Digitized Medicine

Physicians would prefer that an EMR computer screen emulate a tri-fold flow sheet that inspires a "gestalt feel" for the whole clinical situation.   Instead, data split among multiple screens with the need for numerous mouse clicks and excessive scrolling in order to see all the information tends to frustrate clinicians.  The frustration comes as a result of the fragmentation of our thought processes and patterns which have developed over time.    In addition, "user interface" issues can become a patient safety risk.  The following case study provides an example of such an issue.   

Case Study:   A small practice had their EMR vendor develop a custom report called the Patient Summary that “pulls in” EMR data including lab and radiology results.  This report is useful while on-call because they can remotely access it from home over the Internet.   Over one weekend the on-call physician discovered that some of the lab results for one of his patients did not show up on the Patient Summary.    He checked several other patients and found another example of missing results.   He decided that this report was unreliable and that he would not use it until the issue was understood.  The EMR vendor was notified on Monday.   They quickly determined that the report was working normally and no data was missing.   They suspected the doctor forgot that this report displays a little <+> symbol at the bottom of the screen if there are additional lab results that will not fit in the space provided.   At that point the physician realized that he had used his new iPhone for the first time and that the small viewing display had made it very difficult to for him to see the small <+> symbol.

Key Points:

• Displaying medical data on a computer screen in a manner that meets the cognitive challenge at hand is difficult and can fragment a physician’s thought processes

 Since tri-fold computer screens are not a practical solution today, physicians need to involve themselves in the development of the EMR content screens that they will be using on a day-to-day basis.  For most EMR products these screens can be configured and used in various ways.  In this case the small display inherent to smart phones was inadequate when viewing the report.   But this is because it was not designed with the small viewing area and resolution of a smart phone in mind.  If presenting this data on a smart phone is a requirement for this physician, a new report will have to be developed in order to meet that new need.  

• “Think simple” when developing EMR screens that clinicians interact with

Complexity makes EMR use even more difficult and less safe.   Most EMRs allow modifications to some EMR screens that physicians interact with such as documentation templates, reports, orders and order sets.   Good advice to those who develop these screens is to “think simple”.   For example, if six gastroenterologists decide to each create their own list of 25 order sets to accommodate personal preferences, their EMR order set screen will display 150 order sets.   Creating a screen that displays 125 choices for physicians has a pre-determined fate.  They would be better served by collaborating on 25 evidence-based order sets to share.   The result will be less frustration, less variation from evidence-based medicine and less work when annually reviewing the content of order sets.

Inadequate training of staff and clinicians is a common problem encountered when an electronic medical record (EMR) is implemented.  The following case study illustrates how poor training can impede the operations of the physician's office and even lead to patient safety issues.   A successful training strategy will avoid these types of problems by ensuring the staff and clinicians are knowledgeable about proper EMR use and that the staff who are responsible for configuring and maintaining the EMR are skilled and working as a team.    

Case Study:   A multi-office practice has used an EMR for 18 months.   Mary, an office manager with project management and IT experience, is the primary caretaker of the EMR.  She has struggled with two other office managers who want to have the same access she has to configure the EMR.   Their argument is that they know what’s best for their offices and that Mary is too busy to meet their needs.   Mary argues that she is not too busy, but that instead it takes time to properly test and manage changes made to the EMR.  Their arguments are taken to the physician’s EMR oversight group who ask Mary to just “let it go” and provide the access those office managers need. 

One week later Mary was suddenly inundated with trouble calls.  Physicians were unable to enter diagnosis codes and their staff unable to work claims.  Mary called the two office managers who swore they had not done anything wrong.  One of them, whose office was having no problems, admitted that she did add several diagnosis codes to a template because her doctor wanted them.  Mary subesequently discovered that she had failed to link all of the other physicians to the new template which is why her office was the only one with no problems.  Mary fixed this issue, but then decided to run an audit to see if any other changes had been made to the EMR without communication or notices.  She indeed discovered a change the other office manager made to a parameter called “Allergy Severity Default” with the default answer changed from “Severe” to “Mild”.  Mary knew about a “quirk” with this EMR whereby it fails to trigger an allergy alert if the allergy is entered as “Mild”.  She had previously taken the issue to the physician oversight group who determined that the answer in this field must default to “Severe” when physicians enter an allergy.  They felt it was a patient safety risk if every time a physician entered an allergy they also had to actively change the default answer to “Severe”. 

When Mary explained this, the office manager replied that her physician claimed a “Severe” allergy is one where anaphylactic shock occurs and that he was tired of always changing the answer from “Severe” to “Mild”.   Mary changed the default answer back to “Severe”, asked the EMR physician oversight group to re-educate the physicians and began working with the EMR vendor to completely remove “Mild” as an available answer.  The vendor complied promptly.

Key Points:

• The most common source of problems with using EMRs is inadequate training

Case studies of EMR implementations, whether successful or failed, consistently list “training” as a key factor for success.   Ongoing educational reminders, especially for “work-arounds” and unique issues as exemplified in this case, are often useful.

• Be wary of “work-arounds”

“Work-arounds” are encountered because people creatively develop ways, especially manual ones, to work around technology when it obstructs them from doing something.   Be wary of EMR work-arounds and make sure they are the best solution to the problem. In retrospect, completely removing “Mild” as an option in this case would have been a better initial solution instead of the work-around that was developed.

• Resolving EMR-related patient safety issues is a shared responsibility between the physician and EMR vendor

Work collaboratively with the EMR vendor and prioritize issues for them.  If ten issues are reported but only one of them is a patient safety issue, prioritization will focus the vendor’s resources on the important issue.   Mary immediately notified and educated the vendor about the patient safety issue in this case.

• Proactive management of EMR changes will reduce the number of EMR-related problems

There are standardized “change management” practices that minimize the risk of unexpected EMR problems.  These proactive practices ensure that each change is adequately tested, approved and communicated in advance.  Advance communication of an EMR change should indicate who, what, where, when and why changes are being made. This provides an opportunity for critical feedback.

• Effective communication is essential for safe patient care

Poor interpersonal relationships and the lack of effective communication among staff directly contributed to the problems in this case.  Discussion of the proposed changes would have allowed Mary or others to intervene and avoid the problems entirely.