An Open Letter to the New National Coordinator for Health IT: Part 3 — Certification As The Elephant in Health IT’s Living Room

6a00d8341c909d53ef01157012476e970b-pi In the first and second parts of this series we talked about how and why there is no universal definition for the term “EHR.” Instead there is a legitimate, growing debate about the features and functions that “EHR technologies” should offer physicians seeking to qualify for HITECH incentive payments. We explored the layers of network technology, suggesting that federal regulators should “separate the data from the applications.”

We also argued that there is much to learn from development platforms, recently and in the distant past, that have used standards to open the aperture of innovation. The best of these standards have reflected the experience of what works rather than specifying how to make it work. Defining the standards for data, devices, and network technologies too restrictively could choke off innovation, rendering HITECH’s offerings whose expense and complexity are a barrier to, rather than an incentive for, adoption by physicians. Incoming National Coordinator for HIT David Blumenthal, MD seems to have been considering just this concern when he recently wrote:“… [M]any certified EHRs are neither user-friendly nor designed to meet HITECH’s ambitious goal of improving quality and efficiency in the health care system. Tightening the certification process is a critical early challenge for ONCHIT.”

We’re not sure what “tightening the certification process” means. But if the new Administration hopes to entice physicians to embrace health IT, we’ll need a different process entirely than the one developed through the Bush Administration’s sole-source contract with the Commission on Certification for Health IT (CCHIT), an organization originally founded and staffed by HIMSS (The Health Information Management Systems Society) and dominated by large, legacy-based technology firms.

Concern about whether the current certification process is fair and configured to promote the common interest is hardly isolated or out of the mainstream. Last week the Markle Foundation issued a report – both of us served on the panel that developed it, but there were also many representatives from prominent industry groups – with this comment:“A broader view of IT would seed innovation rather than lock in adoption of technology based on what is available today. Health information services and technologies need to innovate and evolve rapidly, as other sectors have transformed themselves by embracing and building upon the internet…To support meaningful use, HHS should endorse a simple specification for a minimal set of open technical standards for secure transport as well as a core set of data types. By creating an obvious and achievable starting place, HHS will enable many options for clinicians and consumers to retrieve and use information to accomplish the meaningful use objectives.

And in a slightly blunter and more acerbic assessment in a Healthcare Informatics interview, Intermountain Healthcare’s CIO Mark Probst, newly appointed to the HIT Policy Committee formed by HITECH to advise the National Coordinator, said,“I mean it’s sure nice for Epic or Cerner or Eclipsys to tell their clientele that if they want to add new functions, you’ve got to go through them. So my guess is standards are somewhat threatening to them. Do we want 15 different gauges of railroad going around the United States or half the country driving on the left and half on the right? I mean you’ve got to have some standards if you want to get some of the benefits out of the systems.


HITECH lists the following “meaningful uses” of EHR technology:

  • The ability to do ePrescribing.
  • Engagement in health information exchange to improve quality of care, e.g. care coordination.
  • Reporting of quality and performance metrics, in a manner to be specified by the Secretary of HHS.

The common link between these three seemingly different uses of EHR technology is connectivity of health data to improve service quality.

E-Prescribing is essentially designed to promote care coordination between patients, doctors, and pharmacists.  It uses EHR technology that is dedicated to exchange of data between physicians ordering medications, pharmacists who are filling these prescription orders, and patients who request refills and are dispensed medications for treatment of their conditions and diseases.  All of these processes are easier, safer, more convenient, and less costly to perform using EHR technology than by paper or fax, and therefore we agree that this is a “meaningful use” of such technology.

Health information exchange between and among providers, especially when these providers are independent entities or exist in separate geographical locations, helps create continuity of patients’ experience by providing continuity of information flow and access where once there were only isolated silos of health data. There is widespread belief that health data sharing could improve care, safety, and decrease waste and duplication.

And quality reports are, in essence, statistical analyses of patient experience, sorted across many different variables: e.g., condition, acuity, physician, location. Providers submit the raw data for analysis and feedback, another kind of care coordination and communication activity, although the results are removed in many cases from direct patient care. Here too, we see that this feedback holds significant potential for improving care and eliminating unnecessary costs.

