[This Transcript is Unedited]
Marriott Wardman Park Hotel
2660 Woodley Road, NW
Washington, D.C.
Agenda Item: Welcome, Roll Call and Introductory Remarks
MR. REYNOLDS: Let's go ahead and get started please. Good morning. I would like to call this meeting to order. This is the second day of hearings of the National Committee on Vital and Health Statistics, the Executive Subcommittee, focusing on meaningful use of health information technology. The National Committee is the main public advisory committee to HHS on national health information policy.
I am Harry Reynolds, chair of the committee. I work at Blue Cross/Blue Shield, North Carolina.
I want to welcome committee members, HHS staff, and others here in person. As we did yesterday, again I welcome Dr. David Blumenthal, as the head of the Office of the National Coordinator. I would also like to welcome everybody who is on the Internet. Yesterday I know we had a lot of listeners.
I would like to remind everyone, because of that, as well as the smallness of this room, anyone moving around or anything might make it harder to hear, so if we could keep that to a minimum, that would be good.
I would ask anybody on any of the panels and the committee to speak clearly into the microphone, so that it does project. I know some people had some trouble hearing a little bit yesterday, based on people's distance away from it.
I would like to now have introductions around the table and then around the room. For those on the National Committee, I would ask, if you have any conflicts of interested related to any issues coming before us today, would you please so publicly indicate during your introduction.
I have no conflicts.
MR. J. SCANLON: Good morning. I'm Jim Scanlon. I'm the acting assistant secretary for planning and evaluation at HHS and I'm the executive director of the NCVHS.
MR. BLAIR: Jeff Blair, director of health informatics at Lovelace Clinic Foundation, member of the committee. I'm not aware of any conflicts.
MS. TRUDEL: Good morning. Karen Trudel. I'm the deputy director of the Office of E-Health Standards and Services in CMS and liaison to the full committee.
DR. WARREN: Judy Warren, University of Kansas School of Nursing. I'm a member of the committee. No conflicts.
DR. W. SCANLON: Bill Scanlon, Health Policy R&D, member of the committee. No conflicts.
DR. FERRER: Jorge Ferrer, Veterans Health Administration.
DR. FRIEDMAN: Charles Friedman, deputy national coordinator for health information technology.
DR. SUAREZ: Good morning, everyone. I'm Walter Suarez. I'm with the Institute for HIPAA/HIT Education and Research, a member of the committee. No conflicts.
MR. TRENKLE: Good morning. Tommy Trenkle from CMS.
DR. BLUMENTHAL: David Blumenthal, national coordinator for health information technology.
DR. MIDDLETON: Blackford Middleton, Partners Healthcare and member of the full committee. No conflicts.
DR. STEINWACHS: Don Steinwachs, Johns Hopkins University, member of the committee. No conflicts.
DR. TANG: Paul Tang, Palo Alto Medical Foundation, member of the committee. No conflicts.
DR. CARR: Justine Carr, Caritas Christi Healthcare, member of the committee. No conflicts.
DR. FITZMAURICE: Michael Fitzmaurice, Agency for Health Care Research and Quality, liaison to the committee.
DR. FRANCIS: Leslie Francis, University of Utah, member of the committee. No conflicts.
DR. OVERHAGE: Marc Overhage, Regenstrief Institute, Indiana Health Information Exchange, member of the committee. No conflicts.
MR. HOUSTON: John Houston, University of Pittsburgh Medical Center, member of the committee. I have no conflicts.
MS. AMATAYAKUL: Margret Amatayakul, contractor to the committee.
MS. GREENBERG: Marjorie Greenberg, National Center for Health Statistics, CDC, and executive secretary to the committee.
MS. BUENNING: Denise Buenning, senior adviser, CMS Office of E-Health Standards and lead staff for the NCVHS Standards Subcommittee.
MR. REYNOLDS: For those of you who are usually at our meetings, we usually introduce the audience, but we usually don't have this many friends that join us, so we are not going to put you through that today.
Again, one thing I mentioned yesterday. I would really like to once again thank Chuck Friedman, Karen Trudel, Denise Buenning, and all of the NCVHS support staff. Those of you who were here yesterday, they have put on a magical opportunity in a very short time. I would again like to make sure that everybody clearly recognizes that. We showed up to do it; they made it happen. So I thank all of them again on behalf of the committee.
As I said yesterday, we have 40-percenters in two days. We did 24 yesterday, so 16, I guess, would be "Meeting Light." We heard a little bit about "light" yesterday. So it seems a little more surmountable than yesterday. But I appreciate everybody staying on time.
We will have again a limited open mike, on a first-come/first-served basis. We will, in fact, start from the list yesterday. I know that one of the people who was here yesterday is not back today, but, in fact, did confirm that they are submitting written comments. Again, the most important thing is, whether you get to speak or not, whether you are on the list or not, we will be adjudicating all of the written testimony. You will hear how we are going to do that and how we are thinking about that as we wrap up later today. You will be heard, either with your keyboard, your pen, or your voice. We look forward to it.
There is information on the table as to how you submit the testimony.
Again, I would like to reiterate to the panels, please be ready to start on time. Respect those after you. We really need to hear from everyone. I think that's key.
We will prepare a meeting and a written comments summary over the next two-plus weeks and send it to ONC and post it on our Web site. It will be a public record.
Media inquiries need to contact Nick Pappas at 202-690-6343 at HHS. They will coordinate any requests and use members of the committee or government officials, as needed. I know yesterday we had some. Those are handled by HHS, not by the committee. I think that's good.
As we think through what we heard yesterday, the information was excellent. It was a really diverse panel, diverse thinking, diverse opportunities. But as we look at today, one of the things that I would ask members of the committee and that I would ask those that are testifying -- I know you have prepared your testimony. Nobody is asking you to think any differently about your testimony, because we do want to hear exactly what you prepared. We want to hear that clearly.
But we heard some themes yesterday that, as we go through them and as we look at -- in other words, whatever we hear here is what we will write up. So it's important for us to hear, maybe a little more in-depth, as we ask questions and you cover things -- we heard things like a discussion of light EHRs. What would be the difference between a light, moderate, sophisticated? Then, if you can, be specific about functionality, standards, level of adoption, outcomes, and so on. Again, in the end, we are talking about how we can pragmatically approach this, especially when you start talking about incentives and other things. We will have to get to a pragmatic level at some point. So any help you can be in answering the questions, any help you can be in moving forward would be good.
We heard the need for registries and registry functionality. What's the difference between those? How does a registry fit into things like the EHRs? What exactly does that mean?
The next thing to consider is equating data exchange with the existence of an HIE. As HIEs are going to continue to expand -- some are in place; some are going to continue to expand -- if you are looking at a provider and it's their responsibility to show the ability to do data exchange, the ability to exchange information, and there is not an HIE by 2011 or so in their area, what would they need to do to be able to say that they, in fact, have meaningful use -- if they don't have partners and they don't have an HIE and they don't have some of these other things that need to occur?
I think that's another thinking that we need to all come up with now or will have to be thought of at a different time.
If you were responsible for distributing incentive money, what are the three to five ways you could document meaningful use to satisfy those requirements? What are the kinds of things that you would think about?
We also heard that the majority of standards for interoperability were available. Assuming this is true, what is preventing their full-blown adoption by vendors in a product or by providers in their RFPs and other things? Again, how do we move this?
Again, if the goal by 2014 is to significantly improve outcomes, health, and all the things we heard yesterday -- and everybody kind of played that theme throughout the day -- what functionality is missing today and where would you see the bar for functionality as we take this journey? We have 2011, 2012, 2013, 2014. How do you see that journey?
I think we heard clearly from the vendors yesterday the possibility of a need for a 24-month cycle -- what they were saying as far as how they would roll out changes. That means, if you are in 2009 now, you have, probably, three major delivery cycles between now and the end of 2014. That is, again, why it's so important to back up to today and think about how you play it out. If those cycles stay, then you really only have a couple of shots to start making sure that things happen as they should.
Again, we are not doing this to change your testimony. We are just trying to drill down. We heard a lot of information from 24 different people, plus a lot of others yesterday in questions and so on. Now, as we move it forward, I would like you, after you do your prepared remarks -- and maybe I'll give the panel a moment after each time before we start asking questions, if you want to weigh in on any of these things. If not, then we may drive the questioning to that a little bit to try to get some more insight from you. That might be key to do.
Again, we want to be able to move forward to ONC, move forward to CMS, HHS a document based on the input from you -- that is, you who are in here talking to us. That's an important thing. That was our charge, and that's what we would like to try to be able to deliver. You are the experts that can deliver it for us. So we look forward to hearing that from you.
Unless there are any other comments from anybody, we will proceed with our normal thing. We will be ahead of schedule, and that will be a good thing, especially since we gave people a bunch of new things to answer -- probably not a bad place to be.
So if the first panel would come up, I would appreciate it. That would be Meaningful Use and Population/Public Health.
I am seeing a number of new faces up there on the panel, so I would like to go through our process. You each have 10 minutes. There's a light box on your table. At the end of eight, it will turn yellow. When it turns red, we need you to close, please. Again, we need to hear from everyone, and we want to make sure that we have enough time in this period to do that. So if you will help us with that, that would be a good thing.
As I'm going to ask each panel today -- I know that for some of the others, it matters -- is there any order that you would want to go in? You are listed here as Carol Diamond, Dr. Hacker, Dr. Lenert, and Dr. LaVenture. Are you satisfied with that list?
After deliberation, it appears we are going in order -- after appropriate deliberation, which we always like. Carol, please.
Agenda Item: Panel 7: Meaningful Use and Population/Public Health
DR. DIAMOND: Thank you. Members of the committee, Dr. Blumenthal, thank you for the opportunity to testify today.
This is a really critical moment. I think we are sitting at a really unprecedented opportunity to jumpstart health IT in the right direction, and that is to improve health, to slow growth in costs, and to protect privacy.
At the Markle Foundation, in our Connecting for Health Collaborative, we have been working with a very broad group of health-care leaders over the last few months to develop views and strategies that are critical to achieving the objectives of this public investment under ARRA. We have really been focused on three critical areas: trying to think about what meaningful use is and how it should be defined; trying to outline a forward-looking strategy for standards and certification; and third, really describing a technology approach that will stimulate much-needed innovation.
Tomorrow we are going to host a forum where experts in health-care information technology will sit down with political and policy leaders to talk about how the new law will impact ongoing health-care reform and the best ways to ensure that the money improves health-care outcome, secures privacy, and reduces growth in health costs.
The forum dialogue will center on a document that is strongly supported by a wide array of health-care leaders. Many of them are here today or have testified over the last few days, and we'll make it public tomorrow and share it with the committee for your future deliberations.
What I would like to do today is really make some comments about some issues that were outlined in the opening remarks and that I think will be critical to some of the deliberations that you are going to have.
I think it's an understatement for us to say that the landscape has changed for health IT. This opportunity was borne out of the economic crisis. The role of government shifted overnight from trying to motivate private-sector adoption of standards and technologies to an environment where there is significant public-sector investment in health IT. That will bring new scrutiny and demands for accountability, and rightly so. I don't need to tell anyone here that the stakes are very high.
This may seem a desire to really try to develop rigid requirements or detailed technology specifications to ensure that the money is well spent. I think, given the nascent state of the health IT industry, it's important to resist this impulse. What we need are crystal-clear goals, crystal-clear goals for what the technology is to accomplish. To create the right market innovation and spur innovation, both in care delivery and in technology, we should not over-specify specific technology at this juncture. People will respond with better and more functional products if we introduce a new driver, and that driver should be a definition of meaningful use pointed to health improvement, not specific features and functions of technology.
We really have to shape this around what those goals are. We need IT to improve health-care quality and to control growth in costs and to protect privacy. These expectations have to be set from the start. If the goals are not set from the start, it really will be very difficult to offer the kinds of results that government is seeking after technology has been commissioned and incentives have been offered. The government risks wasting its valuable resources and losing support, quite frankly, from both providers and the public. So this is a critical moment, and the goals are not the purchase of technology, they are not the installation of technology.
To that end, the definition of meaningful use means to center on improving patient outcomes, improving health status, improving the delivery of care, and controlling growth in costs.
Where do we focus in the near term? I heard some of the discussion yesterday, and certainly in our own deliberations people are very concerned about what's possible today, what's possible in the near term. This leads you to say, let's focus on two things: What has the highest potential for impact in health improvement and reduction in growth in costs now? What information can be made available to point of care in electronic format in the near term?
Based on these criteria, there are at least a handful of areas that offer near-term opportunities. There is mediation management and care coordination. These goals are broadly applicable, and they will have a strong impact on health improvement and controlling growth in costs. They rely on information that is available electronically, including some of the information that was discussed yesterday. They fit and are very consistent with the meaningful-use framework outlined in the law and with the new national quality priorities. The definition should over time, as the law states, expand to include additional things. But there does need to be a clear and focused starting point.
How do we know if we have achieved meaningful use? We distinguish two terms and have distinguished two terms in our work. I will explain that, because I think it's helpful.
Certification is about demonstrating whether or not a system has the capability to be used to achieve these objectives of meaningful use. Validation, on the other hand, is being able to demonstrate that it was, in fact, used to achieve meaningful use.
There has been a lot of mixing of those terms. I would just caution against using "certification" as a proxy for the goal. The goal is health improvement. I want to be clear about that distinction.
The law envisions an evolving set of requirements. The requirements have to enable a broad range of providers to participate, through a variety of mechanisms. As you know, providers are on a wide spectrum of levels of adoption and readiness for adoption. So there will have to be flexibility in this approach.
There has to be something demonstrable in the near term. The validation requirements or the metrics have to be simple and achievable to demonstrate, preferably automatically from technology.
There does have to be a clear motivation in these goals to use and share information to improve health-care quality, but there does not need to be an over-specification about all the things and technology features and functions that need to be specified on how to get there.
We hope what this can also do is stimulate innovation, again, both in health-care delivery, finding new ways of delivering better and improved care, and in technology. There are lots of studies that show that even in small practices, where there is no IT help desk to call, there is no department to call when a system doesn't boot up, costs can be very significant, innovation, especially in that marketplace, for new tools that offer near-term value and can achieve the objectives of meaningful use is very much necessary.
We also would like to say that meaningful use should dictate the technology approach, the approach to standards and certification, and not vice versa.
I think we envision certified or qualified technologies accommodating a wide array of health IT tools. We must reward the meaningful use of these tools, and not the specific purchase of hardware or software products.
Let me spend a minute on standards -- not to belabor the issue. I have said many times to this committee and in the Health Affairs paper, standards are not created through committee consensus. Standards are created through adoption. There are protocols that are created, but they become standards when they are widely adopted and used. This law has the opportunity to create that motivation for adoption.
Similarly, certification is critical to prevent fraud, to prevent faulty products, or reward superficial uses of technology, and clearly, whether or not the technology can achieve the objectives of meaningful use and meet the technical privacy and security requirements of both HIPAA and the new law. But it should be open to a broad range of applications. The question we should be asking today about certification criteria is, will they enable providers to achieve the health improvements and cost-control growth objectives of meaningful use? That's what we should focus on.
In terms of certification, let me just say that it is very clear that there should be clear testing criteria for interoperability, for transport and security protocols, which have to be defined, and they could be validated using Web-enabled testing environments. We encourage a pluralistic approach to applications, not a narrow definition of what an EHR should be, and also a pluralistic approach to certification. If the requirements are clear and specified, then innovation can take place from the marketplace to help purchasers find value in different products.
Let me make a comment about the population health objectives, from my Health Affairs paper. The latest issue is here. Suffice it to say that the opportunity to use health IT to improve population health goals is not lost on any of us. This is clearly something everyone is interested in.
The urge that I would make the committee is to really think about ways in which the paradigm for how population health is measured and analyzed and achieved is shifting and changing in very important ways, from centralized structures and centralized data-collection efforts, where the people who generate the data and who are in a position to make the decision using that information are separate and removed from that process. It has not served us well.
In fact, health care is not unique. This process of trying to create large, centralized, or big-bang approaches to efforts is true in the FAA and the IRS and other large-scale failures, like government IT projects.
I think what's critical to think about are distributed ways of analyzing and measuring population health goals. There are a few examples I will just mention briefly.
One is the DiSTRIBuTE model for flu surveillance, which does not collect detailed, identified, or de-identified data, for that matter. It just collects summary statistics. That project is spearheaded out of the New York City Department of Public Health; also the New York City Department of Public Health's approach, the PCIP project, for quality measurement, where the quality measures were embedded in EHRs and the provider is able to assess in real time and see where they are in terms of performing on the indicators that are necessary, but centrally only summary statistics are collected. Again, my paper here, which is available, offers a more elaborated view.
Thank you.
MR. REYNOLDS: Thank you. Appreciate it. I'm sorry I did not recognize you as Dr. Diamond, excuse me.
Dr. Hacker, please.
DR. HACKER: Thank you very much, Mr. Chairman, members of the committee, Dr. Blumenthal.
I'm William Hacker. I'm the commissioner of public health for the State of Kentucky. I'm the chair of ASTHO's Informatics Committee. I'm taking this opportunity to address the committee from ASTHO's point of view and the state health officer's point of view, in regard to the term "meaningful use."
The nation's public health community is united in the belief that meaningful use of electronic health record systems must include bidirectional exchange of data between public health departments and private providers. I will focus my remarks to try to address the issues related to cities, counties, states, and territories in the United States and how we can improve health across the country.
Public health is an essential component of successful health reform, to transform the United States health system. It is a system that prioritizes prevention, supports healthy environments and lifestyles, provides at a minimum preventive and primary health care for every person, eliminates inequities in health status, and protects the people in communities from existing and emerging health threats. We should work to build a health system, one that is oriented toward health rather than sickness and is able to effectively address known determinants of health, whether they be environmental, social, behavioral, or medical.
Public health enables this kind of health-system transformation. Public health agencies already touch the lives of every American to promote health, prevent disease, and to protect us from health threats. The public health needs to be strengthened and empowered as a significant foundation for health-system reform.
In public health, we recognize that electronic health records and health-information exchange are essential tools that can help us to be more effective at carrying out our jobs and fulfilling public health's mission to improve the health and well-being of all Americans. Public health practice will significantly transform, if we are able to fully leverage EHRs to enhance our data-collection needs to support disease surveillance and population health improvement.
Don't forget that we are also providers of health care. Many of the vulnerable populations in our communities, women and children, seek health services through the local health departments.
With the use of electronic health records, we can strengthen our efforts to ensure that children appropriately receive their immunizations. We are better able to manage chronic conditions and coordinate care, while employing community-based methods to support individual self-care. We can fortify our alliance with medical-care providers and ensure that we conduct screenings for newborns. We can better assist our medical-care colleagues to more rapidly translate evidence-based medicine -- what works -- to inform an effective clinical practice. And with electronic health records, we are better equipped to promote healthy behaviors and healthy lifestyles.
Public health seeks a future where it can easily consolidate and integrate information from a variety of disparate systems, sources, across multiple settings, across clinical and business domains, and across technologies. To serve the public well, the governmental public health agencies must be linked to inpatient, outpatient, home, and industrial settings. We must be able to receive data from clinical settings, as well as provide information and decision support to clinicians.
That is a vision for a transformed health system operating in an environment where the use of electronic health records and health-information exchange is a standard way of practice, not the rarity, and where public health agencies work in close partnership with clinicians to promote and protect the health of all citizens.
There are three high-priority and high-impact public health data needs that can be advanced through electronic health records: immunizations, electronic data reporting, and notifiable and non-reportable disease data. Let's start with immunizations.
Vaccinations are a proven public health tool to prevent and reduce disease. If all providers collected immunization information electronically on each of their patients and were able to send the data to public health, we would be better able to identify poorly immunized populations and groups, targeting interventions and evaluating program efforts. We would be able to use the information to control outbreaks of vaccine-preventable diseases and would be better able to measure progress toward goals of vaccine coverage and disease control.
In addition, we would be able to assure accountability for effective vaccine use.
Immunizations are mouse-ready. Technology is available. Data are currently being captured by some providers electronically today. But immunization data need to be part of all electronic health records. Capturing this information needs to be part of the provider's standard practice, not something that is collected on a separate registry or something that is not an expectation for medical practice.
I am aware that certified electronic health records have adopted core immunization information. We need to assure that electronic health records seeking certification are interoperable with public health immunization information system national standards. At the same, public health agencies need support in maintaining public health systems to be interoperable, through the standards, back to the private sector.
Public health labs are also mouse-ready. Our nation's public health laboratories have worked diligently to implement the laboratory information-management systems needed to exchange data with many stakeholders. The interoperability standards for electronic reporting of laboratory data exist today. Being able to receive electronic requests from specimen analysis and report results, we will be better able to enable public health to monitor communicable disease and control or contain their spread.
Communicable disease reporting has been largely paper-based, relying on faxing, telephones, and mail. We are vulnerable to public health threats on a daily basis -- for example, the swine flu issue we are dealing with today. How can we quickly and effectively protect the public from acquiring preventable communicable disease when we are dealing with a slow paper process today?
Electronic lab reporting from clinical providers to public health can significantly improve the handling of high-volume, time-sensitive diseases and enable public health to quickly conduct assessments of complicated cases and determine the best course of action to efficiently and effectively respond.
Third is the notifiable diseases, as well as non-reportable disease data. Electronic reporting of notifiable diseases -- infectious diseases, communicable diseases -- can significantly be more rapid and accurate.
I believe you understand all that process, so I'm going to skip my words in your handout to save some time.
But in addition, electronically capturing non-reportable disease information, like diabetes, heart disease, and asthma, will better enable public health to develop and target health-promotion interventions to improve population health. This is putting the power of prevention, linked to clinical care. I can say with great confidence that public health agencies are good at this. Public health is effective in studying the data, designing community interventions, communicating the message, and working with their local colleagues to make sure we impact individual health, as well as population health.
The public health community commits itself to linking its shovel-ready -- to use the older term -- components of electronic health records and health-information exchanges by 2011. Specifically, I would like to suggest that as part of the definition of meaningful electronic health record use, immunizations, electronic lab reporting, and electronic reporting of notifiable and non-reportable diseases be included. Real-time public health reporting and bidirectional data exchange should be an expectation in standard practice, if we are to achieve health-system transformation and a healthier America. With bidirectional exchange, we mean clinical reporting to public health and public health exchange with clinical care.
I would like to offer some implementation considerations with respect to the inclusion of the high-priority data needs in the definition of meaningful electronic health use.
As I mentioned previously, immunizations, electronic lab reporting are mouse-ready and could be captured electronically. These should be required by 2011. Electronic reporting of notifiable and non-reportable diseases may be phased in over time.
Electronic health record systems use different methods to capture data today. Some are diagnosis codes, procedure codes, or even medications. The use of different approaches to capture data requires manual review to filter erroneous or non-relevant information, making electronic reporting more ineffective than usual. Efforts are necessary to standardize how disease data will be captured in electronic health records to make it less burdensome to collect and analyze. Electronic reporting of notifiable, non-reportable diseases will be prioritized diseases. Non-reportable diseases should be considered as part of those that are listed and supported by the Prevention and Wellness Fund.
Public health wants a seat at the table as these priorities of diseases are fleshed out in the national debate.
In closing, I would like to point out that incentives for provider adoption of electronic health records and inclusion of public-health-data capture are significant, but still only initial steps. Public health, to be truly strengthened, requires a transformation of its own to be ready to receive, filter through, and effectively translate and utilize the massive amounts of data from clinical systems. We will need support for this transformation, given the limited staff and funding resources available for state and local public health.
We stand ready and willing, however, to work with the committee and our federal partners to assure the successful implementation of the HITECH act and continue the consideration of public health's role and value in health-information exchange and health-information technology.
Thank you.
MR. REYNOLDS: Thank you. I am sure the vernacular of "mouse-ready" will now be ingrained in the conversation going forward. We appreciate that.
Dr. Lenert, please.
DR. LENERT: This is testimony from the National Health Center for Public Health Informatics, NHCPHI, one of 11 centers at the Centers for Disease Control and Prevention in Atlanta, Georgia. NCHPHI protects the public's health, promotes health equity, and transforms public health practice to advance the science of biomedical informatics and public health practice and collaborative development of information systems for public health.
Our vision is to promote innovation and provide leadership on a national and global level to transform the public health system to informatics.
NHCPHI is heavily engaged in the design and development and testing of systems that use the national health information network standards and infrastructure to promote collaboration between the clinical care system and public health. Programs funded by NHCPHI that are focused on this collaboration include BioSense, the National Notifiable Diseases System, and the Public Health Information Network.
I will review these programs briefly as I go on, but I'll cover them in more detail in the written testimony.
BioSense is a program designed to provide real-time situational awareness or national monitoring of health through data generated in emergency rooms throughout the United States. Approximately 12 percent of emergency-room visits to -- hospitals and 100 percent of those to governmental hospitals in the United States are monitored through BioSense.
The National Notifiable Diseases System provides support to states for linkages to public health reporting for notifiable diseases to clinical care, notifiable conditions or diseases where there is a statutory reporting requirement within a state for providers, for laboratories, for the purposes of prevention of the spread of disease. Notifiable diseases are typically infectious diseases. Each state has its own list of notifiable conditions and diagnostic criteria that are active. CDC is informed of states of a case that has been investigated and confirmed.
The NNDS program supports state health departments through direct adjunct funding, through development of software systems for managing notifiable conditions, and through software programs that link clinical care sites and laboratory systems to public health.
The Public Health Information Network is a set of programs designed to produce interoperability between public health departments and state, local, and national-level protection, intervention, and notification. The PHIN program promotes beset practices for interoperability through communities of practice for selected issues.
NCHPHI is highly involved in the standards process, participating in the AHIC subcommittees in HITSP, in standards development with HL7 (phonetic), and in promulgation of standards through PHIN. Our philosophy is that standards development is essential but not sufficient for public health capabilities. Standards are the link between clinical and public health infrastructure. However, without investments in public health infrastructure, there is no capability for collaboration.
An example of the type of collaboration between clinical care systems and the public health department is work on public health alerting and decision-support systems for electronic health records, performed in collaboration with our centers of excellence within the Integrating the Healthcare Enterprise organization. Working with participants in the IHE, we developed a standards-based approach to notify clinicians of public health alerts on their desktops, through publication to a centralized database.
This approach notifies clinicians when they are seeing a patient who resides in a zip code with an ongoing public health problem and has demographics and symptoms potentially consistent with the alert. It also informs clinicians of the types of actions they should take -- restricting food handling, culture, other types of interventions that are frequently ignored in clinical practice.
This type of system demonstrates the degree of collaboration between the clinical care system and the public health system that is inherent in the term "meaningful use" of electronic health records.
Responses to the committee's specific questions follow below.
With regard to the vision for public health practice in an era when all Americans have electronic health records and the priorities for functioning: NHCPHI's vision for population health practice in an era where all Americans' health care is supported by EHRs focuses on two areas: public health-clinical care collaboration, and public health empowerment through access to clinical data.
Broadly speaking, the goals of the clinical care-public health system collaboration fall into five areas:
Systems that assure that every clinician in the community is an astute clinician, one that recognizes public health notifiable conditions, reports unexplained illnesses, follows public health recommendations for outbreak control, and is assisted by decision support at the point of care in order to make this happen.
Second, the infrastructure must be in place for rapid two-way communications with public health about notifiable diseases. Diagnostic laboratories should be able to report conditions to public health departments, all EHRs reporting of public health notifiable conditions to state health departments, helping providers fulfill their statutory obligations.
