Mark Mench EHR System Selection

From Clinfowiki
Jump to: navigation, search

For the sake of argument I will equate EHRs to EMRs. I could write a list of properties that an EHR needs to have, but that is best left to subsections in this category of the Wiki. This writing is about the basics.


The first question that you must ask yourself, for obvious reasons, is “how much does it cost and can we afford it?”


The next question has to be, “Do we have the necessary technical infrastructure and leadership?” There is no point in going past the point of product research until you know you have the money necessary to fund both the IT staff and the IT architecture necessary for such a large product. Reading the KLAS report suggests to me that many times when a hospital blames the EHR vendor the problem is in reality due to problems with its own IT infrastructure.


After deciding that you have the money and the IT infrastructure, or can get them, you have to ask yourself if you have the personal leadership abilities, and medical and nursing staff leadership to pull it off. Make no mistake, if you do not have the above you will fail. If you do not have the personal leadership necessary to get the right people for this job then you should hire someone who does. For an EHR to succeed it cannot be sent down from the almighty on high in the administration penthouse. It must be accepted, owned by, worked and even struggled with by the people who are going to use it. That takes leadership on all levels, and that leadership must be imbued with the vision that the goal of an interoperable EMR is necessary, even vital and good. A broad reading of the literature, and my own experience tells, me that unless you have clinicians who are not only in leadership positions but who also are enthusiastic about the goal of an EMR you might not get there.


Next, you have to map every process in your hospital/clinic that will be replaced or affected by the EMR. Without knowing what your processes are you cannot know were you are going. This is not to say that you are going to automate your prior practice, it is just to say that without knowing your current systems you can’t know how to use the technology to change your system for the better.


Next, the team has to research every piece of information that is available about the features necessary to have a good EMR, list them, internalize them, and match them to your current and future needs based on the systems mapping that you have already done. Then you can start narrowing down the field of vendors, all of whom will love to lavishly wine and dine you, and promise you the world.


The expression, “Trust everyone but cut the deck yourself.” applies here in spades. These companies are not above shading the truth, even outright lying, in order to make the sale. We’re talking big money here, millions of dollars for an enterprise solution. You cannot accept anything they say as the truth until you verify it by site visits and actually see that what they say they can deliver has been done somewhere else and is working. I know people who had been told by the vendor that a certain site had x, y, and z up and running and available for inspection only to find once on the site that it wasn’t true. The institutions where you do site visits may be remunerated directly or indirectly by the vendor and the facility administration may therefore have a stake in making the vendor’s product (and themselves) appear good. For this reason you must insist on being allowed to talk in private to the troops in the trenches, the grunts who have to use that product day in and day out, and who pay for the administration’s mistakes if the system is poor in design, or has problems interfacing with legacy systems. These are the people from whom you will find the truth.

Legacy systems

Almost every institution has legacy systems that it doesn’t want to, or can’t, leave behind. You must be absolutely sure that the prospective EMR can interface with the legacy systems. Again, do not take it on faith. You must ask the vendor to show you an instance where its product and the same legacy system are already interfaced and working. If you can’t do this and you still want that EMR be certain that the contract stipulates that it is void if the vendor can’t interface when it has promised to do so.

Obviously there is an incredible amount more to choosing a system, but unless you learn the above lessons the rest of it may not count for much. I wish the best of luck to those who undertake the journey, because it is a more of a journey than a goal. Mark Mench, M.D.