As National Coordinator David Blumenthal has pointed out, the current CCHIT certified products were not designed for these purposes. And that begs several questions:

  • Should already certified products be de-certified unless they can demonstrate their ability to meet the new HITECH criteria of meaningful use?
  • What would health IT that was designed to carry out these tasks look like?
  • How might it be distributed and sold?
  • Should pricing criteria be included in the certification process?
  • How might it be able to accommodate new features and functions as these become desirable?
  • What tools do the nation’s best performing groups provide to their staffs to empower them to provide high-quality and efficient care?

A new certification process could be streamlined in ways that encourage rather than stifle innovation. Certifying entities should be neutral, dispelling the perception of many in the industry that CCHIT’s ties to HIMSS are conflicted. (Note that we are not arguing for disbanding or dismissing CCHIT. We are simply suggesting that it should not have a monopoly over the specification of certification criteria. Like other organizations, CCHIT could choose to apply to become one of the certifying entities under the new process.)

Most importantly, the criteria for achieving certification should be closely linked to the “meaningful uses” specified by Congress in HITECH as ways physicians and hospitals can demonstrate improved performance associated with the tools, as justification for HITECH subsidies.

This could be easily achieved. ONC could interpret EHR technology as any software with the basic capability to create, protect (privacy and integrity), store, interpret, and exchange (i.e., import and export) a designated health data set, using existing, tested, and appropriate standards for this purpose. The designated health data set would be initiated with a small number of data elements that are already widely digitized and coded, such as problems and diagnoses, medication list and history, vital signs, and laboratory test results. Over time, and as exchange of this summary health data becomes routine, additional data elements could be added, as could new capabilities (e.g., decision support) for using the data.

Begin with a technological crawl, then walk, and eventually run.  Build a platform capable of future extension beyond current transactions and technical specifications.  Leave a lot of room for innovation.


We believe that the market is moving inexorably to answer these questions, but that consideration of them by Dr. Blumenthal and ONC is a rare opportunity to accelerate the market response. By doing so, serious “new thinking” would likely be introduced into health IT. One of the consequences might be an entirely new process of qualification or certification of EHR technology from that currently proposed by HIMSS and CCHIT.

That “new thinking” would reflect the changes that have occurred in computing over the past few years since CCHIT defined EHR technology based on a client-server model that was dated even in 2004. For example, we have seen a major trend towards Internet-based applications, the so-called “cloud computing” revolution. In essence, this is the idea that one can access software applications as a service available over the Internet, instead of having to put the software programs on one’s computer. Web-based software applications mean that customers need less specialized hardware and software to get more functionality at lower cost. Cloud computing allows us to make airline and hotel reservations over the Internet; to run word processing and spreadsheet applications, email, and contact database applications from a thin laptop computer or a cell phone; and to be free of dependence on particular devices or brands of hardware in order to participate in data exchange and communications.

A model of computing is emerging called Software as a Service, SaaS, in which the technology provides a platform into which multiple service applications can be “plugged” or “added” — and often from competing companies that are also not the same as the company that owns the platform. Google Apps and the iPhone are the two primary examples of platforms that allow independent developers to create applications that can run on the platform, and in some cases interact with other applications. These applications may even be substitutable and be replaced by the user who is basing his/her choice of which app to use on the basis of pricing and value. Users of a Google home page can populate it with widgets (e.g. apps for weather, calendaring, email) from Yahoo.

Finally, the World Wide Web is increasingly being used as social media. From blogs, to Wikipedia, to Facebook and Twitter, online tools for communication and social interaction are transforming the way business is conducted and how society gets its information. Group efforts that used to require the filters of relatively rigid institutional structures, due primarily to the complexity of managing groups, are now as easy to organize as hitting the “Reply All” button on an email. We would guess that the number of physician exchanges taking place within Sermo and Ozmosis, two of the leading physician-based social networking sites, exceed by an order of magnitude the communications that take place through medical specialty societies taken all together. These new communications tools are creating unprecedented opportunities for people to express themselves, and medicine/health care is a primary cultural area being affected. People are regularly immersing themselves in virtual communities, like Patients Like Me or Diabetes Connect, organized around particular diseases; cyberspace is used to provide medical advice and visits with clinicians (see American Well or TelaDoc); and more and more patients/consumer are expecting their doctors to have an online presence through web portals and secure communications channels.

By contrast, the CCHIT-certified EHRs are overwhelmingly practice- and physician-centric software applications that pre-date the Internet. They were not designed with participatory medicine or consumer-generated health care in mind.