Public health should work with the clinical care system to help ensure appropriate care for all. Registries should monitor patients with serious illnesses to make sure they receive appropriate care and that none with serious illness, especially those with infectious illness, are lost to follow-up.
Alerts should notify clinicians when individuals with untreated, contagious illnesses come in casual contact with the health care system, so that appropriate action can be taken.
Public health can work with clinicians to manage the growing problem of antimicrobial resistance in communities. It is our vision for resistance to be actively managed through surveillance of resistance patterns and improved prescribing practice delivered through or advised at the point of care.
Public health should also work to promote health in the community. Public health systems should ensure that tailored health-promotion and prevention programs are available to all persons in the community, regardless of insurance and socioeconomic status.
Public health activities and public health departments will be transformed through enhanced access to data from the clinical care system. Routine activities that are now slow and inefficient will be automated. Activities include integration of vital registration tasks with clinical care, 100 percent situational awareness, with appropriate sharing of detail across levels of government, creating an infrastructure that allows the national to rapidly characterize outbreaks and precisely control efforts in pandemics.
Data should be automatically from the clinical care system to automate public health practice. Disease-control programs should spend the majority of their time on management of cases rather than data collection. The clinical care system will also generate data on communities at a highly detailed level that is not practical to obtain with survey methods. This will allow public health officials to identify needs and target health disparities.
Communities will have capabilities to assess the risks of local environmental threats and to monitor health in conjunction with those threats.
There will be widespread automatic surveillance for adverse effects of pharmaceuticals and other therapies. Systems will ensure the first million patients receiving a drug will have special monitoring, and active surveillance to ensure overall efficacy and safety.
The support for monitoring rates of hospital infections will ensure that infections are accurately tracked and data are widely available to the public.
Specifically, what do EHRs have to do to achieve meaningful use to allow this to happen?
In short, we believe that clinicians should put all their patients into an EHR system and should use that for routine care. In addition, they have to collaborate on exchange of information for patient and population care. Information should be delivered through links to the clinical care system and should impact patients' care. Meaningful use should include a requirement for clinicians to respond to specific notifications about a patient or requests for incoming information into their EHRs, including medication issues and errors, abnormal lab results, public health alerts, and requests for medical records from authorized institutions, including public health-oriented requests.
Meaningful use should also include sharing of health information in the EHR for collaboration on patients' care, particularly information that is mandated by state statues for public health reporting. If a patient using an EHR has not implemented features that allow sharing of information, he or she cannot be using that EHR in a meaningful way.
To sum up, how can public and population health needs and requirements translate into meaningful-use criteria for 2011? I think if we look underneath the hood of the NHIN, we will find that there are serious threats to the validity of all operations and that public health practice is as far along as any other activity in the NHIN. What's needed is investment simultaneously in public health infrastructure, as well as the infrastructure for information exchange for clinical purposes. If we do these things simultaneously, we can achieve meaningful goals for 2011 with public health practice. It is a matter of will, not a matter of ability. But if all the resources are committed to clinical care, this will not happen.
What I urge this committee to do is to recognize that meaningful use should include public health practice, and particularly collaboration between public health and clinical care. This collaboration will open a new window and world for public health operations and for ensuring the health security of the United States.
Today, as we face the threat of pandemics in this world, we can see more clearly than ever the need for investments to enhance our health security. Without these investments, the efforts in 2011 or in 2016 will not produce the type of system that we need.
Thank you very much.
MR. REYNOLDS: Thank you. Dr. LaVenture.
DR. LAVENTURE: Thank you and good morning, Chairman Reynolds, Executive Subcommittee members, and Dr. Blumenthal.
Thank you very much for holding this important hearing and allowing us the opportunity to provide a perspective on the Minnesota e-Health Initiative and on the definition of meaningful use, and specifically how it relates to population health.
My name is Marty LaVenture. I'm the director of the Center for Health Informatics and e-Health at the Minnesota Department of Health. I'm pleased to join you today.
The example of the statewide collaboration in Minnesota is reflected by a public-private collaborative that was legislatively chartered in 2004. It reflects the health community's strong commitment to act in a coordinated, systematic, and focused way. It's unique in some aspects, in that it's comprehensive by way of its broad settings and broad definition of providers. Of particular note for this committee and this panel today is that population health and public health has been included some the inception of the initiative, and it is really part of both the vision statement and one of the four major domains as reflected in the circles on the Minnesota e-health conceptual figure, public health, clinical care, consumers, and policy being the major domains for the focus of the initiative.
In 2007, the State of Minnesota legislature passed a mandate that required all providers to have interoperable electronic health records by 2015. As part of that legislation, a plan was required for how we would meet the 2015 mandate. The plan was published this past June at our e-health summit. You have a copy of the plan, both electronically and the paper copy before you.
Of particular note for this plan is that population health and public health issues are addressed in it, in its comprehensive approach for providers statewide. Along with the legislative action was a grants and loan program to focus on underserved and rural settings in Minnesota.
One of the challenges in looking at 87 counties and nearly 2,000 different diverse settings in the implementation of electronic health records was communications. We developed an approach -- sort of a visual way to communicate this process of how we are going to move forward around getting adoption, effective use or utilization of electronic health records, and then exchange of information across this.
This diagram that you see reflects that approach, where all providers and settings, whether it be clinics, hospitals, public health labs, find themselves at some stage, either beginning or more advanced. With that comes a plan, tools, tips, and resources to support them in advancing in that particular stage.
Of note to this committee is the center section, referring to the utilized area, where effective use is discussed. We'll spend a little bit of time talking about that.
Of note in Minnesota, since the effort, nearly 60 percent of the primary care clinics in Minnesota have reported adoption of electronic health records, and we are moving them through more effective use and exchange capabilities.
In terms of the initiative, meaningful use is seen as a broader part of a framework for effective use of electronic health records. We would encourage you to think of it in the broader sense of where we are heading with regard to effective use of electronic health-information technology to improve care, and really recognizing the real value that is associated with that effective use.
The initiative engaged a workgroup this past year to define "effective use" and is developing a guide to effective use of electronic health records that will be published at our June 2009 summit that you all are welcome to attend.
The model that is reflected so far in the work of this workgroup is as shown in the next slide. The four key components are identified as elements of effective use, recognizing that, pre-going-live, organizational issues are important, but also that post-live, organizational issues are essential, that smart use of clinical support systems is essential, as well as quality-improvement reporting, and, fourthly, health status, including population health and public health, which is of note for our work today.
In Minnesota, we follow a definition of population health similar to the AHIC's definition, where population health is looked at as really improving the health of the entire population. If you will, it's all of our responsibilities -- everyone's responsibility -- with public health being the governmental responsibility. It provides the backbone of public health infrastructure and carries out the legal responsibilities associated with particular geographical areas. Both are important and both are included as part of the thinking and focus on meaningful use and effective use in Minnesota.
As we have talked about effective use in the workgroup and the advisory committee, several key principles for meaningful use were identified. They include that population and public health really is essential to the definition of meaningful use, that the criteria for certification of electronic health records need to contain elements associated with population and public health, and bidirectional exchange is essential. Given that notion, we also recognize that the first year must be achievable in its core components of what is expected for meaningful use, and that a broader set of components and the goals of where we are going need to increase over time.
We also suggest the approach of progressive implementation relative to the first year, so that late starters aren't necessarily penalized.
Similarly, the resources for filling the gaps: If we are expecting to do exchange, we need to make sure that both sides of that digital handshake are completed, and the gaps in the capital funding and the knowledge gaps for informatics are filled in several areas. I'll comment on that in a moment.
The meaningful use should include two key components:
Being pragmatic in Minnesota, we tried to answer the four questions that were given us. We have outlined, in summary here, some responses to the key questions.
The first one is the vision. Clearly, the vision in the initiative for five years now has been for healthier communities. We strongly believe that certified electronic health records can help really complete the -- they need to have complete population health functions in order to do that. We need near-real-time bidirectional electronic communications with the public health authorities, at a high level of interoperability. We are defining that in Minnesota to combine technical, semantic, and process components for interoperability, as well as a highly integrated population and public health network.
Second, the priority data needs are listed here, similar to what you saw on the previous slide. We would be happy to go into whatever detail you would like. Some of those are outlined, including the standards that have been adopted. The commissioner of health in Minnesota has the ability to require that and require the use of those standards. We are moving forward with adoption of those, based on a number of national efforts.
The third area is the requirements for use. We believe that certified EHRs have the function to support population health. They require schedule for exchange that, as we mentioned, is achievable. It increases over time and provides relatively first-year qualification as a meaningful user.
What's our path to 2016, starting at 2011? We boil it down into two key areas. One is to fill the gap in essential capital funding resources, and a second is to fill the gap in knowledge.
The capital resources clearly are modernizing our public health and population health systems to be more effective and support exchange, including reporting the requirements and completing the interoperability of standards that are so essential.
The gap in knowledge is several key areas that you are familiar with that are critical to success in Minnesota, because it will involve the technical support centers being as close to the community as possible:
I want to just recognize as I close the advisory committee co-chairs, Dr. Jennifer Lundbard and Walt Cooney (phonetic), who have been providing great leadership, and the workgroup members who are participants. Literally thousands of hours have been volunteered in a collaborative approach to move Minnesota, statewide, for example.
Thank you very much, Mr. Chairman.
MR. REYNOLDS: Thank you very much.
Based on the comments I made earlier, thank you. I think this panel laid some things out. I heard that immunizations were a must. I heard bidirectional. I heard reporting of notifiable diseases, and that public health infrastructure needs some upgrades.
A lot of our friends around the table have questions. We will begin with Karen Trudel.
MS. TRUDEL: Thank you very much. I would like to dig a little deeper into the capabilities for 2011, 2012, in the short term, being mindful of the fact that an eligible professional or a hospital sending data has to have someone that you send it to who can do something with it and a means to get it from point A to point B. Realistically, when we are trying to make sure that meaningful use is set in such a way that many providers will be able to meet the goals, what do you think are the reasonable goals for 2011 and 2012? I know you mentioned immunizations and labs. But in terms of the ready receiver and the means to get it there, how reasonable are those expectations across the country, as opposed to in certain areas of excellence?
DR. LENERT: The issue with immunization registries is primarily the level of funding for the registries to upgrade so that they can receive messages from the provider systems. We have two years to get them ready for the receipt of that. If there was funding available for them to embark on a program to upgrade it, then they would be able to receive the messages as structured in the existing format.
The point is that we can't have the infrastructure without the investment. Standards are not enough.
DR. HACKER: I would add in Kentucky we are hoping to tap into some of the stimulus dollars to connect several data sets in public health, including immunizations, newborn registries, and so forth, to be able to feed that back out. The technology is there. It's just, what do we have the money for?
The ongoing concern is, what happens after our stimulus dollars are spent? Where's the funding for the ongoing support? I think all of us are saying that public health has to be included in any national approach to health-system reform and transformation, and it takes capital investment to do that.
But in terms of the technology, that's available today and it can be done. We are doing it in limited places around Kentucky.
DR. LENERT: I wanted to comment on laboratories, as the next level. What public health needs is all the laboratories in the country on the NHIN in communication with physicians electronically. If that happened, it would be relatively low-cost for us to support the infrastructure to have them transmit laboratory reports to state. But as it stands right now, in the NHIN and ARRA legislation, there is not a specific investment, except for the provisions to states, to bring laboratories onto the network. Without those laboratories, the vast majority of which are not on the network -- without bring them up on there, physicians won't have anything to talk to for their meaningful use.
Let's make communications with laboratories for order entries and results reporting and for alerting when there are abnormal results part of meaningful use. When you do that, you are going to enhance public health practice.
The third thing I would say is that there are certain select pieces of infrastructure we could build right now -- build once, one way, correctly, and deploy, and have meaningful impact. I believe that public health alerts to clinicians' desktops, using standards like Key 81 and RFD, is something that we could get done right away. It's very important, because those things are largely ignored. If we make a decision on one way to do this and make that part of how we are building things out, we can get public health capabilities out now.
But again, we have to make the decision to do this.
So I really see three different things:
MR. REYNOLDS: Judy Warren.
DR. WARREN: I have become concerned, in looking at some of the testimony from yesterday and then with testimonials today -- both Dr. Lenert and Dr. LaVenture mentioned that you were doing chronic illness follow-up and that you were interested in looking at behavioral changes. Yet none of the data that people are saying need to be transmitted will transmit that information, because you are only looking at physician orders, you are looking at labs, you are looking at immunizations, and you are looking at problem lists. You are not really looking at assessments of patients.
The same thing could be said of some of the quality indicators. We are looking at quality data being transmitted, but never the baseline data or the assessments of those.
Do you have any comments on where you see those data coming from or if it's just implicit in what you are talking about, or whatever?
DR. LAVENTURE: I will start. This is Marty LaVenture.
Certainly we see the chronic disease issues as being an important part of the menu of where we are headed. It may not be available, certainly, in 2011. But unless we plan for that immediately, we won't get there. So that's where the issue, I think, alluded to in the previous question, too, of the collaborative development of the standards and the specifications, is so essential, and it will be built into what is expected from an electronic health record. That includes standardization of the data in uniform ways and collecting and coding of the data in the electronic health record.
We would love to expand the surveillance related to asthma, related to cardiovascular disease and stroke, and a number of areas. There are a lot of interesting projects with the health plans that are doing population health. The standards certainly would be welcome in order to update their particular information systems, so we could then share across the health plans. Right now we have a good set of data being collected in a number of different health plans, but not in uniform ways. That would be a first critical step to achieving that, and then hopefully the exchange, for true meaningful use.
DR. WARREN: Just to clarify, you said data was coming from health plans, not providers?
DR. LAVENTURE: Health systems, the large systems that cover a big portion of the population in Minnesota. They are doing terrific population health work, but to really make that effective across true populations, we will need, obviously, uniformity.
DR. LENERT: I would like to suggest that we take something that we can already do, like a CCD, and use that as a way of tailoring health information for people. We take an existing capability, we push that out using NHIN services, and then we try to bring back the health information that is most relevant to people to change their health behaviors. That's a first step.
It's something that could be done, again, relatively quickly, because it's based on existing standards, if we had investments in tailoring technologies that are necessary. We have a lot of investments in tailoring technologies already, through the MCI and other places. But we have not tried to bring together tailoring and the CCD and existing ways of doing electronic reporting.
Again, I see this as being closer than we think. It's just a question of whether we are able to engineer what we are doing now to reach the cause of individual health promotion. I think we can. We just need to think about doing it.
MR. REYNOLDS: Justine.
DR. CARR: Thank you. Thank you for your presentations. We have covered a huge landscape. One of the themes that we heard yesterday was, from the EHR perspective, EHR Light, thinking about the vast numbers of people who are just getting started and where we can go with the EHR Light.
I want to reconfigure a bit of what we have heard today and say a couple of things.
The goals seem to be to promote health and prevent harm. The sources of data could be the EHR, could be system administrative data, could be direct lab feeds, and then, I guess, e-prescribing information.
But I would like to bring it down to -- if we had three pieces of data or four sources of data, what would they be? And by having them, what would get better, either in terms of promoting health or preventing harm? I just want a real simple thing of what we could do.
DR. HACKER: I mentioned that three --immunization registry, laboratory data, and then reportable disease -- are available, doable, and are being done in some places right now. It's the lack of resources in the public health system, as well as the incentives for the private sector to do that.
DR. CARR: But we have that, and we get that information, and then what gets better? How does it get better? It goes back to the EHR with decision support or it's a public program? What happens?
DR. HACKER: Several, I think, are fairly obvious. You prevent duplication, giving the same immunization twice to the child, if you have immunization feeds, both for private sector and public sector. Dad takes the child to the public health department or a private doctor; mother takes him to the health department. They don't know -- yes, he got a shot, but they don't know which, and so people go ahead and administer it.
Secondly, we are able to identify those populations that are not adequately immunized. Then we can go after those folks to increase their targeting, both the private provider world and to the public health world.
Thirdly, if we have an outbreak and we know we have -- say, pertussis -- we have pockets where we knew we had inadequate immunization. We could focus our public health resources. That's more effective use of our dollars and our staff time, as well as achieving a public health goal of decreasing the presence of communicable disease.
The immunization piece out there is, I think -- I can't see anyone debating the issues of why that's not a good idea. That's a classic one.
The others I think are self-evident, and I would be happy to speak in more detail, but I suspect there are other questions.
DR. LENERT: When we talk about EHR Light, if you are willing to send out small pieces of anonymized data to centralized services for decision support, like an immunization registry, you can get back meaningful recommendations on what the physician should do next. That's the same way our public health alerting technology that we tested in IHE this year works. The chief complaint, the zip code, age -- that goes out to a centralized server. It comes back with what the doctor should be able to do.
A simple addition to an EHR Light can add a lot of capability, if we are willing to change the architecture a little bit and get things that are practical on the clinician's desktop. We think that when the clinician is using the electronic record, they have to get something back from the environment, from the community, for it to be worthwhile. This type of system, where you are sending data, anonymized, getting it back, which is relevant just to your patient at the point of care, that's what is really going to drive adoption in this particular setting.
DR. DIAMOND: I just want to respond and say that I think the end of your question is the crux of the issue: What is the public health goal that we are after and need to achieve, rather than what data is available? The goal changes how you pursue the information, and it changes what you really need. That discipline, in our work anyway, is outlined by a set of information principles. But, for instance, to take flu surveillance as an example, the old way of doing flu surveillance was to try to derive all the underlying data and basically derive who in that population might have the flu and then look at trends.
The model that I mentioned in DiSTRIBuTE is actually low-tech. It turns out that to do good flu surveillance in a timely manner, you don't have to standardize all the underlying data and all the data definitions. You can tolerate some noise there and still develop meaningful trends that enable public health officials to do what they need to do, simply by collecting summary statistics in a robust and comprehensive way, even if that means faxing in those summary statistics on a weekly basis. It is better than sometimes waiting for all the data that's necessary to become available or for everybody to comply with those requirements.
MR. REYNOLDS: John Houston.
MR. HOUSTON: Thank you. This is for Martin LaVenture. I was reading through the document that you provided, the manual. I guess I have one question. Based upon the Minnesota mandate, has the state been able to establish specific measurement criteria for what is described in the manual as "effective use," which, I'm assuming, would equate to the same thing as "meaningful use," what we are here to discuss today?
DR. LAVENTURE: The workgroup of the e-Health Advisory Committee is working on effective use as a second stage. We worked first on adoption this past. They are working on effective use. In developing some criteria for measurement of effective use, we see "effective use" as broader than the "meaningful use," in the sense that meaningful use may increase over time or change over time, whereas we see effective use as being fairly comprehensive, in a sense a part of best practices that needs to be achieved overall for your practice. It may vary a bit by community or region as well.
The group is looking at a number of indicators in each of those areas, clearly more related to outcomes, on decision support, for example, as well as public health indicators. We have not published any yet, but will be, as part of the ongoing development of that plan. The plan will be updated based on the requirements coming out of ARRA.
MR. HOUSTON: Is there a timeframe for those?
DR. LAVENTURE: We hope to be updating the plan in the third quarter of this year.
MR. REYNOLDS: I'd like to go to Marjorie next.
MS. GREENBERG: Thank you, and thank you to all of you.
This is not so much a 2011 question. But when we are talking about population health, we are obviously talking about the health status and functioning and condition of people who aren't in the clinical system. We are talking about the entire population. There are obviously a lot of things that we can do in relationship to sharing information about people in clinical care. But just thinking more down the road -- I think someone yesterday someone mentioned, and today at least one of you mentioned, that there is a potential of getting a lot more depth of information than you can get in surveys. But, obviously, it's only for those people who have the electronic health record.
Have any of you thought about the potential for the personal health record, as tethered or not to the electronic health record, and the potential of that to collect information -- with, obviously, the person's consent -- that could perhaps, if not replace, then supplement the data that we get through surveys?
DR. LENERT: Marjorie, like NCHS, we embrace the idea of sentinel citizens and hope that people will start to contribute their personal health records to research, to population health, and to quality purposes for the country. It's a great thing.
But I would say, let's not miss the other infrastructure that we need here to really do public health. Vital registration has to be tied to clinical care. And you know what? Vital registration needs to be tied to master patient indexes for regions. This is an infrastructure piece. It doesn't come out of an EHR. It's something we need to build that says, when somebody is born, for the region that they are in, there is an MPI index created for that person, and when they die, we note that so that we can prevent fraud. These activities that are tied to the MPI for health-information exchange are an integral part of vital registration processes as well. They need to be all brought together.
MR. REYNOLDS: Jim Scanlon.
MR. J. SCANLON: First of all, I thank all the panel members for your insights.
Virtually all of the core services relating to public health that you are discussing today have the dimension of health-information exchange. What do you see, leading into 2011 and beyond -- you talked about various options for how to support that health-information exchange -- what do you see as the options? Are you viewing a public information network as sort of the vehicle for supporting the public health exchanges? Are you looking at commercial services? Are you looking at local health-information exchanges? What do you do when the areas where a local health-information exchange, in the context we have been discussing, is unlikely to form or be sustained?
DR. HACKER: In Kentucky we have an e-Health Board. Our lieutenant governor was instrumental in passing legislation four years ago establishing an e-Health Board in the state, which is trying to develop a statewide vision. We have released an RFP, which is now being reviewed, to respond as a respondent. We are on the road to contracting with an entity to establish electronic health exchange in Kentucky, basically a pipeline across the state, with Web-based portals, and different providers being able to connect into that, with, of course, all the security and privacy, and all the other bells and whistles that go along with that, with a long-term vision of being able, then, to tap in this pipeline statewide information flow.
Of course, this is still evolving. This is a complicated process. Certainly Indianapolis has a much more robust and more mature system, with Regenstrief, for several years. Marty has a great program in Minnesota. I think other states have the problem, essentially, that Kentucky has. We have a good vision. We know where we want to go. We need resources. The technology is not the issue. It's the relationship issues, the legal issues, and the funding issues. But we are on the road.
I think I'm answering your question there, Mr. Scanlon.
DR. LENERT: I wanted to throw another alternative for you that hasn't been considered enough. It's verticals for health-information exchange. We have seen tremendous success with Kaiser Permanente. One piece of health information we exchange is a vertical at Inland Northwest Health. They provide all the services to the participants, both hospital and individual EHRs for systems.
It would make a lot of sense to me that vendors would get into the business of hooking up all the systems that they put together. Why not have Wal-Mart hook up all the commission offices to that?
What do you need for that to work, if we were to do verticals? We again need to go back to this idea of state infrastructure -- master patient indexes, facility indexes, physician and other provider indexes -- that would allow the health-information exchanges to exist as a vertical, based on product lines or other things, without necessarily having regional coverage.
I think we would move to that. I think you would see a huge acceleration if you changed this to exchanges that were organized along similar products. Why not hook up all the labs together in one big system? These things might work much faster than trying to do this on a region-by-region basis. We don't know this with Visa anymore by regions. We don't buy things by region on the Internet. We buy them by functions.
MR. REYNOLDS: Tony Trenkle.
MR. TRENKLE: I have a question for the two state representatives. One of our incentive programs is going to be for the Medicaid incentive. I want to know how that would best support your public health goals in your states.
DR. HACKER: Thank you. In Kentucky, I'm part of a mega-cabinet, the cabinet for health and family services, Secretary Miller -- of course, governor this year. Within our cabinet we have Medicaid, mental health, public health, and social services. Our IT folks have gone through the process of interviewing each of our divisions, each of our functions within all four of those major groups to see where the commonality is, where the shared interests are.
Certainly, the Medicaid piece and the public health piece are self-evident. But there is also a lot of social service issues out there -- WIC, food stamps -- that cross over with USDA, with public health's involvement.
The mental health world is not adequately supported with the stimulus dollars right now as it stands, but we are still trying to find out how we make sure we don't leave it behind.
The vision we are trying to build for, then, and give it different clinical example -- the Department of Public Health and the Division of Medicaid Services within the cabinet are talking about local public health helping to provide disease management for people with diabetes who are Medicaid recipients. We have the Medicaid list of diabetics. We identify those. We have lots of severe problems, a lot of ER visits, a lot of hospitalizations, moderate, mild, different levels of interventions. The Public Health Department identifies and works with local physicians at the community level and the patient, with patient education and so forth.
Having that link of data and using it in the electronic health record, we will be much more effective in sharing information back and forth, from the clinician -- if they send the patient to an ophthalmologist, did they get to the ophthalmologist or did they get to the Health Department?
There are a lot of opportunities for improving clinical care more efficiently and doing it in a team way, while still maintaining a patient home, a medical home, with one person -- primarily, the primary care physician -- responsible. That's the vision we have in Kentucky, and we are pursuing that.
DR. LAVENTURE: In Minnesota Medicaid is part of a separate agency from the state Health Department, but we work very closely on the e-health initiative. We see the Medicaid portion as being a very important incentive to support activities in Minnesota. The Medicaid efforts and funding will support the Minnesota statewide health-information exchange, but also the outreach things, as Dr. Hacker mentioned as well. In Minnesota there are a couple initiatives related to health reform and medical homes. We see leveraging the efforts there to make sure that whatever we are doing in support of these incentives, it supports those initiatives also.
One of the challenges that I think we all have and that in Minnesota we face is the whole issue of assessment. How do we target resources to those who need it the most? We think that trying to leverage both the state grants and, hopefully, the Medicaid funding to make sure that we are doing assessment that allows us to differentiate adequately the subtleties of where people are at on a spectrum and what they need to be doing -- then we can target the grants and the loans which are going to be so important for Minnesota. Our loan program -- it doesn't have the money now, but we did, and it has been very popular -- has really been effective at leveraging to the rural and underserved community areas. That, we feel, is important to continue in the process.
So wherever we can more fine-tune the focus of the public efforts and make sure they are in concert with the private -- many great private efforts going on around the country, in Minnesota, around the state, are so essential to make sure that we can sort of use a scalpel and not a hammer to how we apply some of the resources, in particular the Medicaid funding.
MR. REYNOLDS: Dr. Blumenthal.
DR. BLUMENTHAL: I want to thank you all for very informative comments. I want to ask a fairly simple question, just to be sure that I understand it. It strikes me that most of what you are hoping to get as part of meaningful use depends on most clinicians participating in electronic information collection and exchange. Is that correct?
(Nods from panel)
And since we know that maybe 17 percent of physicians and 10 percent of hospitals are now IT-enabled in any sense, we are starting from a very basic level at the current time. To have the capacities that you are interested in for bidirectional exchange, does it become useful at some level of adoption or is it useful at every level of adoption? That's a question, in some ways, about timing. If you had 20 percent of clinicians or 30 percent of clinicians with electronic records in 2011, would it be as important to have bidirectional exchange at that point as it would be when you have 50 percent or 70 percent?
DR. HACKER: I will take the first stab and then I'll let Leslie respond.
In Kentucky, I would love to have 20 percent exchange of information between the providers in the private sector and the public sector. My concern is, if it's allowed to be developed without the exchange capabilities, we will end up with a system that never gets caught up, and we will end up with a private-sector system and private vendors having invested so much in technology and purchased so much in technology that when we say, "Well, now let's go with the exchange," they will say, "How do we afford that?"
If we are going to do it, let's do it right. We don't have to do it all at once, but let's make sure that what we do do has that functionality and capability of bidirectional exchange between those two sectors going forward.
