for example, are critical for syndromic surveillance
systems to provide real-time situational awareness.

These data are entered into electronic medical record systems using either
free-text data entry, which is the most useful and is highly preferred for
Public Health, or by selecting a text string from a drop-down menu. Will
certified EHR products send the more useful, unstructured free-text form of
chief complaint, or will these products pull the less useful structured data?

Theoretically, standardizing syndromic surveillance data messaging
nationwide is beneficial. With every health department receiving a core set
syndromic surveillance data, communities at every level could gain access to
more relevant and more local health data. Reaching this dream, however, will
take far more than a messaging standard. Those messages must carry quality
data, and Public Health agencies must understand those data and be able to use
and share them effectively.

To help minimize problems and ensure implementation success, Public Health
agencies and national associations need targeted funds that support communities
of practice for front line workers to rapidly develop and share practical
solutions. Syndromic surveillance also needs resources to further develop the
PHIN Guide for messaging de-identified laboratory result data, write a guide
for messaging ambulatory clinical care data, and assess the meaningful impacts
of health data standards on population health.

This is the state of syndromic surveillance health data standards in the
United States. Please be aware that there are also many health-related data
that Public Health agencies collect for syndromic surveillance. Even with those
data, although I haven’t spoken about them we generally find the three points
that I’d like to make:

One, data standards are used because they build a critical capability or
solve a common problem. Two, surge capacity and baseline capabilities are
necessary for standard development and adoption. And three, non-governmental
organizations and institutions are critical for the success of data standards
at a national level.

Now I would like to leave you with one closing thought: We need to know when
not to impose data standards. One of the strengths of this disease surveillance
methodology is in its ability to identify and track something that you never
thought of, something that you could not have built into a standard case
definition. Thus the critical importance of using free-text to collect chief
complaints, reasons for admissions, and causes of death–not just pre-coded,
structured or standardized representations of them. Sometimes a data standard
does not help Public Health.

DR. SUAREZ: I think we are going to open up for some questions from the
committee first, or do we want to continue? We probably should continue. If
it’s okay with the members of the committee we can continue. Or if you have
questions we can stop here and do some questions before. Let’s do that, I think
we have at least six. Let’s do some questions.

DR. COHEN: Very provocative stuff, thank you all. I’m particularly
interested in a couple perspectives that several of you raised, particularly I
think Charlie and Jim in your later talks. And it’s why in your opinion are
Public Health agencies slow adopters of standards? I guess my perspective,
working in the State Health Department, is we serve a variety of constituents,
particularly focused on community outreach. How do these standards really help
us do our work at the community level, and how do they help communities do
their work?

And I appreciated, Jim, your framework about collecting liberally but
distributing conservatively. Communities need aggregate data, not concerned
about individual data, and not necessarily concerned about standardized data,
Charlie, the point you raised about qualitative understanding of what’s
important to the community. So how can we evoke standards in a way that is
beneficial to communities and their needs for data? I guess that’s my general
focus.

MR. ISHIKAWA: I can speak to the slow to adopt point. From my perspective
and experience I think that resources and training are two of the big reasons
why Public Health is slow to adopt standards. In the world of syndromic
surveillance the struggle that we face with Distribute with sharing the
aggregate ILI data was that once you put data out there live and in real time
for the public to see, the public can misinterpret what they’re seeing, and
then that causes a need for the health department to go out and then correct
the misinterpretation, and that costs resources there. So I think that has a
lot to do with the hesitancy to adopt a standard for sharing data externally.

The other point that I’d like to make is I think systems take a long time to
build and a long time to change, and I think agile development is something
that it would be great for governments to embrace. Finally, the last thing is
training. We were very fast to develop a standard, and we had state and local
health agencies develop the standard with us for Meaningful Use.

One of them was an emergency, we had to put a standard out there where we
had none before. But the other was – and this is my point about philanthropic
organizations – there was an RWJF, Robert Wood Johnson Foundation project
through PHI called Common Ground that taught a lot of leaders in the field in
business process analysis. And that process and those tools were very critical
to help us express the business needs quickly to the technical folks so they
can translate it. And I think that way of thinking and that training was more
widely available to folks, standards development may happen more quickly.

MR. DANIEL: I don’t necessary think Public Health departments are slow to
adopt standards. I think Public Health departments are great with standards,
but everyone wants their own. And I think that’s more the problem that we have
in the Public Health community. And especially now that we have EHR vendors who
don’t just operate in one jurisdiction, they operate across the country.

And when we have unique standards jurisdiction by jurisdiction, and a vendor
works very hard to implement in one state and then goes to the next state and
sees it’s completely different, that’s where I think we come into big problems.
I think we’re data geeks in Public Health, we’re good with standards. But I
think we need to put some of our consensus building that we’re also good at and
apply it to ourselves, and make sure that we can come up with a set of
standards that works for everyone.

And that’s really hard, and I think that’s one of the things that we really
need to do. It’s not the way the United States is built. We’re built on states’
rights, and states have the rights to do anything they want, Meaningful Use or
not. That’s fine, but it may mean that they’re not as interoperable as they’d
like to be with vendors. I think it’s sort of up to ourselves as Public Health
to make ourselves go through that process. And it’s not easy, but I think
that’s where the federal government can help, through that process.

I liked what you said about the community needs as well. I really want to
emphasize for our data access framework that I am really focusing on getting
local Public Health departments excited about this, because I think that’s
where people are really going to see where the type of data that resides in
EHRs can be useful to them at a population health level.

I actually use examples from Massachusetts all the time with the ONC funded
project that you guys have looking at data across Atreus and your community
health centers, where you’re trying to identify potential risk factors or see
if interventions can help, and you’re trying to use the HER data to do I think
gestational diabetes that you looked at first. Those are really exciting
things, and I think we need to educate local health departments about that kind
of concept and what they might be able to start looking for.

DR. SUAREZ: So we heard two things about Public Health reporting. And
traditionally certainly Public Health has been the recipient of a lot of data,
and collecting and really doing a lot of analytics with the data. But now I
think we are hearing, of course through the efforts of ONC, a different way of
approaching also this, which is through querying mechanisms, through sending a
query rather than asking give me all your data and I’ll just do analysis.

So my question is really from a perspective of the developers of the
standard do you see that mechanism applying to all areas of Public Health,
including things like case reporting, which the main word of case reporting
means reporting, meaning sending data, and does it also apply to vital
statistics and other things? It’s more conceptually how do you see it fit
within the space of Public Health and the domains of Public Health. That’s one
question I guess, and I have a second one, but I’ll see what your perspectives
are.

MS. DAVIS, D: From the IHE perspective I would pretty much add that I would
recommend that we take and leverage what the vendors are already building in
their systems for that query model. I do believe it’s advantageous. I do
believe it’s something that could be utilized as long as we don’t try to do it
50 different ways.

The query model itself, if we look at the framework that IHE – and I’m
going to use IHE as the example here – for the nationwide health
information led by Healthy Way Today, those types of transports, we’ve been
talking about being able to leverage for the secondary purpose of gathering of
information.

What we’re finding though is you have to have some place to query. And
that’s what we’re having as the biggest issue, is we don’t have those Public
Health departments in the states having those data marts to query from. So
we’re going after more of the health information exchange venues, either the
privately owned ones or the community based ones.

The problem is they’ve not focused on that, they’ve been moving data,
they’ve not very much made their policies and their governance allowed to query
that information. So that’s been our biggest barrier. I think it’s possible,
the technology is there, I think the policy is what we’re going to need to work
on.

MR. POOL: I think my simplistic view would be that for those things that
occasionally I need to know, but I need to know every time A, B, or C, that’s
the place where the logic protocol healthy decisions triggering to say when
this happens tell somebody. That may or may not be triggering a fixed known
content, so it may just be a CDA, or it may be the structured data capture. Oh,
go fill this form out. I think the query really falls more in the you know,
this happened last week, we need to think about this, this is something we
don’t do every day we need to do this week, or we have a project this month or
this new effort.

So the query really is more those circumstances that you don’t do routinely
but are really important when you do them. That’s where data access framework
is really useful. The healthy decisions is useful to recognize when you need to
do something, and the structured data capture is when you figure out what it is
you are going to convey. So I think the three really, as Jim described, fit
together nicely.

MR. GANESAN: I do want to make one comment. During the implementation, or at
least the development of the Public Health Case Reporting Guide, one of the
challenges we found was a lot of Public Health case report forms, there were
challenges mapping to the information model in the EHR, like for example you
could have complications with associated conditions. You need to know this
data, you can get it from a problem list. That kind of mapping, need to educate
on the Public Health side, vice versa on the other side too.

MR. DAVIS, R: Walter, I would like to answer this question by just sharing
an experience. The experience I want to share is kind of how I got involved
with the X12 organization to begin with. I came into the X12 organization
clearly with a Public Health need. We clearly had state reporting systems that
we felt best fit into the X12 standards. We had had our own vetting process in
terms of what we decided we thought we needed.

But what happened when we exposed ourselves to the X12 process, it’s a
consensus process and people get to voice their opinions every which way but
loose. What I found myself having to do and doing for myself as I made
presentations to the X12 organization, and as I went back to my home back in
New York State, which is where I was working at the time, they started asking
questions that were really the right questions.

The right questions were, what are you using this data for. We had decided
we wanted these data elements, we wanted this stuff. It’s in the stuff
category. And then we never really asked that question of how we’re going to
use the data.

Once we started asking those questions we could then go back to the
researchers and say now how are you going to use this stuff, and then if we
started pairing that with what we were learning from the X12 organization, the
data we thought was collecting, what would actually answer those questions. So
we kind of just learned that you have to kind of reverse engineer the process.

And I think that’s where we are too, I think we kind of get tied up in all
these standards and these data elements and those data elements, and we never
really do the hard work first. And for me the hard work first is to decide what
we’re going to use it for. We know we don’t have unlimited funds, so you have
to do some prioritization along the way in the process.

But if you do that, then these frameworks really work, and the integration
with the standards committees becomes a really successful interchange, and it
gives you a way to open up to opinions. One of the things we’ve said, we have
Public Health but we don’t have a lot of Public Health representation and a lot
of committees. What we really need is a lot more opinions about what we want to
do to make better decisions about what we do do.

MS. KLOSS: Thanks to all of you. My question is really directed to Sundak,
but I would hope others would answer it, if it’s relevant. I was struck by your
slide that lists so many vocabulary registries and their scope, and we didn’t
really talk about whether that’s playing into the difficulty we have. We’ve
talked about free text and structured text, but where are we? Have we
identified this as a need to standardized vocabulary? At least start looking at
how they’re updated and disseminated in ways that perhaps are more standard
than is the case today. Is this a barrier? Is it an identified barrier? Is it
something that should be on the committee’s watch list?

MR. GANESAN: So I think the way implementation guides have been created,
whether it is IHE or HL7 guide, most of the time they usually tie you to one
single registry, so the implementers go to one place and get everything else.
So they don’t want implementers to go to ULMS for a few things, and NCA Cancer
Registry for a few things, and VADS a view things. So they usually don’t happen
that way. So far in the Public Health everything has been collected together.

But I think now we are seeing a little bit more, especially with the case
reporting, a lot of vital statistics being entered into the EHR. That’s where I
think we are entering into the domain. There’s no more distinction between the
HER and Public Health. So that’s why we need to talk together, the help of ONC
and others and make sure we don’t reduplicate or don’t distribute different
versions of the value sets, and asynchronization.

MS. KLOSS: It does make the understanding at the local level so much more
complex, when you’re drawing from multiple sources, each of which has their own
rules and idiosyncrasies and release schedules and so on and so forth.

MR. POOL: I would offer, this doesn’t resolve the battle of words that go on
amongst the various curators of vocabularies and ontologies, but ISO does have,
and I think we’re going to see increasingly, the metadata repository for the
terminology. So you may say each of you may have your own curated vocabulary,
but we can have belief it or not a standard about how you access different
vocabularies in the same way. That’s probably going to be the first place that
pops out.

MS. GREENBERG: This may be more a reflection than a question, or maybe a
question that really doesn’t have an answer at this point. Jim, you sort of
stole an old line of mine, that everybody loves standards, that’s why there are
so many of them. I know there is a lot of truth to that. There are reasons for
having different standards, and then there are no good reasons for having them,
people just have them.

But what you said about Public Health was certainly what we found 20 years
ago. It was the same thing with Medicaid, pre-HIPPA. There was no real
incentive it seems for people. Everyone did things a little differently, and
they said if you saw one Medicaid program you saw one Medicaid program, and the
same about Public Health.

At the same time it has been a really concerted, if small and poorly
resourced, I think there has been a really strong effort to move Public Health,
and the Medicaid program really, to more of a recognition and appreciation for
the need for standards. And not just their own standards, but standards that
work with other sectors.

And I guess what I’m wondering is is this still the view that people have
about Public Health, and is it still true? I mean, is the glass half empty, is
it half full? What I’m struggling with is is it a culture thing, is it an
incentive thing, or is it really what we’ve heard so many times this morning
and afternoon, a resource thing where really the incentives aren’t there
because of the lack of resources to implement.

So you’re too young to probably answer my question as to whether things have
really changed in the last 15 years. In fact everyone probably is. But I am, I
guess as I’m about to walk out the door I’m wondering whether things –
there certainly are a lot of challenges and barriers – whether we still
are stuck in that mode in Public Health or whether we actually have made some
progress, and what would be the next thing that would really increase that.