Shedding the bloated feature set now required for certification in favor of a “thin certification” based on data exchange and management would immediately stimulate the health IT economy. It would also focus Congress’ understanding of “meaningful uses” that it hopes will encourage health IT among physicians and hospitals. Opening the aperture for innovation might easily create new jobs for new EHR technology products and services in e-Prescribing, care coordination, health data exchanges, and quality/cost performance reporting.


HITECH is hugely important because it is the Obama Administration’s first major step toward health care reform. The stimulus funds for health IT aspire to lay in a modern national health IT foundation that can facilitate the better care at lower cost our nation so desperately needs.

If the process moves forward as it is currently configured, a not-for-profit agency that is dominated by industry interests and that promotes technology that is largely outdated will have succeeded through its policy influence in securing much of that funding while holding newer, less costly, better technologies at bay. This would be not only yet another serious compromise for American health care’s future, but would signal that other important elements of meaningful health care change – universal coverage, a re-empowerment of primary care, greater quality/cost transparency, paying for results instead of procedures – are still very susceptible to the industry’s wants, and will remain equally elusive.

David C. Kibbe MD MBA is a Family Physician and Senior Advisor to the American Academy of Family Physicians who consults on health care professional and consumer technologies. Brian Klepper PhD is a health care market analyst and a Founding Principal of Health 2.0 Advisors, Inc. Their collected collaborative columns may be found here.

22 replies »

  1. Live Click Heatmaps
    With Live clicks, website owners can see their website come to life. You can literally see clicks as they happen.

    Each time a visitor clicks on the same spot, the live click heatmap tally’s the total number of clicks and creates a new color coated image with the correlated nnumber of clicks associated with that spot on the webpage.

    This is especially useful when launching new campaigns and landing pages.
    A site that helps you with Website Heatmaps . Provides information for customers and all website visitors. A wonderful resource.

  2. I am a 70 year old full time employed person who has Employer provided insurance with a co-pay, Medicare, & BC/BS from my former Employer. My current employer provided co-pay premiums are $29.00 per month, my Medicare co-pay premiums are $104.00.00 per month and my BC/BS co-pay premiums are $65.00 per month. My annual outlay is approximately $2500.00. Can I drop all this and select the new, wonderful, comprehensive, do all, cure all health coverage that is being rammed down our throats for much less money per month?

  3. This pay, healthcare and retirement arguement is very simple to me. We need to tell our national representatives that we want the same pay scale, the same health care and the same retirement they have consistantly voted for themselves. We deserve no less!

  4. Nowhere in the myriad of national health proposals I have read about is there a clear mandate for health promotion and self care for the privileged to be. To the food gluttons, smokers, chewers, dippers, alcoholics, bike riding no-helmet nuts, seat belt naysayers, diabetics who don’t follow healthy habits, the obese, the exercise addicted, those with heart and lung conditions who eat anything and everything and don’t exercise, those who practice unsafe sex, those who keep loaded firearms in the house with children, the immunization eschewing parents, and the just plain lazy big hot messes who don’t give a rip about their physical health, I say: When my taxes go to pay for your health care, I will have the right to tell you what I think about your habits and neglectful behavior. And don’t even think about saying you don’t know the consequences of your behavior. That will be a 1 in 100 chance. Oh, I can’t wait!

  5. Speaking as a consumer, EMT-in-training, and future PA/NP, I’m pleased to learn of this discussion and some of the developments in EHR technology and standards. Even just a few years ago, when I took a health care policy seminar at the University of Vermont (http://www.learn.uvm.edu/hm/), electronic medical records were mainly being discussed in the context of their (successful) use in the VA medical system. Privacy issues seemed to be the major stumbling block at that time.
    In my own experiences as a cancer patient who has been treated at several major hospital centers in different cities, I have become the most accurate archive of my medical history. I used to carry around a phonebook-sized bundle of fuzzy xeroxes (of unintelligible handwritten records!) but after a while I produced an electronic summary which I now bring to medical appointments. Some of my providers now use EHRs but not all. I have to believe that without complete knowledge of my medical history, my doctors cannot give me the best possible care.