DR. LAVENTURE: I would definitely agree with that. I think it's essential to get started. It may not be bidirectional everywhere, but those first bidirectional efforts are going to really, in Minnesota, lead the way to assure a more rapid adoption. Collaborating on best practices, developing some consensus and implementation guides will help, I think, really distribute that to others in the practice.
There's nothing like colleagues adopting something and seeing the value and demonstrating the value of that effort and contribution to accelerating the further adoption. So getting started is very important, and a reasonable step. It may not be bidirectional, first stage, of everything. Certainly something like the immunization makes a lot of sense.
DR. LENERT: What does that really mean to invert it? Physicians need to know what the history of the immunization is. It takes time. We are actually seeing that the first steps are incentives for adoption. They are not obstacles. They are things that will promote it. Bring up all the laboratories on to the Internet so people can order tests electronically and get them back. That's an incentive for meaningful use. It's part of your definition of what a functional electronic record is. That supports public health. When physicians have to track down the immunizations, that takes time. But if they can get that electronically, that's a feature of the system; it's not a barrier to it. It's something you put in right now.
If they could get public health alerts on their desktops by typing in the chief complaint -- I wish I had time to show you the demo of what we did in IHE -- that's a feature. It gives them an incentive for adoption.
So I think that calling what we are proposing a barrier to adoption is just backwards. We are proposing the types of tools that will make physicians want to use it, by basic communication with outside channels, which makes this a meaningful experience to them.
DR. HACKER: I'd like to add one short thing. I practiced primary-care pediatrics in southeast Kentucky for 18 years. I wish I had known then what I know now about what public health had to offer. Public health has a tremendous amount of data and information and resources that can facilitate private practice. We have not been doing that because we don't have the bidirectional exchange of information. It's too burdensome and too cumbersome.
I have made presentations to the e-Health Alliance for the NGA. I was asked to talk about what public health is getting out of playing in this arena, and my presentation was on what we can contribute to it. But until it's there, primary physicians may never appreciate the value public health can add to the health-care system, the efficiency of their practice, and the optimum outcome of the entire country.
MR. REYNOLDS: One clarification on that. Dr. Lenert, you mentioned the CCD. There are standards out there right now. Are those standards the packages that you would see some of that being done in or do we have to come up with new standards to do it?
DR. LENERT: To the extent that we can use existing standards, we should. CCD is a great thing. If I'm trying to figure out a flu investigation and whether this is striking people who have preexisting conditions or not, and I'm trying to contain it, what information do I need to be able to do that? I need to be able to retrieve the CCD from an EHR to do that and say what the patient's problem is and what the basic issues are that are going on there. That type of capability is the type of functioning that we are talking about.
The immunization registry activity -- there are existing standards for transmitting that information that are just fine. But the trouble is, our registries can't accept the information. They have old hardware. They have no money to upgrade. They have no infrastructure to do that. They would be happy to be able to work on that, but we need to be able to get them the resources to shift that.
So the basic standards work has been done. They are at the same level, I think, as the rest of the National Health Information Network.
MR. REYNOLDS: With that, I would like to thank the panel. A 15-minute break. We'll be back at 10:45. Thank you.
(Brief recess)
MR. REYNOLDS: Our next panel is going to be talking The Glide Path to Meaningful Use for 2011 and Beyond for Other Stakeholders -- a tall order. We are excited for them to begin.
Is there any special order you want to go in, other than what's on the agenda?
I would like to introduce Henry Chao first. Henry, thank you.
Agenda Item: Panel 8: The Glide Path to Meaningful Use for 2011 and Beyond for Other Stakeholders
DR. CHAO: Good morning. Thank you for inviting me to share my perspectives with this distinguished panel, as well as the Executive Subcommittee, Dr. Blumenthal, and the participants that are here in the room, as well as listening on the broadcast.
My name is Henry Chao. I am the chief technology officer for CMS. I'm here today to discussion some of the challenges, concerns, and recommendations regarding meaningful use from the policy and operations perspectives.
I have had the pleasure of being part of several teams of very committed and talented people, both internal and external to CMS, working to implement the Medicare prescription drug program. It was a tremendous challenge to integrate the systems and processes for Medicare, Medicaid programs, health plans, the pharmacy industry, and third parties. I would characterize the experience as extreme systems development. Even though it was an environment of intense and rapid development to meet the deadline, CMS strived to follow as closely as possible the discipline of the lifecycle process.
The vision and the major goal were both very clear for MMA: Implement a prescription drug program so that Medicare beneficiaries are able to receive prescription drug coverage by January 1, 2006. The requirements and business processes, however, were less clear, and the operational requirements and increases in scale and capacity were even less clear.
CMS had to quickly adapt to acclimate itself to key requirements for a 24-by-7-by-365 real-time pharmacy transaction processing model and to gain enough knowledge and understanding of the business process before it could choose best alternatives, given the requirements and constraints.
Public agencies often are given large, complex programs to implement, and generally we do not start from scratch because of the necessary integration and accounting for multiple touch points to legacy and new systems and processes. Almost never are there clean slates which CMS can start from, and I'm certain the rest of the world is not like that either.
Implementing the IT systems infrastructure for a complex program taught CMS many lessons about the operational realities of a massive initial program startup and the effort it takes to stabilize over a short period of time.
Also there were important lessons learned about implementing systems that have program performance metrics and data-quality checks built in as part of the design rather than extrapolating the metrics and quality checks from transactional production data. As we discuss and plan execution of the incentive program, we have to give some thought as to how to build in performance metrics and quality data indicators that help determine, at the lowest level, if there are operational performance issues and, at a higher level, if the data coming in and out of the systems are being processed and representing issues at the program-performance level.
What the incentive program under HITECH indicates to the health-care industry and public agencies is that great change is needed, and a starting place for this particular cycle of change begins with incentivizing physicians and hospitals to increase their adoption rate and meaningful use of EHRs and health-information technology in general. By comparison to what we all have been discussing about improving outcomes in health care and increasing effectiveness and efficiencies, meaningful use of EHRs represents connecting only a few pieces of a vastly larger health-care puzzle.
Incentives for the meaningful use of EHRs are a good, if not great, change for health care, but do they address a solution for the entire puzzle? I think not. The mere presence of HIT and EHRs alone do not speak to how meaningfully it's being used. We have to examine the entire puzzle and acknowledge and make all the pieces visible and reasonably accounted before we apply technology to business problems at the local and the national level.
But do we have to absolutely know everything up front? Not necessarily, but we do need to be smart and proactive in working on the present-day change, while always maintaining a line of sight to the larger vision and common goals. Strong and sustained leadership, common vision and goals will ensure that an initial good enough roadmap is defined, while helping to navigate to a more mature state over time. Without leadership and vision, it is exponentially more difficult to define and measure operational successes.
There are many perspectives from the major stakeholder domains. Patients, providers, payers, vendors, and the government all move at different rates of change, and health-care strategies, policies, business models, modes of interaction, requirements, and technical implementation of systems and data are still somewhat fragment, even within each domain.
There is an overabundance of variability and randomness in health care because, while we all share the desired outcomes, we choose to operationalize and attempt to make real the shared vision in our respective ways, which is somewhat understandable, and not necessarily viable as we desire to move towards greater levels of HIT adoption, interoperability, and meaningful use. Obviously, something is lost in translation if we mostly agree at the vision level and not at the implementation level.
What about the vision for security framework needed to support data exchanges and measures of reporting meaningful use at the HIE level, across HIEs, and nationwide? As major federal health-care agencies strive to create the workable and implementable security framework of public-to-public and private-to-public health-information exchanges, how can we come up with a workable model that keeps data secure yet borderless when needed, private but shared when the proper permissions and approvals are granted, while at the same time, comply with the Federal Information Security Management Act and other laws applicable to federal agencies?
Operationally, the authentication piece of identity management is daunting within the Medicare data-communications universe. How do we operationalize and create efficiencies in authentication identity management for an entire network of millions of individuals and entities needing to send information, receive information, request reports, provide reports, give consent, revoke consent, and the list of functions requiring the application of information security controls and applicable privacy protections as data exchanges scale in tandem with the vision for the interoperable health-care enterprise?
I'm going to talk a little bit about the operational challenges. I have a couple of slides for that, as well as policy.
What are the challenges associated with coordinating meaningful-use policy and operations across Medicare and Medicaid in 2011 and as requirements become more robust over time?
Number one is provider identification. When providers enroll and register for meaningful use in reporting and incentive payments, how do we know who they are?
It requires an increased level of granularity to account for relationships and association between the eligible professional, with practice settings in use of EHR, and yet maintain integrity of incentive payments.
Medicare and Medicaid must coordinate across the continuums of time, location, program participation in order to pay accurately.
Factor in provider workflow and patterns of practice to operate a measures collection incentive payment program that is unobtrusive.
Factor in the integration points between Medicare and Medicaid for policy and operations, accounting for overall industry readiness in coordinating robustness of meaningful use with stakeholders, an ongoing integration of legacy processes, operations, systems, and data with new requirements, incentivizing the use of EHRs as an initial step in payer programs to conduct greater levels of integration between the business process of paying bills and measuring quality.
Medicare and Medicaid IT capabilities and capacities require increased scaling to meet growth in data and data-integration requirements.
The entire health-care industry, along with Medicare and Medicaid, is faced with resource challenges in just maintaining day-to-day operations and planning for major changes already in the works. Examples are 5010 and ICD 10.
CMS must implement Recovery Act provisions without adversely affecting our current operations and programs. Attention to security and privacy rules, regulations, and requirements, and how they may change over time, is critical to successful operations and policy development.
Let's move into the most important policy issues associated with implementing the incentive program.
In closing, I ask the NCVHS, the Executive Subcommittee, to consider how sets of constructs and strategies may be established to focus on common aspects of health-care effectiveness and efficiency that will drastically reduce the risk of incongruous and inefficient health-care operations. The sets of constructs and strategies may be able to establish core common vision and goals, a common information architecture for health care, where participants know respective and collective starting and end points and the progression roadmap of converging on the common goals, account for respective needs and uniqueness for tolerance for rates of change, and identify the logical integration points between the administrative and clinical domains.
Through what I have discussed above, CMS and others will be able to adequately and successfully iterate towards an operational model and meet the intent and spirit of the incentive program, and also be poised to play its role in transforming health care.
Thank you.
MR. REYNOLDS: Thank you. Mr. Kelly.
MR. KELLY: Good morning. I'd like to thank the committee for this opportunity to speak here today on behalf to the payer community. May name is John Kelly. I'm the director of e-business architecture for Harvard Pilgrim Health Care in Wellesley, Massachusetts.
E-business architecture means that I'm the person responsible for getting machines to talk to machines. I do this so that those whose primary responsibility is to improve the general health and well-being of a large number of people can do that as effectively and efficiently as possible.
In truth, I have spent the past 10 years at Harvard Pilgrim and the payer community in New England dealing with the question of meaningful use and technology. In addition to my role at Harvard Pilgrim, I sit on the governing boards of NEHEN, the New England Health Care EDI Network and MA-SHARE, a subsidiary of the Massachusetts Health Data Consortium. In essence, MA-SHARE is a health-information exchange.
The first important thing to understand about the relationship between these two organizations is that they use the same train tracks to deliver two types of cargo. The network that delivers HIPAA transactions also handles discharge summaries, medication histories, and planned formularies. The provider and payer members at these collaborative organizations have jointly developed standards, software, and processes that meaningfully improve health-care delivery in our region.
The second important thing to understand about these organizations and meaningful use is that the tools and methods we use in New England can be replicated and scaled nationally. Harvard Pilgrim Health Care has been named for the fourth consecutive year by U.S. News and World Report and NCQA as the number-one commercial health plan in America.
Beyond just shameless self-promotion, I mention this because the capabilities we have developed and the culture we have sustained, which has helped us achieve that recognition, have very much to do with the information-exchange relationships we have established between Harvard Pilgrim and our network of providers. Most of those exchanges are electronic. Some are still paper-based. But our lesson learned is that we continue to strive as a process community to electronically share more and better information from all of our systems in order to get better outcomes for our members.
So meaningful use: Health care is a process community -- providers, members, patients, plans, vendors, facilities, on and on. As the delivery and financing of care has become increasingly complex, so has the web of interactions that take place between all the participants. It has been said that what makes a community is that you have to need each other, willingly or otherwise. The parties in our community certainly meet that criterion, despite the public perception to the contrary. What we need most from each other is information exchange.
As a group, we have and will continue to work steadily on administrative simplification. Delivery-system simplification is a much more challenging task. Can HIT really simplify the delivery system? The Federal Reserve might call it "decomplexification."
What I can say from my experience with technology is that even if I can't make a complex problem simple, I can use smart systems to mitigate risk at potential points of failure. From the payer point of view, this is meaningful use.
It's a cliché in other industries, but right information, right place, right time, is how smart systems improve outcomes. In health care, the stakes might be higher, but the principles are the same, from the patient's point of view.
So payer meaningful use has to be about outcomes. The validity of using claims data to measure quality is challenged on a regular basis. There are two reasons usually cited. Claims data is not a true representation of what happened in the exam room, since the codes are chosen for billing purposes. More importantly, claims data really can only be used to measure process, not outcomes.
I don't disagree with either of these assertions, but I will add my voice to others and say that right now it's the best data we have. I'll further add that a lot of good has come from smart people using claims data to develop programs that have improved the lives of literally millions of Americans, both sick and well. In addition, those programs have clearly shown that the meaning of quality differs very much depending on whether you are generally well or chronically ill. Claims data delivers fairly good process measures about whether or not healthy people are receiving the procedures that demonstrate and benchmark the best standards of care. Did they get a mammogram? Did they get a colonoscopy? Have they got a flu shot?
For chronically ill people, however, claims data at best determines only that the patient probably needs more attention than they are currently receiving.
Meaningful use of the EHR should necessarily lead to the compiling of data from multiple sources so that the quality and value of delivered care can be measured in a way that can be accepted as valid by all the constituencies. Without the HIT money right around the corner, one might take the position that it's time to thank the payers very much for all their good work on disease management, health and wellness programs, centers of excellence, development of discharge coordination programs, et cetera. Maybe between EHRs and medical home providers, we will be ready, willing, and able to reclaim the role of end-to-end caretaker. Maybe meaningful use will mean that payers aren't needed anymore, and therefore are no longer part of the process community.
I was a recent HIMS (phonetic) event in Chicago, where Dr. Wesley Wong, the regional vice president and national medical director for Anthem Blue Cross/Blue Shield, was speaking on a panel with two physicians who were advocating for a diminished role for payers once EHRs are broadly adopted. Dr. Wong observed that size matters. Payers can leverage economies of scale, capital investment in data-analysis systems, and access to large repositories of data to partner with providers in the best interests of the patient. Payers can also put together programs that cover patients across spectrums of care that are beyond the control of all but the largest integrated delivery systems.
I agree with Dr. Wong and believe that payers do have an important role to play in the future process community that is saturated with electronic medical records.
What our regional experience in New England has shown is that not only payers important suppliers to and consumers of a merged clinical and administrative data stream, payers and large delivery systems provide critical leadership in efforts to put together communication networks that will form the basis of regional health-information exchanges. Without the thought leadership and capital investment resulting from the collaboration of payers and large provider systems, the build-out of the communication grids necessary to support meaningful use will be difficult to sustain.
If you accept that payers do indeed have a role, then, with regard to meaningful use, payers need to be considered when determining the baseline certification standards to be established for EHR data-exchange capabilities. EHRs must be able to exchange data with everybody in the process community. I strongly support making CAQH/CORE level 1 and 2 vendor certification a requirement for subsidy consideration.
As an example, if you buy a PC today, there's a baseline assumption that you get Outlook loaded at the factory. Without it, you couldn't do email. Vendor EHR systems should have such a requirement.
The CAQH/CORE standards have been developed on a national basis, similar to the work done in New England with NEHEN and MA-SHARE. While the current data-content standards address only eligibility in claims status transactions, they will rapidly evolve with the goal of automating the entire payer-provider revenue cycle.
What we have proven at Harvard Pilgrim, however, is that we can use the current core connectivity protocols to deliver any payload the two trading partners agree to exchange. Using the core 2 Web services standard, which is secure and free, I could, with equal ease, exchange a HIPAA 270-271, a CCR document, a medication history, or even today's Red Sox box scores.
With such capability broadly available, the forces of market innovation will quickly yield desktop workflows that will create real value in the health-care delivery system. I can confidently assert this because it's a phenomenon we see repeated over and over when any industry fully embraces the Internet. John Wilkey (phonetic), formerly at Warrick (phonetic), now with SAP, once said the Internet is all about connectivity, collaboration, and self-service. That Internet effect begins the minute lots of point As can connect with lots of point Bs.
Important work being done by HCSFIG, IHC, and others to harmonize standards for semantic interoperability will be greatly accelerated if every machine comes ready to talk to each other.
I know that, to some degree, this committee is faced with refereeing a food fight between those that recommend setting the meaningful-use bar low so as to mitigate the risk of adoption failure and those that want the bar set high so as to ensure the realization of the ultimate goals of increased quality, lower cost, and overall value for every health-care dollar spent. I have also grown to believe over time that the fundamental principle of management science is to measure what's easy, not necessary what's important. Increasing quality, lowering cost, ensuring value are all functions of our ability to assess the collective outcomes of a series of complex processes. This is true in any industry.
Within our regional health-care community, as payers and providers, we have begun to view our transactions and efforts as a supply chain-integration challenge. We question the notion that everything we do is proprietary, that sharing information reduces our competitive advantage. We have found that the more we know about each other's interactions, internal and external, the better we can integrate our activities to jointly create an improved product for the customer we all share -- the patient.
As I understand the scope of this hearing, the committee is gathering input to support the requirement under the HIT stimulus program to define the term "meaningful use." Payers, as stakeholders in this discussion, believe that the standards should be established such that they strike a balance between aggressive support of outcomes management and broad-based provider adoption. We also support progressively higher standards being promulgated beyond the timeline of the recovery package. We believe strongly that there will be an inevitable convergence of clinical and administrative data in support of both quality and efficiency and as the PMS and ERH vendors consolidate their product offerings. The national payer community will support the CAQH/CORE efforts to supply better and more information that could be integrated with provider systems to automate manual processes, increase reliability of information, and drive down the costs associated with the coordination of responsibilities between payers and providers.
We look forward to the provider community reciprocating our commitment to information integration, under a wisely crafted definition of meaningful use.
As an active member of AHIP, Harvard Pilgrim and other member plans stand ready to work with providers as they take steps to take advantage of the new EHR incentive program. We will ensure that our quality-improvement initiatives, such as disease management, care coordination, quality management, value-based purchasing, and streamlined administrative processes, are complementary to Medicare incentives. We also stand ready to work with providers to leverage existing data sources, like claims and personal health records, to aid the transition to meaningful use of EHRs.
Again, I want to thank the committee for its time and for this valuable opportunity. I welcome any questions.
MR. REYNOLDS: Thank you. Anthony Rodgers, please.
MR. RODGERS: Thank you.
I'm here to represent the 61 million members that Medicaid serves today -- our vision for health information technology and how Medicaid is going to be a partner and a driver of health information technology for Medicaid networks.
The vision of Medicaid, I think, is summed up in this slide. We see this transformation of the Medicaid system as creating an informed and activated patient and a prepared and informed clinical team. We see that this productive interaction has been broken down by the lack of information, both on the patient side and on the clinical team side. Our goal is to use electronic health information, the ability to share and exchange data, to provide not only the clinical practice with information, but also to extend that to the patient through personal health records.
Our problem is that we don't have the infrastructure. We have to overcome the infrastructure barrier and get widespread adoption. We have to come up with rapid adoption strategies. The good part about being in Medicaid today is that we are probably the fastest-growing market share of beneficiaries and members, especially today, and so we have a lot of market leverage. That being said, we need to use the dollars and capital wisely and work with our public health partners and our private partners to overcome the infrastructure barrier. But Medicaid has a real stake in this.
Moving beyond the infrastructure barrier, we have to get to the level of performing systems and the competencies to create high-performing systems, systems that use the data to improve performance. But beyond that, once we can get to that level, there is the optimized system. The optimized system is the advanced medical practice or the advanced medical home. It's the broader use of technology for population health management and for community quality improvement.
So what we are talking about here is an upgrade of our health-care system. Right now we are at Health-Care System 1.0. We want to use the technology to upgrade the system, but that's going to take leadership. I'm not a technologist. I am a Medicaid director. My job is to lead and drive the technology in a direction that helps us upgrade the system to a system that can manage performance in a much more comprehensive and coordinated way. But again, the goal is to eventually get to an optimizing system, one that is self-correcting and is capable of extending itself into the community, of identifying root causes to disease, including environmental issues that are causing diseases, and to have an advanced medical-home capability at the physician office.
If we do that, we have to have certain performance measures. That means that we have to look at the wide range of performance measures that we overlay across the system, from medical-cost management to the manpower capabilities we have, the quality-improvement practices, the system-capacity issues, the operational issues, and then the compliance. This is our performance focus in Medicaid. We drive this through our provider communities and through our various networks of programs.
That's the challenge today. We believe that electronic health records will provide us with the tool to help drive those new performance levels. But it's going to require both public and private working together to assure widespread adoption.
So what's our glide path to meaningful use? At the first level, we know that we have to build infrastructure. Between now and 2011, one of our focuses as Medicaid -- and we have a multistate collaboration that's looking at these very issues -- is, how do we get the acquisition, installation, and adoption of electronic health records that have e-prescribing capabilities, computerized order entry, quality reporting capabilities, clinical decision support capabilities into our hospitals, et cetera? How do we get that information exchange, and then, from there, using that data, create the continuity-of-care documents that can be shared at the point of care?
This is going to require clinical data repositories. Right now many of us have administrative data repositories. We see ourselves as adding the clinical data and then providing that back in terms of a continuity-of-care record.
We also have to look at the practices. One of the realities is that, unlike Kaiser or the VA, we don't have many staff models. We have systems that are not integrated. We are trying to integrate them with the data. It's going to take a significant amount of effort to support those practices and reengineer the practice to be able to achieve that next level of performance management. So we realize that our role is going to be going out and supporting practices and helping them reengineer their processes.
The structural requirements: We need to develop a strategic plan, which many of the Medicaid programs are in the process of doing, that will establish the approaches, look at health-information infrastructure that is required, be able to do the environmental assessment of where we are today and where we need to go, and then we will be able to invest the dollars from the federal government and other dollars wisely.
We also have to then assist physicians with the acquisition, upgrade, and integration of their technology through health-information exchanges. Many of our Medicaid provider networks are in communities that are not well connected and that are going to need an infusion of capital to do this.
We also need to provide ongoing technical support and assure that physicians are able to get the technical support quickly, and, as I mentioned, practice reengineering.
Building the interfaces with public health and with our case-management systems and our managed-care organizations and the integration of patient decision support tools that we believe can be an extension of electronic health records -- this will raise health literacy in our population, which is so important to reducing health disparities.
It's also going to require creating new relationships between hospitals and providers. The one thing about Medicaid is that we can be the honest broker that facilitates, through contracts, through regulations, and other expectations, these relationships to form. As a payer, but also as a regulator, we have that ability to say, you shall our data, and you shall share it through these EHRs, et cetera. So we can create these partnerships and relationships.
Beyond that phase, we have the initial phase. I call it the burn-in phase. Every new system in a provider's office or a hospital has a burn-in phase. This is where systems fail. So we are going to have to make sure there is the technical support, beyond just the installation, providing the physicians and practices on how to use the system, how to integrate the system, to focus on productivity improvement. That's one thing we know we are going to lose. We will lose productivity improvement initially. We have to focus on bringing that productivity improvement back, by taking out some of the administrative hassle factors, whether it's prior authorization requirements, et cetera, and installing those and integrating those with the EHR.
Again, we also have to give the physicians a sense that we are there to help reduce risk of failure.
Connecting those systems to the external sources, including public health, including behavioral health, including all the other program sources -- the benefit of us doing this is that many of our physicians practice with private patients as well. As we develop these resources, they will be available to the private sector.
As we look at it, we are driving our physician practices to the first level of medical home, which is a physician that can e-prescribe, that can develop individual patient-care plans, that can do care coordination, by using the electronic health record, which has multiple sources of data, as the coordinating tool and integrating tool. In that way, we can use the physician practice as kind of the command center for how that patient's services will be provided.
Getting beyond that:
This diagram kind of illustrates the transparency that we want to see. We want to see that episode of care at every step along the way. We want to see the cost outcome of that episode of care.
To do this, we need a system that is self-correcting, that is cost- and quality-transparent, that is vertically and horizontally integrated, and that is patient-centered.
This is the diagram of Medical Home 2.0 after we have had that burn-in period.
Getting beyond that is the optimized system. We have to really look at all the areas where we need to optimize patient care and overcome, especially in chronic care management, the problems we see today.
This is a diagram of how we see integrating the entire system around our e-health, within our Medicaid organization.
The core competencies that Medicaid is going to need to have in the future are described on this slide.
Finally, the information technology environment, which will be much more complicated for Medicaid, but it's going to be important. That is going to get us to the advanced medical home.
This is a description of our optimized system:
Thank you.
MR. REYNOLDS: Thank you to all of you. I just want to ask one clarification, Mr. Rodgers. It lists you as the state of Arizona, but are you talking about all Medicaids.
MR. RODGERS: It is Arizona and all the other Medicaids. We have a multistate collaboration, and this is the vision that we have been working on over the last two years, since we got the Medicaid transformation grants.
MR. REYNOLDS: Okay, good. Thank you. First is Jeff Blair.
MR. BLAIR: Mr. Kelly, you introduced a new perspective here in terms of meaningful use -- that meaningful use also be applied for preventive care and wellness, for disease management. When we were listening to providers before, they mentioned priorities in terms of registries or interoperability or decision support or quality measurement. What are your priorities with respect to any of those items as a focus for meaningful use?
MR. KELLY: We all ready have a number of programs in place around things such as registries and clinical management and disease management. I think that our priority is that we be included in the information-sharing infrastructure, so that we can augment the data that we have with more, better data that is more reflective of what's happening in the clinical domain.
As I pointed out, we can do a pretty good job around healthy people, HEDIS measures, things like that -- yes, I know that you're 40 years old and you got your mammogram. That's great. What I can't do is tell if the chronically ill or cancer patients is getting everything that they really could be getting, things that we might cover as benefits that the doctor may be aware of, programs that we may have negotiated with a broad agency in the community that would serve as our population, and we need to be able to connect that up to the doctor. A lot of that we can't do if we don't have a fuller picture.
I think we have looked at this as a two-way street, though. What we have done within our community in Massachusetts, to a great degree, is to look at flopping all the information we have back and forth, so that we can really look at it as a patient-centered process and we can coordinate. We are getting better. It's a rather new idea, but the notion that a doctor might actually recommend to their patient, "You're going to get a call from Harvard Pilgrim, because they have this great discharge processing program that they have put together. I work with them. I talk with them. My nurse is in contact with them."
If we could do that, with our size and leverage, we could actually work as a team member with the physician as part of the process community.
MR. REYNOLDS: Tony Trenkle.
MR. TRENKLE: I have a question about readiness. This is for Henry and for Anthony.
Henry, you talked about the Part D program and the tradeoffs you needed to make because of a short deadline. My question for you is, with this short deadline we have here -- and there will be some tradeoffs operationally -- how can you avoid getting in a situation where you are just trying to make a deadline as opposed to building a platform for the future?