MR. DANIEL: I would like to answer that in the context of immunization. I
think immunization is sort of one of our shining stars in Public Health, and I
think they have done an amazing job. If we start with what the 231 message
looked like in stage one Meaningful Use, and what that changed to in the 251
method for stage two Meaningful Use, there was a huge amount of change between
those versions. And it really was a response to what we found out when we tried
to be interoperable across different jurisdictions.

And I think before Meaningful Use this really didn’t come to light, because
we weren’t working with the vendor community holistically. We were working with
a few sites who really wanted to do data exchange, and maybe a large provider
organization. It was more the internal IT staff that were maybe doing this,
they were one-off solutions. I don’t think it became as apparent until we
really started to try to do single vendor implementations working across
multiple jurisdictions.

And I think the immunization community realized immediately what was going
on, and they did an amazing job I think of making those changes. So I probably
was a little too tough on Public Health when I made that statement, because I
think we have done a great job moving forward where we’ve identified the
issues. There are still some things where there are very good reasons for
jurisdictions to have different data needs, but there is a way to put it in a
standard that still works for everyone.

And I think we’re at that point right now where we’ve kind of fixed the easy
ones, now we need to fix those issues that are a little tougher. State A has to
have data field X, but for state B it’s against state law for them to collect
that. How do we put something like that into an implementation guide that’s
going to work for everyone, or do people really need to have to go back and
make some tough policy decisions about all right, it’s okay for it to come in
as a message to me, but I’m just going to ignore it.

We’ve definitely made progress. I don’t think we’ve made progress in every
area, but definitely the ones who’ve been working in Meaningful Use have
started moving in that direction. Immunization and syndromic have been great,
because they’re not going necessarily through ballot processes with every
implementation guide, so they can move a little faster. ELR is just now moving
to their first post Meaningful Use implementation guide, I’m expecting to see a
lot of corrections in that for ELR as well. I think there are still challenges,
but we are getting better.

MS. BURKE-BEBEE: Along those lines, in my SDO days, which there were many,
but it’s been a while, I remember in a situation where you said state A or X
could do one thing but the state next-door might not be able to do that, I
don’t think that’s a technology problem, that’s a policy problem. So SEOs, as
far as I know, allow within their guides to do either or. And then the policy
drives whether they do either or. So I don’t see that as a barrier. I’m not
saying you thought it was.

MR. DANIEL: I think it could be a barrier if the guides are not written
correctly. There are ways to do it correctly. We have some examples right now
where there are some issues that are barriers to interoperability because of
some policy issues. There are ways to write the standards to get around that.

MS. BURKE-BEBBE: And that is where Public Health needs to be at the table.

MR. DAVIS, R: Let me follow up what Suzie said. An idea that Denise had
earlier in her presentation, clearly in terms of state data collection some
states allow for a personal identifier and some states do not allow for a
personal identifier, that’s just the way it is. Clearly the implementation
guides have to allow for both. But in truth the bigger question for all of us
is that it would be better for Public Health if that identifier was collected
by all.

MS. DAVIS, D: I think we need to do a better job at evangelizing this
framework so people can plan their roadmap for it. That’s been our biggest
barrier. We’re changing the paradigm of how this is happening. No-one has done
this electronically before in a real ubiquitous manner. So I think part of that
is getting out ahead of the curve.

We’re kind of always playing catch-up, we need to start preaching and
evangelizing these different ways of doing what I would call cross data mapping
to find out where those barriers are so that the states who have those issues
could determine how they could change it or if they’re willing to. When HIPAA
came out they did a study right when HITSB first started, 38 of the 50 states
implemented HIPAA differently. That’s what’s happening with Public Health as
well, we need to start harmonizing some of these back to the policy perspective
so that people can plan ahead for that. That’s what we keep not doing, I think.

MS. BURKE-BEBEE: And I would say that federal policy is the floor, and not
more than that. So we can’t become more stringent and get into that arena. So
HIPAA was the floor.

DR. SUAREZ: It is interesting if I may just comment. When you think about
HIPAA, you think about HIPAA privacy. But you’re not thinking about HIPAA
standards. HIPAA standards is not a floor, it’s a ceiling. States cannot change
the implementation guide that every provider on this country has to do. They
cannot modify 837s and other elements.

So in many respects the standards became the ceiling on anybody, including
states, Medicaid, and state Public Health that want to see something in the
standard change, they have to come to the place where the standard is
maintained. It’s very different from privacy, privacy is a floor. But the
standard side is actually a totally different perspective. I think that is part
of what we are all here thinking about, because in Public Health when you’ve
seen one standard you’ve seen one standard.

MR. DAVIS, R: Some people treat it that way, but I’m here to report about
the Healthcare Service Data Reporting Guide. That has become a HIPAA compatible
but not HIPAA compliant guide, because Public Health isn’t bound by the HIPAA
regulation for how it collects. So in essence because of the working
relationship that X12 has formed with its Public Health partners, we clearly
have data supported in that guide that goes beyond just the ceiling.

DR. SUAREZ: It is an implementation of a standard.

MR. DAVIS, R: But X12 as an organization was more than happy to change their
standards to support our needs, even though they weren’t going to be used for
claiming purposes.

MS. GOSS: Let’s take that into a larger context. You came forth with a set
of data requirements, you answered all those tough questions we asked you about
why do you need it, what are you going to do with this stuff.

MR. DAVIS, R: I got marked.

MS. GOSS: I remember, because I was there when the 837R was being built, and
Bob and Suzy helped me learn my world of Public Health thanks to X12 days. But
it’s because you brought together your business need, you defended your
positioning, and then you broadly ran around and informed everybody that you
needed to get onboard. And it goes back to the repeating theme that we’re
hearing, it is culture. It is understanding. We’re trying to build the airplane
while we’re flying down the road. It’s a paradigm shift, and it’s good that we
can set some floor on which we can build.

MS. WILLIAMSON: Just a point of clarification from Jim. You spoke about
structured data capture and the work that you’re doing on taking a forms
manager approach. You mentioned about certification for interactions with form
managers. What I’m interested in knowing is what your plans are for that
certification process. Is it limited to only the domains that you mention
within Public Health? If there are others that are taking an RFD approach, can
they also be certified within this process?

MR. DANIEL: It is not something that is definitely there for certification
yes, but I think we’re moving in that direction. But the approach that we would
take would be more just the interaction between an EHR and a forms manager,
independent of the domain of content. So whatever the case report, whatever the
Public Health use case was that wanted to come into that, as long as they’re
following that standard that would be okay.

I think there might be a little trickiness around pre-population, and one of
the things that we’d need to do, and Sundak brought this up, we tend to ask the
same question a lot of different ways on our case reports. And if we’re going
to do pre-population of any of those we need to go through a little process of
rethinking some of those questions.

I remember we did this with the case reporting system once where we looked
at the CDC questions for travel history and there were 30 different ways to ask
travel history depending on the disease. We came up with three basic questions
that we could ask the same across every disease. I think we’ll have to go
through a little bit of that with pre-population, but for the basic interaction
it should work for anything.

MS. WILLIAMSON: I think this is a key point. I know from the IHE work we
have a lot of programs that are developing RFD based approaches. And so does
each one apply? Are they within the game? I think looking at that is something
that IHE would be interested in.

MR. DANIEL: I think they would. And we’re looking for pilots in Public
Health who have partners in the vendor world to show success, so please contact
me, and we’d love to throw you in the mix. We went through a process of
identifying ones that we thought were good for Public Health purposes, but
anyone who wants to do a pilot I think we’d be happy to work with you.

MS. DAVIS, D: ONC, we welcome them, but at the IHE table we do have in our
acute quality research and Public Health domain we’re working through
documenting some of that for the structured data capture. So I think that this
collaboration will hopefully lead to making that a little more obtainable.

Agenda Item: Panel 4: Software and Application Vendors
and Public Health Data Standards

DR. SUAREZ: One of the benefits of our hearings is that usually it opens
doors and connections between people, so this is really wonderful. I think
we’re going to move to our next panel. Thank you so much again for your
participation, your testimony. So we’re going to move to our panel on software.
So we’ve heard from the developers of standards, now we’re going to hear from
the developers of products that are used in Public Health. And we’re going to
start with EPIC, an EHR vendor.

MR. STAMM: My name is John I am from EPIC. I’m here to give you a
perspective from an EHR vendor on integration for Public Health. So, a little
background on EPIC. We as an EHR developer have a suite of applications that
includes both inpatient and ambulatory applications. We are installed or
installing at 250 or so organizations across the United States, which gives us
representation in all 50 states. We are an active member in a variety of
standards bodies, including NCPDP, HL7, and IHE, and we have built out support
for multiple profiles for Public Health, including vaccinations, infection
controls, syndromic surveillance, birth registries, and cancer registries.

As for my background, I have a background in software engineering. I’ve
worked on installations for EPIC systems across three time zones, and I attend
IHE’s quality research and Public Health meetings to bring the learnings and
the new standards there back to EPIC and disseminate them.

This is sort of the big slide. If there is one thing that I am hoping you
all take away today from the things that I have to say it’s this, and that’s to
make the process easy. I am asked not infrequently by healthcare organizations
why is it so difficult to integrate for Public Health, why is it that EPIC has
gone through Meaningful Use stage two certification, and we know you have an
interface that works, and it still takes us a lot of effort to get up and
running.

I think for those of us here in the room the easiest answer is that Public
Health is a moving target. You’ve heard a lot of testimony today about ways in
which the standards are changing, ways in which there are variations between
states. And it’s all true, but that’s not a very useful answer to give back to
those healthcare organizations.

And so as a response we try to aim at consistency, we try to aim at
repeatability, being able to say that if one healthcare organization has been
able to implement a standard to get that information to Public Health the one
down the street should be able to do the same thing, the one next door should
be able to do the same thing, the one in the next state over should be able to
do the same thing. And we try to aim as well at automation, so making sure that
we’re using up to date standards, making sure that we’re using standards that
are available to the industry as a whole so that others as well can participate
with those same initiatives.

When healthcare organizations look to implement for Public Health there are
usually two areas that they ask about. The first one of the two is content. And
I don’t just mean what the documents are going to look like, really from their
point of view it’s a lot more in-depth than that.

Do they even capture the data today that you’re interested in? Do they
capture it but it’s not in the format that you want? Is it something that if
they need to start capturing that should be captured by a clinical user or a
registration user? There are a lot of questions that need to be answered so
that they can figure out where to put that to make sure it even gets collected
in the first place.

The other one of the two areas which is usually a little more technical is
the communications piece, security. Is this something that we have up and
running today and we’ll be able to troubleshoot? Is it something that we need
new servers for? Is it something that we’ve done before or not? And we’ve seen
in both of these areas a lot of successes, when there are good standards and
good recommendations for the way to approach these challenges, and a lot of
cases as well when there are tools that can be used by primarily states to help
lower the barrier to entry for healthcare organizations and for the states into
these information areas.

So, to echo what you’ve heard folks say several times already today,
immunizations is kind of the shining star for solutions here. At this time EPIC
integrates with approximately 46 immunization registries across the United
States, and almost all of those interactions are using HL& Version 2
standards for vaccination messages. It is a pretty robust standard, and as
we’ve heard it has gone through a few changes with Meaningful Use, but most of
the information that those states are looking for is encapsulated in that
standard.

The communication mechanism to those states is more open. There were a lot
of variations in the industry before the introduction of the CDC WSDL, which
we’ve seen have a fantastic ripple effect throughout the industry. States as
far apart as Louisiana, Rhode Island, and Nevada are all able to use the same
communications mechanism now for submitting that data. And it doesn’t just
include submissions. It’s actually robust enough as well if the state
implements it to include queries against that data.

Syndromic surveillance you also heard a lot of talk about during the last
panel. Syndromic surveillance specification for Meaningful Use is primarily
based off of those ISDF specifications that are a strong baseline as far as
what ADT HL7 Version 2 information needs to be submitted. The communications
mechanism for it to the state is left more open.

In particular the implementation guides recommend PHIN MS, or ultimately the
ability to add your own communications mechanism as you should need. BioSense
2.0, which you had a speaker talk about already, sort of has stepped into that
gap to provide a variety of standardized communications mechanisms that again
make it much easier for both states to get involved as well as for healthcare
organizations across state lines to know what they need to do to actually get
that data submitted.

In this case some guidance is necessary. We are seeing organizations
implementing syndromic surveillance in a few states find that there are so many
options they’re not sure which one is best for them, but in general the
industry appears to be going in the right direction here and getting that
information out where it’s needed.

Cancer registries are new to Meaningful Use stage two, and they are an area
that as well has well defined content for it. Physician reporting for Public
Health cancer registries was an XML specification through IHE that has been
implemented at EPIC amongst other organizations for submission of that data.
And we’re expecting to see big gains in the industry as well with the
implementation of eMaRC Plus, which is another CDC tool for the sake of
digesting those cancer messages and eliminating that barrier for states to be
able to start receiving standardized documentation.

The communications mechanism for cancer registries is actually left
intentionally undefined in the implementation guides. To give a little story
about this, as EPIC was going through and becoming one of the first
organizations certified for this objective for Meaningful Use stage two, we
used a method to write those document out to file for the sake of going through
the certification process. It’s not particularly automated for healthcare
organization, but it is something that helps you get through certification and
enables a certification to latch onto that in any way that they need to
actually submit the data.

At about the same time we were participating in the IHE Interoperability
Event Connectathon, and there the tests for the cancer registry documents were
primarily based around that retrieve form for data capture method, which is
something that we’ve built out for other use cases as well, and we’re able to
support and test here. But it does require more of sort of a front-end
commitment in EHR. You need to work it into end-user workflows that they know
when to press it, so they know that they’ve actually filled out the right data
in the chart before they launch and submit that document.

