The Doc's Doc-Part I | Joe Bormel, M.D. | Healthcare Blogs Skip to content Skip to navigation

The Doc's Doc-Part I

| Reprints
Take a lesson from Outlook or just grab a mop?

Electronic Physician Documentation is quickly becoming recognized as a nightmarishly messy and incoherent linchpin to Meaningful Use. Electronically capturing documentation previously done on paper or through dictation sounds pretty straight forward. It’s not.

Many of us use Microsoft Outlook or something substantially similar for our email. How often do we stop to realize that an email one moment might become all or part of a calendar entry the next? It may contain a “To Do” that we need to prioritize and track. Or it could have contact information we need to preserve so it’s usable in an address book, dialer, or end up in Excel. Are you using folders in email beyond the Inbox and Sent folders to better organize your work?

The point is that email is about more than messaging. It’s about organization in general and organization of work involving varied participants. If all we could do in Outlook is send and receive emails it would be a lot harder to use, because the automation with the subsystems such as calendars would become manual processes. That’s exactly the problem with Physician Documentation today. Too often, it’s conceived as a tool to construct a note. That’s just wrong.

Ideal documentation not only summarizes the patient’s current story, it captures the reasoning behind orders and serves as communication about the patient to other providers, as well as capturing the diagnostic and therapeutic plan. It also serves as a “note to self” that facilitates personal mental continuity. This is the ideal. In the real world, time pressure, energy and skill levels often result in less complete or effective documentation.

This is elaborated in wonderful detail by Mark Hagland in his recent article, “ Balancing Act: Can CMIOs and CIOs Make Physician Documentation Work for Everyone?” Mark has assembled the best current thinking on how leaders are addressing the problem. In this two-part blog, I’ll be giving you my take and invite you to share yours as well. 

Replicate, Innovate, Transform

Years ago, Marion Ball impressed upon me that technology migrations require three sequential phases: Replicate, Innovate, and then Transform. It’s very tempting, and would seem to be more economical and faster, to attempt to skip phases under certain circumstances. That doesn’t work.

Today's Physician Documentation tools focus on replicating the existing note types physicians produce. For the inpatient world, the most complicated and critical of these is probably the discharge summary, often dictated days after discharge. In today’s world, there is no “magic” that automatically summarizes the relevant hospital course or problem lists, and as a byproduct, attesting to comprehensive medication reconciliation. Currently, when completed, these are additional organizational work steps, distinct from and prior to documentation. Clearly, this is more than delivering a working word processor with spell checking.

Before 2006, there was no mandate for concurrent discharge medication reconciliation at the time of discharge. There were no universally sanctioned and nationally defined quality measures to be addressed prior to final discharge. There was no mandate (spelled MU Stage 1) to produce a patient summary in a codified and semantically interoperable form.

With little or no contemplation for this fairly large gap, vendors and providers have thrust forward. They are producing disparate tools designed to reproduce the current process in an innovative electronic environment. Like other aspects of HITECH goals, the requirements, wants and needs, when looked at individually are reasonable. Collectively, the implementation burden can seem insurmountable. And then there is usability for that end user physician. Processes that should become one step, when separated out functionally become five total steps. This creates new potential gaps in care, since the probability of reliably following through on any single step by a rested, conscientious person is never 100 percent. Humans are not robots, the work is not widgetry, and economics leave all of us time and/or sleep deprived at times. Do you want to be our patient at those times? I don’t.

The current state, as I wrote earlier, is not ideal. Known care gaps are often implicit beyond the few common conditions that have received some badly needed attention. Conditions requiring things that should always be considered, for example, managing a patient with heart disease, stroke or diabetes, are often detailed somewhere. So while such knowledge is being transformed to become machine consumable, there is a higher expectation that it needs to transform physician documentation tools such that they guide and mentor.




Thanks John. Your work on adaptation is transformative. It's speaks to the critical nature of the social side needed to bring about the simplification of work you have championed.

The fact is that physician documentation (especially in non-procedure based specialties) is dictated by the requirements set forth in professional billing standards. In modern day EHRs we need to think outside of these constrains in order to document truly what the status of a patient is and the thought process of care delivery is at that moment in time. This all needs to be in a timely and efficient manner. The change in our documentation standard needs to happen on a regulatory level before EHR vendors change what they create in their systems.