Tied to that, Anthony, what do you see in terms of the states' readiness for 2011 and the challenges you are going to need to face between now and 2011 to get there?
MR. CHAO: In my testimony I mentioned getting the good enough roadmap done, recognizing that it's connected to a larger vision of exactly what you have to trade off in order to make the deadline. You have to get the program started. You have to get docs taking on this cultural shift. They are getting some assistance and funding for use of EHRs, acquisition of HIT. But then what will they do with it? How do we measure it?
We need to first take care of making sure the program gets off the ground. So we can't make the program start out with the standard so high that we actually have a lot of folks sitting on the fence.
But we do have to know what's going to be happening in 2015. What does that vision look like? If you don't have that, then you don't know what you are trading off.
MR. RODGERS: In terms of Medicaid, we do have 50 different Medicaid programs that typically have 50 different solutions. What we are going to try to do is to work together to come up with some common solutions. Each state has a different type of network. In our state, we contract with community doctors and most of the hospitals. In some states, they have primarily FQHCs and safety-net hospitals. So the solutions will be slightly different on rapid adoption.
The biggest challenges that we are hearing are regulatory challenges in terms of the role that Medicaid will play. In some cases, the legislature and the governor have to facilitate that this role is appropriate. The incentive program that is anticipated, in some cases, will require legislative authority. So those are the kinds of things that will have to be done.
What we are doing to get ready for this is, we are encouraging states to start their strategic planning process. You want to get the return on investment. Our analysis of our return on investment, if we get widespread adoption by 2011, is $248 million. That's a lot of money to have on the table. This is from the level of duplication in the lab and x-rays and pharmacy, the overuse of emergency room, et cetera, that we believe this technology can help us resolve. That immediately comes back to the state. In terms of sustainability, I think that's where the state will see the benefit of supporting sustainability of these systems as well.
The FQACs, the federally qualified health centers, community clinics, IHS facilities, and some of the hospitals that we work with are already getting grants and have already moved with electronic health records. So we have that to build from. Some states are actually operating, as in the case in Arizona, their own health-information exchange. So we have that to build from as well.
MR. REYNOLDS: Judy Warren.
DR. WARREN: This is for Anthony Rodgers. You are the first person who has talked about patient education and family. What I would like for you to talk about is, what would meaningful use of the EHR look like for patient education and family? Where do you see that on the timeline that we are working with?
MR. RODGERS: I appreciate that. We see that as very important. Many of our programs are family-centered. Engaging our family members and empowering them to make choices that are beneficial to them -- and then to understand the impact of their choices, et cetera -- is so important.
We see the technology as helping us do that, if we use it as an extension of the electronic health record. Many of the beneficiaries today have connectivity to the Web. They don't have any product. We have to develop product for them. We have to bring that product so that it's e-learning products for diabetes education, et cetera.
But the other place where we have missed the opportunity is in the doctor's office. We have to provide an e-learning center for our patients in the clinic, in the doctor's office, very similar what you see with ATMs. They used to be all at the banks and then they were everywhere. We want to see e-learning opportunities for our patients everywhere that they can go.
The issue is, the younger people have no problem with the technology. Our older beneficiaries we are going to have to help overcome some of the initial fear of the technology. But once we do that, we believe they will be active users of e-learning, personal health record information, and become active communicators with their practitioner team.
DR. WARREN: The nurse in me just has to say, thank you for mentioning both of those things.
MR. REYNOLDS: As I mentioned to the group, it is all about nurses.
DR. WARREN: It is.
MR. REYNOLDS: Leslie?
DR. FRANCIS: There have been a number of comments made about the investment issues with respect to achieving meaningful use for underserved populations, the need for extra support, training, and so on of providers. I would like to ask all of you whether you see any substantive issues about how we should understand meaningful use. Should we be thinking about it differently with respect to the kinds of standards rather than just indicating that we need lots of money to make sure that it can happen?
I have in mind things like problems with mobile populations, with patient trust, questions about identification, fear of being identified, or other issues such as that. Do any of them play into how you think we ought to understand standards for meaningful use? Or are there other issues that you see?
MR. CHAO: I mentioned something about -- and Tony addressed this as well -- what will be the tradeoffs? Often, when we are kind of squeezed for time, to get something across the finish line we look at what we have and we reuse it. Often data is repurposed -- the classic example of using claims-based billing information to determine quality outcomes. It's what's available. It's common and it's accessible. We use it. It does have value.
But if you want to run a quality outcomes program, what data are you really after? The same goes for -- if you want to collect meaningful use, and meaningful use changes over time and depends on the practice setting and the type of provider, how do you define that set, instead of going after the low-hanging fruit, because the low-hanging fruit may not suffice to cover your needs three, four, five years out.
MR. KELLY: I think another way to look at the issue is that -- I have had a number of conversations with Marilyn Daly (phonetic), who is the CIO at Neighborhood Health Plan in Boston. They do some very innovative programs with the disadvantaged. The tenor of our conversation essentially is, if you can get good at managing a difficult population, then you should be able to do a crackerjack job with the rest of the world.
So I think the meaningful-use piece of the EHR that gets very relevant here is, how do you stop the ball from getting dropped? When patients move from place to place, setting to setting, physician to physician, program to program, especially programs out in the community where the responsibility and the baton is being handed off, that is a perfect opportunity for a smart system. That's not to say it's easy. But the tools that would be part of meaningful use in an ERH that really address the management of the continuum of care, I believe, would be the tools that you would exercise more effectively with a difficult population.
MR. RODGERS: In the Medicaid program, I think to maximize our return on investment at the provider level is to bend two curves -- the quality curve up and the cost curve down. What we want to do is take out the need for duplication in medical errors, et cetera, due to a lack of information. So meaningful use means that they have an integrated source of data that they can see at the point of care. The patient's health history has the information they need to provide the continuity of care.
Our role, as Medicaid, is to assure that that happens and that our investments assure that that happens. It's not just the physician's responsibility. The incentive program will help with that, but also the infrastructure we create around that.
At the patient level, it is the activation of the patient through the information, to communicate to the patient in a way that they would receive the communication through the telecommunication tools that we have today and through the technology and the Web tools that we have today, e-learning tools, so that they become activated and users of the system.
To validate that, we are asking our members what's valuable to them that they will use this in terms of their health care, in terms of their knowledge. What typically is valuable to them is knowing where to go, knowing what to do about specific situations, knowing where the community program is that they can go to to get assistance. That's where you start. That's what they want. Then you build form there into building their medical knowledge about diabetes or about asthma, et cetera. But you start, as a product, with the questions they have today. That's what we see using the Web for. We can do that. We have many systems that can actually project that out.
The other thing we have to capture the member is, we now have Web-based applications for Medicaid, where you can go online and do your own application, with someone's help. That's introducing them to the technology as well.
MR. REYNOLDS: Mike.
DR. FITZMAURICE: My question is, are there any additional or different basic functions needed for electronic health records to support the Medicaid program goals of CMS and of the states? For example, will existing standards and functions examined by CCHIT suffice to prepare EHR products for Medicaid provider meaningful use and for your vision of what HIT should do for Medicaid patients, as a starting place?
MR. RODGERS: From my perspective, the one thing that is not in the standards is the case-management activities that are so important, and the care-coordination activities. It can be developed and it can be formatted. But the health plans we contract with need to be part of that data flow, so that the physician knows what that case manager is managing that patient for.
We see that. We see EPSDT and the EPSDT forms and referral processes as part of the clinical decision support tools that we need to configure in, and then the eligibility. We see that Medicaid eligibility in the provider office or the point of care -- if we can have a simple way for people to keep themselves eligible in the provider's office, without hassle, and assist that member by giving them connectivity back in to change their information as necessary, this will help us.
Those are things that I think are important in terms of Medicaid.
MR. REYNOLDS: Jim Scanlon.
MR. J. SCANLON: I think several of you mentioned that, in essence, the same electronic rails that transmit claims and administrative data for HIPAA are the rails that will transmit clinical and electronic health record data. Presumably, the HIPAA information-exchange capability is ubiquitous, it's nationwide, in virtually every state and every locality, and has the capacity to do that? How would that fit in? How would it support health-information exchange capabilities for EHR?
MR. KELLY: I think -- and this is my view on it -- that a lot of the work we have done we did by looking at other industries, telecom and banking, to see how they integrated the supply chain.
If you look at the Internet, when most of us in this room started using it, it was a green screen and all you could do was push text. The important thing about it, though, is that it connected point A and point B. Just the connection of point A and point B led to an explosion of innovation, where now you are sending documents and then you are sending music downloads, then you are sending videos, and now you have Facebook and Web 2.0.
If you start by connecting -- and the lesson we learned with NEHEN is that just by starting with simple frame relays -- I'll even go back further than that. We started our connecting with Partners Healthcare, putting a tape of claims in a cab and sending it up the street. That led to collaboration, in and of itself.
Once you connect the points and the dots, innovation starts to happen.
I think the key thing, though, right now is that the Web is available everywhere. There is still competition, to some degree, around standards. What CORE and CAQH have done is to come up with a consensus model for creating Web services, which I'll call "haloed light." On the other side, you have the IHE, which has crafted a standard that they call XDS -- and the NHIN is essentially based on that -- for exchanging much more complicated payloads.
I have been brokering conversations between CAQH and the IHE, with the idea that we can't have two email standards; we have to have one. The standards that exist between CORE and CAQH were built with a lot of talk between the two of them. Our position right now is, a lot of the systems in the doctors' offices aren't ready to do the complicated exchange that is needed with the IHE standard, so starting with the CAQH standard, which is a fairly low lift for anyone with some decent technology tools -- it's free. It's just a standard. Then you start getting people connected.
The other thing that we have realized in our experience in the last 10 years is that there is a lot of low-hanging fruit around administrative transactions that save a lot of money. We have used that money to fund further projects and to increase capabilities, to the point where now we are using a gateway for ERX.
We eliminated WebMD. I pay no transaction charges for exchanging claims, eligibility, anything like that. I don't want to say no WebMD. Sorry about that
We believe that the Internet is about this intermediation; it's about connecting people point to point. We look at ERX right now. We have built the MA-SHARE ERX gateway. So I have the option -- and providers and vendor systems now have the option -- of not necessarily going through an intermediary to get my formulary, my medication history. That's all real hard dollars you can pull out of the system, just leveraging tools that exist today and using the Internet.
I getting to the real meaningful use, all it would take would be a commitment to have the IHE and CAQH sit down for six months and reconcile the standards and come up with essentially one email protocol for health care. We don't need to go develop anything more.
MR. RODGERS: I just want to point something out that's very interesting. We use the MA-SHARE open-source software as well. As we develop functionality or as Massachusetts develops functionality, we can share that. That's one of the benefits of having some common standards and some open-source software applications in terms of the health-information exchange.
MR. CHAO: If I heard portions of your question correctly, you were talking about whether we could reuse underlying infrastructure. I have to make the comment that a lot of the underlying infrastructure is point-to-point negotiated exchanges. What we need is to elevate that up to a level -- the Internet certainly helps a great bit. But then how do we impose the security model that I mentioned on top of that, so that the point-to-point is multi-point-to-point and the network takes care of that for you? How do we authenticate individuals on that entity? How do we trust domains?
Granted, NEHEN and lots of the health-information exchanges are beginning to have some answers for the larger question of how you do it at the nationwide level. But we have to be careful. From an infrastructure perspective, repurpose/reuse is great, but we would not use your water system for an oil pipeline. You know what I'm saying?
MR. KELLY: One follow-up on that. The standards that CAQH and IHE have put into place are based on the same kinds of standards that secure the banking industry and the telecom industry -- X.509 certificates, Web Services. Those are open, free, and all that is required to really make a national network would be some centralized authority -- one or two or three -- that manage certificates, the same way we have a couple of authorities that manage Internet domains.
MR. REYNOLDS: I've got a question. Mr. Rodgers, this committee has done a lot of work on privacy, data stewardship, secondary uses of health data, and so on. We have put forward this philosophy of a data steward.
You made a statement, "They share our data." It appears to me that that's a wonderful step towards you establishing that you guys are the data stewards of the information for your members, which then puts it at a significantly higher and more structured level than every single person that deals with it. Then you are responsible. If people are sharing your data, they are sharing it under your auspices, under your control, and under your evaluation, which I think is a lot of the philosophy that we were looking for and a lot of the thinking, which starts to take this.
If you have fewer entities that step up to be larger stewards, you start driving privacy and you start driving all these other things down, a little bit like Business Associates was originally set up under HIPAA. Is that a fair statement? I don't want to jump too far.
MR. RODGERS: That's a fair statement. It goes beyond that. For that purpose, we want to drive the research agenda. We want to be part of that priority-setting process that says, we would like to see more research in this area -- Native Americans and diabetes.
Within our state, we have an agreement where, to use our data, it has to have value back to society as a whole. What the electronic health record allows us to do is actually push down the consent and open up the door for individuals who want to participate to have other information and to allow community physicians at federally qualified health centers to participate in a much more efficient way, so that we can reduce the cycle time between discovery and practice. That's really our ultimate goal.
MR. REYNOLDS: Then you are also assuring the Native American diabetics or whomever you put in the population that you are the steward, and you know that wherever it goes, it's being handled appropriately.
MR. RODGERS: That is correct, because they are part of our advisory councils. They are the ones that give us that authority. We report back to them.
MR. REYNOLDS: Again, that starts to bridge the gap between the individual and the common good.
Jorge, you have a comment.
DR. FERRER: I have a question for Mr. Chao. Today health services research is predominantly done off Medicare/Medicaid claims-based analyses. How do you foresee, as we move into deeper clinical information, CMS facilitating health services research in that area? How would you envision that? Payment reform really can't happen without having the ability to go to that level. Today you provide a great service by providing that claims base. It's limited, but it at least provides a good service to the health researcher. Can you comment on that?
MR. CHAO: There is obviously a need for a business vision that drives the technology vision. I mentioned that the challenge for large operations, such as Medicare, which processes so many claims and eligibility transactions and is trying to send terabytes of data to hundreds of partners -- you can't really disturb that overnight. You are threatening a program that is in operations for 45 million individuals on the Medicare side, probably 12 million to 15 million people that are duly eligible on the Medicaid side. So you have to gradually do that. The business vision has to be created before we apply the technology vision.
I don't know if I'm exactly answering your question. We have to wait and see. We have to discuss this among HRSA and all the other operating divisions under HHS, about how to work on this together under the leadership of the new Secretary and the Office of the National Coordinator.
MR. REYNOLDS: Okay, I am going to have to end the panel now. I thank you very much. Excellent job.
We are going to move to open public comments. I would like you to queue up back there, please, since we only have two minutes per. First is, I believe -- I believe Jack Callahan was the one who came up yesterday and told me he was not here today. Unless he stands up, I will take that as a fact. We have Michael McCoy, Michael Mitzak (phonetic), and Arthur Gasch.
Again, it's just like yesterday. It's two minutes per, please. We would like you to stick with that, if you would. Mr. McCoy.
Agenda Item: Open Public Comments
DR. MCCOY: Thank you, Dr. Blumenthal, distinguished committee members. I'm Dr. Michael McCoy. I'm an obstetrician by training. I'm the co-chair of the IG Patient Care Coordination Committee. I have served on various things for both HITSP and for CCHIT.
This committee has heard a lot about meaningful use over the last couple of days, but little has been said yet on hospital departmental systems and how meaningful applies to those, as opposed to the enterprise-wide systems. Often the departmental systems are from different vendors than those of the enterprise-wide. Thus, certification currently exists for enterprise-wide systems, but not for the departmental systems, except for the emergency department.
How will the meaningful use apply to people who are using those other systems? If there is going to be a requirement for individual certification for those departmental systems, how does that go about coordinating with the enterprise systems?
I think the interoperability between the departmental systems and the primary system needs to be considered. That's something where direction needs to be given to address those concerns. It causes fear, uncertainty, and doubt in both the hospital vendors and the users as to what's going to happen. So I would urge the committee to consider those aspects.
Thank you.
MR. REYNOLDS: Thank you.
Mr. Mitzak.
(Not in attendance)
MR. REYNOLDS: Then we will go on to Arthur Gasch.
MR. GASCH: Hi. I'm Arthur Gash, with Medical Strategic Planning. We do surveys of electronic medical records companies.
As I was thinking about meaningful use, it occurred to me, meaningful use for whom? In the last day or two, the notion of meaningful use for the physician, meaningful use for the patient came up. It was noted several years ago, when Dr. Baylor was the coordinator, that 50 percent of the systems deployed failed. It has been noted more recently by work at the Institute -- Tupper (phonetic) -- that 20 percent fail to live up to the expectations of those who deploy them. A recent article in the New England Journal noted that the reason for this is that many of the systems are very difficult to operate and, in some cases, too complex.
I think if we are going to have the second part of the national health agenda, which puts 46 million people under medical coverage, they have to be able to see somebody, and most of the small practices that have not yet adopted are closed to new patient admits.
So the question becomes -- I think the first meaningful-use requirement is, how does it become meaningful to the physician who uses it and improves his office efficiency so that we create the capacity necessary to see these new patients under the system?
I think we could be careful not to leverage so many third-party requirements onto the meaningful-use agenda that we balk down the actual performance of health-care delivery by the physician to the patient in their office, by encumbering the system with all of these other requirements.
So I would just encourage you to take that into consideration.
I think also that standards exist already. For example, a number of vendors point to the fact that they could accept LOINC codes. We could enforce those with CLIA regulations and then get the laboratories to support them.
Thank you.
MR. REYNOLDS: Thank you. I'll apologize ahead of time for not pronouncing the next name correctly. If you would please introduce yourself --
MR. CALLAHAN: I'm the first one you called, Mr. Chairman.
MR. REYNOLDS: Jack Callahan?
MR. CALLAHAN: Yes, sir.
MR. REYNOLDS: Then you are ready to go.
MR. REYNOLDS: Thank you very much. Dr. Blumenthal, Chairman Reynolds, members of the committee, my name is Jack Callahan. I'm executive vice president with SRS in Montvale, New Jersey. We have over 4,000 high-volume specialists as EMR clients.
First, I would like to applaud this committee on behalf of all of us health-care consumers and providers, who are depending upon you to get all the perfect answers for all stakeholders.
Our focus yesterday and today has been on outcomes, meaningful use as to what data is to gathered, reported on, networked, how it's all brought together to show evidence of meaningful use. We have also heard how the data will improve our health care, reduce our costs, and result in fewer medical errors, and even enable patient rights and access to the data -- all the hard work of many people, to be applauded.
But what about all this data being gathered? What about those who have to gather it -- the physicians, the providers themselves, the ones that give the care? The more discrete data we ask doctors to gather, the more burden we are placing on them in their jobs as physicians and caregivers at the point of care. We have to keep in mind the physicians themselves.
A couple of questions need to be answered: Why is the EMR adoption rate, as Dr. Blumenthal mentioned earlier today and as evidenced in the New England Journal of Medicine -- why is it so low? Why has the adoption rate of EMR up to this point had such a high failure rate -- in many studies, in excess of 50 percent. Why is that?
All the hard work of this group in bringing these incentives into play -- will that change that whole curve going forward?
Providers, up to this point, have used as the number-one reason for not adopting it as, it's too costly. That has two components: one, the upfront cost; and two, the ongoing productivity costs operationally going forward, in protracted training, long learning curves, fewer patients that they have to see. Many doctors suffer this, especially the ones we serve, the high-volume doctors.
I urge the committee not to ignore the provider in this equation in defining what meaningful use is and the data that is to be gathered.
I make two recommendations:
One, that you include in your thinking usability principles. I know that Dr. Blumenthal has already said he is including this. Include metrics for usability. That allows a comparative sense between methods of gathering it.
Second, I urge the committee to allow providers to choose how they provide the data required. You are measuring outcomes, and those outcomes are the essential part of the equation. Allow doctors to choose.
Look for answers to those "why" questions.
MR. REYNOLDS: Thank you, sir.
Mr. Jagannathan, please. Then it will be Dana Green, then Sandra Fuller.
MS. GREEN: If I am cutting in line, I apologize. My name is Dana Green. I'm with the Altarum Institute.
One of the things that you are intending to do is to predict the future, and so I'm going to quote Dr. Phil, the most accurate predictor of the future, as well as past experience. There are a couple from my past that I would like to bring up. I am no longer the project director for the HSPIC (phonetic) project in Michigan. However, from my experience wit the HSPIC project -- no one has mentioned it before -- there is significant overlap to what the HSPIC project did. I found them handy and cheap. So I thought I would at least mention that this is a resource that you may want to consider for field staff in the field. They have already been designated by their governors. They already have steering committees that have multi-stakeholders. They have already canvassed the projects that are in their state. They have developed proposals and implementation plans. So they are handy and very inexpensive.
The other thing from my past is that I used to work for Disney, and one of the things I found out about Disney was that they are great at traffic control. How they do it is, for example, to get in the Magic Kingdom, there is no way to just drive up, park, and walk in. You have to go through the Disney transportation center in order to get to the Magic Kingdom. That is how they do crowd control getting into the parks.
Your definition of meaningful use will define how much stress you put on the system. It's the only thing you have. I would recommend that you be ambitious. What we have learned from HIPAA is that you have the right to change your mind. If you need to move deadlines back, you have that option as well.
MR. REYNOLDS: Thank you.
Sandra Fuller, then Gavin Lew.
MS. FULLER: Thank you, Mr. Reynolds. On behalf of the American Health Information Management Association and its 63,000 members, I thank you for your attention to this topic and welcome Dr. Blumenthal to his job and thank not only the committee members, but the committee staff and the staff of the Office of the National Coordinator, who have done a fine job and will continue to be challenged by this work.
We will submit written testimony, but I would like to make three main points for the committee's consideration.
First, we clearly agree that this is no simple IT project. The goal of this is to improve health care. Like our vision, we believe in quality health care through quality information.
We challenge you first with the adoption of certified standards-based electronic health records, widespread adoption. Somebody said yesterday, let's raise all ships. We think that's the first goal.
Second, consistency. We would challenge you that the definition of meaningful use has to be certain that it doesn't have the unintended consequences of incenting behavior based on a single payer, a single provider type, or a single type of patient. No patients left on the shelf, please.
Third, an incremental path that takes two forks, if you will. The first is to maximize the benefits of electronic health records for continuity and coordination of care. There are some low-hanging-fruit projects there that we have heard about. But second, we should also continue to work on the relationship between the primary use of data and the secondary use for quality reporting, public health reporting. In fact, when you get to the fork in the road, we have to take both paths.
Thank you very much.
MR. REYNOLDS: Thank you. We will take Gavin Lew, and then we will just have two left, as the sign-up is right now, for this afternoon. Mr. Lew.
MR. LEW: I'd like to thank the Committee for this opportunity. My name is Gavin Lew, from User Centric.
I and my colleagues are behavioral scientists, expert in human interaction with technology. I share the vision of improved quality of care through improvement in HIT. I agree with the panelists on the future benefits. However, the key assumption is that clinicians actually embrace EHRs. To quote Dr. Blumenthal in the NEJM article, many certified EHRs are neither user-friendly nor designed to meet HITECH's ambitious goal of improving quality and efficiency in the health-care system.
My question to this distinguished committee is simple: Shouldn't meaningful use address the problem that poor usability is one of the key barriers to EHR adoption? In short, if doctors don't use it, it doesn't work. Poor usability and adoption issues are universally recognized points today. How will this change if meaningful use and certification do not address usability?
Yes, clinicians can be forced to use EHRs. But if EHRs continue to be difficult to use, the result will only be a bare minimum interaction. Even worse, poor design induces error. When it comes to improving quality of care, it starts with the clinician experience with EHRs. Meaningful use must have a behavioral component, one that measures clinical interactions. EHR clinician performance can be measured. Outcomes can be reliably reported. Criteria can be established. Setting criteria by clinical specialty and measuring post-implementation will make user performance relevant and drive technical innovation from vendors.
Technical aspects have been broadly and thoroughly discussed. I urge the subcommittee to understand the human aspects of meaningful use, how to measure how well EHRs really work to make the stimulus package realize the benefits now and for the future.
A more descriptive written statement will be submitted by User Centric to provide a roadmap for implementing these measures.
Thank you.
MR. REYNOLDS: Thank you. With that, we are adjourned until 1:15. Thank you, everybody.
(Whereupon, a luncheon recess was taken at 12:00 p.m.)
MR. REYNOLDS: We are back for this afternoon. Panel number 9, Certification and Meaningful Use. I would like to know what order you would like to go. I would like to introduce Dr. Leavitt.
Agenda Item: Panel 9: Certification and Meaningful Use - EHR Product Certification
DR. LEAVITT: Thank you, Chairman Reynolds, members of the committee, and distinguished guest. My name is Mark Leavitt, I am chair of the Certification Commission for Health Information Technology, or CCHIT. I hope you will spell it and not pronounce it.
We are a non-profit 501C3. I have a new side job as a lightening rod and it singed all the hair on my head.
Over the past three and a half years, we have worked under a contract from ONC to develop and validate certification criteria and inspection processes for EHRs and other forms of health IT. We have been the federally recognized certifying body since 2006.
I found your questions thoughtful and provocative. So instead of delivering a standard presentation, I'm going to answer your questions. In the course of that, I'll share some information. If you want more, the end of the slide presentation has an appendix with some supporting details. So let's go right into your questions.
The first was, what role does certification play in meaningful use? Let's look at the usual role of certification. The usual role of certification is to reduce risk, whether it's electrical appliances, cars, aircraft. Certification ensures that some type of technology complies with some type of standard.
Certifying bodies may be in the private sector, as in the case of UL Laboratories, purely government, such as the FAA, which certifies planes, or kind of hybrids, quasi-governmental. The funding model varies. That's probably not all that important. But what is important is that certification is very valuable when the risks are high or when the technology buyer can't really evaluate the product's quality and suitability.
Think of the one-doctor office and think of EHRs, and I think you will all agree that that applies. Think of the risks that we are taking here of making a mistake -- not just patient safety, but just the wrong route to adoption. The risks are high.
So I think certification has an important role to play.
However, even though during our initial years certification served as a confidence builder for providers who were reluctant to buy EHRs, they were worried they wouldn't have the functionality or the security or the interoperability that was needed. That role is changing, because financial incentives started emerging. Before the Recovery Act, we found $700 million in incentives, and we were already feeling that increased responsibility. But the Recovery Act has taken us to a whole new level. Now we are all engaged in a very challenging journey together.
You have seen 40 different slides about how to describe the roadmap. I'm not going to go through it. I think of it as three parallel tracks rather than sequential. Adopting ERHs is one, developing and sustaining health-information exchange is the second, and ultimately the system reform and transformation. But I don't think they are sequential. I think they all three go on at once and they are interdependent.
That means certification has, I think, a much broader strategic role. In this case, we need to be a coupling mechanism between policies, which are constantly advancing as we learn, and the real-world marketplace of developers and people buying them and people implementing them. That very messy world often doesn't interface well with policy proclamations. So we are a coupling mechanism. We send signals ahead of time. We figure how you ramp into things and how you motivate the marketplace to move.
Let's go to the second question: What are the strengths and weaknesses of current CCHIT processes, and how do these need to change under the Recovery Act?
I'm going to start with the strengths. Since we were founded, we have recognized that a critical factor for success was credibility among stakeholders. We are a nonprofit, so we don't have stock. That's our stock-in-trade. If we are not trusted and credible, then we really can't deliver what's needed. So we developed both our criteria-development processes and our inspection processes to serve that need. The development is transparent. It is done with volunteers. There are three cycles of public comment. There is pilot testing.
