How to Revive a Failed EHR Implementation

My latest post on  Here’s an idea I think merits consideration.

What do you think?

Is the C-suite fiddling while EHR burns?

There is an adage in the military—different spanks for different ranks.  If speaks to a double standard, the less egregious their punishment for similar offenses, similar misjudgments.

We see that every day in business, and we see it a lot in healthcare, especially in hospitals.  Physicians are held accountable for medical errors.  Hospitals pay millions for malpractice insurance knowing that mistakes will be made and people will be held accountable for their mistakes.

But what about on the business side?  Who is held accountable for business mistakes?  An acquisition that failed to deliver.  An expensive new service offering that bled the company dry.  A decline in the number of patients. The failure of a major IT initiative to deliver results.

Take EHR.  Some of you are saying, “Yes, please take it.”

  • Around sixty percent of the large EHR projects have failed in one respect or another
  • Most will not receive ARRA incentives
  • A large number of hospitals are on their second implementation of EHR
  • Some have productivity losses of thirty percent

Who is going to be fired for the two hundred dollar misstep?  The board?  Never.  The CEO—no.  The COO or CFO?  Unlikely.  The CIO?  That is the safe bet.

Did the CIO authorize the expenditure?  Nope.  Did the CIO get all the dollars needed to be successful, all the user support?  Unlikely.

In most cases the CIO has all of the responsibility and only some of the authority.  There are a handful of people in each organization tasked with the oversight of the large project.  They are the ones who should be asking the right questions, the ones who should be demanding answers.

A failed project, a failed strategy should not come as a surprise.  The only people who will be wearing EHR 2.0 T-shirts are those who authorized EHR 1.0.  How come these individuals are not accountable?

saint Paul M. Roemer
Chief Imaginist, Healthcare IT Strategy

1475 Luna Drive, Downingtown, PA 19335
+1 (484) 885-6942

My profiles: LinkedInWordPressTwitterMeetupBlog RSS
Contact me: Google Talk/paulroemer Skype/paulroemer Google Wave/paulroemer

How the C-suite sees the CIO

This link is to my latest post for

saint Paul M. Roemer
Chief Imaginist, Healthcare IT Strategy

1475 Luna Drive, Downingtown, PA 19335
+1 (484) 885-6942

My profiles: LinkedInWordPressTwitterMeetupBlog RSS
Contact me: Google Talk/paulroemer Skype/paulroemer Google Wave/paulroemer

The large provider business model–The Sky is Falling

This link takes you to my newest post on Anthony Guerra’s site.  I welcome your thoughts.

My best – Paul

saint Paul M. Roemer
Chief Imaginist, Healthcare IT Strategy

1475 Luna Drive, Downingtown, PA 19335
+1 (484) 885-6942

My profiles: LinkedInWordPressTwitterMeetupBlog RSS
Contact me: Google Talk/paulroemer Skype/paulroemer Google Wave/paulroemer

Published on–vendor darts

Below is the full article I submitted to, Anthony Guerra’s outstanding site for healthcare leaders.  As always, I am flattered that he finds my contributions worthwhile.

Is there a best Electronic Health Records system? Perhaps Cerner, EPIC, GE, or McKesson?  For those who have followed my writing, you’re probably thinking my answer is “None of the above.”

I’ll do one better, and I write this with the utmost sincerity—it does not really matter which vendor you select.  As the EHR vendors reading this pull themselves off the floor, permit me to explain why.  Researching the question this is very little information to support the notion that any of the major hospital EHR systems quantitatively stands out from the others.

There are a few sites that offer user assessments across a range of functions, but those have at most three opinions—not enough to consider statically significant.  There are plenty of EHR scorecards and comparison tools, just not many scores.  The vendors’ sites do a poor job of differentiating themselves from their competitors.  Vendors use superlatives and qualifiers in an attempt to differentiate themselves.  When one considers the basic functions that make an EHR an EHR, the top vendors all have them.  No vendor highlights major clinical or business problems that their solution solves that another vendor does not solve.  Instead, they state they do something better, easier, more flexibly—none of which can be measured by prospective clients.

Imagine, if you were an EHR vendor, and you knew that your product did things to benefit a hospital better than the other vendors, wouldn’t you have an independent competitive assessment, some sort of “Consumer Report” chart and evidence to support why you are better?  Of course you would.  You would highlight your superlatives.  I have not seen one that would be very helpful.  The only information I found that might be worth a read comes from Klas Research,  However, the names of the modules rated are vendor specific, and none of the vendors use the same names.  It will give you a feel for how a small sample rated features within a given vendor, but there is no data to suggest how those ratings compare among vendors.

Even if there was a good comparison, the other thing to learn from this is all the areas that aren’t listed imply that the vendor is either no better or perhaps worse than the competition.  Cream rises to the top—we are left to choose among brands of milk.