Dr. Bormel,

We don't know whether a typical implementation of Infobuttons would have helped your dad's doctor. A thoughtfully placed Infobutton might have piqued this doctor's curiosity enough for him/her to click on the Infobutton.

Moreover, future implementations of Infobuttons may involve web services. For example, as soon as a chief complaint or an admitting diagnosis is entered into the EMR, the system might make a web service request to a knowledge resource to retrieve a list of documents describing the current treatment of that condition. The titles of these documents would then appear automatically in the EMR without the user having to take any action outside the normal workflow. One of these document titles might really catch a physician's eye. A simple click on the title would take the physician to the latest treatment guidelines.

Regarding documentation itself, we have ten years of continuous growth in physician documentation solutions, and we concur with the theme of your post. Our latest ProVation documentation solutions use some transformative approaches that weren't possible five years ago.

Dr. Bormel,
I think that Dr. Chelico made some excellent points. I agree that it will take regulation to make Physician Documentation work properly and to ensure interoperability. Such regulation can and should be enacted quickly, and physicians actually in practice must be asked to advise on its content.

Of the three areas — time, energy and skill levels — you mentioned as having consequences concerning how effective the documentation will be, the time factor is the most troubling to me. With the workload most physicians are carrying, the impact enhanced Physician Documentation may have could be counterproductive in terms of amount of care and the quality of that care they can provide. However, there also exists a possibility that by practicing accountable care, one process will offset the other. Have you thought this?

Good post, I'm looking forward to reading your follow up.

Doc Benjamin

Thanks Mark!

When I pick this up in Part 2, I'll disclose a recent, detailed study from Columbia informaticians who found that a high percentage of physician documentation is never read. What does that mean? I'll share my assessment of the positive implications.


Thanks for inviting me to react.  I've decided to frame it this way:

The HCIT Paradox – Is the Solution Disruptive or Adaptive?


Joe Bormel’s Doc’s Doc 5/5/11 blog identifies a difficult subject that most people don’t talk about - the HCIT paradox:

   1.   Many very intelligent, highly motivated people want to improve healthcare;

   2.   Billions of dollars have been invested in a multitude of very, sophisticated technological solutions to improve healthcare;

   3.      But, the combination of intelligent, motivated, talented people, billions of $$ and sophisticated technology has failed to deliver.  That’s a paradox.


Because a paradox does not make conventional sense,  it is often ignored – like the “elephant in the room” that nobody wants to talk about.  Having the insight to identify it is essential.  Then, the next step toward a solution answers the question, “Why does the paradox exist?”


Why has the combination smart people, motivation, money and sophisticated technology failed?  Joe proposes that the answer may lie in the concept of disruptive innovation and refers to the HBR article I co-authored with Clay Christensen and Rich Bohmer in 2000.  Here’s an update from my point of view.


Innovation means “new;” i.e., something that does not now exist.  Joe also refers to the need for a “transformation.”  So the solution to the HCIT paradox not only does not currently exist, it produces results that are transformative, i.e., very different than what we currently experience. 


‘Doesn’t exist’ and ‘very different’ lead to the next question, “Will what got us here, get us there?”  Every great transformational innovator I have studied answered that question, “No.” 


If we say, “What got us here won’t get us there,” we don’t have to worry or feel guilty about the fact that we are faced with a puzzling paradox.  More importantly, we now have the opportunity to choose a new path to a solution.

Since the 2000 HBR paper, my research and experience has focused on how to chose to do what others won’t.  Her are three guidelines to choosing a new path to effective HCIT. 

 1.   The FactsHigh performance, transformational innovators adapt and thrive by successfully making choices that the rest don’t.  That’s self-evident, so why don’t we just make them?  First of all, those new choices are, by definition, not best practices.   Secondly, new choices find it hard to survive in most organizations.

2.    The Problem/Opportunity – Currently successful organizational structures, strategies, methods and, most importantly, mindsets will always slow, stall and/or stop the development of new, adaptive choices.  This is why Clay Christensen called these choices “disruptive innovations;” it’s almost impossible for established organizations to develop adaptive choices.  That’s a problem.


