I came across a little bit older article from the Health Care Blog that asks the question, “What if I Had to do HIT All Over Again?”
I must admit that I was pretty struck by the thought of what could be done with an EMR that was built from scratch. Don’t get me wrong, I still think you’re crazy to develop your own EMR. However, I couldn’t help but wonder what might be done differently today knowing what we know about EMR now and the technology advances that have been made.
My mind then drifted off to think about what it would be like to build an EMR without any special interests. For example, you were building an EMR that focused on improving patient care. You were building an EMR that focused on improving doctor productivity. You were building an EMR that made everyone’s lives easier.
Contrast this with an EMR vendor who develops their software in order to sell more product. There’s a wide gap between the two methodologies. Unfortunately, I’ve heard too many stories of EMR vendors focused on building their EMR software to sell more of it. In the short term, this might be a great business strategy. Long term it will catch up with them.
How does someone selecting an EMR vendor identify this in the company?
This is indeed a problem. I have now been at this 10years and while we have sales, which are increasing, there has been tremendous pressure on us to incorporate a “point and click” templating model. If we had done so it is likely I would be likely semi-retired from the business or sold out to a big corporation. But the game for me has always been to improve patient care by helping the doctor in their patient interaction. Not just sell an EMR to a doctor. For this we have shunned CCHIT, Surescripts “integration” and dozens of other smaller “requirements.” Only time will tell if we are successful.
I went back and read the post from 2006, and many of the opinions were valid back then, but this is 2010 and software packages have become much more advanced and stable since then. The power that one or a small set of programmers has now has greatly increased.
The real question here is “Why am I buying an EMR?” Is my goal compliance? efficiency? safety?
Big EMR vendors cater to the concerns of those wanting to simply comply, while small EMR shops like Ankhos are delivering on real productivity. I could foresee a small industry of software companies starting up simply to implement custom EMRs. These companies will be much more agile and much more attentive to the flow of an office and would create a better product.
We can use natural selection to figure out which EMRs are winners. Good doctors (and nurses and admins) who are able to collaborate effectively with developers will create the superior EMRs.
Design by committee is surely a barrier, but how can you guarantee that software from a large EMR vendor was not designed this exact same way? It is not the avenue of origin that will make an EMR product great. It is the teamwork, attentiveness and expertise of its creators that will.
I think the crazy ones (small clinicians, at least) are the ones who buy the blue-chip software.
My 2c,
Nick Orlowski — lead developer at Ankhos
Carolina Oncology Specialists
Whether to build or to buy an EMR has been a question not unique to the medical or health care industry. It has also been a sticking point in general business.
Although the software development tools have simplified the task of building applications that can run on many more venues (namely something following web protocols), there is still the question of interoperability.
If one’s use of a product is meant for an entity that is an island unto itself, then the design of the application can follow every which way. However, if one of the objectives is to simplify interoperability or sharing of data, then this kind of constraint is one of many that will tie the hands of a custom developer.
But an off the shelf (COTS) software package won’t necessarily ensure an easy path either. As John points out, large software houses depend upon a continuing stream of revenue either by subscription or by new license purchase.
What might equalize the field is the kind of revenue that is being paid to a vendor. If all vendors were paid the same way, I would bet that the software might actually be more usable and more able to share data.
I once sat in a conference call where a very large database vendor said that proprietary standards were necessary to ensure mandatory customer upgrades. I think this model is slowly subsiding as the price advantage from newer kinds of software can almost equal or exceed the feature set of older COTS EMR.
There are vendors such as Practice Fusion (No fiduciary interest) that are ad supported and cost nothing in license fees to the user. Will many more of these kind of EMRs come forward? I think so. The critical mass, though, will only be achieved when even these “free” EMRs can interact with the pharmacy benefit managers, the supply chain and the financial parts of an operation.
In summary, I think it is a slippery slope to ask for software that is free from the profit incentive and yet have something that works. We shall see.
Richard,
There is a whole lot of movement happening in the EMR industry. My only fear is that there is too many vendors worrying a lot about market share and not building sustainable models built on EMR software that people will want to use forever. Certainly this is not new to business, but the race to capture market share is ugly and will result in a lot of doctors using an EMR which they don’t enjoy or the EMR vendor no longer supports.
There is little hope of survival for new EMR entrants to the market now, and custom EMRs are only viable for large hospitals or RHIOs – an EMR now includes at a minimum doc management, e-prescribing, vitals /Hx management plus many PM functions like scheduling. Total code base will be in the 350,000 line minimum range – to produce an EMR in two years this is about 900 lines a day or about 20 pages of code per day. This does not include design, DB layout and a logic map nor the fact that the base standard will include many other features like HL7 within 2 years. Some billing companies think they can step up but they are starting with only 100,000 lines usually. The focus is different as well – would you let the billing clerk tell you how to document? Maybe some could help, most couldn’t. Basically we are going to have fewer EMR vendors going forward not more, and custom is almost impossible for a small practices. But many will think they can do it better, most have failed, and the likelihood of failure for new entrants is now even greater in 2010.
John, how funny.
I read your 2006 entry immediately followed by the 2010.
Hindsight is 20/20…
My review was in the context of building an EMR software package – I’m a developer who knows a handful of doctors, all who hate the packages they have now. They take hours to get things done and there’s zero intellegence built into the GUI design. So sad, while I’d like to make a living, I’d much more like to make Doctor’s lives easier. Unlike other professions, you can draw a straight line from making a doctor more efficient to saving lives.
My question is – is there a standard (HIPAA or other government related) that must be meant to sell an EMR application?
Dave,
That’s kind of a loaded question. The short answer is technically: No. Practically: Yes.
The longer answer is that you can sell whatever you want. There’s no government requirement. However, there are a few standards/certifications that you’ll want if you want doctors to take you seriously. To name the most important…
HIPAA – If you don’t meet HIPAA you’ll never sale a piece of software in the US. It’s essential or no doctor will use it.
EHR Certification (ONC-ATCB EHR Certification) – You don’t have to have this, but you do have to have this if the doctors who use your EHR want to get the EHR incentive money. There are a few specialties where this might not matter, but for practical purposes everyone wants to see that EHR certification.
On a different note, I hope you do help design a better interface. It is noble to be able to help save lives.
I wonder if you couldn’t help by connecting with an open source EMR like OpenEMR. Would save you a lot of time in developing and you could have some impact really quickly.