One vendor may have a better medical dictionary than another, yet that same vendor will lack rigor in decision support.  No single vendor seems to have their customers doing back flips in their testimonials.  Some score high in their ability to deliver a complete inpatient solution and fail in their ability to integrate with other vendors.  Others hurt themselves during the implementation, user support, response time, and the amount of navigation required to input data.  Some EHR vendors posit their systems as being better at meeting Meaningful Use or passing all of the Certification requirements.  Ask them to name a single installed client for which they have met these.

Why doesn’t matter which vendor a hospital selects?  The reasoning holds not because all hospitals are the same, rather, it holds because were one to perform a very detailed comparison of the leading EHR vendors with a Request for Proposal (RFP), they would prove to be quite similar.  You might find significant separation if you only compared ten functional requirements.  You would expect to find less separation by comparing several hundred, and quite a similarity if you compare a thousand or more requirements.  The more you look, the more they seem the same.

Although the vendors will differ with respect to individual requirements, when evaluated on their entire offering across a broad range of requirements I would expect each to score within one standard deviation of the other.

Reason 2.  It is possible to find hospitals who will give outstanding references for each of the leading vendors.  It is equally possible to find users in hospitals who have implemented one of the “leading” vendors’ systems who will readily tell you that the purchasing the system is the worst business decision they ever seen.  More to the point, every vendor A has probably had at least one of its implementations uprooted and replaced by vendors B, C, or D.  The same can be said for vendors B, C, and D.

If this is a fair assessment, what accounts for the difference?  How can we account for why one hospital loves a given EHR system and another one hates the same system?  Chances are they both needed about the same solution.  Chances are they received about the same solution.

Here’s the difference.  The hospital who thinks they made a good choice:

  • Had a detailed strategy and implementation plan
  • Paid as much or more attention to process alignment, change management, and training as they did to the implementation
  • Managed the vendor instead of being managed by the vendor.

Simply put—the problem is not the EHR system.

One other thought.  “Pay no attention to the man behind the curtain—the Great Oz.”  Do not put your scarce capital into a solution just because it offers or promises either Certification or Meaningful Use.  Yes, there is much discussion about both of these.  The industry stops and holds its collective breath each time a new set of stone tablets are brought forth from the ONC or CMS.  You can meet Meaningful Use with a Certified system and still wind up with a system the users hate and that does not support your business model.

Here is something else I cannot explain.  For those hospitals replacing a one hundred million dollar EHR with another hundred million dollar EHR, why do they think the second system will be any better?  If the systems are not materially different, the only way to get a different result is by changing behavior, not changing systems.  Why make the same mistake twice?  What could be so wrong with the first implementation that an expenditure of far less than another hundred million could not solve?

What is the cost of EHR 2.0 not working?

saint Paul M. Roemer
Chief Imaginist, Healthcare IT Strategy

1475 Luna Drive, Downingtown, PA 19335
+1 (484) 885-6942

My profiles: LinkedInWordPressTwitterMeetupBlog RSS
Contact me: Google Talk/paulroemer Skype/paulroemer Google Wave/paulroemer

EHR: How do you avoid failing?


I may have mentioned that I’m a runner. In high school and college I’d run anytime, anywhere. I ran cross country, indoor track, outdoor track, and AAU– kept my hair tied back in a ponytail—I miss the hair.

Those were the days. I was the captain of the cross-country team during my senior year. Behind the school was a long series of hills we used for training, and they were blocked from the coach’s view. I remember that one day I told the team it was okay to walk because we were out of the coach’s sight. I also remember when he took me aside after practice and said he didn’t think it was a good idea for the captain to tell the guys they could walk. He said he couldn’t see me but he sure could hear me. I also remember the time I had my mom dropped me off about half mile away from my girlfriend’s house so I could run, making it look like I ran the entire six miles.

My friends and I ran a few 50 mile races and a couple of marathons. But the strangest race we ever ran was one that lasted 24 hours. The event was a 24 hour mile relay. More than a dozen teams entered the event. Our team had seven runners. The idea behind the race was that each person would run around the track four times with a baton and then hand the baton to the next member of his team. If one member of the team was too tired to take his turn, that team was disqualified. The race started early on a Saturday morning. At the end of 24 hours, my team had run 234 miles. We were proud of what we had done. We were even prouder when we saw the article printed in our local paper the following week that we had set a world record for a seven-man team in a 24-hour relay.

I’d like to believe that the world record had something to do with the fact that we were a great group of runners. However, as I look back on it I tend to believe that the world record had more to do with the obscurity of the event than with the capability of the runners. I don’t know if that same event had been run before we ran it or was ever run afterwards. Who knows, we may still hold the record. I guess what I learned from that event, is that it is easier to be viewed as being excellent at something that isn’t done very often.  Obscure or not, it was a one-time event for us.