The inspection process, which is firewalled from the development processes, uses juror volunteers and staff, but no one is permitted to have any financial interest in the vendor, meeting the federal standards, basically, for non-conflict of interest.
As a result, we have had really great industry engagement. This year we had 640 volunteers come to us. That was probably two to three times as many as we can take. That has increased every year. The 10 largest professional organizations have all endorsed CCHIT.
We have strong compliance by vendors. We started with ambulatory EHRs, followed by hospital or inpatient EHRs. But once it became clear that certification was having an impact, other domains of care came to us, urging us to bring certification to them, so they could see more rapid adoption. So we added emergency department, child health, cardiovascular medicine. This year we will launch personal health record certification and standalone e-prescribing certification. For next year, we have committed to eight new programs. Long-term care is one of those, advanced clinical decision support, and a half a dozen others.
We have also earned the confidence of payers. Put aside the Recovery Act for a second. Certification by CCHIT was the gating mechanism for incentives in the Medicare EHR demo, the Stark law safe harbor, several state EHR initiatives, private-sector pay-for-performance programs, like Bridges to Excellence, and even discounts on liability insurance, malpractice insurance, for physicians.
All of those accomplishments are a credit to the work of our volunteers.
Nevertheless, it's all the more reason why we have to listen to our critics as well and strive to improve our processes. Let's just have a frank discussion about it.
I think the number-one issue is the one that Dr. Classen talked about on the first day: Buying a certified EHR is not a guarantee of successful implementation. There is a wide variation in success. Although we got started just by testing products, we know now that we need to go beyond it. The Commission has approved moving beyond that. We are already beginning investigation into how to measure and rate things like usability, correct implementation and training, literally all the way to success rates and user satisfaction and patient satisfaction as part of that, once products are in the field. I think the Recovery Act gives us the context to get out of our box and start asking these harder questions and measuring them and reporting and rating them.
The second issue -- you heard one speaker yesterday from Open Source. Their licensing model is really different from the licensing model of commercial products, and our policies are not really fully compatible. We think we can change them. We see what the issues are. They have to do with versioning. With a commercial product, you want a stable version. With Open Source, the more dynamic and more quickly the versions are changing, probably the more viable the product. We can fix that, and we are committed to fixing that and updating our policies this summer.
We also know that our fees can be a barrier for, for example, nonprofit developers, for EHRs that serve vulnerable populations. We are looking for some grant funding so that we can use that to defray the cost of certification for them.
Finally, what about the next Thomas Edison? Maybe they are not working in their garage; maybe they are working in their consultation room developing the next great breakthrough. We don't want to be the barrier. We think we can find a way to certify experimental EHRs that are going to be used in a few sites or one site that aren't yet broadly commercially sold, and have lower cost and fewer hurdles for them to jump over, to let those innovations be tested in the real world.
Just to sum up our answer to your question, we understand the higher expectations. We will rise to meet them on all of these parameters -- accountability, transparency, focus, scale, and speed.
Let me get to the next question: How should certification work in 2011? How should it develop over time?
Let's talk about how we get to defining meaningful use. Now I'm going to take off my CCHIT hat and just keep on my EHR wounded-veteran hat.
First of all, I feel strongly that meaningful use should be documented electronically. Of all the places not to use paper processes, this is one of them. The risk of fraud would be high -- not to mention it sort of proves that you are not really trying to get electronic when you file paper forms to prove that you have computers. Let's let that one go.
But here's something that may be news to some of you. Current certified EHRs are inherently capable of collecting many meaningful-use measures. They are required to have an audit trail. It records every transaction -- a log-in, a view of a chart, a prescription, a review of a lab result. If these audit trails can be used properly, I think we are halfway there.
So the concept I'm suggesting is that all certified EHRs in the near future, and probably for 2011, would need to be able to register their existence with an authority, saying, "Here I am. This is this EHR, and it's certified," generate and display a meaningful-use dashboard to the user, display that basically on demand -- this is Carolyn Clancy's "press F7" -- and then sign and securely submit the dashboard statistics to a designated entity when they apply for their incentives, and retain an audit trail for verification.
The dashboard would look something like the dashboards you see in many other cases. It would have different practice models, different years. The provider sees it. The goal is to get it up to green, get your money.
I won't go into this slide about timeline concepts. I agree with the two-year cycle that the vendors -- I think that's probably the right pulse.
I'll stop there and look forward to your questions.
MR. REYNOLDS: Thank you. Mr. Ratliff.
MR. RATLIFF: Thank you very much. I want to thank the NCVHS Executive Subcommittee for the opportunity to testify today.
I am Rick Ratliff. I'm the president of the Virginia division of Surescripts. It is Surescripts now. As most of you, we are the result of a merger of Surescripts and RxHub in 2008. I am also a commissioner on the Certification Commission for Health Information Technology and work with Mark and his team on the certification processes he just described.
My comments today are going to focus on Surescripts' practical experience with regard to electronic prescribing and the application of this experience to standards development, certification, and the definition of meaningful use.
As most of you know, we do operate the country's largest electronic prescribing network. It's used every day by thousands of physicians, prescribers, pharmacies, and other stakeholders in the health-care system, across all 50 states, including the District of Columbia. Our network connects prescribers, through their choice of electronic prescribing software, to the nation's major chain pharmacies, independent pharmacies, and the leading payers across the country. The network is the backbone that facilitates the electronic prescribing. We enable this by providing prescribers with secure access to prescription benefit information, prescription history information, and by allowing prescribers to electronically transmit prescriptions to the patient's choice of pharmacy.
What's real important and relates to the comments that you just heard from Mark Leavitt is that in order to utilize the network, whether you are a pharmacy, a prescriber, a payer, you must use software that is certified on our network. That certification is a process that we manage, that we have been managing for a number of years, that is to ensure, as Mark was describing, that the different systems connected to our network adhere to the industry standards. So everything that we do within the network is all based on industry standards. I'll talk a little bit more about that.
I want to emphasize that from our perspective -- and this really starts to get to where we would like to see the definition of meaningful use go -- electronic prescribing includes not just the usage of a standalone electronic prescribing system or electronic medical records system, but the use of prescription benefit information, prescription history information, and, again, the routing of that prescription to the patient's choice of pharmacy, whether that is a retail pharmacy or a mail-order pharmacy.
Essentially, electronic prescribing occurs when a prescriber uses a certified system that allows them to access the patient's prescription benefit information, with the patient's consent, allows them to access prescription history information -- these are prescriptions that could have been written by other physicians for that patient -- to help enhance clinical decision support, and then electronically route that prescription to the patient's choice of pharmacy.
It's only through the use of these three components within the standalone system or an EMR with electronic prescribing that you can realize the full benefits of electronic prescribing. I'll comment on that in a minute.
First, just to hit some highlights. I know there have been some comments about where we are with electronic prescribing, what the status is. We released a report last week called The 2008 National Progress Report on E-Prescribing. This will provide additional details on exactly where we are with these different services I just described. But just to hit, again, some of the highlights:
We have seen significant momentum. From 2007 to 2008, we doubled the volume of the messages I just mentioned. Again, these are standard NCPDP SCRIPT messages, NCX12 messages over our network -- over 240 million messages processed in 2008.
We also have seen a tremendous growth, due to many individuals' efforts in this room, and working collaboratively with our organization in pharmacies and payers, in driving physician growth in the network. We currently have almost 110,000 physicians that are actively prescribing on the network. It's important to note that these are physicians that are using electronic medical records, that have been certified for electronic prescribing to the NCPDP SCRIPT standards, working in conjunction with CCHIT to certify those according to the e-prescribing interoperability requirements, et cetera.
About 70 percent of those 110,000 users are using an electronic medical record system. The other 30 percent are using a standalone electronic prescribing system. We continue to see growth across both.
Electronic requests for prescription benefit information grew from 37 million in 2007 to 78 million in 2008. Prescription histories delivered to prescribers went from 6 million to 16 million in 2008. Prescriptions routed electronically went from 29 million in 2007 to 68 million in 2008. We have already processed over 134 million transactions in the first quarter of this year, and we have added over 30,000 physicians in the first quarter of this year to the network. Again, these are real users of electronic prescribing and real users of systems that have been certified.
The reason for the success has to do with a number of different factors that are coming into alignment, such as the broad-based support for industry standards. The Medicare Modernization Act put forth the NCPDP SCRIPT standard, and we are certifying against that, as is CCHIT. We are seeing the new incentives start to have an impact as well. The Medicare incentives have really driven a significant amount of activity towards the end of 2008 and particularly, as I have just described, in 2009.
We have over 130 very large physician groups that are primarily associated with large health-care systems across the country that are connected to the Surescripts network. I'm going to name a few of them -- I might leave out a couple, so please excuse me -- just so you have some idea of the reality of where we are: Sutter Healthcare system, Cleveland Clinic, Geisinger, UPMC, Baylor, Partners Healthcare system, the Care Group, et cetera, et cetera. They all have users on our network. They are electronically prescribing to pharmacies, sending prescriptions fully electronically. Those are primarily electronic medical record systems.
In addition, we are seeing applications from Allscripts, their e-prescribed application, DrFirst, RxNTm MedPlus, and others that have standalone systems. We are still seeing growth in that area. I think a lot of that has to do with the fact that CCHIT has moving towards certification of standalone systems and the fact that small and medium-sized practices need a way to easily get into this area.
In our experience, the underpinning of all the success in building this national, open, and neutral network is about working with these agreed-upon and non-proprietary standards and moving through the certification process. Once the standards are in place, we have to have a proven certification process, as was described by Mark.
I'll move on, in the interest of time.
We certify applications, as I mentioned, across all three different services. We have completed over 450 certifications and recertifications of prescriber, pharmacy, and PBM systems. Nineteen of these systems are standalone systems, 165 EMR products. I mention that, again, because of the fact that we are moving into the electronic medical record world.
There are 135 products that are in active use, with users across the network. So we are seeing tremendous growth in this area.
I would say the certification process has been a very collaborative process. It's very critical that it is a collaborative process, as Mark describes. We have to continue in that direction. Certification is important, but I would have to say certification is not enough. That is what Mark was alluding to.
Of the 135 products that actually have active users in our network, many of those are EMR systems, as I described. They are not all certified on prescription benefit, prescription history, and prescription routing. Even with those that are, there is not consistent deployment of all those services into the user base, for a variety of reasons that are outlined in the written testimony.
I think it's very important, when you think about meaningful use, that it is the use of those three services. If you are not looking at benefit data and the history data and the routing component, you don't get the full advantages of electronic prescribing -- selecting the right economical alternative, the right clinical alternative, and then routing efficiently to the patient's choice of pharmacy. I think it's extremely important that certification considers that, but you have to go to go a step further, as Mark was describing, and look at implementation, deployment, look at support, and look at utilization.
If you look at our national progress report on electronic prescribing, you will see some of the details of the real utilization. We can measure usage. We can measure meaningful use.
I'll leave it at that.
MR. REYNOLDS: Thank you. Next, Dr. Lansky.
DR. LANSKY: Thank you, Chairman Reynolds. Thank you, Dr. Blumenthal. I appreciate the opportunity to be with you.
I do have some slides.
I have been with you before. This time I have the pleasure to be with you on behalf of purchasers of health care. I work for the Pacific Business Group on Health, which is a large coalition of employers and public purchasers in California and the nation. I want to share with you our views of meaningful use and how that might pertain to the question of certification and validation that we are discussing today.
I appreciate the opportunity in these hearings to bring other perspectives -- that we don't go too quickly to a narrow definition of EHR or of meaningful use or of qualified EHR. I hope you will take this opportunity to think more broadly about the health system as a whole and our objectives.
Our members certainly are concerned about improving health and improving affordability of health care. Meaningful use ultimately needs to serve those two objectives to be of particular value to the purchasers in the country. While we certainly support the broad applications that we are talking about in these two days, I want to raise some other issues in the course of my presentation today and really highlight the opportunity to evaluate whether the products and services that we are talking about deliver health and affordability improvements to the health-care system.
Purchasers' expectations for the clinical IT infrastructure that we are all laboring to build come from a slightly a different angle than simply and only improving bedside care. Purchasers have a particular interest in some of the drivers of cost in the health-care system and of overuse and misuse in the health-care system, which we have all talked about for 10 years. Here may be an opportunity to take some action, by evaluating and monitoring new technologies, planning and evaluating changes in payment systems, changes in workflow, changes in practice design.
We are particularly interested right now in implementing recognition and payment programs that reward high-quality care, whether through pay-for-performance, pay-for-outcomes, bridges of excellence in other programs. We are looking at how to measure the performance of individual providers, how to create financial rewards for high performers, how to aggregate data that may be derived from the systems we are talking about here to support bundling and other payment mechanisms.
Ultimately, we want to support a flow of information that, as we said before, is not only two-way, but multi-way across the network. We want feedback to clinicians, to patients, to families. We want to capture data from patients in terms of symptoms and outcomes, functional status, and so on, and fully engage all of us in our health care and our health.
So we have fairly broad objectives for this health-care system. I would just highlight that the provisions of the stimulus bill are primarily -- Congress really gave specific attention to these public-interest uses of the information. The definition of meaningful use, in my mind, has to reflect Congress' insistence that information sharing and clinical quality reporting are primary reasons that public money is being invested in what is essentially a private good. So for us to capture the value as purchasers, both public and private, we want to see that meaningful uses are being demonstrated along these lines.
In the short term, that means that we have to be able to measure and document individual physician performance. We have to be very attentive to the differences and similarities between what Medicare, Medicaid, and commercial payers may be looking for in terms of information sharing and measurement.
From a purchaser point of view, I think a very high priority for us -- and I'll talk some more about this are the quality deficits in our system, in our ability to provide feedback and improvement tools that will allow us to ameliorate some of the problems, both in terms of cost drivers and quality failures that we see. So the areas of high variation, inappropriate utilization, high cost, which may be different across different payer groups, need to be accessible to this information infrastructure that we are all trying to build.
A caution, obviously, that we have heard about in these two days is, we are particularly sensitive to the fact that over 20 years we have certified a lot of things in health care, and not always seen performance improve as a result. So we can certify something and still see mediocre performance.
Just to Rick's comment a moment ago, I flag one comparator statistic he mentioned that I know Surescripts has been concerned about for a while. There may be only 10 percent or so of doctors who are transacting a prescription across the network who are retrieving the medical history, and therefore, capable of looking up potential interactions and risks that are afforded by the new prescription. So here's a case where, in rough numbers, a large number of the transactions are not being associated with the use that we think is really essential. For us, we need to be able to test performance in order value.
As Carol suggested this morning, your ability to stipulate what the goals of this investment of public funds are and then assign objectives to those goals and measure whether those objectives are being reached is really vital. Certification is one important step in a cycle of activity, which I will simplify to say, we need to certify that a product can do what we want it to do to meet the goals of the public. We have to validate that it is doing it. We have to measure whether the uses are being satisfied, provide feedback to everyone involved, and reward that high performer, so there is a virtuous circle of improving the products, improving the services, and documenting it.
I want to do a little sidetrack here on a cardiology example of our concern. Purchasers, when they look at interventional cardiology in particular, are concerned right now about the appropriate use of imaging studies, or the inappropriate use of such studies, whether stress tests are being used appropriately prior to interventions, the use of diagnostic catheterizations, when they are indicated, and if they are being used when they are not indicated, and certainly the use of interventions in the cath lab. We are very interested in the outcomes of these services in terms of symptom relief, functional improvement, survival gains of course.
Right now, if we look at the PQRI measures, which are going to be an underpinning for this clinical quality reporting that we are talking about as part of meaningful use, they are inadequate to tell us what we need to know as purchasers about whether the health-care system is performing in ways that we are particularly concerned about, in terms of improving health or reducing and managing cost.
The National Priority Partners, mentioned earlier today, have identified 20-odd categories of overuse as a national priority. I would ask the committee to give some thought to whether the meaningful-use definitions that you ultimately promulgate can address these 25 or so areas of overuse. This would be a very great value to society as a whole, to begin to get a measurement and improvement cycle to intervene in some of these areas. I highlighted some of them that we are particularly concerned about. They come out of this area of cardiologic intervention.
As an example of percutaneous interventions in cardiology, we just had a recent paper showing that 55 percent of the cath-lab interventions in the country apparently have not been preceded by a stress test to document ischemia. We would like to know more about that. We would like a data system emerging from meaningful which will tell us whether these treatments are actually indicated or whether they are being overused.
Fortunately, the American College of Cardiology has published a list of indications for these particular treatments, and they have done us the favor of telling us what data is needed to document the appropriateness of these cath-lab interventions. I would ask again, as we look at the certification criteria that Mark and others may be developing, are we able to evaluate and tell everyone -- the practitioner, the patient, and the purchaser -- that these extremely expensive and frequent interventions are or are not being used appropriately in our society.
In general, I would say meaningful use is to satisfy the needs of the purchaser and payer community for information about not only routine primary-care services, as important as they are, but also the very expensive and highly variable interventions that are happening in the operating room, in the cath lab, and elsewhere. I would suggest for you to consider that where there are significant quality, safety, and cost variations in procedural and specialty care, we do not omit those from the data network; we do not define meaningful use and the use of public money in such a way that 60 percent of doctors and an enormous amount of American health care is not captured and documented on this electronic network. The incentives that we are all working with need to bring those clinical registries and other data systems into the fold.
As we look at certification today and the current CCHIT ambulatory certification criteria, they would, generally speaking, not reach into some of the data areas that we have great concern for. In the areas of cancer, orthopedics, maternity, behavioral medicine, cardiology, I hope we could take a special look at how we tailor meaningful-use definitions to capture critical indicators of high public value in those other areas.
Finally, certification is needed to inspect the capability of product to deliver value. Then the product and services and users must then be validated after installation to ensure that the capabilities have been enabled. I hope we will consider a tailoring, or picking-and-choosing strategy, here, where -- Carol and others have mentioned the importance of innovation -- we allow the certification criteria and the meaningful-use definitions to be tailorable enough that we can allow innovative products and services that bundle in a different way than may be a fully functional EHR to satisfy the meaningful-use criteria, if they are adding to the public good.
Markle will be issuing a report tomorrow that Carol mentioned. That report will identify a number of criteria for certification that I hope you will take a more careful look at. Some of them are summarized here.
Finally, just to represent a purchaser perspective on this, our key question is, is the information being used to increase appropriate use of expensive, dangerous, and variable technology? Certification criteria need to address these kinds of requirements that are relevant to public-interest concerns, outcomes, appropriateness, and efficiency. I think the meaningful-use definition should be construed to capture the use of clinical registries that CMS and others might designate, which have high public value in the eyes of Congress and certainly private purchasers.
Certification can verify that EHRs, as both Mark and Rick have just said, capture and audit the documentation that will allow us to inspect whether the uses are being performed.
Ultimately, this is a virtuous circle. Improving the data will permit better measures. Better measures will drive payment. Payment and feedback will drive improvement. The HIT incentives by themselves will not be the only driver of this mechanism. We have to engage the full payment cycle.
Thank you.
MR. REYNOLD: Mr. Findlay, you would appear to be next.
MR. FINDLAY: Hi. Thanks very much for the opportunity to testify today. You are coming to the end of a long two days, so I want to be brief. I'm at the end of this panel, so I have sliced a few things out here. If you will indulge me, I'll read a statement. It will take three or four minutes. Then I'll be available for questions.
My name is Steve Findlay. I'm here today representing Consumers Union and the consumer-patient perspective on this issue.
In full disclosure, I'm a registered federal lobbyist. I do not have any investments in any HIT companies, outside mutual fund accounts. I'm also on the board of the National E-Health Collaborative, the AHIC successor entity.
Certification has played a really critical role in the advance of HIT so far. It was reasonable in 2004, when CCHIT -- and you have heard an update on that today -- was launched and it is reasonable now, as specified under ARRA, to look to certification as one of the foundational pillars of promoting an HIT marketplace which meets certain standards. In compelling certification in ARRA, Congress aimed to build on work over the past five years. That's good. However, it is now time to pause and reflect on the future role of certification in the context of the changes ARRA calls for, but also in light of the plain fact that EHR certification has not been particularly successful to date at driving adoption.
But I won't waste time on the past. Your mission is to help envision the future use of HIT and the best and fastest way to get there.
So cutting to the chase and being responsive to the formal questions asked of the panel:
First and foremost, certification of EHRs and PHRs should be redirected and adjusted to explicitly promote better-quality and safer care, more efficient care, and new models of care that over time will save money. That's the meaningful use you have been hearing about for the last two days. Technology for technology's sake is just not the point.
But we also believe that meaningful use of HIT should enhance consumer and patient engagement in care and make consumers' access to accurate medical information and data about themselves and their care easy to obtain, understand, use, and act on. Thus, meaningful use for consumers will require some mechanism to present things like lab, blood, and imaging results, diagnosis and clinical status, medication history, and other relevant information to consumers in a way they can evaluate and understand and that permits them to track changes over time.
As you all aware, ARRA grants the consumers the right to get an e-copy of their medical record if it's kept in EHR.
We believe this kind of patient-centered meaningful use should indeed be a criterion for EHRs, as well as PHRs. Obviously, this is a given for PHRs, but it has not been for EHRs. So we are talking about an evolution in certification that would reward EHR systems that translate and contextual part of the content of an EHR for the consumer and the patient. We believe ARRA intended that.
In this regard, some experts, including committee member Paul Tang, have talked about PHRs that are integrated with EHRs. That is one way to go, and it is already being deployed in various settings and creates an attractive model.
We also believe that EHR data and information should be easily transferable to a secure standalone PHR or other e-health tool set up by a consumer. Of course, that's in large part the model of the PHR vendors today.
Either way, integrated or standalone, it's critical that data integrity and accuracy be a key component of both EHR and PHR certification. There can be no meaningful use of medical information that is wrong. Clearly, billing data, as we have learned with a recent case, won't do.
That is why we think EHRs simply must be open to patient input, and certification around the ease of that is essential. A patient's input can come in a variety of ways. It can come from a PHR or it may simply come from a patient asking to see their EHR and reviewing that.
There are two reasons for such patient input -- two main reasons; there are hundreds of reasons. The first is accuracy, as I just mentioned. As everyone knows, medical records are hardly infallible as a source of truth. There has been a flurry of discussion recently about the garbage-in/garbage-out problem stemming from a patient's detailed account on a Web blog and in the Boston Globe of finding errors and misleading information when their data was transferred from an EHR into a PHR.
The second reason EHRs need patient input is that, as we all know, most of the stuff that impacts our health and may influence the treatments we are prescribed happens outside the clinical setting. The self-care revolution is alive and well, and a growing number of people are actively engaged in using e-tools, Web sites, and creative software to track their medical lives, whether that's their weight, their symptoms, the OTC medicines they take, the course of their chronic illness, whatever.
We think that organized notes from a patient can be every bit as valuable as notes from the doctor. We think every EHR should accommodate patient notes if the patient so chooses.
So certification around meaningful use must promote two goals: transfer of information from an EHR to the consumer, and transfer from the consumer into an EHR. It's a two-way street. This has a vast potential to improve a lot of things, not the least of which the doctor-patient relationship.
Again, to repeat, most of us now think of this integration, this two-way street, in terms of PHRs and EHRs. But it's not hard these days to imagine new platforms of such information to flow into the consumer's hands. Is an iPhone or another PDA app that allows you to organize your top-line registry, insurance, and medical information far away? I doubt it. Should that be certified? I don't know.
Indeed, guidance on meaningful use being released tomorrow by the Markle Foundation, which you are now well aware of, notes that the definition of certified or qualified EHR technology should not be narrowly construed within the confines of EHRs and HIEs. Such tools, the report will point out, are not the only way to get to meaningful use for doctors or, more importantly from my perspective, for patients and consumers. Rather, the Markle Foundation report will note, a broad view of HIT and certification of HIT tools would and should feed innovation and not lock in today's technology. Certification must evolve and be adaptive.
Two final quick points. ARRA has propelled us into a new world when it comes to the privacy and confidentiality of medical information. I'm confident that vendors and CCHIT will know that stricter standards must evolve abetted by new technology.
But I would still urge this committee, which has deliberated on this issue in the past, to develop clear guidelines on privacy in the context of the ARRA legislation. For example -- just one example -- audit trails must now clearly be a larger part of certification.
Finally, you posed a question to us about additional EHR certifiers recognized formally by the HHS Secretary. In a word, yes. We do think that would be advisable, though we do think CCHIT ought to be among those groups.
Thanks for your attention.
MR. REYNOLDS: Excellent. Walter?
DR. SUAREZ: Thank you. Thank you for this panel. This is one of the panels that I think everybody was waiting for.
We heard over the last two days a lot about certification. We heard that certification is a key element of meaningful use. We heard that certification is not enough, as has been mentioned a few times today and during this panel. We heard that certification defines a series of capabilities that a product can have, but after installation, there is no assurance of whether those capabilities have been actually enabled or turned on, if you will.
My question goes into that part. It seems that we are now talking about one level of certification or one aspect of certification, which is the product, and then another, which I wouldn't necessarily call certification, but the verification, the monitoring, the assurance process of the fact that after the implementation and installation of a product, a series of those capabilities, which is the heart and soul, in my view, of the definition of meaningful use, now turn on and enable and are actually being used.
So my question is, who should be conducting those types of evaluations of assurance that the types of capabilities that are part of meaningful use are actually enabled? Is that a role of the certification body? Is that a role of an accreditation body, for example? Is that a role of a government body?
DR. LEAVITT: We will all probably take a crack at it. It's a great question. I think one of the important things is to think of the scale of this. We talk about the number of doctors in the United States, the number of offices. I don't believe we are going to be able to send inspectors into all of those offices and accredit every office in their use of the EHR. I think we are going to have to measure it using technical tools. That may be within the EHR, which -- in order to be a medical legal record, it has to have the audit trail. We require it. It's very rigorous, actually. The number-one cause of systems flunking our test was not having an adequate audit trail. So I think you want to use that data.
Some of it you can't get from within the office. You need a system view. I think it's going to take a good dialogue and probably a combination of efforts.
I think we could help. We can make sure that the EHR has an accurate self-instrumentation system that can't be played with. We can make sure that that is submitted in a secure way and it's coming from who you think it is.
But the system measurements -- to get those, you have to reach outside the office. That's what you are trying to do, integrate care. So all we can really make sure of is that the EHR can submit not just the quality data, but the data about what is happening to the patients, to health-information exchanges or other repositories or registries. That's where you begin to measure these larger things.
I think it's important to realize that those won't exist tomorrow. If you want to write checks in 2011, those doctors and nurses need to be buying these systems this year. So I think you are going to have to start from where we are. Where we are is not so bad. But I absolutely acknowledge it's not perfect, and we need to do better.
That's my answer. I think it's going to be a combination of standard certification plus some new concepts about how we measure quality in accountable care-delivery organizations or networks.
MR. RATLIFF: From our perspective we can actually measure, at least to some level, what we call utilization. To Mark's point, there are certain aspects of the system that we can see in the network related to prescription routing, as an example. We can see which prescriptions go electronically. We can determine when an eligibility check is done and when it's received.
To Mark's point, though, we can't tell if the formulary was used to determine the right economic alternative. That is in the practice. That's in the four walls of the practice. That would be in the audit process in the tool itself, in the EMR. Medication history would be the same way. We can see if medication information is pulled into the system. However, you have to have the local view in order to know whether that information was reviewed as the prescription was being written.