  6. David, you did use a standard–it is called “English.”
    That is because the communication we are seeking is between human beings, not computer to computer.
    Communication between the clinician and the pharmacist consists of only two things: 1) the clinician’s knowledge of the patient’s allergies which are hidden in the particular prescribed medication and 2) the medication, dosage and administration order as well as the amount to be delivered to the patient. The pharmacist can get the same information from a Problem/Allergy List and a Medication List available on-line. Formulary compliance is another, non-clinical, issue. The pharmacy benefits managers who contract with managed care organizations to get the best cost medication. This changes all the time and is only important to the pharmacist for payment and the physicians who are at risk for their payment. The real problem is having a real time list of allergies and medications. Right now the patient is the database, but often all I get is, “I’m taking a white pill.”
    I am intrigued by the calls for a “single payer”–have you ever dealt with one that you would like to deal with for the rest of your medical career? Not me.
    Finally, about getting what you pay for, most clinicians I know get paid on many different pay scales. I don’t keep up with it. In addition, alot of patients won’t or cannot pay. That is why in most offices the bill is optional.
    We see patients whether they have the ability to pay or not. The media ploy about turning people away is vanishingly rare. It doesn’t occur, by law, in ERs and that is why they are full of people without emergencies.

  7. This argument will continue but although my sympathies are with the dogs, I think that this “battle” has been won by the Cats already. The real question is now not the $20 billion (or whatever) in HITECH, but how CMS or HHS decides to spend the $700+ billion in Medicare and Medicaid. Commenter R Watkins says it well
    “We are getting EXACTLY the results that we pay for. If we want different results, we can get them by paying for them, and the necessary technological support will be created.”

  8. The current legacy systems promulgated by the big vendors are not only unsuitable for dr’s offices, but also, if one is frank about it, largely unsuitable for even hospitals, without a lot of inhouse tweaking and arguments with the vendors (coming from the perspective of a hospital-based physician). I strongly agree with those who warn against tilting the playing field in favor of those vendors and against innovation – this is one area where innovation and improvement are absolute imperatives. Hello, is anyone in policy-making listening to us out here in the front lines?????

  9. “This is yet another reason why a single payer would simplify life for everybody.”

  10. This post is the best of the series, and the best description of the paths before us in the implementation of the ARRA. It can be used to shackle the health care delivery system with the ball and chain of legacy technology. Hopefully, power politics will not prevail and foreclose a more innovative, more fluid HIT environment.
    Perhaps more importantly, you deconstruct a false choice between performance now (on all the varied objectives of health care reform) or innovation going forward. We can have performance now and innovation going forward. Your argument exposes “performance now” as an accretion of functionality shared across the diversity of health service providers and their information systems–something that is sure to happen gradually. The qualification for funding can be tuned–and continuously re-tuned– to that reality and not locked to the best technology of 1997.

  11. wonderful set of articles.
    Re standards I agree they should have nothing to do with features except ensuring that applications comply to the letter and spirit with statutory security, data quality, utility, exchange, transmission and ownership requirements.
    Current CCHIT certification is a travesty. Their interoperability standard amounts to mandating the ability to record someone elses record/file number. So what does that have top do with technology? I can do that in a paper file with the same level of utility. Existing CCHIT serve only to preserve a predatory business model that forces clients into financial/”technical” extortionate relationships with vendors.
    FOSS is nice in many domains but most HIE functionality standards aren’t about programming,rather they are about implementations of external registries. These are mostly for fee services that support their maintenance. In order to support these functions FOSS is no longer Free but it can be open source.
    The heart of the matter is all about the data. Is it available, is it complete, can it be trusted, can I compile it, can I share it, does it preserve the originators meaning, is it the right information for the right purpose, can I use it in the way like information should be used?

  12. Excellent letter. Your comments validate many of my own thoughts and opinions. The CCHIT model of big legacy EHR systems primarily designed for hospital use is not adaptable for small physician practices. I work for a developer of software for the chiropractic profession. Many of the requirements for CCHIT certification do not make sense for chiropractors. They do not prescribe medication, for example. Yet, they will be forced to purchase and use an EHR system with capabilities that they neither want nor need.