After Connectathon we moved on to be involved at the HIMSS Interoperability
Showcase for Cancer Submissions. At that location we didn’t use either one of
those methods. There the submission happened automatically as part of a
provider’s workflow in the system, and it was submitted using the Direct
Projects XDR Protocol. At HIMSS we talked with a lot of folks that were
interested in the standard, either just about how it worked or about how to
implement it, one group of which was the state of Kentucky, who is piloting the
ability to receive those documents. They have written a web service for the
sake of receiving an MDM T02, it’s a transcription style HL7 version 2 message,
with the content being the CDA from the PRPHCA specification.

So over the span of approximately three months we had come up with four
different communications mechanisms that needed to be supported in order for
organizations to be able to submit the data. And we’re seeing as organizations
start licensing the interface from EPIC and looking to see what they’re going
to need to do to perform those submissions that now they are running into
heralds both with the fact that they need to contact the registry to find out
whether the registry is using EMRC Plus or able to accept the documents, but
also whether or not there’s even a communications mechanism that is supported
or is standard or is something that organization can use to submit the
documents.

So we’re back here again, make it easy to do the right thing. This means
providing good content standards, so that we know what’s being reported, we
know what needs to be captured. It means providing recommendations for the
transport. They don’t have to be as stringent, if there is a good use case for
a variety of different occasions/mechanisms, then really having that guidance
as far as what works best and pointing out the standard communication paths is
sufficient there.

And in a lot of cases we have seen very good progress with the creation of
tools that are lowering that barrier for states to enter and at the same time
lowering that barrier for healthcare organizations that want to be able to
submit that data. It is difficult to integrate for Public Health, but it’s
something that we’ve done before, and we’ve used these tactics to eliminate
much of that difficulty to be successful, and it’s something that we can do
again. Thank you.

DR. SUAREZ: We are going to go next to Ken.

MR. POOL: I am going to take the opportunity to stand up here so that you
don’t confuse me with the Ken Pool who was sitting over there. I have no
relationship with that fellow whatsoever. In this circumstance I get the
opportunity to kind of speak my mind as it were, from not somebody who was out
doing this kind of stuff, but has been an interested sideline irritant in this
process. We have worked for 17 plus years, standing between Public Health and
provider point of care people from hither and yon around the globe, and so it
has kind of given me an opportunity to watch this as the folks that try to make
these connections.

First off, I’ll just kind of with you, not surprising I knew ahead of time
you would be hearing some emerging themes today, like we need participation by
Public Health, we need adoption, we need money, we need motivation, we need
experience, we need to clear the path and get rid of all these blockades that
are getting in our way. Let’s get rid of those silos. I don’t know why
everybody hates silos, but let’s get rid of those silos. Those kinds of pieces.

Being a man of few words I’d like to take a different tact in this next
session. First off, reminding one of my favorite phrases about standards, and
that is think of standardization as the best thing you know today, but it’s
going to get better tomorrow, but you get somewhere if you start doing that.
That was Henry Ford’s perspective on standards. I would say that my assessment
as a bystander is that we now have a store full of standards. All the shelves
are full of them. But you’ll notice the absence of buyers. There is no-one in
that store buying our standards. The point being for Public Health, and being a
clinician Public Health is clinical Public Health stuff, so those of you who
don’t live there, doing Public Health takes two. Somebody has to send somebody
else something, there’s two of us. It takes two to tango. No, tangoing can be
rather elegant, but most importantly anybody can do it.

And so if we’re going to tango in the tango of Public Health point of care
communication, one of the questions then becomes so why would the providers at
the point of care do this. Now, we can go on and on, but the truth of the
matter is this is all it’s about, we want to be good citizens. When is the last
time you saw a physician jailed for the failure to report a chlamydia case? It
doesn’t happen. For the most part it is truthfully, having come from the
provider side, it is because we want to be good citizens, so we want to do our
part.

The problem that we face as a clinician, as a point of care provider, is
that if I find myself on the field throwing my data down the field, I’ve got a
problem. There’s nobody there to receive it. I don’t have a receiver. So before
I can throw my data I need to have a receiver out there. The other thing is
sometimes if I do this, if I go through, does it count? Do I get a penalty?
There are jurisdictions where sending my data electronically wouldn’t qualify
as having done my job as a Public Health citizen. We need to make sure, and
you’ve now heard several people saying we need to make sure our policies align
and allow the kinds of capabilities.

So we need to dance. But if we’re going to, someone has to lead. And that
someone should be Public Health. Now, it’s a little daring. It means you may
feel a little bit like the Wild West trail blazer that’s wandering out if
Public Health has to go first. Public health is not historically a group of
folks who are necessarily daring. But I’m going to tell you to bear up, move
on, because if you do trailblazing can bring glory. And so there is fame that
can come to those of you who move forward in Public Health.

So all we need to do is to get Public Health to get the ball rolling. Now,
as amazing as I believe my colleagues in Public Health are, they may need a
little bit of help. Guess what guys, that’s where you come in. That’s what I
think you guys at the National Committee need to address: how do you help
Public Health get the ball rolling, because they’ve got to go first.

Well, convening these high level groups is your first step. Because if
you’re going to help them get the ball rolling it would be nice to know where
it’s going to go. Because it might not be where you want it to go. But the good
news is once you do that your next task is get everybody pushing in the same
direction, because to get that ball rolling we’re going to have to push
together. See, your job is so easy, isn’t it?

The neat thing is once you finally get the ball rolling some really exciting
things can begin to happen. So where do you start first? Where do we go?
Clearly we need to find the right path. We need the path to glory that helps us
come to the right place. I can’t tell you a lot about it, but I can tell you
first off it’s not the secret path. This has to be public, you can’t keep it
secret, proprietary, local. This cannot be the secret path.

And that pathway probably means you’re going to have to build some bridges,
something to connect pieces together to get small little crossings working.
Those bridges are probably your path to glory. Now, I do have a couple of
admonitions about the bridges. First off, not a bridge too far. Don’t
overreach. Choose brides key and strategically. Make sure they go somewhere. We
don’t want to build those bridges to nowhere. Choose that wonderful little
place you’d like to be. Make a bridge to a wonderful magical place in your
little domain, one at a time.

Now, recognizing this problem is out there, we know that if we can’t start
building our bridge from the provider level, because there’s nobody there, we
can’t start building the bride from the program level, because each and every
one of them would have to build up, but they would lose their piece.

The bridge in the middle is the first step. Each piece at the ends can begin
to adopt, et cetera. Now, what that means is there may be a time when you need
more than one bridge. The new pathway, the old pathway. Data from this, data
from that. You may need intertwining bridges for a while. A bridge for a little
while, a bridge that will last forever. Think about the idea that you may need
multiple bridges. But of course the piece is if you choose the bridge wisely
you’ll end up at some success.

Now, you may not be able to read this quote, this is another one of my Henry
Ford quotes. If we’re going to build Public Health standards and do this, what
are we going to build, what is this bridging capability going to be like? Well,
Henry Ford said: If I had asked my customers what they had wanted, they would
have told me faster horses.

Okay, what is it you’re going to get if you go to standardizations? We
recognized we’re walking a monetary tightrope. We know that one size won’t fit
all. But on the other hand most of us fit pretty comfortably in only a few
variations. Now, that doesn’t mean just because something is standardized
doesn’t mean you can’t have a little personalization to it, but do be careful
and not go to extremes, because that really becomes extravagant, and maybe not
particularly productive.

So the last piece I will touch with you is the issues of scales of judgment.
Who makes a decision? How do we know when something is right, when something
works? It’s important that we have a certifying body that we feel good about.
It’s not just that you’re certified, there is the question of who certified it,
does it really mean something. Clearly a good answer from a physician side is
that it must be doctor approved. I couldn’t find a NIST Approved stamp
somewhere, so FDA will have to do.

So my summary is it takes two to tango and someone has to lead. It ought to
be you guys. It will get real exciting when you get the ball rolling, you’ll
find yourself going to a wonderful magical place, still managing to navigate
that delicate balance, and just buy certified products. Thank you very much.

DR. SUAREZ: We are going to go to our next one.

MR. KHETERPAL: My name is Vik Kheterpal, I am a clinician by background, and
I guess this particular perspective is maybe just a little bit different, as
the sudden note of this particular title slide, Observations from the Front
Lines Trying to Connect Healthcare. It’s a little bit different vantage point
that we bring. We are I guess participating in something that would be
described as facilitating the exchange of health information cross
stakeholders. All of them, any of them.

I didn’t quite know why last week this slide made sense, but I think a
couple of the comments that Charlie and others have made, just instinctively I
thought it was important to sort of think about what Public Health reporting
is, sort of in the weeds a lot of the standards and answers versus immunization
versus ELR versus all of the various use case specific things that people need
to see for promoting the public welfare is all well and good, but we need to
appreciate that the bulk of the times the source of that information is some
sort of a caregiver/patient interaction.

And it tends to happen out in provider networks, inpatient settings, and
ambulatory care settings. And these folks, to be blunt, have a hard enough time
exchanging information with one another as due course of taking decent care of
patients. They don’t know medications, so they don’t know what to prescribe
them, and it has an immediate effect. And in that broader context there are
many others who wish to have that information, but this issue of exchange of
health information is all we see Public Health reporting as, it’s just a
particular use case of that specific thing.

The target happens to be a Public Health department – state, federal,
CDC, you name it. There are many regulatory bodies that want data. CMS wants
clinical quality measures, payers want to do payer for performance,
professional societies, AECC, STS, a variety of infection control, join
commission, lots of people want data. And every one of them seems to come up
with a rally awesome standard by which the providers ought to take what is
supposed to be the byproduct of them delivering care, documenting in their
electronic medical record.

And arguably the holy grail was supposed to be I should take care of
patients, I should document what I need about them, I should have some sense of
my broader responsibilities to the community, and there might be some things,
but by and large if we’re going to have efficient and effective care we don’t
want to distract the clinicians at the bedside with too many secondary use case
kinds of things. SO I worry about RFD and a lot of these things, that they’re
well-intentioned but they’re not the dog, they’re the tail.

And we ought to be very careful when we think about Public Health data
standards and some of the challenges, that part of the reason they may be
challenging is they have a place, but it’s not likely the dominant place. And
on the Maslow triangle of needs the food and shelter that I think clinicians
have is the ability to share data with one another to deliver safe care while
the patient is there, so they don’t prescribe something that a patient is
allergic to. Because they don’t even have good data about that today.

And so in that context it’s very hard to get time and attention from them to
talk about what might happen from a surveillance perspective six months to a
year later. And so I don’t think there are any good actors or bad actors, I
think it is a matter of context that I think is very important. And that’s our
perspective, this is what we do. We have sort of a frontline seat. We have
seven states where we are the statewide HIE, we have several regional health
information exchanges, we have large integrated delivery networks that almost
always are exceptionally running fragmented medical records.

And in that context we have this front row seat in facilitating as a
technology provider how to exchange data with one another. And in doing so,
this issue of what I would call the last mile problem, getting the unstructured
chief complaint as Charlie was talking about in addition to the structured
chief complaint resolves problems versus active problems, meds at home versus
meds currently prescribed that I haven’t filled 30 percent of the time, those
types of nuances become exceptionally important.

And that last mile connectivity, so we love standards, don’t get me wrong,
because that’s the only way we can reduce the friction costs of moving data.
But I do believe we must temper our desire to have lots of standards, one for
each use case. And I think that if anything else is sort of a dominant concern
I would have in our experience in the field.

So, now to switch to something more specific. Our perspective is that the
PHIN, the 2011-2016 strategic plan is thought of, and this is literally stolen
from the PHIN Strategic Plan, cut and paste, is part of the National Health
Information Network. So this coexistence and mutually reinforcing theme is
certainly called for in the strategy, the question is to what extent is it
actually realized in practice. In that way, there were seven questions we were
asked to sort of consider, and we did look at them fairly carefully with the
team who’s out there in the states trying to do this work.

Where are we strongest at, and where are we weakest at? I think the stuff on
the left-hand side, as an example, BioSense 2.0 in our experience is an awesome
project because it reduces variability. It facilitates municipalities that are
unable to maybe have the technology investments to be able to do that in a
first class way. It’s an opt-in model so if you don’t want to use it you have
something else. It has all the right governance ways of thinking about it, the
technology is first class compared to the alternatives available oftentimes.

And then we work up from that, some of the technology block and tackle
that’s been done, the NIST test tools for Public Health, the vocabulary
available on VADS, fantastic. To put all that stuff in one place, it’s a
wonderful resource for the community that’s trying to do this kind of work. And
the HL7 Standards themselves, among standards are pretty robust because they’ve
been around the longest arguably, or at least from an adoption perspective.

Even in those areas, if you move to the right things start to weaken when we
talk about payload and content. There is no question 2.5 is a more robust
standard than 2.3. But that doesn’t by itself make it better. To take Ken’s
thing, it’s a really cool thing on the shelf, but if there are not customers
who are able to afford it, they can’t buy that really awesome thing that you
just put on a shelf, it’s kind of neither here nor there. It’s a tough upgrade
for folks to go to that thing.

And too much of a good thing, and not following the sort of the enemy of
good is better model, thinking that in mind, and maybe having greater adoption
for the current generation of standard before we sort of jump into the next
iteration, because that will give us some corner edge use case that is really
cool for a pilot, that gets grant funding moving while we leave the base of the
population well behind because they don’t even know what’s going on because the
top aspects of the triangle keep moving, I think that’s a real concern we have.

