Back in 2010, Epocrates had its EMR ducks in a row. The company, known best for a very popular smartphone-based drug interaction database for physicians, announced plans to release a mobile SaaS EMR.  While Epocrates was jumping into a market more crowded than a barrel full of monkeys, one could see where leaders might see an EMR as an extension of the relationship it already had with physicians.

Now, Epocrates leaders have said “oops” and announced that they were killing the product,  telling investors and the public that building the darned thing was distracting it from its core business.  It does seem that the company was struggling with the EMR rollout process:  it didn’t roll out its first-phase product until August 2011 and didn’t get its Meaningful Use certification until February of this year. But this is the first time I’ve seen a company kill a product at this stage of development, particularly in such a high-profile manner.

It must have been more than a bit embarrassing to make the announcement during HIMSS12 when, of course, companies traditionally kick off products they’re planning to sell vigorously. As Epocrates was making plans to dump or sell their EMR, the company’s CMIO, Tom Giannulli, MD, was pitching the company’s new iPad EMR to editors.

As Epocrates itself pointed out, there aren’t too many dedicated iPad EMR offerings out there. So in theory, this should not have been a waste of the company’s time.  On the other hand, with the iPad still a new frontier for EMRs, we still don’t know whether it will ultimately work as a platform of choice for physicians.  As we’ve previously discussed on this blog, the iPad seems to be a pretty good medium for reading data but a very awkward one for entering data. Whether that’s a fatal flaw remains to be seen.

Truthfully, this looks like a failure of execution from start to finish, rather than a product that couldn’t possibly work. But these are tough times. Even the best execution may not work; and if so, Epocrates was probably wise to fold its cards before further damage was done.

I try to avoid navel-gazing, which to me means commenting on someone else’s commentary – a practice all too commonly relied upon in the healthcare IT blogosphere. How many blogs, articles and rebuttals have been generated, after all, as a result of the Health Affairs/Mostashari back-and-forth in the past few days? Quite a few, and yours truly happily participated in the fringe commentary. So as you can see, sometimes a topic already covered by someone else just begs for a second opinion, which I’ll happily give if the context is right.

Yesterday I came across two pieces of online content that I couldn’t help but draw correlations between. The first, a blog written by Dr. Rick Weinhaus entitled “Humans Have Limited Working Memory,” tells the tale of our poor ability to retain information, made all too obvious by a common EHR design feature – the utilization of a row of clickable tabs at the top of a dashboard to designate the different categories of data that make up the patient visit.

Dr. Rick laments that since humans are capable of only retaining four to five unrelated elements in working memory, the row of one-click tabs, though logical, doesn’t work very well. In fact, it drives him “crazy.”

I certainly believe in our limited capacity for remembering unrelated things at any given time, and I’m sure other working parents will agree. Our capacity for keeping everything straight is finite – the more kids, colleagues, coworkers, patients, tabs, bells and whistles you add, the more likely you are to forget something, leave something behind, or, if you’re like me, leave your car door wide open in a parking lot while grocery shopping with two kids in tow. But I digress.

The second piece of content revolves around the results of a survey put out by CDW Healthcare on what clinicians find frustrating about implementing new health IT systems. Surprise, surprise, “too many passwords to memorize” came in at the top, emphasizing what Dr. Rick pointed out in his unrelated blog post.

So what’s a clinician to do? Especially those that work in multiple facilities on different EHRs? Are you like me, scribbling down usernames and passwords on a master paper document, which just screams “privacy breach waiting to happen?” How are vendors helping to address these issues – single sign-ons? Better, overall design? Whose doing it the right way when it comes to designing an EHR, or as Dr. Rick says, designing one “based on what humans are good at — using our visual system to make sense of the world?”

Please let me know in the comments below.

For quite some time, talking heads have noted that EMRs will be an essential part of ACOs, so much so that most doubt you can have a successful ACO organization without one.  What I don’t see asked as often, however, is whether EMRs are shaping the future of the ACO movement, both negatively and positively.

What would an ACO look like, if it could exist at all, without an electronic record or HIE in place?

* There would even more mistakes and delays in sharing patient records, as one can hardly expect a larger group of institutions to make *less* mistakes

*  ACOs could launch without having to spend millions of dollars on EMR software, hardware, training and support

*  Clinical workflow would remain the same, generally, even if doctors were forced to include larger numbers of co-workers in their network

And how are ACOs working with EMRs in place?

*  Aside from limited case studies in individual institutions , it’s not clear whether EMRs are turning large, newly assembled care organizations into safer places to get care.