  13. We use Nextgen which uses Surescripts to e-prescribe. And I stand corrected. The service that you mention is available. However, it is not mandatory and we don’t use it. We don’t collect PBP data from our patients and we have no interaction with the PBP when it comes to determining which medications to use. It would decrease the number of med change requests we get from the pharmacies and be a nice service for our patients, if we did have that connection, but then we’d be doing part of the pharmacist’s job while he gets to collect all of the filling fee.
    It’s not clear to me why you would want to prescribe medications that are not covered by a patient’s insurance. I would just as soon know in advance when I have to fill out that stupid paperwork or change to a nearly identical “me-too” version of a drug.
    This is yet another reason why a single payer would simplify life for everybody. I would have one formulary to memorize, the patients would know in advance what their co-pay would be and avoid embarrassment when they can’t afford expensive medications, and far less insurance data would need to be collected!

  14. “Almost unbelievably, most doctors have no way to know how well they are managing their panel of patients.”
    Why is it unbelievable? No one pays a doctor to manage a panel of patients, so why would we expect them to do it for free?
    We do pay for lumbar laminectomies for non-radicular pain and MRIs for acute knee strains, so a highly sophisticated system has evolved to maximize the number of those procedures being performed.
    We are getting EXACTLY the results that we pay for. If we want different results, we can get them by paying for them, and the necessary technological support will be created.

  15. Drs Klepper and Kibbe hit on very important points. To improve the care of chronic conditions like diabetes and coronary artery disease, you need to use evidence based protocols consistent with best practices and you need to know how well you are managing the blood pressure, LDL cholesterol,triglycerides, and blood sugar in your population of patients. Even if a doctor understands that and is very determined to learn of the performance in his group, it is a very cumbersome process using the current EHR products generally available. Doctors need feedback that is much more easily obtained so that it can be provided more frequently. That is critical to improving care. Almost unbelievably, most doctors have no way to know how well they are managing their panel of patients.

  16. My suggestion would be to limit “certification” to documenting that a system produces correctly formed continuity of care documents (CCD’s) and can parse incoming CCD’s. Any one who genuinely believes that as critical an element in their practice as an EHR could be selected without personally trying it out, having some of one’s office staff try it out is not some one whose judgement I would trust to my health to.

  17. J Bean:
    Not paranoia at all. If your E-prescribing system works by sending faxes to the pharmacist, this is not considered “true E-prescribing” and will not qualify you for the Medicare reimbursement increases.
    “True E-prescribing” means that the prescription is sent electronically to a clearing house (such as SureScripts) where it is then processed according to the demands of the PBP, and is then forwarded on to the pharmacy.

  18. It’s paranoia day here at THCB. I’ve used e-prescribing for years. It’s much better than scribbling on those teeny pads of paper. The PBP is not involved at all in the prescription although it sure would be convenient if the e-prescribing software would give me some help on the formularies instead of demanding yet more information from my over-taxed brain.

  19. “E-Prescribing is essentially designed to promote care coordination between patients, doctors, and pharmacists.”
    No, it’s not. True E-prescribing is designed to completely prevent direct communication between the physician and the pharmacist.
    With E-prescribing, the physician is forced to submit every prescription to the patient’s pharmacy benefits plan (or its surrogate); only after the the insurance claim is filed and approved is the prescription forwarded to the pharmacist. Thus, the physician in the exam room becomes an insurance clerk and must deal with all the demographic issues that insue (have you ever spent 30 minutes trying to send a prescription when the patient hasn’t told the PBP of their new home phone number?). Of course, the pharmacist and the PBP, who are making the profit from the transaction, are thrilled to have the physician do their clerical work for them!
    And this is the true elephant in the room with EMRs. Do we want them to improve patient health, or do we want them to be instruments of insurance payment and review? I think the two positions are incompatible, and the second position is winning.

  20. david:
    You’re wrong about the lack of standards. There are a large number of very basic standards that facilitate movement of your example e-mail; TCP-IP or 802.11n are allowing your computer to access the network, ASCII codes are being used to change your thoughts into numeric values, URLS are assigned and decoded internationally, and so on and so forth. Mostly those standards are managed by IEEE. If those standards didn’t exist, you wouldn’t have a prayer of communication whether you were using big company Microsoft operating systmes or freeware BSD Unix. You would simply have chaos.

  21. The purpose for creating more rigorous standards is specifically designed to make it more difficult for smaller software houses to compete with the larger shops who can afford the extra development bandwidth to jump through the hoops created by the “standards”. Simple.
    Funnily enough I can send you an email or even navigate to this website and noone had to set up any “standards” to enable me to do this independent of platform.