The Meaningful Use issues, and sort of the lack of cohesion between
Meaningful Use, VDT, and CDA requirements, everybody else has talked about it,
but that just needs to be fixed. It’s inexcusable on a going forward basis. And
the idea that the requirements themselves in the EHR aren’t tested, it’s okay
to think of it as we don’t want to test each EHR vendor during the four or five
certified ATSBs, but then we need to have a content certification when the
templates are done, because at the end of the day if that doesn’t happen it’s
not going to work, and we’re going to have the same kinds of things that John
was talking about.

You can have a certified project, and every deployment, you go from one to
the other, both it doesn’t work because of how they configured your product at
their EHR, and it doesn’t work because the Public Health municipality did
something different. So the standards themselves are pretty different, if
anything we keep changing them too often before they get widely adopted because
somebody comes up with a really cool idea in a conference room like this,
unfortunately, while the marketplace out there tends to really be struggling
with something from three years ago.

So, just a little bit more. And I’m not sure there is anything new here. If
I can describe it as maybe there’s two themes here. One is too much of stuff.
So there are too many content standards, there are too many vocabulary
standards, there are too many transport standards. And it’s done in the context
of people should have choice. But it’s like the old analysis, you’re trying to
buy jam – I think HBR does this awesome study – and people freeze.
You’ve got 16 jars of jam, you can’t pick which one it is that you want to get.
Consumers can’t make a choice. I think the ecosystem in healthcare is the same
way.

You put 25 standards in front of us, we just freeze, because no-one has
emerged as a winner yet, so why waste our time trying to pursue all of them? So
somebody has to pick winners or losers, otherwise we can’t get adoption. So I
think we might have to be a bit more brave in picking a standard two rather
than waiting for the marketplace to do it, because it’ll take 10, 15, 20 years,
it’s an expensive way of getting there.

And then in some cases there’s not enough, and so the other kind of thing
that we do is we have not enough testing I believe before we certify products
or we certify the vendors of the Public Health system, the HER vendors, the
template itself. And so we have this expectation that plug and play, my USB
thing should work, and it’s anything but. You need to download a special
driver, and you need to upgrade your operating system, and then you might be
able to save on that brand with that laptop on that OS. It reminds me of USB 15
years ago, USB 1.0.

And then maybe a couple of analog implementation challenges out there. The
Public Health departments have an analog heritage. That’s their base, they’re
amazingly hard workers, they have MPHs, they have PHDs, they are folks trying
to do real works to say is there an outbreak. That’s the core part of the work.
IT is not supposed to be the core mission, it’s to actually intervene and say I
think somebody has lousy cabbage out there that they’re selling and people are
getting diarrhea and vomiting in our EDs. The information is just a means to an
end.

So in general I think what we find in practice is that the testing
harnesses, the transport, the technical infrastructure, when you slice and dice
it down to each state, each municipality, it’s just not an efficient way. There
are things like economies of scale, so the concept of a BioSense 2.0 which
tries to roll those kinds of investments up in some way and still allow
decision making capabilities is a very effective strategy to sort of address
this kind of thing.

And I think we’ve already talked a little bit, but there’s a little bit of a
strained relationship from our vantage point between Public Health and health
information exchanges. There is this concern that maybe the HIEs that are
intermediating and making it easier for providers to get through a gateway to
Public Health might dis-intermediate the Public Health, and it sort of becomes
the fiefdom problem. And I think when this kind of stuff happens you can’t let
the market forces rule, because that’s when policy is needed so that this kind
of stuff doesn’t happen, because otherwise those politics on the ground can
really be devastating to making progress.

So some things that have not come up. I think a couple of other people have
talked about bi-directionality. It’s not just Public Health reporting. I think
one of the reasons as a provider I might be intrigued in investing my energy to
connect might be to answer the question what do I get back. What are you doing
for me?

And I think to be honest there is a lot of protectionism once the data is
aggregated in getting it back out to the very people that in many cases are the
source of that information. And so that’s a policy issue, and the standards
also tend to be quite weak when it comes to that, and certainly the
requirements around bi-directionality. What are the obligations of the Public
Health department once they achieve something to get something back out to
folks?

Latency frequency, this gets to a trust fabric issue. There’s not, other
than biosurveillance to say within 24 hours, there is not great guidelines to
say how often am I supposed to send immunization. So if I’m a pediatrician out
there worried about VFC, and I really want to deliver awesome care to my
patients, am I going to really rely on my state immunization registry, even
though that is about as good as things get?

Because if I have concerns that I might be a week or a month off because not
everybody is complying with the same thing, it reduces the trust in the very
thing that you’re achieving. And so we think maybe there are some governance
and non-technical issues to wrap around standards implementations that might be
relevant here.

And finally, just a list of maybe recommendations. They’re in the hand-out,
I won’t go through that in the interest of time, very specific potential
thoughts to consider for a committee like this. And the last thing I would say
is it’s very interesting. Everybody else maybe is thinking about data
aggregation not on a use case specific basis, if you look at Twitter or Google
or other fields.

The notion on the very long term, whether Public Health reporting should be
by disease state, by use case, and then compelling that the transactions be
sent that way, versus you should have patient data and you should mine patient
data, because you don’t know what you’re going to need so you don’t have to
send out case report requests after the fact, I think there’s a scalability
problem of trying to come one off standards per use case on a long-term basis,
that’s not how the rest of the non-healthcare industry is thinking about doing
things like Public Health surveillance. Thank you.

DR. SUAREZ: Thank you. WE have, last but not least, Genesis.

MR. HUBER: Good afternoon, I am Rick Huber, I’m CEO of Genesis Systems Inc.,
and on behalf of Genesis and all of its employees I’d like to thank the
committee for the opportunity to present this afternoon. Just as a rough
generalized statistic, genesis supplies approximately 30 percent of the birth
data received by NCHS.

With regards to standards development and Genesis’s role, a huge opportunity
exists to dramatically impact the quality and cost of Public Health data and
healthcare in general. This opportunity must be pursued, as the benefits are
too great to ignore. As we all know, the ability to enter data once and have it
proliferate throughout the healthcare system almost instantaneously is
unbelievably important to Public Health data collection, evaluation, and
potential intervention, as we see it.

The technology does exist today that allows us to be able to provide for
these exchanges. The good news is that these exchanges can occur
instantaneously. The bad news is that these exchanges can occur
instantaneously, and bad data can proliferate the system that much faster. This
is why data exchange standards are so vitally important. Without promulgated
standards, there would be countless variations of data exchanges.

Without documented standards everyone would be developing their own
standards, inconsistency and quality of data could not be assured. Standards
development in any industry is challenging. I’m a CPA, and in the accounting
profession the common answer to the client question of how much is two plus
two: what would you like it to be? We must not allow that mentality to occur
within the vital records and statistics professions.

Vital records and statistics present unique and different challenges. It is
our belief that development of standards within the vital records and
statistics profession must be done by vital records professionals with
significant experience. If standard development is not done correctly data
quality will suffer and impede, if not eliminate, comparability and statistical
inference.

As to Genesis’ activities, we continue to develop interfaces in cooperation
with CDC and other partners to facilitate the electronic exchange using XML and
HL7 protocol, among others that were discussed earlier. Our development is
primarily focused on HL7 Clinical Document Architecture, or CDA. However we
find that not all facilities are utilizing CDA currently. Accordingly, Genesis
is also exploring versions 2.5 and 3.0 messaging models to assist with quicker
adaptation by facilities, leading to the eventual adoption of CDA.

The current flow of information has been accomplished utilizing ante-partum
and post-partum summaries, which are CDA documents, along with the appropriate
birth/fetal death profile. Utilizing these documents appears to be
accommodating the need, providing the appropriate content. However, we continue
to review these documents to assure we are using the best possible data source.

Genesis will continue testing the interfaces with EMR/EHR vendors who are
providing the content for us to assimilate. We will also continue participating
in standard-setting meetings, and to provide input regarding reconciling
interpretations. Wherever and whenever possible we continue to promote the
implementation supporting the standards. We continue our participation in IHE
and HIMSS events in 2014, as we have in 2013 and 2012. Effectively best said,
we continue to eat the elephant one bite at a time.

The challenges we face and have faced in developing standards include
accumulating and developing knowledge of HL7 CDA coding and protocols,
developing the appropriate interfaces to integrate seamlessly with our vital
records module. The major challenge for everyone of course is trying to
determine what data source within the EMR/EHR provides the appropriate element
to facilitate the intended definition. We also find that EMR/EHR data may be
stored as PDF, JPEG, or other non-usable data structure files.

We continue to ponder and discuss what should happen if a data element in an
EMR/EHR is changed after it’s assimilated into our system. Considerations
include obviously what element is changed, why has it changed, what to do if it
happens before registration, what to do if it happens after registration. As we
have in the past, Genesis continues to place significant emphasis on ensuring
confidentiality and privacy of vital record data.

Other challenges include how to handle multiple births. What data, if any,
should be manually entered, as discussed earlier? The development of the
appropriate business associate agreement along with the technical
infrastructure and transmission protocols to interface with our partners.
Refining existing security parameters, and I’m sure many more yet unidentified
issues.

Commenting on client interest and efforts to promote the standards, we feel
that there appears to be growing interest in pursuing the model. Clients are
starting to recognize the benefit to implement both in quality and timeliness
of the data. Genesis continues to promote the use of the model and standards to
healthcare facilities, HIEs, and state vital records clients. We have made
preliminary contact with HIEs, and see this interaction to be of significant
benefit to everyone. Genesis will continue working with public and private
healthcare partners to identify the route towards implementation.

It goes without saying that Genesis will also continue assistance in the
standard setting efforts through attendance at IHE and HIMSS events, continued
attendance at standard setting meetings, suggesting and support early
implementation ideas. We look forward to continuing expansion of the model to
include other Public Health programs. We will, this is one of my personal
passions, continue to promote the use of GIS as an add-on.

We also believe that the volume of information that’s being gathered in
today’s applications will ultimately drive implementation from a financial
perspective. Genesis has systems that gather up to nearly 700 individual data
elements. That alone should be a force which insists in implementation.

And I’d like to again thank you for allowing Genesis to share its
perspective, and we look forward to the continued discussion today. Thank you.

DR. SUAREZ: Thanks everyone for the testimony. So we have some time for
questions before we go to a break.

MS. KLOSS: For Rich, but others also: What are the issues you are finding in
terms of data quality as this data is moved from the electronic health record
to the birth records?

MR. HUBER: I think the issue is defining the data element. The quality of
the data seen, once we receive and where we have determined it to be
appropriate, then we can assimilate it. The issue by and large as we see it is
just defining which element within the EMR/EHR to populate the various data
elements, that’s probably the biggest challenge that we face.

MR. KHETERPAL: I would agree with that. We would describe that as a last
mile connectivity problem. So you’re running, you’ve made a huge investment as
a provider in a Meaningful Use 2014 addition certified software. IT complies
with Public Health reporting requirements, it’s gone through the validator, and
yet I would agree with Rich’s comment, you connect it up and pump some stuff
out, and it’s not exactly what you would expect.

And so each implementation then requires this mapping exercise. And so this
would be an example of we invest in a mode of saying that the standards must be
complied with, so we incur a lot of cost in this process through the entire
ecosystem. We deploy the technology, but it’s not getting us to the finishing
line. So we then start incurring the service cost after implementation on the
part of the providers and sort of the project management.

That’s where having one foot in either side is I think what’s not viable
long-term. You either need to abandon having highly specific standards with
content standards and vocabularies so that you don’t have that and so that you
can invest your energy into mining things and doing things that others outside
healthcare do after your receive the data, or you go all the way and you ensure
that the template for certain key things, that they in fact must produce
something that is reliable.

DR. COHEN: I want to continue the conversation a little bit about vital
statistics. There has been resistance in the vital statistics community to
being drawn into Meaningful Use stage three, and vitals are conspicuously left
out of Meaningful Use stage three. I wanted to get your general sense of the
issues surrounding that.

Also, Larry Nielson for NAPHSIS spoke earlier about some of the issues
related to the lack of interoperability between vital statistics requirements
by state law and regulation, and what we see in electronic health records. For
instance, the name. What’s the best source of information for a person’s name
is a classic one that the vital statistics community deals with. I wanted to
get your sense of the issue around bio-directionality in terms of populating
EHRs from vital statistics as well as populating vital statistics records from
EHRs.

The third issue I’d like you to address has to do with data ownership. The
vital statistics systems are independent of other data collection systems based
on state regulation and a variety of historic precedence. How do you see
potential conflicts between who owns the data being resolved with respect to
EHRs and vital statistics?

MR. HUBER: Clearly those are all relevant and significant issues that exist
within the vital records/vital statistics community. Some of it, I’ll go
backwards in the sense of the ownership of the data, truthfully a lot of that
centers around financial implications. We have a lot of jurisdictions that
basically have to fight tooth and nail for any type of funding. A lot of their
funding goes to the general fund, and therefore they’re fighting to get it
back. That’s one reason clearly that impacts that.

The other reason that I see is the fact that the registrars by and large are
pretty passionate about the use of the information and afraid to have it
generalized and publicized. Clearly we’re aware of PII and the confidentiality
that exists around that. So I see those two issues driving the ownership. As
far as bidirectional, I think that’s also part of the issue was financial. I
think there is a willingness to share the data where they can do it, and I
think it can be done. Although it’s anecdotal on my part my sense is that if
it’s de-identified data that will be good.

Now, when you talk about bidirectional clearly that poses a problem. If we
can get the public vital records community to understand the important and
need, which they do to a degree, I just think there has to be more coaching
that has to occur. And I’d say as far as the Meaningful Use stage three and
being left out, my sense is, and again anecdotally, my sense is a lot of the
registrars don’t fully understand the impact and the benefits of being involved
in Meaningful Use, and I think the fact that they’re not included, they’re not
involved, so it’s meaningless to them.