*  ACOs are forming more slowly than they might be, arguably, because a comprehensive EMR is part of t he cost of doing business

* New clinical workflow patterns are being forced upon clinicians, cutting across multiple institutions. While this might ultimately increase efficiency, it’s hard to ignore how many human hours are being invested (or wasted, depending on your position) on new technology.

As you can see, I come down on the “EMRs may not be all they’re cracked up to be for ACOs” side of things. Now, I’d concede that I haven’t been completely fair — I know EMRs have yielded great benefits for some groups of institutions– but I’d say the jury’s still out overall.

In my conversations at HIMSS a few weeks ago with providers and vendors, I heard more than a few references to user-friendly EMR design, easy-to-use dashboards and the bar that has been set so high by Apple and the iPad. I had a chance to chat with David Carleton, VP and CIO at Heritage Valley Health System in Pennsylvania, about the adoption of the iPad in the clinical setting, particularly with regard to EMRs. Carleton, with the assistance of dbMotion, helped a team of docs and IT staff at HVHS in Pennsylvania develop their own EMR iPad app.

In a nutshell, the internet Clinical Access Portal (iCAP) app organizes and harmonizes data captured and stored in various systems – including its Allscripts Enterprise ambulatory solution and its soon-to-be-completed Allscripts Sunrise Clinical Manager, as well as the ClinicalConnect HIE in western Pennsylvania – and delivers Continuity of Care Documents (CCDs) to HVHS providers via the tablet. Named to the 2012 Top 100 Integrated Healthcare Networks, HVHS seems to be placing a high priority on enabling its facilities to be truly interoperable with one another. It made sense to me that the hospital would want to better enable its physicians with a handy iPad app, but I wondered why they took the in-house development route.

Carleton explained to me that one of the reasons was physician buy-in. (You can view more of our chat in the video below.) Apparently, the key to getting physicians to adopt and consistently use the tablet and app was to have them on board from the very beginning. Involvement in the design process let them have a say as to what would best fit their workflows.

With the release of the iPad 3, the details of which were announced yesterday, I’m willing to bet we’ll see an up tick in clinical interest in the iPad and a corresponding surge in app development – in-house or otherwise.

Are you aware of other facilities getting into the EMR app game? Please share the details in the comments below.

Today I got a call from an executive recruiter who wanted to know, in essence, where the EMR market was going.  Aside from the usual chatter about Meaningful Use, talent shortages and HITECH, one question she asked made me think: “What do you think is the main thing someone like me should know about the health IT market.”

Having pondered this for a while, I realized that the answer is fairly simple. Above all, anyone who wants to understand health IT needs to know two things: a) That health IT leaders need to be change leaders, more than ever before in the industry and, more importantly, b) that the EMR is at version 0.5 when it comes to maturity and integration into the life of most hospitals.

Yes, I mean version 0.5. We’re talking barely in beta, when it comes to solid integration, staff training, enough institutional knowledge so people can share and learn and a high-performing system that doctors love.  Sure, a few hospitals (1 percent, as I recall) have reached that legendary HIMSS Analytics stage 7, but most are lucky to have gotten their Meaningful Use Stage 1 payment into the door.

When you consider that a large number of CIOs doubt they have the man/women power to complete their Stage 1 implementation, the picture looks even grimmer.  Not only are the EMRs immature, they’re largely being implemented and run by consultants who will cut and run with their experience bank, as they have little ability to share it other than in (to staff and doctors at least) boring reports.

Bottom line, I’d argue that it will be a whopping five to seven years, at least, before EMRs meet either HIMSS Analytics criteria for maturity or my personal Zieger seat-of-the-pants model.  I hate to say that it could even be 10 years, but I see it as a possibility.

The reality is, government can be powerful, and big financial incentives are tasty, but you can’t force an industry to change overnight just because it would be really, really cool.

If you don’t check out the HIMSS group on LinkedIn from time to time, you should. I always pick up something to think about when I visit, and this time was no exception.

A group of IT pros, most of whom seemed to have plenty of institutional memory of EMRs gone by, were talking about whether the current leaders of the EMR vendor pack would take over and most of the rest fall away.  The consensus, not surprisingly, was that hospital CEOs are herd animals, and that a few leaders are likely to take most of the market.

As things stand today, even EMRs that seem to be a better fit usually lose to the Epics, Cerners and Meditechs of the world, writes Richard Rauber, FHIMSS.

“Let’s say the preferred EMR has 10 clients similar to their facility, and the second choice has 75 clients in the same bed range with a high level of user satisfaction. Is the risk/reward ratio low enough to go with the smaller vendor? It today’s market it would be unlikely.”