So has to be within the certification process -- it does have to evolve to the point that you have the right audit capabilities and the right capabilities to report out, if that's the approach that you go through, to demonstrate real use, at least in the context of prescribing as we have been describing it, in medication-information sharing, as an example.
Who does that? There are certain capabilities that an organization like Surescripts has. Certification bodies could do that. There are accreditation bodies out there as well that would have that ability, with the right definitions.
DR. LANSKY: I think parsimony is important. That is, the criteria that will be used for this kind of inspection should be very limited and focused on high-value considerations for the definition of meaningful use and for the public interest. It should be electronic, as Mark said, at the outset, not through some other mechanism. Some of it may be done through attestation or other certification techniques.
The design questions I hope this committee can take up and assure that the certification criteria permit the discernment of meaningful use. Maybe the audit trail is sufficient. Maybe there's some other kind of data capture that, as Rick suggested, has to happen. That's part of the definition of meaningful use. But the execution might happen through independent parties and so on. At least the design piece, I think, is a critical feature here.
MR. FINDLAY: I would only add one point, and that is that there would seem to be an important role for some applied research here, as the health research community gears up to track this stuff. I think that's going to be important.
MR. REYNOLDS: Jorge, you have a follow-up?
DR. FERRER: Mark, do you think the certification should reside under a federal agency? Would that help or hinder your efforts?
DR. LEAVITT: I think it is best placed in the private sector, but with federal funding. We have to move pretty fast. We have to be able to do outreach very effectively and quickly when new standards come through. I think people would mostly agree that we have moved very fast. One of the number-one complaints is, you move too fast. We are moving faster than the vendors can keep up.
So I think the private sector is good because of the speed and the connection to the market. If what we do doesn't match the market, we fail. So we have to have a very good market sense.
But I think it needs a core of federal funding. If our contract runs out -- and it's in extension right now so that all of our revenue as a nonprofit comes from the fees we charge vendors to certify, then even if we do our best to avoid vendor influence, there will be a perception that the vendors who are paying the fees have undue influence.
So I think it should be federally funded and operated in a public-private partnership.
MR. REYNOLDS: Blackford.
DR. MIDDLEFORD: Thanks for a great panel. Nice to see you all again. I have a question for David and Mark.
Interestingly, we seem to struggle still with certification between functionality and feature sets, if you will, versus the overall objective of our HIT adoption, et cetera, in terms of value in reducing inappropriate use and misuse and increasing appropriate use, et cetera.
My question is, should we reorient certification around value clusters of a kind that actually are more targeted toward the payer's objective, whether it's public or private? If we had this orientation, what might those value clusters be, in your mind? Do we consider this kind of certification both within a product -- so the Cerner suite of products, if you will, might allow certification of value clusters within a set of products, so it's not inconsistent with the physician adopting a technology -- and could we consider certification of value clusters across products?
It's a fundamental kind of relook at certification around value as opposed to feature function.
DR. LANSKY: I like that idea, Blackford, and I like the terminology. I think it also helps to support this pathway toward innovation that many of us have talked about. We could imagine a new player introducing a radiation therapy management technology EHR, which is a value cluster for a certain application and discipline that is very high-value and could be hypothetically certifiable, but doesn't have all the capabilities that a contemporary ambulatory EHR might have. The same in oncology or other disciplines.
So I think the opportunity to think through -- I don't know the answer offhand, what those buckets are. But I have been thinking, for example, about the Medicaid meaningful-use application and how different it may be, given the difference in PQRI measurement philosophy in a Medicaid population, for example. We do need to think about classes or value cases of patient requirements, as well as clinical settings, that will accommodate more diversity in the product mix that we try to certify.
DR. LEAVITT: I would respond as well. I think there is actually a role for both. One of the reasons that I think you still need to be certifying products is that it's a long leap to get to that quality improvement. I think we can see that providers don't know how to get there. So you say, you have to get a tool that you know will get you there. It's 10 steps to this quality level. When you do step 1, you don't know enough to pick the right tool. So we'll make sure you are not going to get a tool that -- like a ladder that won't reach as high as you need to go.
So you need to certify the product, but you can't stop forever on that, absolutely not.
If you look at programs like Bridges to Excellence, they sort of phase out that pay-for-adoption into a pay-for-measurement and, finally, pay-for-performance. I think we need the same model here.
MR. REYNOLDS: Justine.
DR. CARR: Thank you. Thanks for a great presentation.
I would like to go back to the e-prescribing experience and what we can learn from there, in the context of how much of that is transferable to EHR. With e-prescribing, we identified a tremendous patient safety opportunity. We also identified that there was efficiency for the provider. We had standards, developed certification, and then the incentives.
As we look now at EHR, are we going to have a doubling of use? If you look at how quickly e-prescribing doubled and tripled in a year, what is the piece that made it happen? Was it just the incentives or are there fewer obstacles in e-prescribing than in EHR? Can we somehow capture that experience to get the same level of success?
MR. RATLIFF: Let me clarify a couple of things. One is, the significant growth that we saw last year was with e-prescribing in the context of the electronic medical record. What happened was, there are a large number of users of existing electronic medical record systems that are writing prescriptions electronically in the system, and they are printing that prescription or faxing it. So the way that you can get very fast adoption is to certify those systems, make sure the practice upgrades to that certified system that is connected to the network. Those prescriptions that were printed and faxed to pharmacies, as an example, are now routed electronically.
That was the low-hanging fruit, if you will. There is significant uptake, if you will, within existing EMR usage.
The challenge at this point, going forward, is, as new EMR systems are deployed, we still have the same challenges and complexities of the change management, et cetera, involved in implementing the electronic medical record system. There is the cost and the complexity. What you can do with electronic prescribing is, as a first step, put in place the medication-management component. You already have connectivity and the interoperability.
The key is that, as we do that, we need to make sure it's not just prescription routing. We saw growth across all different types of transactions, but you are not going to get to the outcomes and the value-cluster kind of approach unless the electronic medical record systems are installed with all the functionality. So every new system that goes in needs to have that capability.
What I will tell you is that, at least for small physician practices, there is a question: Do they need a fully functional EMR? I suspect there's -- unfortunately, I haven't been able to participate in all the discussions. But is there something short of that that you could implement in the practice, particularly a single-doctor, small doctor practice? That's where we see a great deal of success with standalone electronic prescribing. There are a few of those systems that went out last year. That automates the prescribing process. But you need to move towards lab information delivery and sharing of other information, towards something that looks more like an electronic medical record.
I don't know if that happens.
DR. LEAVITT: Can I add to that for a moment? That brings us to the concept of Light EHR, which was discussed. It's very valid to give people an easy first step. You don't eat the whole cow at the same time when you install an EHR. What's critically important is that that first step isn't a dead end or a trap. I will also tell you right now, it has been, in some cases.
But if it's a first step, so that it just whets their appetite for more -- and then, when they want to move past it, they are not locked in. One of our requirements in certifying a standalone is that you have to have a migration path to a more complete EHR. You have to have interoperability.
MR. REYNOLDS: Before I let Blackford follow up, I think another thing that would be helpful is that we have a definition of e-prescribing. People that are buying it a lot of times think one thing and people that are selling it think another thing. I think, as we get a clear definition of what that really means to the industry -- I'm not asking a question. I'm making a statement.
Blackford, your follow-up, please.
DR. MIDDLETON: A new topic, if I may, briefly. I'm going to ask you to put on your crystal-ball hats -- to mix a bunch of metaphors. I sort of wonder where does certification end and regulation begin? There is some concern that medical-device regulation types of policies and procedures may eventually apply to medical software. This hasn't been the case historically, for lots of good reasons. Is there a scenario in the future, as we think not only to 2011, 2014, but beyond, where certification runs out of gas and we need to think about regulation of the medical-device type?
DR. LEAVITT: Regulation is needed if certification is not sufficiently rigorous. So if you have cases where it's a sham and you have a risk to the public or to the country, then it's really appropriate to have it be a government function. But I think, in the absence of that, if it can run well in the private sector, because of various factors, then I think you probably get a little bit more for your money and probably more responsiveness by having it on the edge of private sector or a public-private quasi-governmental function.
MR. REYNOLDS: Leslie.
MR. FINDLAY: Can I respond to that?
MR. REYNOLDS: Yes, sir.
MR. FINDLAY: Just very briefly. That brings to mind the JCAHO issue. I don't think any of us think JCAHO got us where we wanted to go. I'm not sure certification will either. I think at some point we are going to deal with your question.
ARRA sets up a lot of regulatory teases. I think we are going to be grappling with it in the not-too-distant future.
MR. REYNOLDS: Leslie?
DR. FRANCIS: Steven, in your closing comments about privacy, you used the example of audit trails for the protection of privacy. Audit trails are a way of identifying when records have been accessed, giving the potential of notice to patients, for example. Could you comment on whether you see other directions -- for example, a consent process -- as also necessary for the protection of privacy, and thus functionality for this to be included in certification?
I noted, for example, that in Rick's written testimony -- although you didn't mention this in your discussion -- that there was mention of "with patient consent" to access to medication history in the Surescripts process.
I would like to hear all of you comment a little bit more on whether audit notice or something more is what you see in the certification.
MR. FINDLAY: Clearly it could be other things. ARRA just made a huge leap in the privacy space, more than many of us thought was going to be possible. It was really dynamic and great. There are no fewer than 15 or 16 specific privacy, confidentiality, and security measures in ARRA. Not all of those are going to be amenable to certification, but probably half a dozen or so will. That's going to take some think-through that I haven't done about how those get incorporated in EHR certification.
But I'm sure Mark has done some thinking in that area.
DR. LEAVITT: No. We asked our legal counsel to do some thinking. It's sufficiently complex.
MR. RATLIFF: Relative to patient consent, we actually do require contractually that vendors have a process in place, not only contractually with their customers, but technically, to capture patient consent before a medication history is pulled out of our network for that particular patient, to be used in the context of delivering care. So we not only require it contractually, we do certify against it. Then the technical standard actually has a flag in the transaction that says whether patient consent has been captured or not, which is another audit trail.
We feel that's very important. At the time we actually implemented this process, HIPAA did not require us to do that, because it was for treatment purposes. However, we thought it was an appropriate approach to take, in anticipation of the future, such as we have just described.
MR. REYNOLDS: Last comment.
MR. FINDLAY: I would just add to what I just said. You asked for another example. One would be that ARRA specifies that patients have the right to edit their record as it's released from one physician to a health plan, if there is a business-associated relationship there. I think that's an area that would be very important for certification around, as an element of EHR.
MR. REYNOLDS: With that, that ends this panel.
If the next panel will please come forward, we would appreciate it.
(Pause)
MR. REYNOLDS: Our panel is on measuring meaningful use. Dr. Eisenberg, welcome.
Agenda Item: Panel 10: Measuring Meaningful Use
DR. EISENBERG: Thank you. I am Floyd Eisenberg. I want to thank Chairman Reynolds and members of the committee.
I am senior vice president at Health Information Technology at the National Quality Forum. I also do wear two other hats. I'm a co-chair of the HITSP Population Health, now Quality Tiger Team Committee, and also of an IAG Integrating the Health Care Enterprise Quality Research and Public Health Technical Committee.
Just to get started, I want to give a short background on NQF, National Quality Forum. We are a nonprofit, with the mission of setting national priorities and goals for performance improvement, endorsing national consensus standards and measures, and education and outreach concerning the priorities and consensus standards.
We were pleased to convene the National Priorities Partnerships, which is the first dimension of two areas of significance that we look at. This was a 28-multi-stakeholder group looking at priorities for care in the U.S., which includes engaging patients and families, well-coordinated care, improving health, improving safety, guaranteeing appropriateness of care and compassionate care, and eliminating waste.
Our other dimension has been to look at high-impact conditions -- an example provided here for acute MI care, trying to look at episode care across different settings of care and across the continuum, the example here given with acute myocardial infarction, showing care across acute, post-acute, and secondary prevention settings.
We also do not develop our own measures. We put out a call for measures, based on these priorities and based on these high-impact areas. We have fairly strict evaluation criteria for the measures that we evaluate. What I want to really focus on here is the fourth one, which is feasibility. Note that all of these evaluation criteria have been updated as of August 2008. Much of the reason for the update was the result of work we did to identify information required to address data captured within electronic data sources, including electronic health records. So we have strict criteria.
The concept is to try to connect information across clinical guidelines, physician support, and quality measures, using Paul Tang's example that the sweet spot is where all of these connect in the middle. That's often with the data flow and the data set that creates the connection. So that's where much of our attention has been.
Addressing your questions, you start with limitations of various methodologies. What I want to address are the essential components of the EHR used to improve to care as it's delivered.
In 2008, chaired by another member of your panel, Blackford Middleton, we pulled together a panel to look at consensus standards for health-information technology structural measures. As was said many times in these presentations, certification is necessary but not sufficient. As presented by Dr. Classen yesterday, many of the certified components were not necessarily implemented. The intent of this group was to identify the measures of implementation that need to be addressed to see that the components could be effectively used.
In such, there were five basic areas: e-prescribing, EHR interoperability, care management, quality reporting, including registry reporting, and addressing the medical home.
Much of the output of this group dealt with attestation by a practice that each of these was implemented. That was a good first step. Attestation is helpful, but, like certification, not sufficient. The intent of the group was really to look for the fact that key functions are actually implemented at individual sites. But the first was attesting they are there.
Where CCHIT certification had not been identified, the key areas to be attested to were use of essential components of the record for good medical care, including medication lists, problem lists, laboratory results, privacy and security, key elements for direct patient care, and making sure they are utilized. Those are key elements as a start for identifying that it's being implemented.
The next step, as in your next question, is, how do we implement the F7 key? Thanks to Carolyn Clancy for coining the "F7 key."
Our current process, as we look at this area from NQF, is that we do endorse measures based on standard specifications and standard endorsement criteria. We have a group called the Health Care Information Technology Expert Panel, which is now in its second iteration, looking at developing a quality data set. We developed some of the framework in our first iteration. Now we are identifying a set of data to be used for measurement. It could also be used for decision support and direct care.
That information moves to HITSP, where health IT standards are selected to address the areas of the quality data set. Based on work already performed in HITSP, some of that data set can easily be addressed using existing standards. Those standards are approved by the CCHIT and then that enables new standards to help us create a new measure-authoring tool so that quality measures, as we move forward, will be able to be authored using the components identified within EHRs, in the same format, in the same terminologies, in the same vocabulary.
To do that, that means taking the source of the measure in a standard format, which is progressing now to be developed, making sure the data are collected, and receiving the report on the other end. The two ends are orange because these are the new pieces being worked on this year, through the National Quality Forum, through HITSP, and through IHE.
What's very important is the collaboration of many industry groups to make this happen. One alone is not sufficient. All of the groups together can make this occur.
Our next question is based on various methodologies for use. As we identified in the structural measures, some of the areas of concern that can be identified for measures of meaningful use are use of problem lists for every patient, allergy lists, medication lists, orders, prescriptions, medical summaries, and creating a continuity-of-care document -- many of these already certified in systems, so now measuring that they are used; using some of the audit capabilities discussed in the past panel, to be able to determine that they have occurred.
To give an example, a measure of tracking medication interactions for patients with a history of falls, to see if they are receiving drugs that may make them more confused, dementia, drugs for confusion -- basically, to measure the process. Is there success for decision support? No treatment over the number of prompts. What is the outcome? How many ADEs were there? So you look at the process and the outcome together.
In summary:
Thank you very much.
MR. REYNOLDS: Thank you. Dr. Rapp, please.
DR. RAPP: I'm Michael Rapp. I'm director of the Quality Measurement Health Assessment Group at CMS. I'm involved in developing, helping measures get endorsed, and implementing programs using quality measures in a variety of settings, including being involved in the Physician Quality Reporting Initiative, the Electronic Prescribing Reporting Initiative, and measures for the hospital programs.
We have some questions that we were presented. Floyd has already gone over the basic questions.
The first one is about the strengths and limitations of methodologies for demonstrating meaningful use. This is the "how" of meaningful use and what is most opportune.
There are several statutory options that were provided, including attestation, claims, survey, reporting quality measures, and, of course, other. The statute gives the Department a lot of options in terms of how to approach this.
In terms of demonstrating being a meaningful user, there are basic requirements. The first is the "how." This is the "what," which is what this discussion is about. What is a meaningful user? Electronic prescribing, being connected, and quality-measure submission, when the Secretary can accept those. It becomes a policy question of what should be a meaningful use at any particular year. We have heard a lot about that.
But in terms of the mechanism, how the professional or the hospital will give that information, those are the options that we are going to talk about.
One of the things that's important to note is, it's not only what they assert, it's not only how they do it, but it's the fact that you have to tie it to an individual, ultimately, and make a payment based upon that.
So when you think about attestation, survey, or other, it would be fairly easy for one to certify or attest that they are doing a certain type of use of an electronic health record and get that information. Then you could even tie that to the individual. But what's missing in that is that you have to end up tying it to a sum of money that will be distributed to the party to whom that money is owed. Within the statute, we have statutory maximums required to coordinate the incentive, limitations among the practices, and a requirement that no incentive will go to hospital-based eligible professionals.
So how do you bring that all together? As I mentioned, basically you have to make that determination at the individual level and ultimately tie it to claims. We do have existing structural measures that are used in the PQRI program and the e-prescribing program. They are basically collected using claims. As I say, the attestation survey and other would make it possible to make that determination at the individual level. Then, however, one would have to go to the claims to deal with the 75 percent.
Many of these things we have in our current programs work through the Physician Quality Reporting Initiative and the Electronic Prescribing Reporting Initiative, in particular. Things that we have not yet worked through are how to apply these maximums. In the PQRI program, the electronic prescriber incentive program, there are no maximums. So that adds some complexities to this.
So our current structural measures -- a structural measure, I would say, is a vehicle, and one would plug into that structural measure, if one decided to use something of this sort, the various items of meaningful use. So it's really a plug-and-play option. Claims are one vehicle where that structural measure could be used.
The elements of a structural measure: If one talks about attestation or survey, you get a qualitative determination, but you don't get a quantitative determination. If one used claim codes, one would get that quantitative determination, both on the things that one might require for meaningful use and for the things that aren't required, but one wants to track over time in terms of seeing the bar being raised and where one might want to go.
The second main piece in the requirement for meaningful use is reporting quality measures from EHRs. Here are a variety of ways that we collect quality measures at CMS: from claims, from augmented claims, like in PQRI, chart abstraction measures from patient-assessment instruments. In PQRI, we also use clinical registries and have 31 of them that participate in that.
There are considerable advantages to quality measures being collected from EHRs. We have looked forward to the possibility of this in terms of trying to develop this in our existing programs, the Physician Quality Reporting Initiative in particular, and looking for it with hospitals. It's primary-source data. It doesn't have to be segmented by pair. There are standardized data elements, much more timely data available. We have talked about audit output files that would be available, an improved relationship between population measures, which have been discussed as being highly important, and accountability measures at the various payment settings or provider settings, potential to aggregate clinical and claims information, the potential to use that data as an element to consider in payment.
So clinical quality measures -- again, HITECH makes that one element, but only if the Secretary can receive that information.
In terms of what is in a measure, Floyd that -- numerators and denominators, the specifications, the certified systems, and, of course, the transmission of the data, getting back to that point that the Secretary has to be able to receive that data.
So we have looked to see how we might work to develop that, not for this program, but for our other programs, the Physician Quality Reporting Initiative and the hospital. There are two basic ways. EHR could produce an output file, and you just take that information. It does come from the electronic health record. You have to have the elements in the electronic health record. But it's basically a file output. It's not the same as direct data-element transmission, which we have talked about, such as through health-information exchange.
Those are two basic approaches. But certainly, as one looks to the future, the possibility and the promise of being able to query data elements from this health-information exchange, as opposed to just pushing it out from a system, is something that we have significant interest in, as we look toward these other programs that I mentioned.
To show you that this is not something that only ARRA or HITECH has stimulated interest in, CMS, back in the PQRI program in 2007 -- even though that was initially a claims-based system and is today for reporting -- we announced our plan to test EHR submission. We posted the electronic specifications for 10 physician quality measures. We coordinated this with the HITEP panel that Floyd was talking about. We have worked to make sure that we follow standards. Now we are looking toward carrying out this testing, and in so doing, to not just look for an output file, some way to get the data to CMS, but look to see how we can pay attention to the transmission using health-exchange standards that exist. We have the potential, possibly, to implement this in PQRI as early as 2010, which could provide a foundation for what might be implemented in the HITECH legislation.
Again, we would look to expand over time the measures. Part of that is thinking in terms of having them be able to be collected in the electronic health record.
As far as hospitals, in our hospital reporting program we have stated an interest in moving to EHR reporting. We have had quite a limitation on our abilities to expand this because of the chart abstract and nature of measures. We have worked with HITSP to get sets of measures that have not been implemented yet in the hospital program specified in a way that we could have them collected in the electronic health record and ultimately move toward that.
There are a variety of challenges for use of EHRs for clinical quality measure in terms of the unstructured data, the dynamic development of quality measures. You have heard people about existing quality measures and what they would hope to see in quality measures. One has to think about the physical quality data set. We have talked about initial measures set, the inpatient quality data set. There is a care data set that we have been working on as well. There is a process needed for regular update of EHR certification requirements that would help these data elements be captured and then used for the reporting of quality data. We have heard about the two-year cycle for that.
We believe that, at least for our other programs, it's important to pay attention to health-information exchange and not to have the quality measures reporting be something separate from that, but have something where these two items work synchronously, at least as far as our testing. There is a significant relationship, we believe, to interoperability.
Several conclusions here for you:
• Quality measurement can support measurement of meaningful use progressing from structural measures to clinical quality measurement.
• On the clinical side, once CMS has the ability to collect the quality measures, then those audit output trails we talked about could come with it.
• Structural measures are based on claims submission.
• Clinical quality measures require the Secretary to receive the data.
• I have talked about a number of these advantages, standards, and existing structural measures that could provide a foundation for EHR submission.
That's a quick overview of a few things. I'll stop at that point.
MR. REYNOLDS: Thank you. Dr. Tripathi.
DR. TRIPATHI: Thank you, Chairman Reynolds and the committee, for inviting me to be here today. I really appreciate the opportunity.
I would also like to thank you for not going in alphabetical order, so that it's Phyllis and not me who has the distinct honor of being the very last person to testify after these two long days. I'm very grateful for that.
MR. REYNOLDS: If you make her late, we will hold it against you. (Laughter)
DR. TRIPATHI: The clock hasn't even started yet. I'm very happy to see that, too.
Let me just dive in and talk about measuring meaningful use. I'm going to have somewhat of a different take than some of the previous speakers and very much welcome the discussion afterward on this.
The first thing I would like to talk about is just to point out that the government has very high expectations for the Medicare and Medicaid incentives that are being rolled out. What I have done here is just graphically depict the CBO scoring of the incentives and the money that's flowing, to just point out that the expectation is that we are going to spend something on the order of $35 billion -- $33 billion in direct incentives, $2 billion discretionary that Dr. Blumenthal and the ONC team are going to be deciding -- with an expectation of a financial benefit of something on the order of $16 billion.
The reason I like to point that out is that that is a pretty high standard that we all need to hit here. I think, as all of us have been hearing over the last couple of days, this stuff doesn't just happen; it gets done. In particular, my concern about this is that too little, I think, is being invested on the infrastructure needed to meet those high expectations.
I think in two slides, I'll come to how this relates to measuring meaningful use, which you may be wondering.
But as you can see here -- this is just blowing up that $33 billion versus $2 billion -- you have the vast majority of it going into direct payments to individual providers to just take technology and put it in their offices, and very little money dedicated to all of the support infrastructure that is really needed to make that happen. I think it's wildly disproportionate with the need there.
As I said, meaningful use doesn't happen; it gets done. In the current approach, just to put a fine point on it, it funds the pieces, but it doesn't connect them. So we have money for incentives, as I pointed out, we have money for RHITECs, the regional health IT extension centers, we have money for the health-information exchanges, which is all good. I would argue that there is too little money given to the support infrastructure. But they are completely disconnected in the law -- completely disconnected. There's no connection between those incentives and the HIEs and the regional health IT extension centers. What we need is that connection, I would argue.
So why am I talking about this? Because how you have to measure depends on what it is you need to measure. Assuming that meaningful use is going to have some of the following core elements -- and as I went through this -- and I think probably other speakers went through this as well -- how do you talk about measuring it without talking about what meaningful use is supposed to be? That was panel, and I didn't want to dive into what they are doing.
But certainly it's about interoperability. We know it's about electronic prescribing, assuming now that it's about electronic laboratory and radiology results and order entry, assuming that some type of clinical summary document exchange will be a part of that, and then quality data reporting.
I think one question to ask is, what is the best way to measure these use elements, with an eye toward trying to make sure that the cost of monitoring and enforcement doesn't exceed the benefits of higher compliance?
Just to get to this point now, the reason I mention that is that I think that the depth of meaningful use is actually going to depend on our ability to connect the support infrastructure with the incentives. I think if we don't do that, we are actually going to have to set a very low bar on meaningful use, because the vast majority of physicians won't be able to accomplish it. They won't be able to accomplish just getting the implementation to have the structure data elements that Floyd was talking about to get out the door. The example that he used on adverse medical events, I would argue, is actually a very, very high bar for the vast majority of practices out there, even the practices that have been on EPIC for 15 years. I would argue that that exact measure that he was talking about would be a very high bar. So when you talk about the onesie/twosie physicians who are 80 percent of the physicians out in the ambulatory space now, that's really going to be a big challenge.
In thinking about this, what I did was to sort of say, if we set that low bar, these are the things you could probably cross off the chart, I would argue. If you don't have robust regional health IT extension centers connected to those incentives and implementation, and if you don't have health-information exchanges that are trying to lower the cost and raise the effectiveness of interoperability, you can probably cross off order entry as a meaningful-use standard, because most people aren't going to get there. You can also probably cross off clinical summary document exchange as a meaningful-use standard, again because most people won't get there. Finally, I would argue that you can probably cross off electronic quality data reporting -- not process measures, like PQRI, but actual outcome data. I would argue you can probably cross that off, again because most physicians aren't going to get there.
One thing I would like to point out is how we do it at the Mass eHealth Collaborative. The Mass eHealth Collaborative was founded in 2004. We implemented roughly 550 physicians across three communities. We implemented them on electronic health record systems, created three health-information exchanges. We have a quality data center that sits on top of that that extracts quality information out of the health-information exchanges, and we use that for outcomes analysis and benchmarking.
Where we measure our version of meaningful use -- and we can argue about whether that's a good way to measure we actually measure down here. I'll show you this in a second. We measure it down at the EHR level to see how physicians are using the individual EHRs, and we measure up here, up at the quality data center level.
What I'm going to argue, actually, is that I think the vision ought to be to have interoperable systems where we measure up at this level, on the assumption that if you are getting robust reporting up here, you don't have to worry about what they are doing down here, because they must be doing it right.
But, obviously, there is a lot that has to happen between where we are now and being able to get to a ubiquitous, uniform system that has this kind of infrastructure in place.
Just to give you a sense of what we are able to measure, down at the practice level, what this shows is one of our usage reports, from January 2007. Every bar is a practice. We actually are able to see for what percent of the encounters that they documented in their system did they actually enter something on the clinical portion of the chart -- not just use it for billing and scheduling, but for what proportion of the time they actually used the clinical portion of the chart -- which is one indication that they are actually doing something meaningful in the chart. They could just be entering a text blob in the review of systems. So there is still that second question: Is it structured data? But we at least know that they are using the clinical side of the chart.