They don’t really understand the import of being excluded. And I think in
that regard better communications, more public dissemination of the information
of what it is and how it can help them and how everyone can benefit out of that
would be beneficial to the industry. Does that answer?

DR. COHEN: I am interested in other comments too. That was very helpful.

MR. KHETERPAL: Certainly from a provider’s perspective the absence of the
vital statistics as a requirement in Meaningful Use in practice for us means it
just doesn’t come up. The CIOs and the healthcare facilities, they’re not
talking about it as a key mission from our perspective. The second comment,
related to the potential comment that I did not get a chance to -– Vital
statistics, this is part of the whole Department of Health and sort of the
jurisdictional issues sometimes.

Public health is in one area, Medicaid is in one area, foster care is in one
area. So what we often will find within the geographic public sector, there is
a lot of shall I say silos. I won’t say whether they’re good or bad, but that’s
what they are. And vital statistics would definitely always be standing on its
own in our experience. And so that results in I think both a policy and an
oversight in governance and funding and all the things that tend to follow.

MR. STAMM: To echo some of those comments, we’re seeing a lot of healthcare
organizations are just so focused today on large initiatives that are dragging
out a lot of their resources, like Meaningful Use implementations, like ICD-10,
that for things like the birth registry it’s so difficult to convince them that
that’s going to be worth their time when they know and they have either very
visible incentives or very visible penalties for the things that they need to
get done over the next year, two years, five years.

In the birth area as well we’re really stepping in to trying to send data in
an area where all that information has been captured for the sake of submitting
to a variety of states on paper for such a long time, that it really is going
to take some work and some time, and a lot of cooperation to agree on what the
standards are going to look like.

You can between two different states have elements that look similar but not
identical. So the difference between a breech birth versus a non-vertex
presentation, or you can have elements that are just completely different,
things like were they having seizures, was there steroid use during that
period. And it is going to take incentives, and it is going to take cooperation
and time to get to the point where those elements converge.

DR. SUAREZ: Thank you. Two more quick questions.

DR. FRANCIS: I am going to ask my privacy question again, which is: Do you
have any thoughts for recommendations this committee might want to make with
respect to state privacy floors. I have in mind when information comes from
clinical records. So whether you think it might be appropriate to have HIPAA
standards.

And I have in mind for example open vital record statistics of the kind we
were talking about earlier. Also, nobody has brought up the question of
de-identified data, and whether there should be any attention to that whole
nest of problems in this Public Health arena for this committee. But I know
that’s a much larger question, and we’re very much out of time.

MR. HUBER: With respect to security standards, I can tell you that we’re
very aware and very concerned with them. We are receiving from states that we
deal with more requests for encryption of data, and particularly PII. It has
been our recommendation as we discussed earlier, there are certain databases
out there which are marketed as encrypted which may or may not fit the bill. We
tend to lean more to actually encrypting the data prior to transmission, and
then encrypt the data at rest. That gives you a slightly better security
protocol while the data is at rest.

MR. KHETERPAL: I think one of our recommendations of the eight was some sort
of a position paper helping stakeholders dealing with Public Health balance
their responsibilities towards maintaining privacy with the very nature of this
directionality of actually facilitating the exchange and dissemination of the
information. Our observation would be that the pendulum remains swung and stuck
very much to data becoming hostage in Public Health.

So on the one hand you get it out of one site. So unlocking that data and
getting it back out of a Public Health department to various sources, the
presumptive question is prove that you need it. So it seems to be very much in
that light. Obviously this stuff, there are some severe penalties under HIPAA
and state statutes for violating those things, this is all PHI so all of those
security standards go without saying. But it’s more of a posture and sort of
framing of the role of once you have the data, how can you maximize its
utilization while maintaining public trust.

MR. POOL: I would augment those pieces. I agree, first off, security, yes.
That’s a given piece here. I recognize that we have enjoyed and provided to
Public Health special context because of its unique position. It has been
having helped folks address this to some degree for some time. Just because we
have the special circumstance doesn’t mean we should abuse it or be lazy about
it, and in fact I think Public Health should be more vigorous about maintaining
privacy. Even though it gets to see stuff, it should be that more vigorous.
There are complex pieces of pseudo-anonomization versus truly anonomized versus
aggregate data reporting, all of those pieces come in. But Public Health will
lose its special status with a couple of really ugly stories on the front page
of the Washington Post. We need to be careful to preserve that wonderful gift
we’ve been given.

MS. WILLIAMSON: I think Vic, on your slide you mentioned not enough testing,
and Ken you talked about the need for needing two to tango. So I know that it
has been a huge challenge for us in a lot of the vital record standards work of
getting those partners together, especially when we go to vendors we hear is it
in the Meaningful Use requirements, we say not yet, they say talk to you later.

So this is a struggle. We have really valued the partnership I have to say
from EPIC and OZ in working with us to test. So what I’m looking for is if we
had to say something, if there is a recommendation, what would drive the
marketplace? Is it that it has to be a Meaningful Use? Is there another way to
get more vendors at the table so that we can support the pilot testing that is
definitely needed of these data?

MR. POOL: I kind of skirted it a bit, mostly because I ran out of time. I
think that if you want to get more vendors as it relates to certification an
testing, the day that I see federal and state procurements coming out saying we
want to procure part Q-17 that has been passed and has a seal of approval from
Bozo Limited, it has been tested and certified, that’s what I want to buy. The
day I see a procurement is the day you will begin to see vendors come in to
testing, because then it makes a difference in your market. Until our buyer,
Public Health and its related agencies, begins to expect that, demand that, and
purchase that, it’s not going to make any difference.

MR. STAMM: I think what you heard from a variety of folks here is very
similar, which is that really the first thing that happens is there is a need
between providers and states, and there is an identified use case for
identified pieces of data. That then moves through the creation of standards
for how you can move that data from location A to location B and actually get
the things that are necessary.

But even at that point those extra steps to do that implementation, to make
sure that you’re capturing it, really comes down to when incentives start
popping up so that it can be prioritized by those organizations against all of
the other things that they have to spend money on today.

DR. COHEN: So you are proposing in essence a certification for vital
statistics, for instance systems? And if you are proposing that kind of
ubiquitous certification, how do you see that happening, who would do it, and
how would it role out?

MR. POOL: First off I am going to make clear what I’m suggesting is that if
you want to buy a part that is a one inch screw of diameter one eighth inch,
that is what we are looking for. Here is the standard, here’s the parts, here’s
the things that it takes to do it. Do I think those can be tested, and truly
tested where they are truly plug and play, they can in fact walk, sit down
– Yes, that’s very possible.

Is there a testing agency out there that does that today? No. And so I think
we have some increased expectations we need to place on our testing facilities,
which if any of them you talk to, whether it be from NIST and Connectathon and
its variations, HL7 and its variations, every one of the organizations is
looking at how to up that piece. So yes, I do think it is possible. I don’t
think we have the testing in place today, but I think it is —

DR. COHEN: Do you think vendors would respond if they needed to be certified
for developing EBRs and EDRs?

MR. HUBER: I think the primary driver is inclusion in Meaningful Use. There
are testing regimens already in place, we’re participating in it. I think
having the monitor resource of Meaningful Use behind it certainly doesn’t hurt.
I think that will help drive the industry towards that end. And by inclusion in
Meaningful Use you’ll get the financial benefits obviously from the
organizations, as well as the overarching standards developments. So I think
you already have the mechanism, just include us.

MR. KHETERPAL: If I may split hairs a little bit between those elements that
are not yet in Meaningful Use versus those that have been there since the new
one, ELR, immunization, biosurveillance. My comment related to more testing was
there already has been some time, and our technology stack is Meaningful Use
certified for those three use cases. And yet it’s not quite there on plug and
play.

So the question is more to tease that out to say providers often feel like
it’s a head fake of some sort, what did I actually get, because I thought I
bought that. And there are some things that say yes, you need service hours to
do this, but we often find in the EHR community the relationship with a vendor
and the expectations that have been set, still under Meaningful Use stage two
as to what do I actually possess, own, and just implemented, versus when I hit
the button to say go connect.

Does it, and does it have the right payload? And the answer is not quite
yet. So just tightening that last gap. And that’s completely independent of
aspects of Public Health that are not even in Meaningful Use. And I would say
we ought to perfect the first one before we add more things on there.

DR. SUAREZ: Thank you so much again for the testimony. We’re going to take a
ten minute break. I know we have one more panel to go and at least one more
person is on the phone, and we very much appreciate the flexibility with the
schedule, we’re running about half an hour late on our start. We’ll be back in
ten minutes and we’ll start with the last panel, thanks.

(Break)

Agenda Item: Panel 5: State Initiatives and Readiness
for Standards Implementations

DR. SUAREZ: So we want to welcome our last panel of the day, the panel
involving the state initiatives and representatives from state and local health
departments. We thank you again for your patience with us here today. And it
couldn’t have been better to finish really and wrap up our hearing with the
perspectives of states and local Public Health agencies. So thank you so much
for bearing with us, and we’ll turn it to John from Tennessee.

MR. ROBERTS: I discovered today that I agree with a lot of people more than
I had in the past. For instance, I agree with anyone who makes a reference to
Public Health that is at the table, I agree with that. Unless they say whether
or not Public Health is at the table. John Linsk, his remarks on Public Health
case reporting and the core message, Denise Love, the reliance on volunteers
not sustainable, I sort of agree with that, but I think I really agree with
voluntold volunteers from state and local Public Health, I agree with that.

Phil Brant, he nailed it. If the healthcare system fails Public Healthcare
has to pick it up, you got that right. Roland, value set maintenance and the
RCMT. Wesley, the stress on ELR. Janet Hamilton’s remarks on almost anything,
but certainly categorical funding. And John Stamm of epic, who by the way, I
haven’t met him at the HL7 meetings, but EPIC reliably sends someone to find
out what we’re up to and stay ahead of us, and that’s kind of unique among
ELRs. And as usual, I agree with Ken Pool, I think.

Here are my main points. It’s real simple. At the table, you can’t predict
what Public Health is going to do. There are 3069 counties and 57 states or
state equivalents and all sorts of things in between, and even with Public
Health at the table you can’t predict what’s going to happen on a given day.
But if Public Health isn’t at the table you can predict a bad outcome. We
should all commit to using standards, and I think Public Health does. It’s
using standards that are inappropriate or un-useful is not going to happen. And
we have this little trick called regulation. We write the rules in our states,
and if something isn’t useful we write the rules to say use something else.

Thank you, CDC and NCVHS for keeping us at the table, many voices of Public
Health. Thank you, Michelle. And a challenge to do even more of that. We need
more Public Health people who work 80 percent or so of their time in Public
Health, and maybe have 20 percent of their time to go to HL7 meetings or other
meetings.

Someone who spends 100 percent of their time going to standards meetings
doesn’t really fill that bill. And I don’t know how we find them, but we need
to find them. Tennessee is lucky, I have a commissioner that allows a few of us
to spend a little bit of our time on this effort. He knows that nothing
directly comes to Tennessee for it, but he allows it anyway.

We’re the bad guys. We have so many agencies, yet we do have a fair amount
of authority and responsibility. Few resources, and so many ways to ignore
standards, and we need them so badly. My line on standards, and those of you
who have heard it, I apologize for it, but there are hospitals in Tennessee
that talk to 1000 doctors. There are hospitals in Tennessee that talk to five
labs. There are doctors in Tennessee that talk to three or four hospitals. I
talk to 150 hospitals regularly, I talk to 400 clinical labs. Potentially I
talk to 15,000 doctors, but thank goodness not all at one time. And that’s just
the ones in Tennessee, and I border eight states, and I have probably that many
or more in the neighborhoods around Tennessee. So who is interested in
standards in Tennessee? I am. I have a lot of potential trading partners, I
need standards.

Silos, we have got to get the silos knocked out. CDC has come a long way in
my 20 years in Public Health. We’ve gotten rid of the micro disease
classification silos, TB finally went away, it’s part of NBS in many states,
and it’s been merged in a lot of other states. Watch out STD, you’re next. We
will get rid of STD and its progeny and move it in with communicable disease
surveillance.

At the end of the day we’ll get the CDC’s categorical funding and silos
wiped out, and then we’ll discover we have this thing called WIC and the
nutrition programs, and they’re still a silo. We’ll merge them in. And then at
the end of the day we’ll discover the last silo left is Public Health and the
healthcare space, and we can go to work on the big one then. We have to get rid
of these silos that treat our data and our patients as if they belonged to us
and didn’t have anything to do with any other part of the healthcare system.

Here are the bright spots. This has been a really cool ride. This is the
best time in history to be in informatics, and particularly in Public Health
informatics. More is happening now and will happen over the next few years.
It’s really remarkable, we’ve come a long way in a short time.

Here are the main points. Keep Public Health at the table, thank you,
encourage Public Health’s participation in standards activities. And there are
lots of ways of encouraging. I’ll share one of them with you that works. Every
now and then for some strange reason my management gets a comment in passing,
thank you by the way for having something to do with ELR, some standards.

That goes as much as almost anything in encouraging participation on the
part of state and local Public Health, just knowing that it’s useful and has
been useful, it really is. That’s the challenge, to do more of it.

And I didn’t mention earlier but pilot demonstration projects, we need to
make sure that those things actually will survive the grant funding, we need to
get a good business model, get enough parties together to really make something
happen that will live and work. And if it can’t survive the grant funding,
either change the business model so it will or don’t do it. It’s that simple.

