The Case for Dropping MU Stages 2 and 3

Physicians are lining up against Meaningful Use.Dale Sanders

In a detailed letter sent this week to CMS Administrator Marilyn Tavenner and National Coordinator Karen DeSalvo, MD, the American Medical Association presented a long list of ideas to make Meaningful Use better for doctors.

The AMA warned that “unless significant changes are made to the current program and future stages,” doctors will drop out of the meaningful use program, patients will suffer as existing EHRs fail to migrate data for coordinated care, thousands of doctors will incur financial penalties, and new delivery models requiring data will be jeopardized.”

All of which is true. But the AMA didn’t go far enough.

Meaningful use is well intentioned, but like a teacher who “teaches to the test,” the program has created a byzantine system that might pass the test of meaningful use stages, but is not producing meaningful results for patients and clinicians.

A formal study published in the April 2014 issue of JAMA Internal Medicine reveals there’s no correlation between quality of care and meaningful use adherence. This study validates what common sense has told many of us for the last few years.

Meaningful Use Stage 1 was a jump-start for EMR adoption in the industry. That’s a good thing, I suppose, although meaningful use also created a false economic demand for mediocre products. It’s time to put an end to the federal meaningful use program, eliminate the costly administrative overhead of meaningful use, remove the government subsidies that also create perverse incentives, and let “survival of the fittest” play a bigger part in the process.

Let the fruits of EMR utilization go to the organizations that commit, on their own and without government incentives, to maximizing the value of their EMR investments toward quality improvement, cost reduction, and clinical efficiency.

When I arrived at Northwestern Medicine in 2005, it was clear very early that our EMRs (Epic and Cerner) were not being used in a meaningful way; this was several years before any broad discussion of meaningful use in the industry. Many Northwestern physicians were still using paper charts alongside the EMR, especially Epic in the clinics, thus creating a fragmented and dangerous medical record for patients.

Using the log and audit files in Epic and Cerner, we created an “EMR Utilization Dashboard” for each physician that also rolled up to the organizational level. The data was revealing. Outside of General Internal Medicine and a few other spotty areas, the medication list was not being used. The problem list was not being used. Order entry (CPOE) for medications, prescriptions, and tests was not being used.

Templates for documentation efficiency were not being used. Clinical alerts for best practices were not being used. Many patient encounters were not being documented in the EMR, indicating the continued use of paper records. In short, these very expensive EMRs were being used only occasionally as expensive word processors and dictation systems.

With input from all physicians, Drs. David Liebovitz, Phil Roemer, Gary Martin; and Tim Zoph and I decided to develop a simple document, describing the core principles of EMR utilization. Sarah Miller, the director of clinical applications, also played a huge part in this project.

We declared that it had to be constrained to a single page, normal spacing and font, and that we had to be data driven. It was a big success. Over the next two years, our rudimentary EMR Utilization Dashboard showed steady and significant improvement.

I showed the dashboard and the core principles to John Glaser (then at Partners HealthCare) while we were both speaking at a conference in Victoria, British Columbia. A few years later, when John went to the Office to the National Coordinator (ONC) to support David Blumenthal, John took the influence of those core principles and dashboard with him.

I’m not exactly sure what role the dashboard and those principles played in seeding the federal meaningful use program, but I suspect they had some degree of influence. By the way, we (Northwestern) offered to give the code and dashboard to the EMR vendors so that all clients could benefit, but the vendors declined.

Below are the simple but effective “Core Principles of EMR Utilization that we developed.”These principles played a huge part in the progressive value of Cerner and Epic on the Northwestern campus and laid the foundation for a relatively easy qualification of Northwestern under the federal meaningful use program (Thanks to Garima Sharma and others on the Northwestern Enterprise Data Warehouse team, which played a critical role in MU Stage 1).

Core Principles of EMR Utilization


· All patient appointments/visits are to be documented in the EHR as an encounter.

· Visit encounters should by closed by the attending physician within 48 hours of the patient visit.


· All medication prescriptions and refills must be documented in the EHR, including those ordered in a telephone encounter.

· Medications are to be reviewed at every patient encounter, in accordance with the individual specialty’s standard of care.

· Every effort should be made to maintain a valid and complete list of patients’ current medications in the EHR, including end dates, discontinuing medications no longer being taken, and removing duplicate medication entries.

Problem Lists

· All chronic, persistent patient diagnoses or complaints should be documented on the Problem List in the EHR, with the exception of highly sensitive diagnoses such as those associated with mental health care.

· Problems should be documented using the most specific term applicable to the problem, ex: mild intermittent asthma vs. asthma.

