EMR Design

Southwestern Medical Center University Hospitals continues the work of implementing an electronic medical record for our 450-bed, two hospital system. We are on track to implement an electronic results repository, emergency department system, pharmacy information system, eMAR, clerk order entry system and basic nursing documentation in 2008; 2009 will include CPOE and full nursing documentation.

We are fully staffed and heavily engaged in the design of Year 1 modules. Though our implementation of EMR modules will be staggered through the course of Year 1, we are performing an integrated design at the beginning of our project.

It is worth noting at this point that deciding on the staggered approach has in and of itself, been a difficult, gut-wrenching process. Three natural forces were pushing us towards a big-bang approach: the force of clinicians, who would like more functionality sooner; the force of our own IT staff that would prefer all new systems so as to minimize the dependence on legacy systems and interfaces; and the force of our vendor who believes that life would be far better if we implemented six of their modules at once instead of three so clinicians wouldn't have to look into multiple places and processes.

Ultimately for us, however, the deciding factor was that both our clinical operations and our IT staff have very little operational experience in an inpatient EMR environment. Our clinicians currently operate in a world of paper orders and documentation. Our IT staff is well versed in supporting best-of-breed ancillary systems in the inpatient environment. In the end, the workflow and technical implications of a “big-bang” were far too great a risk for our organization.

It is important that we take meaningful pieces of the EMR enironment and implement themgradually, giving our clinical staff the ability to assimilate the automtion and adapt processes to optimally serve one another and our patients. This approach will allow us to focus our training and support on one module at a time. We have recognized the importance of moving quickly and have planned four go-lives in one year: Results Repository (Q1), Emergency Department (Q2), Order Entry (Q3), Pharmacy, eMAR, Nursing Documentation (Q4).

The design process has been at times exhilarating and at other times, devastatingly painful. The positive moments have come from the tremendous promise of information technology to improve clinician communication, provide relevant clinical reminders, and improve the quality of documentation. The pain has come from the complexity of inter-related technologies and workflows and the newness of partnering with clinicians on the design.

During the design, three common themes have consistently drawn a great deal of attention: the medication reconciliation process, clinical handoffs from one unit to the next, and the ease of use for physicians. We have not finalized our designs in these areas, and most likely we have not finished collecting all our questions.

In the area of medication reconciliation, we are focused on three major areas of design:

Structure of the eMAR. The transition from the paper world to the electronic could not be more relevant than in the physical screen format of the eMAR and how it will be viewed in each setting of care. Our current experiences in the OR as well as the clinics have shown us the value of having the same information everywhere; however, the visual display has been tricky.

Inpatient-outpatient flow. Being an integrated healthcare system, medication reconciliation from the clinic setting to the inpatient setting and back to the clinic setting has dominated our consciousness. The devil is in fact in the details: the level of standardization of processes in this arena is significant to achieve the desired benefit.

Medication review by pharmacists. The literature is overwhelmingly in favor of pharmacist review of medication orders. The technology is a lot less clear on how to make this easy to use for both pharmacist and physician. This is further complicated by specialized needs in unique settings such as the emergency department, operating room or ICU.

In the area of clinical handoffs, we are focused on three major areas of design:

Short reports. It may sound simple, but not to today's technology vendors. Our current operating room system (from our enterprise vendor) produces a 25-page, poorly formatted tome that is a small pillow for the patient being wheeled into the ICU, which is currently in a paper environment. Aside from the obvious fact that the CIO needs to be fired (sadly, that would be me), this report demonstrates the need for a move away from the traditional EMR report that prints every field that has been documented. We are working on a two-page report that has only the information necessary for the accepting unit.

Dashboard view. When we go-live, we realize that we need something better than a two-page paper report: we need a dashboard view for the clinician. This view, which can be customized by service or unit, would present key clinical information on the admission, including key results and discharge notes.

Physician consult workflow. The consulting physician has traditionally been overlooked in many EMR installs. The unique workflow and screen requirements of consult services are a major area of focus for UT Southwestern. It is also an area where we hope to heavily involve our physicians.

In the area of physician workflows, we are still very early in our progress. We have done great work to standardize our nomenclature for orders and results. Though this effort has been rewarding, it has demonstrated the tremendous need to “correct” reports from our current ancillary systems that do not map to “appropriate newly decided” nomenclature. This has been the cause of enormous unplanned work, but will ultimately benefit us tremendously. The other major area of focus has been to work with physicians who are savvy clinical users of the EMR, and asked them for guidance on screen design and workflow for the inpatient side in an effort to make the system more easy to use.

We still have a long road ahead of us, and we are excited about the prospects. We've developed a new graphical status report that we plan to distribute to our key clinical stakeholders on a weekly and monthly process. We do indeed have a lot to talk about.