Thank you very much for letting me attend this, this has been a real
experience. I really appreciate what you do, you’re really cool people. Thank
you.

DR. SUAREZ: We have next Marty.

MR. LAVENTURE: My name is Marty LaVenture, and I’m Director of the office of
Health Information Technology and e-Health here at the Minnesota Department of
Health. We want to thank the committee cochairs and members for the opportunity
to speak today regarding standards.

I’m sorry I’m not available to attend in person, I just returned from
Guatemala early this morning. But I’m pleased and thank the committee to have
the opportunity to express a perspective from the needs and opportunities of a
state Public Health department. I’ll be focusing my comments on what I see as
recommendations relating to key policy guidance and strategic action that the
subcommittee should consider in their recommendations related to standards.

As you all know and understand, Public Health informatics is a team
activity. Sometimes it seems like a team contact sport, but I want to
acknowledge the work of many of the MDH colleagues that are shown in this slide
who provide Public Health informatics leadership in our department. And from
the commissioner level all the way down to frontline staff, they all play an
important role in both informatics as they relate to standards, and we’ll talk
a little more about that later on. And I especially want to acknowledge Dr.
Priya Rajmani, our standards interoperability director who played a key role in
preparation of this presentation.

Development of standards are critical, but implementation and the use of
standards is really vital to aspects of establishing an information
architecture at a state health department. And this implies I think the use of
standards needs for the improvement of Public Health informatics systems and
infrastructure.

So we sometimes speak of it as to improve the use of standards. To address
standards we need to speak about interoperability. If we’re going to also
improve the use of standards and address interoperability we need to speak
about information systems and standards, and to address information systems we
need to speak of our workforce. And ultimately to address workforce we need to
speak of an information savvy organization. So thus we look at standards in the
broad sense with a systems view, and a Public Health informatics perspective.

We think the S&I framework from the Office of National Coordinator
provides a good example of this, where individual standards are really a vital
element. But what’s critical when we look at it from a state health department
perspective is to consider the context of that interoperability, where a
portfolio or information stack if you will of specifications for Public Health
to achieve specific Public Health interoperability is critical.

We need our standards and guidance we believe to create and maintain our
Public Health portfolio to achieve Public Health specific bidirectional
transaction. Thus the notion of standardization extends far beyond the single
vocabulary standard, and are not isolated but really connected in a purpose of
interoperability. For example, LOINC and SNOMED are essential to have in place
for electronic lab reporting, but guidance of the entire interoperability stack
is essential to achieve the transmission of that information.

In Minnesota we extend the notion from the S&I framework model further
by adopting an HL7 version that looks at not only the technical and semantic
layers that are reflected in the S&I framework, but add the concept of
process interoperability, or if you will integrating the data into our Public
Health information systems and information workflow so the information can
actually be used and put to work as necessary added value and return. So
standards guidance and policy should address the needs for process
interoperability as well. For example, the notion of standards reflect the need
to integrate this data into the Public Health workflow for immunization
registries or disease surveillance systems as examples.

Like we’ve heard I think earlier today, in Minnesota immunization provides a
good example of some of this progress, but clearly gaps still exist.
Immunization registry in Minnesota gives us room for some optimism, and is an
instructive example. One of our I think more advanced and mature activities
regarding standards is reflected in the graph on the left, where from 2004
almost 90 percent of our transaction data was entered via humans directly into
a system. By 2012 that figure is under 10 percent, and batch entry and
real-time entry has increase dramatically.

I think that the further good news is that this is occurring across the
state from small to large organizations and systems, not just some big health
systems. So we have seen, and I think one of the recommended actions that are
important lessons learned here is that focused collaborative efforts and
resources directed at a particular type of use case can make a big difference.

A few of the lessons that we’ve learned along the way so far are related to
a number of the topics that I believe you’ve already heard, and I’ll try to go
through these fairly quickly to emphasize them, with a couple of key elements.
And that is that standards development and policy and guidance we believe needs
to address some critical issues around our information standards on
specifically dealing with structural differences, systems that understand the
unique structure of the specific terminologies.

Our scope differences, where we have content overlap. Specifically an
example would be nursing terminologies, where we have Omaha versus SNOMED, or
one of the other 17 nursing terminologies that are currently dealing with of
what to use.

Coordination issues, the whole schedule release, the conflicting licensing
agreements and managing those. The terminology and value set creation, use and
alignment. Net harmonization, we’ve heard a number of people identify that,
that’s certainly consistent in our experience as well. Guidance for both the
people issues on policy and the technology, the support for infrastructure for
centralized access to disparate content, as well as the advanced education and
training.

We also need to look at standards in the context of our information systems.
Otherwise we are at risk of really creating the equivalent of information
bridges to nowhere, where an electronic message actually comes to our door but
we cannot efficiently manage that message in a timely, efficient manner. It may
sit in a queue or require a disproportionate level of IT resources to integrate
it into our information systems, and the standards guidance and policies
associated with that are essential.

And we believe that effective processes should be part of the
recommendations dealing with the whole notion of a standard approach for our
collaboration and our business processes, including sharing those. An example
of course is the Public Health informatics institute business process
laboratory, and others. Coordinate the business information architectures, and
the whole notion of not technical architecture, which we have enough of, but we
do not have enough of information architecture models in which to really
integrate our information system, and collaborative approach to looking at
those.

Support for the internal and external sharing capabilities, guidance and
gateways and hooks, for example, as we receive that data. Fostering
collaborative partnerships with application vendors. We cannot go to the next
level of sophistication without greater involvement by our vendors in the
community as partners in meeting our information requirements. And then a much
greater effort on policies and recommendations reflated to leveraging open
source and cloud based services, these are vital to our next steps.

This slide represents an example of the Minnesota e-health standards process
in Minnesota. From a state perspective, we’ve been talking about a lot of
isolated types of activities. But when we bring it together, what needs to be
done at a state health department, there are a lot of activities that are
necessary. For example, under the e-health initiative in Minnesota the
commissioner of health can actually recommend and if needed require the use of
e-health related standards.

This model shows the process that we use to monitor the standards in the
left column, the input for example, related activity, process this information,
and then provide good guidance recommendations and support for the use and
implementation of those standards. The input on the left shows the complexity
that a state faces. We monitor nearly 70 different organizations and activities
related to national standards, standards development organizations, and local
and regional activity that’s going on. That requires I think a great deal of
effort.

And in the middle that effort reflects filtering, analysis, evaluation, and
validation of what is either recommended or emerging activities, and determine
some readiness for standards use within the community and the state. The output
involves significant communications related to those efforts and the standards
in the guidance and policy as they are implemented in the process.

This slide reflects in Minnesota the sort of four key factors we believe are
important as we look at information systems to improve practice, which are
vital to us implementing standards. We focused a lot on technology, that
remains an IT focus. But we are seeing that the content, the processes, and the
workforce, we consider the informatics focus is a critical area that we need to
devote attention if we’re going to have more successful use and implementation
of better integrated information systems across our applications. So use of
standards certainly will increase at the speed at which we can increase the
informatics and standards related competencies of the workforce, and the
workforce is a critical element for us in the use of these standards.

We also believe that a critical step to improve the adoption and use of
standards is the training of the workforce in understanding and the use of the
standards. And this can be done in a number of ways, and I know that you’ve
heard about some of this as well. But we have found that the increase in
workforce informatics companies need to address at least four key roles. And we
like this model developed by the Public Health Informatics Institute in
conjunction with CDC. It has a framework of incorporating knowledge and skills,
including related to standards from a foreign executive in a leadership role,
what do they need to know? For the professional role, as in the lower left part
of this slide. And their more advanced understanding of standards as they fit
into the activity for their particular project. The manager role, who has to
make key decisions at a program level about use of standards.

And then the clinician, clinical informaticians and nurse or physician
example, that plays also a vital role in standards. Increasing the competency
of this workforce we believe is essential to anything from expanding the CDC
informatics fellowships to all states that support Public Health informatics
institutes type of training program, academic and technical project, any
projects of which you’ve heard a number of today are all going to be vital to
increasing the level of competencies in the workforce.

We also see that training individual Public Health professionals is
absolutely necessary, but this is not sufficient to make implementing the use
of standards sustainable. We must grow more informatics savvy Public Health
departments.

This includes at least four items: Strong leadership and related
informatics, including teams related to standards. Effective governance
structure for setting priorities and making decisions related to standards. To
assign areas of responsibility, the roles that have the authority and
responsibility to make decisions. For example, epidemiology teams as they
relate to the adoption of standards. Establish and sustain informatics
capability and capacity in monitoring those. A crucial role is the adoption and
use of standards requires a really growing informatics savvy organization, and
leverages the use of these standards.

In closing, as this slide implies Public Health informatics and standards
are vital to our ultimate goal in the upper right, and that is to support
healthier communities. We base that in new standards and use the science of
informatics to support the use of standards along this continuum, from data
collection to creating information in the form of graphs, tables, charts, maps,
to generating new knowledge created from bringing a variety of data together,
the wisdom that we may learn in community context by adding value.

And also ultimately leading to our goal of not only healthier communities
but fewer health disparities and greater health equity. We believe we have a
great opportunity with Public Health right now, and we encourage the committee
to consider a broad notion of standards related to Public Health along this
entire continuum and in the context of Public Health informatics. Thank you
very much.

DR. SUAREZ: Thank you. We are going to go next to Joe.

MR. GIBSON: I am Joe Gibson, the Director of epidemiology at Marion County
Public Health Department. We serve about 900,000 people in our county. We serve
Indianapolis, we have a pretty big health department as local health
departments go. I also chair NACCHO’s Informatics Workgroup and the BioSense
Governance Group. People have talked a little bit about BioSense. And I do a
lot of work with the Regenstrief Institute and the Indiana Health Information
Exchange. But I’ll be mostly talking about the perspective on standards I have
as a local health department epidemiologist.

John Kennedy said all free men are citizens of Berlin, and I kind of think
that all Public Health epidemiologists are citizens of the Balkans. We develop
our datasets, our data systems, and we are allowed to develop them in our own
little way and make our little idiosyncratic data systems. In Public Health
across the country the local health departments do the same sort of work, and
we have the same sort of data systems.

So we have birth certificate data systems, we have the STD, the HIV, we do
population health surveys. And all these data systems, most of these data
systems could at least share common elements. We could have the same variable
names, the same variable types, the same coding systems, but we don’t. And
nobody encourages us to do that.

Recently we revised our birth certificate system, and I went online, I
looked for birth certificate systems that I could use so that maybe I could use
it and the state could use it, so when I wrote a program they could use my
program to analyze their data, vice versa, we could share it with other health
departments. I’d like to migrate towards those things, but they aren’t out
there. At least I couldn’t find them, so they’re not out there obviously enough
that somebody who is interested can find them.

If I could share tools like that, or if I could borrow other people’s
analyses, I could be twice as productive. My group could be twice as
productive, we could get twice as much work done in Public Health. We could
take the tools that are high quality around the country and use them all over
the place. But we don’t have those standards.

And it’s not that big a deal to say ender, we’re going to call the field
gender, and it’s going to be coded M and F, or whatever other enders people
want to put in there. But it’s not that great a challenge, but we just haven’t
gotten it done. And I’m confused about why we haven’t.

The recent IOM report about data and Public Health accountability talked
about the need for a core set of measures for Public Health. That would be
hugely valuable to me. We’re doing a community health assessment right now, and
it’s not too hard for me to compare Marion County and Indianapolis, essentially
a medium to large city, to the United States overall. But those aren’t
appropriate comparisons. I need to compare Marion County to other cities.

The question is always, what can we do better. And to know what we can do
better we need to compare ourselves to other people who are similar. And the
state of Indiana is mostly rural, Indianapolis does not. The United States
overall does not have the demographic profile and the problems of an urban
area. I want to be able to compare it to other urban areas. If we had a core
set of measures that was used around the country to describe Public Health, the
general status of the population, then I could do comparisons. Right now I
can’t.

And just from my day to day work, we get hundreds of data requests per year
that we respond to, and we just keep reinventing different measures to respond
to those requests. Fairly customized, and yet it’s good for that specific
response, but it’s more work than if we had some standard measures that we
applied. Standard measures would be imperfect, they wouldn’t fit everybody’s
needs, but they would decrease the amount of work and they’d probably create a
lot more value than we get with the customized work that we do.

Which leads me a little bit to, why don’t we have standards? There are
standards floating around, why aren’t people implementing them? It’s always
easier not to implement a standard than to implement one. Within our health
department we try to standardize a lot of the work that we do, and the
challenges we have is that for that specific task it’s easier for my programmer
to write a program to solve that task and not take the extra time to make that
conform to some standard.

He can more quickly fulfill that immediate need can he can fulfill that need
and comply to a standard. In the long term he may know that there is more value
in complying to that standard, and me as the boss and the other people in my
group may be telling them you have to go to these standards and they may get
dinged for not doing it, but they can get more work done in a small way by not
complying to standards, by not doing things in standard ways. In the long run
we spend more time, but in the short time it’s more quickly rewarding or the
specific tasks. So it’s not that people don’t understand the value of
standards. There is definitely a push against standards. There are some
immediate needs that are easier to fulfill if you don’t use standards.

So, nothing is perfect but nothing is worse. There is going to be
disagreement about what the standards can be, and we need to get past trying to
make really good standards and just get something out there, because not having
anything out there, again we’re living in the Balkans, we just have morass of
this and that, and on the local level there is a lot of value missed because
there aren’t tools we can share and measures we can pull up to do the work we
need to get done.