If these posters are right, the hospital market is going to standardize on a dozen or so of the most successful vendors. Unfortunately, that’s likely to lead to some really nasty implementations, suggests Terry Montgomery, PMP: “I had such a project last year. They had to move the go live date three times and there were still bugs they had to fix.”

That being said, I think there will be a lot more dancing when it comes to the physician EMR market.  You’ve got breakout models like the no-cost Practice Fusion — and its bundle of VC cash to fuel the fire — iPad-based DrChrono, Free Mitochon PMS-EHR-HIE and a growing number of elegant, doctor-crafted implementations like SOAPware and Amazing Charts.

While the dynamic of hospital IT purchasing is to standardize on the big boys (the old “nobody gets fired for buying IBM” syndrome), physicians can’t afford to buy a system just because the practice across town thought it was cool. Not that such doesn’t happen, but it’s less likely.

I predict that doctors will have some great options to choose from when they hit HIMSS13 next year, systems integrated intelligently with revenue cycle needs but also cleanly designed and physician friendly.

The smaller EMR companies focused on doctors are just doing a better job of mirroring a doctor’s process, there no doubt in my mind. If only such logic would float upward to the billion-dollar boys behind the hospital giants.

Full Disclosure: Practice Fusion, Mitochon, SOAPware and Amazing Charts are advertisers on this blog.

Over at the august halls of the National Institute of Standards and Technology, researchers have been compiling data on what makes EMRs usable. A year ago, in April 2011, NIST presented a draft set of usability standards. At the same hearing, a wide range of academics and scholars got up to talk about what they saw as they key issues — including whether EMR workflow should be changed to make it cost-efficient.

Since then, from what I can tell, there’s been a lot of noise but little light shed on the design specs a truly usable EMR should adhere to.  There’s been some progress in the development of HIE connections between EMRs, some worthwhile work EMR return on investment and even some improvements that might leverage EMRs to help doctors collect more from patients.

Talk to many doctors, and they’ll tell you their EMR stinks. Why? Largely because workflow is still inefficient and the “click burden,” which can drive doctors through a dozen steps to get tasks handled, hasn’t been reduced any too much. Some older docs I’ve spoken with even pine for the rough-hewn EMRs of 20 to 30 years ago more, which were at least built by their colleagues.

Honestly, I don’t expect the “awkward interface” problem to go away anytime soon. But while we stew on this issue, you might be interested to learn that NIST is taking over a few related problems in which it could conceivably make a real difference.

A few months ago, NIST released the 16th and final draft of its recommendations on definition of cloud computing. (Talk about insisting on getting it right!)  Not everyone in the health IT industry is even aware that NIST has kicked out a cloud standards document, which our friend Shahid Shah, “The Healthcare Guy,” is urging people to get onto their radar.  Maybe this time, NIST has a chance to actually standardize before an industry runs while with its own implementations of key technology.

I think I’ll finish with Shahid’s comments on the subject, as I think he’s pretty clearly got it right:

My strong recommendation to all senior healthcare executives is that we not come up with our own definitions for cloud components – instead, when communicating anything about the cloud we should instruct our customers about NIST’s definition and then tie our product offerings to those definitions. The essential characteristics, deployment models, and service models have already been established and we should use them. When we do that, customers know that we’re not trying to confuse them and that they have an independent way of verifying our cloud offerings as real or vapor.

If someone asked you how long it’s been since someone lit up an EMR, what would your guess be?  Five years?  Ten? Even 20? What if I suggested that the first EMR was installed 50 years ago in an Akron hospital?

According to IBM, the first EMR was rolled out at Akron’s Childrens Hospital in February 1962.  In Big Blue’s own words:

Though Dr. Lawrence L. Weed is credited for developing the first electronic medical record, the so-called Problem-Oriented Medical Information System (Promis), starting in 1969. But IBM, working with Akron’s Children’s Hospital, implemented a system years earlier that would be the grand-daddy of today’s EMR.

Other early players in EMR evolution included doctors at the University of Vermont, whose PROMIS system and later the POMR (problem-oriented medical record) followed in the late 1960s, as well as the Mayo Clinic. Still, it seems we may have IBM and a pioneering children’s hospital to thank for much of what we discuss so passionately here today.

By the way, it’s interesting to note that while the technology has evolved in astounding ways, the EMR as a concept hasn’t changed nearly as much.  For example, even back then execs were noting that nurses were spending far more time handling paper than they needed to (and that one patient could generate 50 forms, a number which I’d bet still hasn’t changed). It’s amazing that a problem we defined 50 years ago still defies easy solutions, but there you have it.

