May 2017

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      

« Sometimes, Not Knowing is Better! | Main | How Many Different Clinical IT Systems Do You Use? »


Dave Sellers

Alan, I think the guys in the trenches are ready to go, it is the provinicial College of Pharmacists that are holding up the works. In Ontario, the Pharmacist will be fined if they accept a document without an "original signature".

Raymond Simkus

Warning! I am a practicing doctor who sees real patients using an electronic medical record (EMR). This is the first line of a blog ( The article went on to talk about various issues related to EMR use and specifically mentioned the McMaster paper that showed a lack of evidence based benefits for computer-based prescribing. Although this conclusion seems to get generalized to EMR use in general. One explanation could be that the systems that were being used were not really what I would call well integrated high performance systems. Most of the data that was looked at was hospital based.

Looking at ePrescribing from a high level perspective does not reveal the difficulties when it gets down to implementation. A physician will do about 10,000 prescriptions per year using about 500 different drugs. There are a few things that have been a headache for many years. It is unfortunate that physician users of these systems are not aware of the difficulties and those that do speak up are ignored. FIRST IS THE UNWARRANTED USE OF ALL CAPITALS FOR DRUG NAMES. Another issue is the fixation of using the full drug name which typically includes the brand name, the strength and the form of the medication. This is not how physicians think of medications. I would like to be able to search for medications by indication, drug class, what the patient has had before and then the name of the drug and after that I would like to see and select the available dosages. Why can't EMRs follow this sequence or at least make it possible for users to search for a drug on these criteria. Another sore point is that if you go to change a dose you generally have to discontinue a prescription and then start a new one having searched for and selected a new drug. Wouldn't it be nicer to let the user just change the strength of the drug? Trying to read and understand the sequence of previously prescribed drugs is usually a miserable exercise because of the way previous prescriptions are recorded. Doing refills of previously used medications can be a pleasure if the EMR has things set up right.

There is continuing pressure for EMRs to use DIN numbers as the identifying code for medications. Staying at a high level I have no hesitation in saying that DIN numbers are the wrong identifier to use because they refer to a specific formulation and package. What is prescribed is not necessarily what is dispensed. There are a number of other choices for identifiers but it should NOT be the DIN number.

While this could be a really attractive function in an EMR the EMR developers have many different and partially functional solutions. Physicians need to recognize the problems, speak up and get the point across that things are just not good enough.

The comments to this entry are closed.