· The Problem List should be reviewed and updated at every patient encounter, in accordance with the individual specialty’s standard of care, and problems not currently clinically relevant should be filed to history and marked as resolved.


· Allergy lists must be actively maintained for validity and completeness for all patients, including marking as reviewed when no new allergies are reported. The allergy list must be reviewed during any encounter in which a medication is ordered.


· All patient orders must be documented in the EHR.

Progress Notes

· All patient encounters should have an accompanied progress note that appropriately documents the history, physical, and decision-making in a way that is succinct and minimizes redundant content.

· If dictating, notes must include the patient’s name and medical record number, the date of the encounter, and the attending physician’s name to ensure timely documentation.

In Basket

· Patient results and messages should be reviewed within 72 hours of receipt, and In Basket coverage should be assigned when clinicians are unable to respond within that time frame.

In the early days of EMRs, the pioneers like Intermountain Healthcare, Vanderbilt, Duke, and HealthCare Partners differentiated themselves by developing their own proprietary EMRs and then using them in a meaningful way, without any financial incentive except their own to do so.

Meaningful Use Stage 1 served a valuable purpose — it jump-started the adoption of commercially supported EMRs in an industry that needed jump-starting. Now it’s time to cancel Stage 2 and Stage 3, spend some of that money on seeds for true innovation (think DARPA for healthcare), and let survival of the fittest decide which organizations will utilize their EMRs, and subsequent data, most effectively to improve healthcare.

In addition to serving as a CIO in healthcare and the US Air Force, Dale Sanders has been one of the most influential leaders in healthcare analytics and data warehousing for the past 17 years. He currently serves as Senior Vice President for Strategy at Health Catalyst. Dale blogs about health IT at callitanything.blogspot.com.

Spread the love

31 replies »

  1. We adopted a Cerner system in December 2017, which replaced several discrete McKesson modules.

    As far as I can tell, the major purposes of our EHR are systematic upcoding, payment farming, and excessive documentation.

    When was the last time you saw a younger doctor actually do a 10-point ROS?

  2. I completely agree that MU 1 had its time and place, but the complexity of MU 2 and beyond is now preventing EHR vendors from being able to focus on the EHR features providers want. We have moved away from a market-driven development process to one driven by regulation.
    Welcome to the machine. EHR vendors need to learn how to obey what they are told by regulations, not to offer objections. Put on the white coat, and welcome to our world! What, you think regulations exist to permit you to make a useful or profitable product? Who told you that? And the Government knows what the markets are all about nowadays -MU 2 and 3 is, by definition, market driven. Are you volunteering to lose your certification?

  3. I completely agree that MU 1 had its time and place, but the complexity of MU 2 and beyond is now preventing EHR vendors from being able to focus on the EHR features providers want. We have moved away from a market-driven development process to one driven by regulation. Current MU2 requirements have added little new incremental value while creating a significant burden for vendors and end users. I believe that without MU, many EHR features would be similar, but there would be notable differences resulting from the focus on user feedback versus government direction. We could spend more time looking at how we could use the practice data to highlight workflow problems or areas where the practice isn’t using best practices. By leveraging our large pool of operational and clinical data we could generate more recommendations for practice optimization and patient care. These are very high level concepts that we’d love to explore more in depth, but we don’t have the resources to do that and prepare our software for MU requirements.

    • Thomas, as you know, I agree, and encourage all of us to lobby ONC and, eventually, Congress, to recognize the significant downsides to MU 2 and 3, not the least of which is what you mention– the legislation unintentionally removed the motivation and the bandwidth among EHR vendors to be creative and innovative. I met with a CIO and CMIO from a Cerner-based, academic medical center last week, well-known for their innovative use of EHRs to improve quality and efficiency of care. They achieve and exceed the spirit of MU 2 and 3 right now, under their own initiative, but technically, they would not qualify for MU2 as it currently stands; they are really frustrated by that possibility.

      It’s time for a course correction with MU. EHR vendors would benefit, as would their clients. And nobody in Congress or ONC should be offended or ego-reluctant to embrace a change; adaptability is what America does best.

  4. Perhaps the VA system – which has been constructed 30 years ago and spans the entire history of EMR – should just be adopted as the gold standard, and whoever cannot approximate it in quality measures should be Federally decertified.

  5. Meaningful Use is conceptually well-intended. As my blog mentioned, we took it upon ourselves at Northwestern to increase the value of our investment in Cerner and Epic by using those tools more “meaningfully.” But, as my dad would regularly remind me as I tried to explain my way out of a predicament, “The road to hell is paved with good intentions.”