We are also able to dig down into the various elements of the clinical side of the chart. This shows it for a community perspective, but obviously this is a roll-up, so we are able to see it physician by physician. We get monthly reports send electronically via FTP. Then we process the data and then we deliver these reports back to physicians, so they can benchmark themselves against how other physicians are using the chart.
Some of the noisiness of that kind of data -- you can see here that it's hard to know what the denominator ought to be. For example, for medical history here, 45.9 percent of the time, that's 45.9 percent of all of the encounters that occurred in these practices. We don't know if it should be 100 percent. It's not clear that you should have taken a medical history 100 percent of the time or documented it -- if it was a quick follow-up visit, for example. So that's some of the messiness you get into with this.
The other thing I want to show is, just in the quality data center, we measure it at the highest level as well. We can see the data that's flowing through all the way to the quality data center. That's a good representation to us that the physicians are actually using the system. If they actually are entering data in structured fields and then appropriately triggering the data so that it goes through the interface and is then delivered to the quality data center, we will see the data up here, and we will see it in a pretty robust and uniform way, physician by physician. That allows us to identify gaps. Then we do go back to individual practices and try to identify, how come we are not getting anything on smoking status from you? We should be getting measures from you as a primary-care physician. Then we will walk through that a little bit.
One of the things that I tried to do here is just think about, what is the spectrum of measurement approaches that we might think about? That can go all the way from out-of-station and audit to HIE activity measurement. I would argue that unless we are going to raise the bar on meaningful use, we can just get away with some of these easier ones -- out-of-station and audit and surveys -- because, frankly, if all you are going to be doing is looking at e-prescribing and lab results delivery point to point into a practice, we are not going to get a whole lot of value out of that anyway. So why spend a whole lot of money on monitoring it? That would be my argument.
But if we do have this vision of a robust infrastructure for it, we can start getting into what I would argue is the HIE activity measurement piece of this, which I think would be the best. Let me just quickly flip through.
What I did was to try to look at some criteria of valuation that you might use for each one of those measurement approaches. I'll quickly go through my little bubble chart here.
Basically, what I would argue is that the goal of getting up to this HIE activity measurement is actually a very good goal to have and that ought to be the vision that we have. It would suggest that the focus of ONC discretionary spending really ought to be on making these HIEs and quality data aggregation points more accessible now, sooner, and also more geographically available. That really ought to be a primary focus.
I'll just make this my last slide.
The notion that I would propose is that HHS actually contract with three sets of organizations to get this done:
• The regional health IT extension centers, so that they are actually implementing according to meaningful use to begin with, so they have a clear vision of what meaningful use is and they are actually required, as part of the contractual obligation, to facilitate implementation at the practice level, so that they are able to certify that a practice got off on the right foot.
• Next would be the nationwide health information network nodes, which would be the nationwide health-information exchange, which is able to monitor and enforce data exchange according to HITSP-approved exchange standards and a particular reporting physician-level use of the HIE exchange.
• Finally, having quality data entities and public health entities.
All of these could be wrapped up in a single organization that allows you to start looking at the semantic standards. Are physicians actually entering codified data into structured fields so that we are able to measure things, and, by implication, they are able to use decision support as well?
Thank you very much.
MR. REYNOLDS: Thank you. Since this is playoff time in a lot of sports, you get the last shot.
MS. TORDA: I am very cognizant of getting the last word here.
I'm going to try to use my time to both reinforce some points that I think you have heard from many speakers over the last couple of days and also to focus a bit on the specific experience that NCQA has that we think is relevant to your considerations today.
NCQA, the National Committee for Quality Assurance, is a private, nonprofit organization. Historically, we have been best known for accrediting health plans. We also have a growing business recognizing physicians and physician practices, including medical homes. We develop and maintain the HEDIS set of performance measures, which has accompanying audit and certification programs that we think are relevant here.
Our program, physician practice connections, the patient-centered medical home version, was really the first program to very explicitly link the use of HIT to what is required to delivering high-value care. That's where we think the most relevant experience is. The requirements for that program are very deliberate about when HIT is needed and how it is related to delivering quality care. I'm going to focus on that.
I would also say that, thanks to Mike, we are a recognized PQRI registry. We received 560 submissions in 2008 -- a number which surprised us, frankly.
To reinforce some of the themes that I know you have heard already from many of the speakers over the last two days, let me start by saying that we envision that one of the important tasks related to the Recovery Act and to the meaningful-use provision is, at least for 2016, defining a set of measures of meaningful use, specific measures that relate to the overall objectives of health reform. We envision in the measure set for 2016. That measure set needs to be very deliberate about what they are going to do to improve the quality of the nation's health, the safety, and reduce unnecessary costs.
We recommend, strongly, working backward from that vision, recognizing that what we are going to be able to measure and report and pay on in 2011 is very different than what we are going to be pay on in 2016 and beyond, but that everybody needs to see that trajectory in those relationships.
Multi-stakeholder input into what that measurement set should look like is needed, as you have heard over the last couple of days. There are very important perspectives from the different stakeholders.
The phase-in strategy needs to be realistic.
Individual practices need support, and that support ranges from financial support to thinking about how to redesign their programs, what systems to buy, how to use those systems, and how to engage in quality-improvement activity.
Federal leverage is needed for standardization and progress at a faster rate than we have had so far.
Let me turn specifically to our experience with the medical home program. We have recognized through that program to date 370 practices of 4,000 physicians.
The first point is that the link between HIT capabilities and practice design must be made clear. One of the most common reactions that we get to our requirements from practices trying to meet them is, "I didn't realize that I could use my EMR to count across patients and come up with a performance measure." Really making clear what it is that HIT can do that goes beyond taking paper records and replicating them in electronic form is needed.
Our own published research shows that the presence or absence of an EMR per se correlates only weakly with clinical measures. Again, practices need help with thinking through how it is that HIT can bring them into the 21st century and lead to improved care.
Implementation is disruptive and takes time. We know this through much of the feedback that we get. But also simply the time between when a practice purchases our standards and then submits them for our review can often be long. We get lots of phone calls during that period about what's going on.
Small practices need assistance. We have recognized a lot of small practices, and we don't think a lot of this is impossible for small practices. Sometimes it's more possible, frankly, than in big practices, where you have a bureaucracy of one sort of another to deal with. But it certainly helps to be able to provide external assistance.
Financial support: Practices don't go through our programs unless they believe that there will be financial support of some kind or another, or financial rewards or recognition for doing so. It takes money. Particularly primary-care practices don't necessarily have that disposable income.
Getting to your questions about attestation and surveys, we absolutely believe that documentation of some sort is necessary, but that the reporting effort, of course, should be minimized. Again, our own published research shows that practice self-report is not reliable as a way of getting information. There are a number of reasons for that that we can go into if you want to.
There are many aspects of current EHRs that do not support the kind of reporting that we are talking about now, even for basic sets of clinical measures. So there is a pathway that is needed to get from here to there.
I just want to give you a few examples of some of the measures in our medical home program that might be, quote/unquote, shovel-ready, if you will, or will show at least some measures currently in use. They get at some of the areas anticipated by the Recovery Act.
We have a measure that is percent of patients seen in the last three months, with seven specified data fields entered. I think that builds on some of the ideas that you have heard here. It goes a little bit beyond just looking at whether one cell is filled in. It has to be searchable information. There are specified types of information that count towards the measure.
Moving up from that, use of registry functions: Does the practice use electronic data to generate lists of patients needing specified services? You are moving from, did they simply enter the data in a searchable form, to showing that they used it to reach out to patients.
Test tracking, electronic systems for lab and imaging we have talked about.
Use of e-prescribing: There is a progression here. We reward along that progression, going from standalone all the way to linked-to-patient-formulary, for all the reasons that you have heard about.
We do look at whether the practice can electronically report on nationally approved measures, a small subset.
Let me talk about what we would call validation. Our understanding of the use of the term would be, in Recovery Act terms, are the eligible professionals actually making meaningful use of the system? I'm going to describe how we currently do it and then where we hope we might be able to get, with the input of everyone over the last couple of days and the leverage and incentives offered in the Recovery Act.
Under our current process, we have an electronic, Web-based data-collection tool. That data-collection tool allows the practice to attach documentation, just like you attach a document to an email. But the practice has to create that documentation. It can range from a screen shot to a report, to having to physically abstract data from individual charts. There is no F7 key to push and get the reports that we need to document the measures that I talked about. We certainly hope that we could get there in the future.
So the practice creates the documentation. Our trained reviewers review that documentation. We go further. For a sample of practices, we do a physical audit onsite, whether the federal government requires that or not. We can talk about why it might or might not need it.
So we are hoping to get where others have described, that certified electronic health record technologies would be able to generate standardized reports, that that capability would be certified. We envision that the eligible professional would then authorize electronic submission of that report to CMS or whatever branch of the federal government is appropriate, through something like the registry function that we have now for PQRI.
We could talk about whether the federal government wants the assurance that comes with doing some random audits onsite. We hope that you would consider whether practices that are meeting the medical home standards -- once we know what the measures of meaningful use are going to be, look at the overlap with our current medical home requirements or our future medical home requirements, and whether there might be some kind of deeming that would be appropriate.
I'm not going to talk too much about electronic quality measurement, because Floyd and Mike have already talked about that.
I think, in summary, we have laid a good foundation for standardized reporting directly from electronic health records of clinical quality results. We are not there yet. Most systems can't do it yet. But hopefully we can get there a little faster than maybe we anticipated before the Recovery Act.
Just to reiterate, our recommendations would be to:
Important considerations -- the practical considerations:
MR. REYNOLDS: We'll need to wrap up, please.
MS. TORDA: -- specified to be reported through the EHRs.
With that, I'll stop and turn it over to you.
MR. REYNOLDS: Judy Warren.
DR. WARREN: This is for Phyllis Torda. On one of your slides, you said that current EHRs do not generally support reporting. My concern is that most EHRs have reporting functions that are fairly robust. Are you talking about the EHR software not supporting reporting or the implementation of that EHR that cannot populate reporting of the details of the elements of your indicators?
MS. TORDA: That specific statement referred to the kinds of reports that would be necessary to document the measures on my next slide. Percent of patients seen in the last three months, with seven specified data fields entered -- that's our measure. But those kinds of reports and certainly any standardized reports -- some can do it, to some degree, but most of it is pretty ad hoc across the different systems. The same goes for actually reporting standardized clinical quality measures.
DR. WARREN: So it's the implementation of the EHR that doesn't support the reporting.
MS. TORDA: I think it's both.
MR. REYNOLDS: Marjorie.
MS. GREENBERG: I wondered if you had any thoughts on how the general public might receive information about progress being made towards meaningful use of electronic health records, either in the aggregate or physician-specific, in the context of the transparency about how funds are being spent under the stimulus bill.
DR. RAPP: I believe the statute requires the Secretary to publish the names of the individuals who are qualifying for the incentives. That's something that is required in the statute. It's required in the Physician Quality Reporting Initiative, in the electronic prescribing, and in the HITECH Act.
There are two authorities. One is at the individual level and the other is at the group level. Depending on whether the Secretary would implement the group level, then -- but that would come out.
As far as so-called performance -- in other words, if there was reporting on this function, that function, that function, which ones those particular doctors are using or not using -- I don't believe that's in the statute specifically.
DR. EISENBERG: Can I take a stab at that, at some risk, having been a vendor, about six months ago? Vendors are often very attuned to something called KLAS rating, K-L-A-S. I could see that if there are measures of use which their customers are required to report on, perhaps KLAS rating could collect that information, at least for vendors and public reporting of use of vendor systems. That could be another mechanism to say how these systems have been used in practice.
DR. TRIPATHI: I would certainly agree with that. You could imagine all sorts of other kinds of ways to market this that would raise the visibility, I think.
But the other piece that we haven't really talked about is whether patient-facing applications ought to be a part of meaningful use, and that physicians use systems to populate patient-facing applications. It would certainly raise the visibility of it, if all physicians are going to be required to do that at some level.
MS. TORDA: Let me add my two cents. One of our research findings, actually, is that positive patient reports of experience do not correlate well, for whatever reason, with EHR use. There are a lot of potential explanations of that. But I think that there is room, in the sense of public education, to make for the public the connection -- and this goes to having the broad vision -- the connection between improved quality, improved safety and use of the EHR. Why is my doctor using this? How can the doctor make it a more patient-friendly experience?
But there is really room to make that case better. I think there is also evidence that the public doesn't understand why this is in part of the Recovery Act and how this money is really going to benefit them, so that there is an opportunity to do that.
We do know that the systems that are patient-facing -- and certainly the most well publicized is the Kaiser system, which is wildly popular with Kaiser members. Certain key functionalities of that system, which allow patients to make their appointments and see their results, are really wildly popular. I would certainly recommend that those features be given and consideration.
MR. REYNOLDS: Blackford.
DR. MIDDLETON: Thanks for a terrific panel and for your endurance.
For my question, maybe Micky would start, then Michael and Phyllis might pick up the second half.
Micky, I was impressed with your vision, the notion that from a disparate set of EMRs, potentially we might gather data into a reporting resource of some kind and from that, then, meet reporting needs of a variety of forms across a region or across a state, as you have described. My question for you is, how much variance do you see in the report types that might be expected from such a resource, given the needs of NCQA or PTE or PQRI or meaningful use, et cetera? How much variance or overlap is there in those sets of data requirements?
Then, Michael and Phyllis, if we see a way to actually gather these data into a reporting environment, how can we clean up the front end, if you will, where the data comes from, into the EMR? Can we actually design standardized documentation templates or minimal structure documentation requirements that make the backend reporting work so much better?
DR. TRIPATHI: That's a great question. I'm completely going to punt on it. On my slide I said a report on a national core set. Mike and Floyd and Phyllis, I think -- not to be so flip. I think that's a big part of the conversation, to say what we would start with as a core set of measures that are going to apply to everyone. I think that's the unique opportunity we have with raising the national consciousness around meaningful use. Perhaps the federal government can at least start that conversation, to say there surely has to be a small set that all of us can agree on that will apply to all primary care and none of us will be surprised around what those things are -- get a common definition of those so that, as you said, then we can roll those out into training, as well as the kinds of templates and the implementations to at least make that happen.
That would be a start, just a start.
For some of those -- for example, if you are doing diabetes and you are collecting glycosylated hemoglobins, it isn't as if someone is going to set up a lab interface to just collect those. Once you have that, you know they have a lab interface set up. Similarly, if you have a measure that has something around prescription history, again, that means they are getting all the prescription histories; they are not just getting it just for that measure.
So I think that can go a long way, if those are carefully selected.
DR. RAPP: I am not sure which of the two issues you are raising, whether it's the clinical data elements or the data elements that relate to so-called meaningful use, functionalities in the EHR itself.
DR. MIDDLETON: Yes.
DR. RAPP: Both. Going to the clinical data elements, I think the hospitals are one area where I think, at least in this area, it's not as challenging as for physicians. In the hospitals, most hospitals are pretty general, and so we have a set of three measures -- emergency department throughput, stroke measures, venous thromboembolism -- that basically all hospitals take part in. We have worked actively to try to see if we can get those data elements in an electronic health record at the inpatient level for the certification and so on and so forth. So that, I think, we can get to, and we will have at least a start.
One would to then go to all of the other measures that one is interested in at the hospital level. But I think once you get to the point that you can put these in the electronic health record, then things are just a lot easier and we are not limited to the measure set that we have for hospitals now, which is really held back by the fact that these are chart-abstracted.
Somebody said that you can't get outcome measures from claims. I disagree with that, because we have quite a few outcome measures that we, in fact, get from claims. But that's the one thing. On the doctor's side, this is a completely different kettle of fish. This was our challenge in the Physician Quality Reporting Initiative. Congress wanted all doctors -- and not only doctors, but every other kind of practitioner, from therapists -- and they even added audiologists at the last legislative go-round in MIPPA.
When you deal with that, there is no single data set that would apply to all doctors. So the way we have tried to tackle this is to start with a set of measures, 10 of them, that apply to primary care, prevention, chronic care, that will hit a lot of doctors and the issues that you are interested in.
What PQRI has been successful with is getting a lot of measures developed. The problem is, these measures are not really those that we have had a lot of experience with.
I think we can get somewhere -- we work with Floyd and others to get this basic set of data elements in the certified products, and that way we can start with collecting these measures, whether you do it through some output file or we try to take advantage of the health exchanges, which I think would be the best way, as I have indicated.
So that's the clinical aspects of it.
As far as the EHR itself and the use of the EHR, as we heard about the audit trails and so forth, if and when the Secretary has the capability of collecting data elements from EHRs, presumably those same audit trails would be able to be collected as well.
I'm not techie enough to understand the ins and outs of this higher level, probably. But certainly if that was a vehicle to do that, then that would be good, too. It has to be in the EHR first, though, before you can pull it out of it.
MR. REYNOLDS: One last comment.
MS. TORDA: Let me just put in my two cents. The short answer is yes. The cacophony that exists today exists, I think, for two reasons. One, fools rush in where angels fear to tread, and we are all fools that have rushed in. But we are certainly, I think, willing to harmonize our activities, particularly in the face of strong national leverage. That's what has really been a missing piece. I think the Recovery Act provides the potential of that national leverage to bring us all to the table and do it.
Just a word about the audit trails. Maybe this gets at the earlier question. My experience in working with some of the systems and the audit trails is that they are really hard to get to. They are very deeply in the structure. To get reports out, the document to use is a pretty difficult task at this point. So it's not just that you have an audit trail, but how easy it is to get to it and generate a report.
MR. REYNOLDS: First I would like to apologize to Jeff Blair. I have had to cut off the conversation the last two times. Sorry, Jeff.
That ends our last panel. Thank you very much. We'll take a 15-minute break and then we will be back to hear public comment and then wrap up. Thank you very much.
(Brief recess)
Agenda Item: Open Public Comments
MR. REYNOLDS: Let's go ahead and get started. Let me tell you the order. I'll apologize again. Mr. Jagannathan and then I would like Mr. Nayak, Dr. Rose, Robin Raiford, and then Tony Schueth. Two-minute limit, please. We'll get everybody in, and that will be a wonderful thing.
MR. JAGANNATHAN: Thank you for giving me the opportunity to provide some comments. My name is Juggy Jagannathan. I head up research at MedQuist, a transcription company. Hopefully, I will bring a unique perspective to this august group here.
Contrary to popular belief, we are not Luddites. We transcribe 35 million documents annually. We have about 1,000 computers in our data center. We do a lot of high-tech work in providing a solution.
One of the points I wanted to get across here in this audience -- I want you to do a thought experiment. What is the vision you have for what the physician interaction with an EHR should be or needs to be? Just think: Do you want to use a mouse-ready system to interact with an EHR or would you rather have a voice-ready system which would allow the system to interact with the system with speech?
Now think of the context of what's happening around you -- Google; IBM just released the Watson system, which can play Jeopardy. Look at where technology is going.
I also want to underscore that natural language is probably one of the best ways to document clinical care. It is arguably the best way, and it also passes the smell test that 100 years down the road, people can actually understand what was said.
I hope I have convinced you that it's an efficient way of documenting care. Now I will also go into why it is also one of the most effective ways of documenting care.
All of the transcription companies who are in the forefront of the transcription systems -- and I represent about 100,000 transcriptionists, who are actually editing speech-recognized documents. We are using speech recognition. We are using natural-language processing technologies --
MR. REYNOLDS: Sir, you need to summarize, please.
MR. JAGANNATHAN: We can create CDA now. We can create CCD now. We are using a combination of speech recognition, NLP, and human oversight in our role. Physician adoption of EHR should not preclude creating documentation, creating CDA content for distribution to EHR or HIE or PHR. That is one of the best and most effective ways, and the technology and the industry is ready to be part of the solution and not part of the problem.
Thank you.
MR. REYNOLDS: Thank you, sir. Mr. Nayak, please.
(No response)
MR. REYNOLDS: Dr. Rose.
DR. ROSE: Hello. I appreciate the NCVHS holding these hearings and the opportunity to comment on the meaningful-use issue.
My name is Eric Rose. I'm the product manager for McKesson's Practice Partner EHR. I'm also a practicing family physician and have been a CCHIT volunteer since its inception. I have co-chaired the CCHIT Ambulatory Workgroup for the last two years.
I have two very focused suggestions regarding the development of a meaningful-use definition.
First, the definition should not presuppose system capabilities beyond those reflected in the certification criteria for a qualified EHR. This may seem obvious, but since to do so would create situations where a physician who invested in a certified EHR system might be unable to use it in a manner deemed meaningful to HHS, unless the utmost care is taken to ensure harmonization of the meaningful-use definition, with certification criteria, there is a good chance that some subtle inconsistency could make it into the final version.
My second suggestion is that ONC, in defining meaningful use, consider the full spectrum of providers and avoid requirements that would compel any provider to use technology not relevant to his or her practice. For instance, while the value of immunization data flow between EHRs and public health systems is beyond question, it makes little sense to require a podiatrist who doesn't even stock immunizations in her office to invest the time and money to implement the technology required for such data exchange.
These sorts of considerations could be handled either through a lowest-common-denominator approach, where only those requirements appropriate to all Medicare providers are implemented, or a segmented approach, where the requirements vary by practice characteristics. But they must be addressed. After over four years of developing and refining HR certification criteria, these issues are at the forefront. The CCHIT volunteer community should be kept in mind as an expert resource to draw on in the development of the definition of meaningful use.
MR. REYNOLDS: Thank you, sir. Robin Raiford, then Tony Schueth, and then Rodney Peele, and then that ends the list.
MS. RAIFORD: Hi I am Robin Raiford. I'm on the HITSP Education Committee.
At the urging of Walter, I share with you the largest single-page submission to NCVHS. It's a summary of the 7,000 pages of HITSP. It's three feet by five feet that I'm holding up in front of you. I'll send it to you electronically. It's all the components of HITSP, much of which is the continuity-of-care document and those pieces.
I just want to point out that a silent voice that has kind of been here is that of a consumer. In the HITSP webinar series, there is a story called "Steve's Story," which is, what would it be in an interoperable world from a consumer perspective? I wrote "Steve's Story." It's the story of my nephew, with a brain tumor, who has to repeat his history over and over and over again, who has had numerous CAT scans and whatever. His grandmother died, who was my mother, because they couldn't trend her lab results across visits, across doctors, which is a sad way to lose your mother.
I would just encourage you, when you come up with meaningful use, even though it's tough for the doctor, it's nothing like for these consumers. I read letters at the White House for a year and a half in the Clinton Administration -- how these people get trapped in the system and get dropped. It's a preexisting condition. Then they are Medicaid. Then they get disability. Then they make too much money in disability, and now all of a sudden they have to wait for Medicare. Whatever a doctor goes through to get to meaningful use, it's nothing -- nothing -- to what these very sick people go through in Medicaid to get through this. Who holds their hands? There's a nurse out there somewhere listening to them and going through that.
So I urge you to strongly consider the continuity-of-care document, even if it's just level 1 and a summary, so these poor people do not have to repeat their problems, their meds, their allergies, and their past medical history to the one of 20 doctors they see in the course of a year.
Thank you very much.
MR. REYNOLDS: Thank you. Appreciate it. Tony Schueth.
MR. SCHUETH: Hello, everybody. My name is Tony Schueth. I'm the CEO and managing partner of Point-of-Care Partners, which is an HIT consulting company.
I have testified to the NCVHS in the past. You know me as an expert on electronic prescribing and electronic prior authorization.
I want to bring two thing to your attention. First of all, I'm not an attorney, but as I understand it, if a physician accepts MIPPA money, they are statutorily unable to accept ARRA Medicare incentives. But there will be documented-care pay(?) beginning in 2012, if I understand this correct. I checked with a few attorneys and a few others.
As someone who has his ear to the ground, I can tell you that there are a large number of physicians that are considering that the bucket for ARRA is much bigger than that for MIPPA. Despite what you might have heard yesterday, EHR purchasing is slowing right now. We are predicting that the rapid growth in e-prescribing that Rick talked about this morning is going to sort of level off as some of these physicians wait for the ARRA money.
So what's the implication of this? 2011 is going to be a big year for EHR adoption and electronic prescribing. Also the nation's largest payer, CMS, is not going to realize the kinds of savings that they might have realized as the adoption levels off, in my view. I'm not sure if this has been communicated to you, but in SEMI, where I'm the project manager, the autos have found a savings per prescription of $3 to $5. So that's a meaningful amount of money to CMS.
I have been here before. I know you don't want me to just raise problems. You want me to sort of mention solutions.
One of the things that I would like to suggest is, if the bar were to be raised on electronic prescribing so that it's --
MR. REYNOLDS: Tony, you'll need to summarize, please.
MR. SCHUETH: -- higher than Medicare, possibly the statutes could be changed. Some of the bar raising would include some of the things that Rick talked about earlier. I guess I can't go into detail. I'll send it to you electronically.
The other thing I really want to talk about and mention is, I'm not sure if you are aware, but Surescripts right now is asking for activity reporting from its EMR and e-prescribing vendors. This activity reporting would be very similar to sort of meaningful use. They have a format for the submission. It goes electronically to Surescripts and then on to the payer, who can then see this information. There is a certain level that can be seen at Surescripts as well.
So the point is that activity reporting can happen today. It is happening. You may want to actually --it's too bad Rick didn't talk about that or isn't around to give you a little more detail on that. But that is happening today. It's happening through an intermediary. It could also go directly to some sort of data repository.
Thank you for your time.
MR. REYNOLDS: Thank you. Rodney Peele.
MR. PEELE: Good afternoon. I'm Rodney Peele. I work for the American Optometric Association. I'm making these comments also for the American Academy of Ophthalmology and the American Society of Cataract and Refractive Surgery. We represent more than 50,000 optometrists and ophthalmologists, and we have tens of millions of patients.
We are concerned that we will be shut out of this process, no matter how meaningful our use is, because there is no certification available yet for eye-care EHRs. We continue to pledge our support to help develop a certification process for eye-care EHRs in 2010, so that our members can be using certified products. We ask that you recognize the limits of product certification on meaningful use and to assist us with rectifying the situation so more physicians qualify for recognition of their meaningful use that benefits our patients.
We'll submit more extensive comments in writing. Thank you.
MR. REYNOLDS: Thank you. That completes everybody who signed up for public testimony. The time limit was good.
Again, we would recommend that anyone who wants to submit written comments for us.
With that, that ends the prepared portion of the program. Now we have to figure out what we do.
Agenda Item: Subcommittee Discussion
MR. REYNOLDS: Since Dr. Blumenthal is here, I asked Chuck Friedman whether or not they want to make any comments, and I think he want to make a couple brief ones. Then we'll continue on.
MR. FRIEDMAN: Sure, Harry. Thank you.
Very, very briefly, on behalf of David Blumenthal and myself and the entire staff of the Office of the National Coordinator, I want to issue a great deal of thanks to a great many people for these wonderful two days. I have to say, thinking back on a very long time, this has to be the most informative two days I can ever remember. Even though this is on paper, it is an extremely valuable collection for us as we go forward.
I would like to thank you, Harry, the NCVHS members, the NCVHS staff, who worked so hard in preparation for this, the 39 panelists, who gave us wonderful material, everyone who attended, and among those who attended, those who provided public input. This was marvelous. Thank you so much.