And the other big value of a core data set is that you can rank against
other people. The question is always why aren’t we better than somebody else,
or what are we doing poorly compared to other people, and a core set of Public
Health measures would allow us to rank ourselves and see where our greatest
opportunities for improvement are.

Data sharing policies. As important as technical standards are, I think
policy standards are much more important in terms of getting data flow to
happen within Public Health, and from Public Health to other areas. People have
already mentioned confidentiality rules vary by disease. We have some kind of
whacky rules around HIV in terms of what can be shared and what can’t be
shared. WIC has always sort of been a lock box where we can’t even get to the
WIC data. We have over 100 people providing WIC services, but we actually can’t
get to the operational data for that system because it’s under USDA rules that
goes off to a vendor, and then to a state, and the state has restrictions on
what they can share with us, so we can’t even get the operational information
we need to be more effective in running the WIC program.

With all these different rules for different diseases it gets very
confusing, and so the tenancy of people is to go to the most conservative
denominator in terms of sharing data. They don’t want to break any rules, so
they look at the most restrictive rules that they know and say if I comply with
this I know I’m not violating anything. So it really decreases the amount of
data sharing that can go on.

So what we need is some national respected organizations endorsing some data
sharing policies, that this is a reasonable standard for sharing data, this is
a reasonable process around sharing data, and NCHS would be a great
organization to endorse something along those lines. If you’re in a local
health department, if you’re in a state health department, you don’t have the
expertise or the time to develop that refined understanding of what’s okay or
not. But if somebody you respect says that this is reasonable, then you’re in
good shape.

And about data sharing, we have to protect privacy, and you won’t find many
people more concerned about that than I am, and more careful about that. But if
by not sharing data I’m hurting the health of somebody, if by not sharing data
I’m damaging the health of the population I serve, then that’s a problem too.

And we have to think about the balance between protecting health and
protecting privacy. And there are going to be tradeoffs there, it’s going to be
a grey area. There are lots of rules around sharing data for research in our
legal system. There are not a lot of rules around sharing data for Public
Health. And so often when I ask for data from the state from people I get put
into the research bin, because that’s well defined.

How do I share data for research, what’s allowed, you have to be very
specific about how you’re going to use the information, and it doesn’t fit
Public Health use very well. Research also has good training about protecting
the confidentiality of data. There is human subjects training that researchers
have to go through. The state is reluctant to share data with me sometimes
because if they share it with me they have to share it with all the other 90
counties, and not all the 90 counties have 600 people in the health department,
10 people in the epidemiology group. They’re health departments with four or
five people, and they might not be able to protect that information. And I
wonder if we can’t do some sort of training like human subjects training to
protect confidentiality of data, do some sort of certification that here is an
organization that has been certified as being competent in protecting
information.

In general, about participation of Public Health in data standards, I have
two really bright able people who were trying to participate in the Public
Health reporting initiative, which is an important initiative and potentially
very valuable to Public Health, like the S&I initiative that Jim Daniels
runs as well, very important initiatives, we’d really like to participate in
those. But they get on the calls, and they just get lost.

They’re very smart, able people, but one business analyst who knows IT very
well got lost after a couple calls, and she has a day job, she just doesn’t
have time to spend in that development, because it’s all an extra thing that
she’s going to do. I don’t think you’re going to get the participation we need
in developing standards from Public Health practitioners unless there’s some
offset, unless there’s money and some responsibility for that person to be on
the phone. Especially standards development. It’s very intensive, it takes a
lot of time, you have to understand things, you have to be in the conversation.
You can’t just step in and step out. It’s not going to happen unless people
have the time, and people don’t have the time for free. We have lots of other
stuff that we have to do. It’s too hard. But if we sign a grant that says we’re
going to spend ten percent of our time on this thing, we’ll make the time to do
it.

In readiness for standards, about one percent of local health departments
serve about 25 percent of the US population. About five percent save about 50
percent of the US population. The largest health departments probably have the
capacity to implement standards fairly quickly. Medium health departments our
size, we have the talent to do it, we don’t necessarily have enough critical
mass to have the wiggle room to get the standards implemented. The smaller
health departments, a lot of them don’t even have data systems.

You could probably get a lot done by working with 100 health departments
around the country, state and local, in terms of implementing standards. It’s
not that big a number. But again, there are the competing priorities. The
talent is there, a lot of us have good data systems. We would love more talent,
we would love better systems, but there is a baseline.

And you hear some of the people talking here, there is a lot of talent in
Public Health, but we don’t have the time. And we aren’t going to have the time
unless there is some offset for the other things we have to get done. And
again, it comes down to funding people’s involvement in some of these efforts.

We can do something about Balkanization by having some incentives for us to
collaborate. A lot of the funding for Public Health comes from the federal
government, and it usually comes with no strings attached in terms of working
with someone else doing the same thing. But it could.

You could tie strings to that, and force collaboration. You could give more
money if there is collaboration. You could not give enough money for some
things unless they’re working together with other people. There are definitely
ways that funding can be used to promote the collaboration that we need that
leads to the standards.

We talked about core measures. It would be great to have some sort of
endorsement of some model data sharing policies. Not that people are going to
immediately implement those, but if it’s out there we can try to migrate
towards some standard for sharing information. It’s going to take years to get
there but we need the thing to migrate toward before we can migrate toward
anything.

And again, we do have a lot of talent and tools. We have enough talent and
tools to implement standards and to participate in standards development, but
we don’t have the time, we don’t have the resources, and it’s going to take
somebody putting money up front for us to get the resources we need to keep our
day to day jobs going and participate.

DR. SUAREZ: Thank you. We have Scott.

MR. AFZAL: I realize we are after five and the folks over here don’t look
too amused. I am going to move somewhat quickly and maybe skip over a few
things, feel free to ask questions at the end. I am not a Public Health guy or
expert, I’m the Program Director of the State of Maryland’s Health Information
Exchange. SO hopefully what I can provide is at least a lens through which we
look at these issues, the things we’re focused on, and how we think about
Public Health, how we think about population health, and maybe the difference
between those from our perspective.

We are the single state designated entity for health information exchange in
Maryland. I started working on CRISP about six years ago, we’ll talk a little
bit about progress in a moment. If you look at the mission up here, a little
bit lofty. And there is a temptation to be idealistic with a mission like that.
But we’ve really tried to focus on being painfully realistic and pragmatic
about what we pursue.

And that kind of leads to this next slide here around how we think about
what to pursue. It has to benefit the patient as a starting point. Indirect or
directly, there has to be a financial benefit, and it has to be technically
feasible. That financial benefit hopefully has direct financial benefit to the
participant in the service. Those things have to be true to make a service
work. Too much of what is happening in health information exchange around the
country is idealistic without considering each of these three points.

A few quick stats here, and these will come into context as we go through
the next few slides. In Maryland there are 46 acute care hospitals, some
non-acute. All 46 are at least connected for admission/discharge/transfer
messages, and I’ll talk to that in a minute.

That’s a powerful concept, just to have the ADTs connected. Most health
information exchanges use ADT to build a cross-entity understanding of a
patient, to know that the patient who went to hospital A and the patient who
went to hospital B are that same patient, so you can present clinical content
in one view. There’s a lot more you can do with ADTs that I’ll talk about.

When we think about clinical data feeds coming from hospitals beyond ADTs,
that’s lab results, radiology reports, clinical documentation typically is
transcribed documents, discharge summaries. There’s a world of about 134
possibly hospital clinical feeds in hospital, three per hospital. We have about
103 of those live, we expect to have the rest in about six months. There are
six million unique identities in CRISP today. So again, when you’re thinking
about two member identities from two hospitals, the unique identity is one,
there are about six million of those.

We have a query service that I’ll talk through in a moment. That’s kind of
how most folks think about health information exchange, the ability to access
information from a wide array of participants when you’re treating a patient.

It’s happening about 500 times a day in Maryland. That’s not a lot, there
are workflow barriers there. But I’ll correlate this back to the Public Health
side. And then the other number I’ll point out here is the 90,000
notifications. We trigger out real-time alerts to PCPs, to Public Health when
folks are arriving at hospitals, whether it’s an ER registration admit,
admission, or discharge.

One other relevant bullet here is we have started working with the eight
acute care hospitals in the district to onboard them directly into the
infrastructure. There will be some cool cross-border opportunities there.

I’m going to stick on this slide for a minute. It’s not meant to be overly
technical, but it’s meant to convey four key points. One: That yellow box is
that master patient index. That’s the most critical thing we do as a health
information exchange, and I’ll have a slide on that in a minute. The other
three areas really outline out broad-based services.

Up top there, that’s our query infrastructure. In the middle, our
notification infrastructure. And in the bottom a reporting infrastructure
that’s homegrown, relatively basic out of the gates, but really is focused on
population health measurement, intervention, management, and I’ll show you some
output pictorially in a moment.

Ironically, some of our best progress in Public Health has not been on those
things in Meaningful Use. It’s not around immunizations or syndromic
surveillance or electronic lab reporting.

We operate the state’s prescription drug monitoring program, so every
schedule two through five drug dispensed from Maryland pharmacies is available
through the query infrastructure. That’s a significant Public Health
opportunity, and that’s live today.

When we looked at our query solution it was targeted at clinicians treating
patients, but we expanded the appropriate use to include cancer registrars at
hospitals. We didn’t realize, going back a year ago, that there is an archaic
process that cancer registrars use to get information to report to the American
College of Surgeons, it was wild.

And just by allowing them access to this solution, and there are some tight
controls, it has really changed a lot of the way they do business at hospitals
and how they report to the American College of Surgeons. Just enabling local
health department access, it’s beyond again those three Meaningful Use
concepts, it’s the ability for a treating provider that happens to be out of a
local health department to use the same tools that we provide any other
provider.

And then EHDI. When you think about that ADT again, and I want to emphasize
the importance of basic ADT messages, because they’re so easy to work with
hospitals to get. The first thing that most HIEs start with that again they
usually limit what they use it for, you can look at the data of birth and if
it’s under 72 hours make some assumptions around it being a newborn, and we can
route to programs like EHDI. And we’re doing that in production as well. We’re
also just starting down the path of integrating with ImmuNet so we can pull
immunizations into this single view for providers.

This is the patient identity management concept. The only thing again here
that I want to emphasize is that doing this well is the predicate for doing
almost anything else we do, especially in the population spaces. We start to
think about trying to string together super-utilization numbers or
understanding where a particular cohort of patients is having a cluster of
visits. This is entirely necessary, and this incorporates street level address.
The ADTs are carrying street level address, which I’ll come back to in a
minute. When you look at most data sets that are available they’re carrying zip
code. So you can do a lot more with the ADT messages.

This is a quick overview of kind of the active projects. I forgot to put the
cancer registrar I run up here. But these are the things we’re working on with
our Department of Health and Mental Hygiene in Maryland.

You see the three Meaningful Use oriented measures, we’re working on those
mostly with hospitals. That’s just starting to come into play in terms of us
seeing increased interest from the hospital community in getting these live. As
you’ll see in a moment we have almost no progress in the ambulatory space,
which is particularly challenging as we look at immunizations.

So, our broad role. There is a lot of text here, so I’m only going to hit a
few points. We have positioned ourselves, particularly as it relates to the
three Meaningful Use criteria, as a conduit. Most of what we do as a health
information exchange in Maryland, and I would suggest that most HIEs around the
country, is not interoperability.

And we’re not really trying to solve interoperability. We’re doing
connectivity. We’re doing so that when you look back at that previous slide
around message interoperability and semantic interoperability, we’re taking
messages, transforming them, and then sending them where they need to go, and
that person’s standard. Interoperability would be if the sending system and the
receiving system already communicated, and that would reduce what we do.

And if we’re just applying an ever increasing patchwork of maps that’s going
to become unsustainable for an organization like ours. So we are positioning to
be that conduit with hospitals to say hey, we’re already getting ADTs carrying
the chief complaint, right? So put it in the right spec on the outbound and we
can route that to the Department of Health, and we’re working on that.

The other example would be we’re getting all of these lab results flowing
out of hospitals today from some of the independent labs as well. If you can
put an indicator about it being a reportable value we don’t have to read the
result and see if it’s reportable. If you just put a binary yes or no we can
route that. That’s what we thought we could do.

When we actually looked at the data and started working with it and trying
to route it we realized our vendor was taking microbiology results, converting
them from discrete to text because it displayed better in the viewer, and you
can’t go back from text back to discrete, which is what the NED system in
Maryland needed.

So you start to run into some of those kind of pragmatic issues, and you
address them when they come up. But there are opportunities I think as a health
information exchange to leverage relationships more than us solving a
connectivity problem. We’re not creating necessarily new efficiencies, we’ve
established relationships that are relevant to making progress with the
provider community and the department of health.

Our ability I think to act in the Public Health or population health role is
predicated on having other valuable services that we’re offering. That’s not
the primary thing we’re pursuing. I think there is great risk in health
information organizations focusing on Public Health reporting as the thing that
they offer, because it’s missing part of that financial circle. It might be
technically viable, it might benefit patients, but how do you actually sustain
that? How do you charge enough to keep the infrastructure that supports that
operation, the people that supports that operation going?

The last bullet here on the bottom here, this has been echoed I think I
heard from the three presentations I heard, that people resources are a
challenge. Getting expertise in the space is really difficult. The ambulatory
space, as I mentioned, has been very difficult from a connectivity standpoint.
When we think about immunizations the hospital world is very addressable.

From a population health management perspective, again when you think about
the context of all those ADTs coming in creating a unique ID, you can start to
create really interesting views on the data. For those of you familiar with
Jeff Brenner and what he did in the Camden Coalition, that’s really where we
started modeling some of our capabilities.