    Stage 1 encouraged EMR adoption and achieved that goal, though at the expense of innovation that comes from free market competition– why take risk and be innovative when you are an EMR vendor in a market that is pounding down your doors to spend $25B in government money? We need to take a bold step to either let the free market play a bigger role or fund innovation in EMRs with federal funds via something akin to DARPA for healthcare. Dangle millions of dollars in front of software engineers, not thousands, and you’ll see a mad scramble of innovation and healthy competition.

    Stage 2 and 3 could be collapsed into one requirement for the vendors: Interoperability through granular data exchange, including portability of a patient’s record. But, the ONC should not be the organization that takes on the responsibility of creating that interoperability standard; they should represent the world’s largest healthcare customer– the US government– and as that customer, they should express the requirement for interoperability, and then leave it up to the vendors to figure out how to achieve it.

    BTW, I am not a “get the government out of our lives” poliitical radical. I’m a centered pragmatist, with a big dose of laziness– win the game in as few moves as possible. MU Stage 2 and 3 are going to be littered with wasted moves, and will likely not win the game, either.

  6. And they call the tale of the Tower of Babel in the Bible a silly myth. If the goal is meaninglessly useless, how can the process be usefully meaningful? I suggest that Curly, above, most certainly rocks in his exposition of the matter.
    Meaningful Use is about Shifting of Blame – Whose fault is it that there is a dead raccoon floating in the swimming pool? For there certainly seems to be one, that’s for sure.
    In our culture, when given the choice between a shoddily-designed machine, and a characterologically-deranged individual, we’ll blame the individual.

  7. @John O’Brien, I’d be glad to chat about our dashboard. How about connecting with me on LInkedIn? https://www.linkedin.com/in/dalersanders.

    @John Irvine, I agree. We can work with the good folks at ONC to sponsor a better alternative. I don’t know anyone in the industry who believes the current trajectory of MU is achieving what we all hoped. It’s time for an adjustment and there’s no need for anyone to be offended or defensive about the changes. We should take pride in our ability to learn and adapt.

  8. @Dale

    Then again, the best ideas are almost always “unrealistic.” If we stuck to the realistic ideas we wouldn’t get very far.

    Instead of taking meaningful use away, I think we need a better alternative.
    Admitting defeat can be difficult, moving to a superior option is much easier.

  9. Dale,

    After reading your article “The Case for Dropping MU Stages 2 and 3”, I would appreciate the opportunity of presenting your dashboard to our leadership.
    Keith Melton and I have been producing custom caché code (Epic) for some time now and have some applications that may be of interest to your organization as well.


  10. I can be an odd combination of naive and doggedly determined, both of which might apply to this situation.

    I agree that politicians being what they are, they are unlikely to pull the plug on MU because that means an admission of fault and a loss of money to the constituents that are benefitting from MU. But… I’m naive enough to believe that, with enough groundswell, we could do something, even if not outright cancellation, that would improve the Frankenstein that we created, especially if we redirected the money to better HIT uses and sustained the appeal to constituency.

    I’ve always dreamt of an EMR that was designed from the beginning to support clinician efficiency; quality of care; and cost of care. And then rolled all of that together into something that looked like a project management tool, like Base Camp, that recognized healthcare as a long term project involving several project teammates that need to interact and communicate. Dropping a bill would become a natural functional outcome, but wouldn’t be the primary motive of the design.

    It’s amazing to me that those of us who procure EMRs don’t insist on a downloadable, transferable patient record. How did the music industry manage to pull off the MP3 standard without a government mandate? Maybe there’s a lesson in there for us, somewhere.

  11. MU is meaningfully useless for patient care,. No, it is worse than that. It is an additional impediment to patient care.

    Medical care is about ambiguity and shades of gray. EHR systems depreciate the nuances of care, and meaningful ruse destroys care processes by focusing on the irrelevant.

  12. I was one of the leaders in the EMR arena for many years, and was initially really excited about meaningful use. Yes. I admit that with some embarrassment now. I even was part of a CDC public health grand rounds regarding meaningful use and why it would be a good thing. Over time, however, I saw what you see now: meaningful use is not a definition of using the EMR productively; it is simply another bureaucratic layer doctors must get through before they can focus on patient care.