Meanwhile, courtesy of Scribd, check out the actual IBM press release  on the subject (typed on an oldie-but-goodie typewriter):

Doctors love them, but don’t think the iPad is ready to play a major role in clinical practice, as Apple hasn’t done enough to optimize it for healthcare, according to a new study by Spyglass Consulting Group.

According to a new report by Spyglass,  doctors don’t feel the iPad is ready to have an impact on care delivery. While 80 percent of physicians responding predicted that the iPad will have a positive impact on future care, it’s just not ready today, they said. (Most doctors I’ve talked with agree, noting that while it’s great for reading data, it’s extremely difficult to use for data entry.)

We’re not at all surprised to hear this given some of the iPad horror stories traveling around. For example, when Seattle Children’s Hospital pilot-tested iPads for its doctors, the result was a complete flop. Doctors there complained that that it was just too awkward to enter data into the otherwise sexy device. Shortly thereafter, IT switched its plans and rolled out a zero-client set-up.

So, what will it take to make the iPad clinically useful? To be successful in healthcare, Apple and its partners need to rewrite and optimize clinical apps to include gesture-based computing, natural language speech recognition, unified communications and even video conferencing, Spyglass research concludes.

I’d add that EMR/EHR vendors need to create native front ends for the iPad; given its penetration among doctors, I’m baffled by vendors who demand that doctors use their system via Citrix or the Web.

Unfortunately, with the exception of Epic’s Canto, few vendors offer a fully-fledged iPad app as a front end to their system. (One of few examples of a native iPad app from a smaller EMR vendor comes from Dr. Chrono, which, perhaps not so coincidentally, just got $2.8 million in venture funding.)

What’s more, Apple will have to do something about iOS security. It’s little wonder that 75  percent of doctors said that hospital IT departments weren’t eager to support mobile devices on corporate networks. While any device exposes networks to additional threats, Apple seems to have some particularly difficult problems, especially where its Safari browser is concerned.

Like the doctors surveyed by Spyglass, I have little doubt that iPads will end up assuming an important role in healthcare.  But given the snail’s pace at which native iPad apps are being launched, it may be a long time before that happens.

When a company like Greenway Medical Technologies (NASDAQ: GWAY) goes public, most of the post-IPO talk centers on what its leaders will do with the money.

Ideally, the newly-rich EMR vendor will do customer-friendly stuff like improving their product and strengthening their technical support organization. In reality, though, public companies have a different focus; their job is keeping the Wall Street folks who own their shares happy.

Since happy largely means only one thing — increasing profits and earnings per share — that vendor isn’t likely to take on new expenses. No, it’s more likely to find ways to charge more and sell more, rather than doing a better job of showing love to its existing customers.

SRSsoft’s Evan Steele has shared a nice analysis of  how KLAS customer support ratings (for companies serving the 6 to 25 physician practice) compare with the vendor’s financial status.  While they’re not exactly scientific, Steele’s conclusions are still striking; he concludes that five of the top six vendors are privately owned.

Now, I’m not sure how that correlates with another KLAS data point, in which publicly-held EMR/practice management vendor athenahealth (NASDAQ: ATHN) was named as top-ranked provider for its cloud-based EHR in December. Its stock has also been on a generally upward climb for the past 12 months, ranging from $39.87 to $72.70 per share.

Is it possible athena is managing to please both its customers and its investors? Well, if the typically nasty gossip you see on athena’s discussion board is any indication, no. It looks like grouchy insiders are shorting the stock, which some expect to plunge below its starting price to $30/share or so fairly soon.

That being said, one particularly intriguing comment suggests that Cerner (NASDAQ: CERN) is eyeballing athena, which observers think would be a good fit.

Cerner fits the profile I’ve outlined: it’s huge, profitable and what’s more, in need of a product to fill the physician niche it doesn’t own. If you think Cerner could just build its own physician presence, look at GE’s decision to drop doctor-oriented Centricity Advance. Clearly getting doctors to buy was  much harder than it looked at first glance.

Cerner doesn’t need athena to build its margins: analysts expect it to see sales growth of 13+ percent this year, to $2.5 billion or so. It should also see earnings growth of 22+ percent to $2.25 per share.

Buying athena would give Cerner a critical medical practice presence, and at the same time, let athena keep its customers happy without forcing it to play only for a Wall Street audience. In this situation, at least, maybe an EMR vendor can have its cash and eat it too.