That Problem becomes an Opportunity when you ask one more question.  “If it’s almost impossible, it’s possible; so “What’s possible?”  My research and experience over the last 12 years has focused on those few organizations succeeded at making new adaptive choices and led to:


3.     The Solution – If current organizational DNA is perfectly designed to stall adaptive choices, develop new DNA that is designed to adapt.


You can’t build that DNA with data in a meeting room because you can’t think your way to a new way of acting.  Instead you have to act your way to a new way of thinking by creating innovation incubators inside your organization, designed to make new choices.   That means designing adaptive innovation in real time at the point of care.


Adaptive Design is not vaporware.  It has generated logarithmic improvement in performance in multiple healthcare environments.  The book Designed to Adapt: Leading Healthcare in Challenging Times documents the facts.  In the Foreword, Clay Christensen wrote, “Adaptive Design gives the original, core business far greater flexibility in adapting to the demands of the market that I had ever thought achievable.”  That flexibility creates the solid base on which to build the systems that eliminate the paradox because it redefines the role of HCIT.  


Instead of “THE SOLUTION,” HCIT becomes the enabler and accelerator of flexible, responsive systems that continually provide more and better care at less cost.   Joe described how his dad received less than what he needed for his colon cancer. Replicating lab results and progress notes in an EHR or CPOE are not the solutions to that problem.  Simplify the point of care and provide his dad’s physicians the flexible, responsive decision support they needs to make the right choices – that’s the answer. 


Solving the HCIT paradox does not start with technology; it starts by simplifying and improving the work of patient care.  Then HCIT can help us deliver on the promise of more and better care at continually lower cost.  That’s an adaptive, not disruptive, transformation.  It’s not rocket science, just a different choice. 

Joe Bormel, John Chelico, Doc Benjamin, and Howard Strasberg,

Thank you all for your contributions to this wonderful dialogue! Each of you has provided extremely insightful and helpful comments in this area!

And Joe Bormel, I have to thank you very sincerely for your kind words regarding my May cover story. I'm delighted that the core theme of the story made itself clearthat we really are on a conundrum around physician documentation, freighted down as it is with so many contrasting (and sometimes conflicting!) tasks and responsibilities.

Fundamentally, I think that this is one of those areas in which it's clear that the "replicate" functions of health care and health care IT industry processes seem to be working relatively well, without the "innovate" and "transform" elements as readily following suit, to reference what the esteemed Marion Ball (one of my very favorite colleagues) told you years ago, Joe Bormel.

What is clear to me is that the health care industry is filled with really, really, really smart peopleand yet this area remains something of a conundrumwhich shows how truly complex and challenging it is, of course!

My fervent hope is for the very smartest of the smart in health care to come together to strategically work out new solutions in this daunting area. And I agree with all of youwe will need the best and the brightest from among practicing physicians, clinical informaticists, non-clinician informaticists, consultants, and vendors, to participate, in order to move towards success. I absolutely agree with Doc Benjamin that the time (of physicians) issue is one of the most crucial issues doctors in practice simply don't have any more minutes in their workdays to give over to added tasks. So we somehow have to make this work in a better way for them, as for everyone.

Again, thank you all for your terrific comments! And Joe Bormel, thank you again for initiating this great discussion! I sincerely encourage other readers to join in and offer their thoughts, to keep this dialogue going. Again, thank you, everyone!

Thanks John. Both for your industry leadership, clarity and inspiration to see Physician Documentation in the larger context. Your insights mirror those of other leaders and policymakers who have concluded that Payment Reform, Clinical Integration/Coordination, and Electronic Enablement/MU need to occur concurrently. Attention to workflow, usability and semantic interoperability improvements can only be achieved if the incentives are better aligned. The regulatory experiments are underway. It's our challenge to make them successful.

Thanks Doctor Strasberg.  I was thrilled to get your comment.

The work that you, Dr Robert Jenders, Dr Robert Dunlop and others working with various HL7 subgroups has been an instrumental force for great innovation and transformation.  The range of range of Web Services that you have been pioneering, both commercially and through the standards bodies is exactly what needs to be accelerated.  I elaborated that a bit in the prior comment.