Doing something once makes it difficult if not impossible to prepare for the gotchas that lay in wait.  There are healthcare providers who are on their second and third attempt at implementing their electronic health records system (EHR).  This is not the type of event where practice makes perfect, far from it.  If you don’t get it right the first time, you’ve probably already laid waste to your most important stakeholders, the users.  They are difficult enough to get on board the first time.  The second time it becomes much more of a fool me once shame on me, fool me twice, shame on you.

How do you avoid second and third attempts of something as difficult as a full-blown EHR?  For some providers, it’s even worse in that they probably have multiple dissimilar instances of EHR already in place in parts of the hospital, instances that will have to be integrated to the corporate platform.  If you let the clinical side run the project, you run the risk of losing the IT side.  If you let the IT side run the project, you run the risk of losing the clinical side.

Who do you trust to run what could amount to a few hundred million dollar project, bring out the best skills of the team members, and make sure the vendor is operating in your best interest?  It’s a difficult question to answer.  The good news is that if you get it wrong you probably won’t have to worry about doing it over, that will probably be your one-time event.


Healthcare Informatics: one time at band camp…


Here’s a response I posted to a Healthcare Informatics article, by Mark Hagland, “Revenge of the Clinical Informaticists”.

The link is:

I wrote the following:

I wrote on this same topic yesterday, albeit with a slightly different bent.  Like you, I see two distinct groups who do not play well in the same sandbox—clinical and IT.  Having one group go to the other’s summer camp to pick up a few skills is not the same as pulling a few costly and hairy projects from the bowels of project hell any more than it would be to have an IT executive take an EMT course and then assume that person was qualified to perform surgery—this one time at band camp…

Before I get up on my stool and knock myself off, I know CMIOs and CIOs who have made HIT and EHR very successful.  To them I ask, do not rake me across the Twitter coals as I try to make a point.

There’s knowledge, and then there’s qualified.  Doctors do four years of medical school, they intern, and if they specialize, they throw in a few more years before they become the in-charge.  Years of training and practice before the doctor is allowed to run the show.  Why?  Because what they are about to undertake requires practice, tutelage, and expertise.  Most of the actual learning occurs outside the classroom.

There are those—not Mr. Hagland—who suggest that the skills needed to manage successfully something as foreboding as full-blown EHR can be picked up at IT Camp.  They do a disservice to seasoned IT professionals.

Most large IT projects fail.  I believe large EHR projects will fail at an even higher rate.  Most clinical procedures do not fail, even the risky ones.

What’s the spin line from this discussion?

  • Rule 1—large EHR projects will fail at an alarming rate
  • Rule 2—sending a doctor to band camp probably won’t change rule one

Don’t believe me?  Ask friends in other industries how their implementation of an ERP or manufacturing system went.  There are consulting firms who make a bundle doing disaster recovery work on failed IT projects.  They circle the halls like turkey vultures waiting for CIO or project manager carrion.

Back to Rule 1 for a moment.  How can I state that with such assurance?  Never before in the history of before—I know that’s not a proper phrase—has any single industry attempted to use IT to:

  • impart such radical charge (patients, doctors, employees)
  • impart it on a national basis
  • hit moving and poorly defined targets—interoperability, meaningful use, certification
  • take guidance from nobody—there is no EHR decider
  • implement a solution from amongst hundreds of vendors
  • implement a solution with no standards
  • move from an industry at 0.2 to 2.0 business practices
  • concurrently reform the entire industry

Just what should a CMIO be able to do?  What are the standards for a CMIO?  To me, they vary widely.  Is a CMIO considered an officer in the same sense as the other “O’s” in the organization, or is it simply a naming convention?  The answer to that question probably depends on the provider.

Here’s how I think it should work—I realize nobody has asked for my opinion, but this way I’ll at least provide good fodder for those who are so bold as to put their disagreement in writing.

I love the concept of the CMIO and think it is essential to move the provider’s organization from the 0.2 model to the 2.0 model.  Same with the CIO.  However, getting them to pool their efforts on something like EHR is likely to fail as soon as one is placed in a position of authority over the other.  It’s sort of like getting the Americans and French to like one another.

I liken the CMIO’s value-add to that of the person providing the color commentary on ESPN—it adds meaning and relevancy.  The CMIO owns and answers a lot of the “what” and the CIO owns and answers a lot of the “How”.

Still unanswered are the “Why” and “When”.  A skill is needed that can state with assurance, “Follow me.  Tomorrow we will do this because this is what needs to be done tomorrow.”  That skill comes from an experienced Project Management Officer, the PMO.  It does not come from someone who “we think can handle the job.”  Nobody will respect that person’s ability, and if they can’t lead, yo can plan on doing the project over.

Oh, if anyone is still reading, here’s my original post;