We received information, knowledge, and wisdom over these past two days that will be of enormous value, as I said to us going forward. We will use this material in very, very important ways. Again, I can't thank all of you enough.
MR. REYNOLDS: Thank you. Tony.
MR. TRENKLE: I want to echo Chuck's thanks to the committee, to you, Harry, and to all the staff who helped support this effort and to the speakers, who came in on short notice, as well as the people in the audience, both here and those who listened remotely.
I think we have made a lot of strides over the past couple of days. I was sitting next to Dr. Blumenthal, and I know it was very helpful to him. It gave him some insights that brought some different perspectives that he hadn't, probably, totally thought through before, as did all of us.
I think it also shows the value of the NCVHS that you can bring experts together on very short notice and come up with some hearings that are meaningful and useful to us. We look forward to moving forward and working with the committee as we begin to figure out how to actually implement this over the next several years.
MR. REYNOLDS: Thanks to all of you.
Before we start the deliberation, I want to thank the committee and the staff around the table. You kept the energy the whole time. There was plenty of opportunity to lose that.
The other thing is, what I'm excited about is that every seat participated. That's what has always been good about our committee. That's what has always been good, and that's why we have such a strong group. Everybody played. We got a lot of great perspectives, and that's one of the things we try to do.
With that, I'm going to give you an update on what we are going to be doing. I'll reiterate to everyone that our goal is, within the next two-plus weeks, to make sure that we have a document that clearly describes what went on here. We have not been asked to give specific recommendations. However, we will be. Margaret Amatayakul is already working on a set of draft categories that we want to kind of think about.
Back to Chuck picking up the paper, if we just give him the paper, that wouldn't be very helpful. We're not going to go there.
We are actually going to go through, and a number of us have touched base at different times about some possible categories of how to think about this. So Margaret is going to produce those. By the end of the day Monday, we will have a draft out to the full committee on what that would look like.
In the meantime, when we open it for discussion here in a few minutes, I would like you to talk about categories that you might have heard that you want to make sure are included. Understand, it's not directing the conversation. It's grouping the conversation. That's what we were asked to do, to report out, not to push it in one way or the other and not to add any value-add, until such time as we would be asked that, possibly. That's part of the whole process as we move forward.
So that will be happening.
We will get copies of all the submitted comments to the committee as soon as possible. We will communicate with you quickly as to how fast that would be out to everybody. It will be out on the Web site. You won't get a UPS package. It won't fit in your front door. We'll make sure you know where they are and you will have access to look at them -- all the written comments from the people that didn't testify, because we want to go through those.
I will be requesting -- and some of you have already stepped up -- a member and/or a staff lead on each of the categories that we come up with, so that we make sure, as we look through it, we are calling out the right things that we need to add to the extensive testimony that we have heard. I will ask each of you to step up in those ways. If we have a category, as we start putting it together, if you read something different in the written testimony or something that adds further clarity to it, then that's how we would adjust it. So we will look at all of it, but again, we are doing a readout by category, making sure that we make the points that we heard, clearly, in there.
With that, let me open it up for any comments or any discussion by the committee. Again, the purpose today is to gather as much as we can from each other as to what we thought, what we heard. Again, it's not that that's going to be exactly what we do, but it's very important that, if you have anything to add to the thinking that we are going to do as we put these categories together, you are able to say it while we are together. Then, obviously, we will start our process.
Again, to those of you in the audience, our document, when it is completed, will, in fact, be posted to the Web site and it will be delivered to ONC. So it will be returned to the public environment, as it should be and as is always our practice.
With that, let me open the floor for discussion. Paul and then Justine.
DR. TANG: As the committee's lumper --
MR. REYNOLDS: Yes, we ought to get clearly divergent views over the next two comments. Go ahead, please, Paul. You first.
DR. TANG: I thought it was a really good two days. Thank you to the staff for organizing this all and for bringing so many people together, and for the clear presentations.
I thought of three categories to look at meaningful use. I'll just throw this out.
One is to have the system that interacts with the human professionals and the human patients be meaningfully usable. That's very broadly encompassing. That's the feature set, the way it presents itself -- i.e., the user interface -- to the humans interacting with it, et cetera, the coding. It's the meaningful usability.
Then there is the sort of adoption, which is the meaningfully used. That speaks to the adoption and getting it in, getting people interacting with it.
I think the endgame, which I heard the strongest from these two days, is meaningfully useful.
That's a bit of a play on words, but there are different ways of characterizing what we are looking at. Then we have to decide what is the most important. One of the things that struck me about Micky Tripathi's -- near the end, where he talked about if you get the meaningfully useful part right, then the other two probably happened, and you don't have to micromanage those pieces, or even to instrument them very finely, which all costs money.
At any rate, that's one suggestion.
MR. REYNOLDS: Justine.
DR. CARR: I echo what everyone said. It was an amazing two days.
I actually agree with everything that you just said. But I wanted to add some granularity to it. I think that there are just a couple of things that are probably worth -- kind of as background information.
We heard a lot about, meaningful to whom? I think, as we put these things together, we want to think about what's meaningful to the patient, the provider, the payer, to public health, and, really, to the nation.
A second area, again sort of in the category of definitions, is starting with exchange. I think I understood something that I didn't understand before. There are large groups, like Partners or Kaiser, that unto themselves they exchange, but they don't exchange, let's say, across the street to another entity --
MR. REYNOLDS: Only using them as examples, correct?
DR. CARR: Only as an example.
So understanding what the meaning of exchanging is -- and I think the fact that Amazon doesn't exchange regionally -- one Amazon, one exchange -- I thought that was important.
The second thing, of course, is, what is EHR? I think we need to pay close attention to the concept of "EHR Light," in the context of meaningful use. We heard a bit about the fact that an EHR Light can have a big return early on -- thinking about that and sort of getting everybody engaged.
I'll stop with that.
MR. REYNOLDS: Could I add one thing to that comment on EHR Light? Would you take the amendment that maybe it's "EHR, the Journey"? If we just use one word, it starts to --
DR. CARR: Right.
MR. REYNOLDS: I think, as Paul said, we have heard the endgame. The endgame is much more than EHR Light.
DR. CARR: Right, but I think we need to define it in terms of the endgame and where we start.
MR. REYNOLDS: Yes, right. But it's within the journey.
DR. CARR: Absolutely.
MR. REYNOLDS: Good.
We have a list. We're going to get everybody, I promise. Paul is playing off that one, and we have everybody else on the list.
DR. TANG: I think this was an important point, what you just said. I think what Justine is saying is, the concepts should no longer be tethered to the old labels. EHR isn't a thing; it's a set of data that is useful in care or health. Similarly, we may have come into the meeting thinking of HIE as either an organization or a set of operations, but it's really an exchange of data in a meaningful way.
I think Les Lenert mentioned a vertical exchange. That's just as valid when that's where the data has to go to support health and health care. We need to recognize meaningful exchange. The whole "meaningfully useful" is the bigger term. I think that's a big thing we learned in the past couple of days.
MR. REYNOLDS: All right, we've got Marc.
DR. OVERHAGE: This kind of plays off of the lumpers and splitters a little bit here. I'm a lumper. In some ways, as I listened to what we heard, we heard a lot about where we need to end up -- too much about where we need to end up. I think some of the things about meaningful use from multiple aspects and so on belong there. That's where we need to go.
We heard some about our current state, what some of the constraints and limitations are. We heard vendors says, two years to get the next version out; we don't have two years. So there is some current-state information.
Then, in between, we have to navigate a path. To me anyway, there are three major categories of that navigation:
• One is the data evolution that we heard about, starting with targeted things, evolving onward.
• I couldn't come up with a great word for this -- sort of "value realization," the EHR journey that you just described. That's somewhat orthogonal. It's related. There are certain functions you can't do until you have certain data, but it's somewhat orthogonal.
• Then measures and measuring. How do you identify and report progress along that continuum?
Those seem to me the three buckets of the journey.
MR. REYNOLDS: Good. Blackford.
DR. MIDDLETON: I will resist the temptation to be categorize myself as a lumper or a splitter. But just to add on to the comments so far, I guess if I was to distill what arise in my mind as kind of the essential buckets for thinking about the testimony, they sort of revolve around system capabilities and functionalities, which was the focus of a lot of our conversation:
There were a couple of other thoughts, just to go on for a second, if I may.
I wonder if it would be useful to organize the testimony, what we have heard, as sort of a cross-tab of stakeholder and goals by stakeholder, by criteria or features or capacities or functions of EMR. In my own mind, I would love to see that, because I think it will help us analyze the great array of testimony that we heard and see where there are clusters of related concepts that we may not be able to get to in any other way.
I think we heard a little bit about value and utility as well. Once we have a list, how do we actually prioritize or rank it in a way that's meaningful, so that our assessment or the interpretation of these testimonies is not only accurate, but also useful itself? I think we have to do some work there. Part of that work might be to suggest to the national coordinator that, possibly, testimony is not going to be sufficient, that testimony needs to be substantiated with a careful review of the evidence to substantiate the testimony so that we have a clear idea and a defensible position going forward -- or they do.
I think part of that evidence review might suggest also these value clusters, this idea that we came up with, that might help us organize or prioritize these criteria.
Lastly, I would just suggest that the "light, medium, and heavy" discussion of the technology -- we have to be very careful that that doesn't become a distraction. All of this needs to be technology-independent at a level. There will be various forms of light, medium, and heavy each and every time, if you will. I would almost propose that we avoid that language. While we are thinking about technology independence, if you will, that's one of these overarching principles that I also suggest we try to arrive at. There may be a small number of them -- utility, usability, value, and then these process ideas -- to sort of rank the testimonies.
MR. REYNOLDS: Leslie?
DR. FRANCIS: I want to raise a couple of additional or different overarching themes.
One, given that John and I are co-chairs of the Privacy and Security Subcommittee --
MR. REYNOLDS: We would be disappointed if you didn't.
DR. FRANCIS: Yes, you would be disappointed if I didn't mention it. I do think we should simply pay attention to what was said about the role of privacy and security in facilitating all of this, what needs to be done and so on. There were suggestions, for example, about standards.
The second thing is that a theme that went along in a number of points was whether the question about how to address underserved populations is essentially a resource question. So getting more in the way of knowledge, getting more in the way of funding out to Medicaid providers, safety-net providers, providers who serve the underserved -- whether that's really just a matter of getting more money to them or something else -- paying attention, for example, to what certification ought to look like for varying practice situations.
My third observation -- and I don't quite know what to do with it, but it's an observation -- I think there is a fundamental divide between people who think that the way this is going to need to be approached is basically in the private sector, with federal support of money, or whether the way this is going to need to be approached is much more in the public sector, and what that would look like. I think we are going to need to pay attention to -- maybe this is borrowing a word from John -- the governance and whether that's really a public-sector issue or public/private, and exactly how that will look.
MR. REYNOLDS: Jeffrey.
MR. BLAIR: Those of you who know me recognize me from my engineering background, where I always, typically, wind up structuring things. I'm going to do exactly the opposite this time.
MR. REYNOLDS: We won't recognize you.
MR. BLAIR: I know. I'll have to have a different name.
There were several things that came up where I feel like, as Margaret begins to structure the information for us, we really need to give her the freedom and latitude to be able to modify that structure, to incorporate, for example, the fact that if we look at certain data elements for meaningful use -- and I'll just give an example here, fill notifications in e-prescribing. Is that a measurement of the health-care provider, the clinician, or the pharmacist or the patient? It may be all three. I could start to think of many, but time won't let me list those.
But those are really important, because this gives us the chance to begin to move beyond just simply saying meaningful use from the standpoint of the clinician to meaningful use getting closer to the patient. I don't think we want to lose those.
So I just want to make sure that Margaret has the flexibility to recognize areas where data have multiple use. That same data element may have use for better patient care, but it may have use for public health or clinical research.
MR. REYNOLDS: Jeff, one of the testimonies talked about acceptance of new responsibilities by clinicians, payers, patients, et cetera. I think what you are saying falls into that category.
MR. BLAIR: Let me add one other piece here. Some of the people that spoke at the open mike -- there are some outliers there, and I don't want to lose those either. One of those that kind of wasn't in our framework really struck me as something that I don't think we want to lose. Somebody wound up saying that of all the areas of meaningful use, we never included a measure of having more accurate and complete diagnosis. EHRs and CCDs absolutely will contribute to that. That is a huge area for savings in health care. Those are the types of things that I just think -- we need to make sure we don't lose those.
MR. REYNOLDS: Good. Marjorie.
MS. GREENBERG: I want to thank everyone as well. Particularly -- always at risk of not mentioning somebody -- I want to thank Denise and Karen, who just, I thought, did an amazing job in their lead staff role, and, of course, working with Chuck and the ONC group as well, and then also the NCVHS team. We keep referring to them, but I would like to mention by name those who are here
MR. REYNOLDS: Please do.
MS. GREENBERG: Debbie Jackson and Katherine Jones and Marietta Squires and Cynthia Sydney and Jeannine Christiani.
I want that on the record. The presentations just kept appearing on the slides and we just kept getting copies of things, and everything was amazingly smooth. I really appreciated that.
My reaction to this whole buildup was to leave the country for 10 days. So I really take no credit, except that, hopefully, maybe in the past for something I have done. But I really am very grateful to the all the professionals who helped put this together.
As is my wont, I sort of agree with what everybody said. But I was really attracted to Blackford's suggestion as one of the ways of looking at the information, because I'm confident that Margaret will not feel bound by what people have said, but will be enriched by it, and then we will be able to react to her drafts.
I think there are several different ways you are going to want to cut it. But the idea of looking at the stakeholders as one cut and what would be meaningful to each of those categories -- I think we heard a lot of interesting things from employers. We just received a few presentations on that, but it definitely had some different things, and I think it's important that we not lose that.
The consumers -- just in the last panel, I guess, we had an official consumer representative. We are all consumers in some regards. Because we didn't have a lot of consumers, I think we should look back, possibly, at the hearings we had on medical home and just what all those features of meaningful care are that we have been talking about. Our theme for this past year and the coming year, without knowing that this was going to be on our agenda, is the person-centered health. I think we did hear a lot of people referring to that. It was perhaps most eloquently captured by one of our last public speakers. What really would be meaningful to patients and to people with a lot of chronic conditions, et cetera?
I think, although we will summarize the hearings and the data from the hearings, there are some other areas that we can draw on that the committee has been looking at and discussing probably over the last 60 years, but I would say, more over the last year. I just want to suggest that.
Also the provider perspective that we heard in public comment -- the whole issue that if this doesn't make workflow better, if it makes it worse, then we are probably not going to get meaningful use either.
Although there seemed to be a lot of consistency across a lot of the testimony, I think there were some really different stakeholder perspectives. I just want to endorse that concept.
MR. REYNOLDS: Walter.
DR. SUAREZ: Thank you. I think we heard a lot during these two days, as everybody has said, and I think we heard a lot from things that were much broader than the actual focus of trying to figure out what to include in a definition of meaningful use, which was great. I think what we heard were a lot of innovative and more challenging things about health-information technology in general.
I want to make a few comments.
The first one is the definitional issue and the scope. We started with a hearing on meaningful use of health IT. What is health IT in that respect? Is it EHRs only? Is it EHRs and PHRs? Is it EHRs and e-prescribing? Is it HIEs and EHRs? We have to really, at the end of the day, concentrate and focus the scope. It seems to me that we were mostly talking about EHRs and how EHRs relate to e-prescribing functions or relate to HIE connectivity or relate to PHR interoperability or things like that.
So that's one issue we have to start with.
I would group the things that I heard in four areas. The first one is the purpose. I think we heard a lot of reasons about meaningful use around the four or five major things that we always talk about -- quality, safety, access issues, cost and efficiency issues, and value, ultimately, the concept of value. All those were kind of the framework for the purposes of meaningful use.
But I think we heard more specifically about -- and this is a second element -- the features and capabilities. I would separate internal features and capabilities from external features and capabilities, internal features meaning the internal characteristics of EHRs and the things that need to be enabled, not just that the system is capable of. Those include things that we heard from the very beginning -- registries, clinical decision support, privacy and security elements, and specific applications of internal features.
For external features and capabilities, we heard about interoperability and being able to exchange messages, comply with interoperable standards, quality data collection and reporting -- that is the second one we heard -- and a number of other features and capabilities related to external characteristics of an EHR.
The third element I heard is the application areas. Here what I heard is that there are differences, of course, between meaningful use in hospital EHRs, ambulatory EHRs, EHRs implemented by others. Then there are also the application areas for specific sectors of the industry, like public health, Medicaid, the medical homes model, the underserved. What are the elements that, in a meaningful-use definition, we should consider recommending to be included in order for the EHR to communicate public health messages or Medicaid --
MR. REYNOLDS: Let me jump in. We will not be recommending.
DR. SUAREZ: Oh, I'm sorry.
MR. REYNOLDS: Since this is a public record, I just want to continue to remind everyone --
DR. SUAREZ: I used the wrong word.
MR. REYNOLDS: I'm not worried at all that you used the wrong word. I'm bringing it back to make sure that it's a public record -- you don't need to apologize. That's my job, to make sure that the record sticks as it should be.
DR. SUAREZ: So noted, then.
The last area I wanted to highlight is operational issues. Here is where we started to hear about the idea of what the requirements are that we would include in a definition of meaningful use -- the requirements in terms of going beyond the certification of "capable of" to actually expecting that the system will have this or that and then balancing the bar between setting it too low and making it useless or setting it too high and making it discouraging for the adoption of EHRs, and the metrics that we would need to use in order to assess and ensure that systems and organizations are implementing EHRs and actually are complying with this meaningful-use expectation.
Another very critical element was scalability and the ability for those measures and metrics to be scalable. When we talk about EHRs in a single-doc/solo-practitioner clinic, like in New York City connecting to public health, versus a very large system implementation, we are talking about different levels of implementation of the metrics. So scalability in those metrics is something important.
Then we talk about three other things inside the operational issues: Assurance and the way we can look at assuring the meeting of those; enforcement, as potentially the next level of this assurance; and then the timeliness and the timeline itself of the transition. There are a lot of questions and challenges about being able to do all this in the timeframe between now and 2011. Many of the things that need to be done between now and 2011 have to happen in the next few months in order to get a system and be able to, by 2011, demonstrate that it is meeting the meaningful use.
So that's my summary. I'll stop there. I have a number of other comments, but I'm sure we will be able to exchange more offline. Thank you.
MR. REYNOLDS: Judy.
DR. WARREN: As I thought back about the testimony we heard, it seems to me -- I'm almost with Marc about lumping here -- that there were two issues that we really heard about. One was about the technology and its ability and readiness to serve the purpose, and the other, which to me is probably a more important function, is trying to get the clinicians to exhibit behavior of meaningful use. It really is a person thing -- and not only the clinicians, but patients and families as well.
So it's the behavior of people using the system to be sure it's meaningfully used and it's the behavior of the systems to make sure that technically they can support that use.
MR. REYNOLDS: Jorge.
DR. FERRER: My observation is very global, but I think there are themes that we heard. It is a phased approach. It is not a one-fit model; it's really phased.
Then there is also the progression path. A lot of the testimony also is progression path, based on the sector which the testifier is coming from. As fragmented as the testimony is, when you really delve into it, you still are going to view a very centric view of what that individual has given testimony on. So you have to be sensitive to that, because if they are testifying from a provider's standpoint versus a plan, their testimony will really reflect that doctors want to provide great clinical care, but plans really want to evaluate the cost-effectiveness of that care.
MR. REYNOLDS: Chuck.
MR. FRIEDMAN: Thank you, Harry. I'm really hesitant to say anything, but I feel like I should make a comment along these lines. It's a comment in the direction of conservatism in preparing the summary. This comes from a couple of perspectives.
The first would be a conservatism in the sense of suggesting that you might want to stay close to the structure of the law, because that's going to be our foundation as we move forward. The closer the structure of your summary is to the structure of the law itself, all other things being equal, the more helpful I think it will be.
This is sort of derivative from that first comment. I think it was that kind of conservatism to stay close to the structure of the law that suggested the structure of the panels for the session. So derivative, then, from my first statement is a suggestion that you might want to be sure your summary preserves some of the constructs that were implicit or maybe even explicit in the structure of the panels themselves.
The three things that a meaningful user has to do: make meaningful use of a certified product, exchange health information, and report quality measures.
Another construct along those same lines would be what we called the glide path. It was used in the names of the panels. I realize the "glide path" is a bit unfortunate terminology, in the sense that a glide path is a way down, and I think we are talking about a way up. So we are open to suggestions. But "glide path" is a good metaphor for other reasons, and I think we introduced it into the structure of these sessions for those reasons. It would be good if that were preserved in the summary as well.
MR. REYNOLDS: Good. Paul.
DR. TANG: This comment is to follow up Marjorie's comment and Judy's about the patient. Just to remind ourselves -- I know it's in the transcript, but I want to make sure it gets in the summary -- the first panel responded to the direct question, "Should patient access to and use of be part of the definition," and they unanimously said yes.
MR. REYNOLDS: Mike.
DR. FITZMAURICE: On the definition of meaningful use, a lot of what I heard, on the one side, could be the characteristics of the definition itself versus the expected impact of the definition. I was pleased to hear that several of the speakers prioritized what they wanted to see. They prioritized the electronic health record functional priorities and the public health functional priorities. I think that could be a great help.
I heard, "Avoid provider burden." That, I guess, falls under characteristics of the definition.
Look at the difference between capabilities and our vision -- that was guiding a lot of comments -- versus comments on the other side, "Look at what we have in actual practice today, and don't push it too hard."
The use of structured data came through strongly. even when people didn't use the term "structured data," they talked about the need to get the data elements, the picture of the data elements, for labs, for example.
I thought they really did us a good service by giving us some criteria for what the definition of meaningful use should have. What are the properties of a good definition of meaningful use?
Also there was comment on how to move from a claims data focus to a clinical data focus for measuring the same things. That is, how are we improving patient outcomes? How do we move -- and it's very hard to move from a claims data focus, which we have, to a clinical data focus. Do we do it piecemeal? Do we build that piecemeal into the definition of meaningful use or not?
The importance of connectivity as a part of interoperability keeps coming back. Many places aren't connected to do something, whether it's rural areas or not. We have to be aware of that. The Department has to be aware of that when we come out with definitions of meaningful use.
I didn't hear any comments -- or maybe I just wasn't listening at the time -- on different requirements for different kinds or categories of providers, except for ophthalmologists, and that was through the benefit of a public comment. I don't know if public comments would guide us not to try to categorize priorities and give different definitions of meaningful use for them or to make it as general as possible and leave it to the implementation.
DR. REYNOLDS: Okay.
DR. FITZMAURICE: I guess I could make one more comment, if I may. That is, I didn't hear how ARRA constrains us, constrains HHS, in developing an incentive reward system for providers over time, how it affects the timing of what we can do, how it affects the amounts that we can do, and how it affects the relationships of the timing and the amounts to the functions and the outcomes that we seek. That might have been a good conversation to have.
MR. REYNOLDS: Okay, excellent.
Again, playing off of Chuck's comments, we want to be very pragmatic. We want to tell the story of everything we heard. We want it to match pragmatic, and then we may have some other things that we put in there that are other things we heard that may need to be considered, whether they be other aspects of what we do or others do. They could be things that we consider doing as we do this.
Marjorie?
MS. GREENBERG: I got at least one e mail sort of asking, what next? Of course, we know what we are going to be doing next, with the report. But what is the kind of timeline for the Department to process all of this, to begin to identify how this is going to play out?
I just wonder, since this is a public meeting, if there is anything that anybody can say about that. We know that the first incentive payments are -- what? -- 2011. So sometime between now and 2011, this is going to be worked out. Is there anything more specific or are there milestones in particular that people might be aware of? If not, that's fine. I'm just trying to be responsive to questions I have received.
MR. REYNOLDS: If there is a comment on this, I would rather it not be from NCVHS.
MS. GREENBERG: No, I don't think it could be from NCVHS.
MR. FRIEDMAN: I was going to say, I don't think there are any specifics.
MR. REYNOLDS: Okay. We have been asked to produce our document, and we will do that.
DR. SUAREZ: Could you explain to me a little bit more about the process we are going to follow as a committee?
MR. REYNOLDS: I just went through that.
DR. SUAREZ: I might have missed it, I'm sorry.
MS. BERNSTEIN: Margaret A. is going to produce a summary and then?
DR. REYNOLDS: We are going to produce a summary and, as I have said six times in the two days, it will, in fact, be put out on our Web site and it will be delivered to ONC.
MS. BERNSTEIN: In between those two things, I think there is some -- is the committee required to meet again in order for that to --
DR. REYNOLDS: No, we are not. We are not required to meet again. We are doing nothing more than summarizing it. We are just coming up with ways to summarize it.
DR. SUAREZ: So this is not a position statement where the committee approves --
MR. REYNOLDS: We are not making -- we are reading this out in a way other than taking this stack of paper and handing it to them and going, "Thank you very much. We had a great hearing. Use it."
MS. BERNSTEIN: There will be some review process.
MS. GREENBERG: The Executive Subcommittee essentially will be the review group.
MR. REYNOLDS: That is exactly correct.
MS. GREENBERG: And so there will be --
MR. REYNOLDS: I will go through it again. I'm tired of going through it. After two days, I'm tired of going through it, but I will go through it again.
Here is the point. By the end of the day Monday, we will have some idea of the categories we are going to use. Each of you, by Tuesday, will be asked to pick a group that you will be involved in and help go through the testimony and make sure that you call out any things related to these things. We hope to have some kind of summary document by the end of next week. It gives us part of the following week to review it and turn it in to ONC.
Again, we are not making magic. We are not taking it to the committee and having arguments over our position or anything else. We are trying to tell the story of the two days in a meaningful way that goes against meaningful use and the ARRA, period.
The other thing is, we have never done this before. We have never had this much information to do in this fast period of time. If anyone wants specifics, get over it. We're going to work this out in the next few days and we're going to make this happen.
DR. SUAREZ: Harry, will you define "meaningful way"?
(Laughter)
MR. REYNOLDS: At this late date, if you walk out with me, I'll explain "meaningful way." It has been a long two days, guys.
Leslie, the last comment.
DR. FRANCIS: This is just a process question. When will the submitted documents be available for us to look at?
MR. REYNOLDS: They will be on the Web site by the end of tomorrow. We have announced that over and over again. They have to be in to us, and they are submitted to the Web site by the end of the day tomorrow. That's it.
DR. FRANCIS: So they will be up on the Web site --
MS. GREENBERG: They won't be posted by the end of the day tomorrow.
DR. FRANCIS: That's the question I had.
MR. REYNOLDS: I don't know. You have to tell me that.
MS. GREENBERG: People have until the end of the day tomorrow to submit them to us.
MR. REYNOLDS: That is correct.
MS. GREENBERG: Then we will have them all posted on the Web site within, hopefully, 24 hours after that.
MR. REYNOLDS: We should all have access to those by the end of the week.
DR. FRANCIS: And I take it those will be included in our summary.
MR. REYNOLDS: Those are absolutely equal to anybody that spoke to us, in any way, shape, or form.
Again, we are making this up. I ask you to play. We played the two days. I ask you to play hard between now and then. We may redo, we may rethink, but I'm asking you to play. We have never done this -- fast-track, hang on. If they could pull the hearing together this fast, we ought to be able to put a document together.
Thank you very much. I appreciate everybody. We are adjourned.
(Whereupon, the meeting adjourned at 4:30 p.m.)