This is just a basic view of county of residents by county of hospital
visits, and you can drill down into this. This is something we created as part
of our state innovation model work. And then you can start to create some
pretty cool utilization work. This is obviously all test data you’re going to
see here, but take one patient, if I’m a target hospital responsible for a
population, show me any patient who has had at least X number of visits at my
facility, but show me all the other places they’ve had.

And it’s astounding when you stop to look at that data how diverse,
particularly in the metropolitan areas, those visits are across facilities. I
mean it’s really wild when you look at the data. You can start to produce
inter-hospital readmission reports, which is something we distribute to
hospitals every month, to show of all the patients you discharged in a 30 day
window, how many were readmitted, not only to your facility which you probably
knew, but to other facilities as well, and who were they, expand and see the
MRNs in the context of your facility.

You can also start to map at a much more granular level. Again, this is all
just based on ADT data, and look at utilization normalized by the underlying
population. And you can start to drill down further and further. I didn’t
include one map that actually puts a pin on the address and shows how many
visits, what’s the overall length of stay, as this particular map is showing.
And then lastly this is again sample data, but you can start to do analysis at
a really granular level. You could drill in all the way down to a block group
and look at super-utilization just based on ADT data.

DR. SUAREZ: Well thank you so much again for the testimony. I think we have
a few minutes for questions. So if anyone wants to ask a question. And I’ll
jump in as the co-chair and ask the first question. Throughout the day we have
heard a lot of very critical messages about what can be done, what should be
done.

And my question is really for each of you, if you had one priority to choose
from in terms of the areas for Public Health data standards, we heard
consistently issues around resources and personnel and funding and support, I’m
thinking more about priority areas as in case reporting or cancer reporting or
lab reporting. In terms of advancing the adoption of standards, which would be
that priority area that you would recommend to focus on? And if you can say one
or two words about why and what can be done about it.

MR. ROBERTS: I would say for sure work on the informatics people and state
and local Public Health. Get them involved in the process. I recognize the
issues. The compensation, these are the busiest people in most health
departments, yes. But figure out a way to get them involved, because they will
take the standards message back into their health departments also.

MR. GIBSON: For me, maybe it’s a little selfish, it’s getting some model
data sharing policies out there. The knee jerk is to be careful about sharing
the data, it’s to be overly careful. There are often messages from the federal
level that more sharing might be okay, but by the time it gets to the ground
level and they actually have to make the decision people get conservative again
and they pull back. There needs to be clear messages that this sort of sharing
is okay, and we need to describe what that level is.

MR. AFZAL: From our angle, the S&I and NIST people could fill in the
blanks on this, but I think it’s having test harnesses that folks can use to
actually send immunization tests from my EMR, and I know it will pass and I can
just start sending it to the Department of Health. Having a set of test
harnesses that enable you to know that it’s going to pass would be huge, and I
think that states are dealing with that on a localized basis and building tools
for their particularly implementation of Immunaid(?) or NADS or whatever it may
be.

DR. SUAREZ: Marty, do you want to comment?

MR. LAVENTURE: I think I would tend to focus on the issue of saying we have
a great opportunity with Meaningful Use if we take advantage of it. And so how
do we pick your choice? Say one of the challenges is electronic laboratory
reporting, and go across the spectrum, a set of policies that gets back to
policy guidance and the standards associated with that, the consensus related
to interoperability, related to the workforce that’s needed, related to the
technical capability, related to the content, and really advance the
implementation of something that we know we need to do and we’re quite far
behind on. Test the model, use it as a way to test the model, that this can be
implemented as successfully as immunization has so far.

MS. KLOSS: I was curious whether there are any best practices or models that
exist that show where the resources let’s say of your entity in Indianapolis is
leveraged to help the smaller local communities. It sounds like everybody may
be working in their own offices, but how are those resources being leveraged?
Because the locals are never going to have the kind of ability to do some of
these more complex without sharing the resources of the whole.

MR. GIBSON: It certainly varies by state. Florida has a model where the
local and the state is tied closely together, I think they’re able to share
resources and knowledge a lot better than we do in Indiana. We do it fairly
informally in Indiana. We try to share but we don’t have any kind of formal
model that we follow. Our director hears about a need somewhere, or we have
some tool that we’d like to share with the state because we think it’s useful
if they can apply it to their data sets. But again, a lack of similarity in the
data sets often gets in the way there.

MR. ROBERTS: We have a regionalization, the six large counties are health
departments, they operate one county at a time, the others are split up into
groups. So we end up with not a huge variation in size from largest to
smallest. It’s not perfect, but it works.

MS. KLOSS: So there are models, and I presume that may be an area for some
resource stretching that could help.

MR. GIBSON: There is a lot of talk about regionalization, and there are a
lot of people trying to figure it out. I don’t know how developed the models
are yet.

MR. ROBERTS: Our smallest region is big enough to employ an epidemiologist,
and our largest reason has two.

MR. LAVENTURE: I think we have looked at the idea of not only having an
epidemiologist regionally, but have started some proposals towards trying to
find an informaticians regionally that would also support collaborative
efforts. We have some work from, I think it’s about 24 health departments,
they’re not in one region but they are organized around a particular
application for Public Health, and it’s the collaboration that that group
brings, they’re really developing a consensus towards use of their system that
is I think one of the valuable lessons, they are I think a good model for how
to share both resources across small health departments. Most of them are very
rural.

MR. GIBSON: One more comment on that. Most of our small departments depend
on the state’s data systems because they don’t have their own. We had a grant
where the state was able to go around to those health departments and do a lot
of development, because they developed their communicable disease online
system. And that system went over really well. But they only had the resources
to do the requirements gathering to get to a very effective system because of a
one-time grant. So good requirements gathering of the frontline people is very
effective if you can fund it.

MR. SOONTHORNSIMA Early on, I asked a question around some common priorities
across the different domains. I’ll ask a different question this time: At the
local level, what are some of the competing priorities you’re seeing locally or
nationally that are drawing away the valuable resources from the Public Health
agenda?

MR. ROBERTS: We have so many competing priorities. The preparedness
programs, I will single them out as being a unique challenge, because they tend
to be well funded and operate in isolation and not always get lined up with
peacetime activities. That’s in my view a mistake, not to have wartime programs
that operate in peacetime. You never use them in wartime if you don’t use them
in peacetime.

And they tend to go off in the corner and do their own thing and become
their own silo. You know what the problem of silos is. In any empire that gets
built, whether a silo or not, is a problem. They tend to grow a life of their
own. Priorities, gosh, I don’t know where to start. I’m glad I’m not the
Commissioner of Health.

MR. GIBSON: Some of the things that come to mind are system upgrades. They
take a lot of resources and they’re always going on. We have good systems but
they keep having to be upgraded, and a lot of RT just goes to that. Public
health accreditation, it’s very important, very valuable for Public Health, and
it takes a lot of resource.

And I’m glad it’s there, and I don’t want to pull resources away from it,
it’s the direction we need to go. Grants, some big grant comes through, you
spend a lot of time applying for it, and then you get it and you spend a lot of
time answering it, and it draws resources that could go other ways if you
didn’t have that grant. But most of our time is spent under just day to day
operations, at least 80 percent of our effort is just the day to day
operations. There is not a lot of extra capacity to put toward all these other
things.

MR. AFZAL: I would just add from an HIE perspective and my vantage point
working with the Departments of Health that it’s resource issue. They don’t
have that many folks working on these issues on their side, at least from a
technical connectivity perspective. So resourcing continues to be a major
impact. And then the other side of this is what is required by regulation and
what’s permitted and what’s not required, and where does that authority exist
to make it required. And beyond that, the enforcement challenge of if it’s not
being complied with, it is really hard to measure and enforce.

MR. LAVENTURE: I would add to those. I agree and add with those the whole
siloed funding issues, funding streams direct short-term implementation, which
really addresses the tactical needs but rarely the strategic architectural view
of information from a health department perspective. You can’t really do longer
term planning, which is needed. So it puts a lot of focus on the notion of
maintenance of existing system as opposed to building towards where we want to
be. And that’s a negative incentive from trying to take the long-term view.

We’ve noticed in our state and in a few states that there is a large move
towards, and it actually happened here, separation of the IT staff, the
technology staff, from the program staff. So it becomes a gap, really no-one in
charge of the business requirements in having that enterprise view. Those are
some of the examples of barriers that currently exist.

DR. SUAREZ: Leslie, you have a last question?

DR. FRANCIS: This is a request actually. We on the privacy subcommittee have
been attempting to flesh out a data stewardship framework for community health
data use. We have struggled because there are no models out there. With the
request for some kind of model policies maybe for data sharing, I’d love to
hear from anybody out in the big bad world, all of you, but not at 5:30.

If you know of any good suggestions for stewardship practices or for what
the actual content of data sharing policies in your view ought to look like,
people who have some really good examples of how to do it, I’d love to have our
committee know about them, because they’re very hard to find. Everybody thinks
it’s really important, but the trick is trying to figure out how to do it. All
the insight we can get the happier we are.

MR. ROBERTS: I don’t know myself, but I know someone who will. I’ll put him
in touch with you.

DR. SUAREZ: So with that request, so of course we are going to go to a
public comment. I want to say thank you to this panel, I think this has been
terrific, a great way to conclude our hearing. We’re going to go to open mic
for any public comments before we close off. As a subcommittee we’re not going
to hold any more time here with discussion, we’re going to have our discussion
a little later this week and over the next few months. We wanted to open the
mic for any public comment, anyone on the public that wants to make any
statements, or any of the members that have testified, if you want to add
anything or make any comments we’ll appreciate that.

MS. STRIKER: I am Laura Striker, with the International Society for Disease
Surveillance. I just wanted to reiterate one thing that came up about data
quality. And I think that when we think about data standards, you can’t think
about it separate from data quality, especially when we think about data
standards, it implies a certain automation also which is also something that
things are moving towards.

With the demand for things more real-time that gives you less time to check
that your data quality is right, and also that we need frameworks for ensuring
and monitoring data quality, and that if people are using the same standards
and have the same assurances for data quality, that that can really help,
especially with data sharing.

What we’ve heard from some of our members is they’re hesitant to share their
data because maybe they’re not quite sure about the quality, and they’re also
not quite sure about the quality of the data that is coming to them and
sharing. So I think that that’s an important element when thinking about
developing standards and data sharing. Thank you.

DR. SUAREZ: Thank you for that comment.

MS. ORLOVA: I would like to add to the presentation that John Stamm from
EPIC had today, very wisely separating the content from the transport. Though
he didn’t comment on the privacy and security around this transporting data,
but this is just not the point of my comment. I would like just to ask this
committee to consider not to specify a particular transport standard for Public
Health, like it was done in Naming and Full Use Regulation for the Messaging
Standards.

We know right now after being for several years in implementing the
messaging standards that the CDA standard is on the horizon. CDA standard can
be sent by three different ways: direct connect, which is secure email, or
reliable data interchange in IHE terms. It’s cross document sharing, the full
infrastructure that would support document registry and the access to the
registry. And the infrastructure that had been developed by NIST for the past
10 years, successfully tested by many EHR vendors in the IHE Connectathon.

And the third transport mechanism for CDA documents is a retrieval form for
data capture, with the emphasis that S&I framework is right now putting on
the structured data capture document just to not want to make an impression
that this is another panacea for sending standardized reports. I think those
transport standards, as John Stamm put them, or maybe we should call them
technical implementations, should be left in the way for communities to decide.

If today in a particular community we see that direct connect is the
preferred way of sending information, then we should be revisiting this issue
in terms of dictating how they should be sending data to Public Health. Please
be open to that. And what this gentleman said today from Care Evolution, he
mentioned it several times. And maybe that’s what ONC means by standardized
queries, or whatever they call this initiative. Data mining is the direction
that the developed world has taken to process data and get access to public
health data.

Last year there was a publication issued by a French group called the
National Disease Surveillance System, and they are calling for standardized
access for Public Health data. Standardized access, standardized data sharing
of those opened Public Health data sources. Data mining is the technology that
is on the horizon, and I think it has been used in the business world very
intensely.

I suggest that we will begin looking into this experience and see to which
extent it is not applicable or maybe very much applicable to Public Health,
because cornering ourselves into these forms, structured documents in terms of
receiving information, may not be wise in the long-term when we actually would
like to interoperate as the users of the systems we create.

DR. SUAREZ: Thank you. We would love to get a copy of that French National
Report. Is anybody else on the public site that wants to make a statement?
Hearing no-one, I only want to say three basic things in less than ten seconds
if I may, given the time. I want to thank everyone, all of the panelists. This
has been really an amazing way to get this first truly national hearing on
Public Health standards going.

It is a journey as I said at the beginning, a journey that the national
committee is going to be engaged in and involved, and now in partnership
certainly with this team of conversions across our subcommittees, the privacy
and the population group, we’re going to be certainly engaged in additional
conversations.

Secondly, I want to thank Michelle and Kamahanahokulani for putting together
this hearing. As Marjorie mentioned, even as the government was shutting down I
was working with Michelle and with Kamahanahokulani just as they were grabbing
their computers literally from their hands, so thank you so much.

Third, I think in terms of next steps our subcommittee, working in
partnership with the other subcommittees in the national committee, will be
looking at summarizing over 25 pages I have in notes and everybody else on
extracting key themes and messages that we heard, observations, and
particularly identifying these areas of recommendations for the secretary for
HHS, for different agencies within HHS, and for the industry in terms of how to
pursue enhancement in the adoption of standards and working standards in Public
Health. On behalf of the National Committee thank you so much again for coming.
Thanks to our staff on the hearing for holding with us here, really appreciate
it.