You brought up two further points that I strongly agree with.  One is that "We don't know whether a typical implementation of Infobuttons would have helped your dad's doctor."  That's accurate, loaded with humility, and speaks to the need to design our experiments such that we learn.  That concept is elaborated in Lions and Tigers and MU Betas, here (see Designing for Failure).  Measurement and evaluation is an important component of the ONC's framework and it's important to call that out.  Thanks for that.

The second point you brought up is that there are multiple points during a Physician Documentation interactive session to imbed clinical decision support.  The guide,

Improving Outcomes with Clinical Decision Support: An Implementer's Guide by Jerry Osheroff, Eric Pifer, Jonathan Teich and Dean Sittig, does a terrific job of elaborating that point.  It's possible to deliver CDS through intelligent menus, displays and User Experience/Interface techniques, as well as through interruptive rules, as well as through policy, profiling, order set construction, etc. 

In the same spirit of "We Dont Know" above, the authors warn to guard against instrumental approaches.  For example, adding a rule that fires and messages the wrong person is clearly not a good thing.  Unfortunately, it's been tried!  We need to think about adaptive design approaches more than we do.  This gets back to John Kenagy's leadership work, cited in the original post.  More here:  Designed to Adapt: Leading Healthcare in Challenging Times.  Incidently, this book was awarded 2011 Book of the Year - American College of Healthcare Executives (ACHE).

Thanks again for your multiple, insightful perspectives, Dr Strasberg.

Doc Benjamin,

Thanks for your comment.  Regarding your question, ".... one process will offset the other. Have you thought this?"  Yes, I and many others have.

One school of thought I find interesting is the notion of retiring E&M-based payment systems.  The best elaboration I've read comes from Peter Basch, MD, FACP, a medical director of MedStar Health in Columbia, Maryland.  He has written about this in Medical Economics, as well as presented plenaries at HIMSS/AMDIS physician symposium.  Peter has been using ambulatory Physician Documentation, as well as leading care transformation within a provider organization, as well as leading policy development for ePrescribing, as well as documenting clinical and financial and satisfaction impacts of EHRs for two decades that I know of.  Here's what he has written:

"... These suggestions misread the "perfect storm" of events and incentives, and ignore a far simpler and more obvious solution: Replace the E&M documentation-based payment system with a system that rewards service and value, results in sparser and more useful documentation, and eliminates unnecessary complexity and burden. ...

"Another little-known fact: The Medicare Modernization Act of 2003 recommended to the Secretary of HHS that pilots of alternate payment systems be conducted. Unfortunately, this never happened." ...

"This is the right time for the Centers for Medicare & Medicaid Services to reissue its call for pilots of payment and documentation schema without E&M coding requirements. This is the right time for physicians to reject the shameful organizing metaphor of E&M coding—"it's not what you do, it's what you document"—and replace it with a renewed focus on what our patients deserve: better healthcare."

(source:  Quotes from "Viewpoint:  The End of E&M-based Payment -- Modern Medicine, Medical Economics, 2009.)

The second school of thought is going to be covered in Part Two of this post.  If we are really going to transform Physician Documentation, we need to capture the interactions between the physician and EHR, and, in the background and at the appropriate time, help them complete the documentation on things that they were silent on. 

The silence could be time pressure, ignorance driven, or simply that note construction process today is too onerous.  So, for example, completing the Medication Reconciliation should involve only touching those drugs that haven't already been reconciled with documented and current reasons for the continue, discontinue or change reasons. 

Another example is getting real time clinical decision support on opportunities and deficiencies, just prior to signing of the in-progress documentation , using robust, web services to ensure Physician Documentation is not inappropriately silent. 

Neither of these outcomes (Med Rec or Contextual/Comprehensive real time review) in design are possible or likely by simply replicating or innovating on current paper or EMR-based approaches.  And, there are vendors and academics, large and small, building these products and services to transform.  Others are more focused on the innovate path.  That is absolutely necessary, but wont address your concern. 

Peter and I agree on one other observation - the current regulatory requirements are centrally based on distrust of the physician.  We need to get beyond that.  Clinical Integration, care coordination, and gap closing are critical steps in that direction.

Stay tuned for Part Two!