    I do agree with items on your list, but the real benefit of the EMR is not one of documentation, it is about work-flow. Computers are good at remembering things we don’t remember, and are good at organizing information more efficiently. I would add several things that would make EMR systems more meaningfully useful:
    1. Task managment. Why don’t any products focus on team management of tasks, as it is clearly one of the bigger barriers to good care. I believe that a system that focused on this would gain adoption without incentive, as it would actually make doctors’ jobs easier.
    2. Information prioritization. It’s not what is put into the system that is important, it is what you can get out of it. Most EMR systems are a jumble of useless information that hides the useful information.
    3. Better communication tools. We are using iChat in our office (locally hosted) and have found it to be incredibly useful to answer questions while the patients are on the phone. We can handle problems with fewer steps. There are many tools out there to make this kind of thing work. Patients could, for example, record MP3 files on their portable devices and have that upload to an EMR for handling by the office staff (in lieu of the overworked phone system).
    4. Risk assessment and reduction – this is the overall goal of care: to make patients healthy and prevent problems from happening. The problem is that risk assessment tools are scarce in most EMR systems.

    Our success at EMR implementation was due to our focus on it as a tool to improve patient care by transforming our workflows. As the burdens of meaninful use came on, however, the ability to do that was hampered enough that I not ony abandoned Meaninful Use, but I left the system altogether. My home-grown EMR is far more useful than anything I could find on the market.

    • “2. Information prioritization. It’s not what is put into the system that is important, it is what you can get out of it. Most EMR systems are a jumble of useless information that hides the useful information.”

      Agree strongly. I almost feel like a shadow chart needs to be developed so I can see what is important. What a mess!

      • “I almost feel like a shadow chart needs to be developed”

        It has. It’s called “paper.”

    • MU is just a way to get money Rob it doesn’t stand in the way of patient care unless you bought a crappy system and fail to change your workflows.

      Can you tell me a single MU measure that impaired your ability to do your job (vs how the EHR you had designed something)?

      You also no longer care for medicare or medicaid patients and have a panel half the size of more docs so it sounds you are really interested in maximizing your income vs helping to solve the health care crisis

  13. Dale–You can cancel Stage 2 AFTER I get access to downloadable data from every part of my (and my wife & daughters) EMR and/or other records. Till then, I want MU 2 to keep up the pressure

    • Matthew- Have you ever actually seen your electronic medical record?
      If so, you’d be stunned at how much excelsior there is in it, and how much you have to work to find the meaningful stuff. You don’t want to be able to download the whole thing. You’ll just end up deleting most of it. It’s a big source of wasted clinician time.

      • I dont want to read it. I want a machine to read it and format it into something useful (and prettier) than the version Epic now shows me AND combine the other stuff (random lab tests, Rx from Walgreens, etc)

        If the MU 2 thing works, then the incipient market of little vendors who are starting to do that will get real….

        • “If the MU 2 thing works, then the incipient market of little vendors who are starting to do that will get real….”

          Not if Epic and Cerner can help it.

          • Epic dominated the market long before MU so don’t blame this program if you don’t like them blame the free market that created them.

            1 in 4 docs now use Epic but the bulk of ONC program like the regional extension centers (vs cms payments) programs were focused on helping small and rural practices get an EHR and what works is that 147,000 docs now do (and very very few of that category use Epic)

          • George – There’s certainly plenty of blame to go around. And remember that the MU program wasn’t government operating in a vacuum (i.e. the ONC Policy Committee).

            That Epic remains the “cream of the crap” among EHRs speaks to any number of failures in this marketplace.

            As Matthew describes, making it easy to get data out of monolithic systems – like Epic and Cerner – holds great promise. But don’t underestimate the amount that it also threatens the incumbents.

  14. My patient was on lovenox, plavix, and aspirin for acute vascular illness. The MU of the EHR has a DVT prophylaxis form that takes a few minutes to fill in the boxes. I was called as an emergency twice today to fill out the form, despite the maximal anti clot therapy in the orders and MAR, that never was populated on the DVT form.

    That is meaningfully dumb and disruptive. From Dallas to Damascus, most half wits would code the system to file the form based on the patients’ anticoag therapies.

    Ban this meaningful ruse!

  15. There are many in Washington who would privately agree with you; however this is an unlikely outcome. The reason? It is almost physically impossible to cancel a government program. Even a small government program, Let alone a program on the monster scale of meaningful use program. Why? Doing so would force officials to concede that a mistake may have been made. We can argue that plugging in the healthcare system was worth the expense, relative to other major government pushes. But publicly acknowledgement of defeat? Not happening.

    Cancellation of MU2 and MU3, would invite scrutiny of the overall EMR incentive program and the billions of dollars of payments that have flowed from Washingon to vendors and consulting companies. Given the close call with Healthcare.gov – and the federal dollars spent on the exchange – this is not going to happen.

    Instead look for the meaningul use program to take a feel similar to the politics of the doc fix; everybody and their dog will come out against the thing, there will be votes, but nobody is going to be cancelling anything.

    There will be great drama and speechification, but in the end nothing will change.

Leave a Reply

Your email address will not be published. Required fields are marked *