It is inherently obvious that the patient information such as previous visit history, laboratory data, and prior EKG should be readily available to the practitioners. However, many current formats use multiple data silos—not integrated repositories—requiring multiple separate passwords and access steps inhibiting efficiency.
When the patient goes through the registration process, the patient’s information should be actively provided to the department via a secure data terminal that includes past visits and discharge. A more modest program can begin at least with retrieving the most recent EKG when a new one is performed, as well as the old medical records present at the facility.
There has been a great deal of excitement concerning physician order entry. However, detractors suggest this is another efficiency workaround, enlisting the highest paid personnel to perform clerical activity. This shifts responsibility to another cost center—from nursing to the physician area—while the unit secretary is shifted to patient care responsibilities associated with a nursing budget seeking cost savings in that area.
While proponents suggest this approach was adopted to decrease medical error and more recent evaluations; however, some have reported no improvement at all or, in fact, worsening of medication and administration errors.
Medication errors are a major concern. A prospective pediatric trial of computerized physician order entry (CPOE) examined adverse drug events (ADE) that were then separated into two subcategories: medication prescription errors (MPE) and rule violations (RV).1
The results were favorable with a decrease in ADE to 40.9% (2.2 to 1.3) RV from 97.9% (6.8 to 0.1) and MPEs from 99.4% (30.1 to 0.2) per 100 errors from an overall error reduction rate of 95.9%. It is important to note the greatest effect in so-called “writing” errors, and less so with adverse drug effects with potential patient harm.
Another trial of CPOE in a critically ill patient cohort found a significant increase in mortality from 2.8 to 6.6% after implementation.2The inference is that while physicians perform data entry tasks, more timesensitive patient care therapy may be omitted.
A prophetic review entitled “Computerized Physician Order Entry: Helpful or Harmful?” suggested the apparent dichotomy of results.3 They suggest that the process variables associated with CPOE have been improved, but the change in patient outcome is lacking. They also document a significant increase in costs and a potential for a parabolic increase in ADEs during the implementation phase. The optimum scenario for physician/secretary communication is to have dedicated data entry personnel with physician on-line clarification ability.
Another impediment to care involves the ever-burgeoning documentation requirements which involve multiple patient-focused but non-integrated items mandated by both external (JCAHO) or internal sources (hospital information dashboard).
One approach is to establish a centralized committee to oversee all documentation requirements to ensure feasibility, efficiency, and lack of redundancy. It is essential to ensure a properly functioning electronic interface and not hand-written documentation as the electronic data processing program—the Electronic Medical Record (EMR) is being developed.
A common problem is that data processing tasks begin with system capability and are not necessarily task specific. Ideally, data processing system development should begin with the end-user needs as a basis of operation.
Therefore, assessment of data capability and needs should begin with clinical staff and then ‘constructed’ by the information systems (IS) afterward for best use.
1. Potts, A.L., Barr, F.E., Gregory, D.F., Wright, L., Patel, N.R. “Computerized physician order entry and medication errors in a pediatric critical care unit.” Pediatrics 2004; 113(1 Pt 1): 59 – 63.
2. Han, Y.Y., Carcillo, J.A., Venkataraman, T.S., Clark, R.S., Watson, R.C., Nguyen, T.C., Bayir, H., Orr, R.A. “Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system.” Pediatrics 2005; 116(6): 1506 – 1512.
3. Berger, R.G., Kichak, J.P. “Computerized physician order entry: helpful or harmful?” Journal of the American Medical Informatics Association 2004; 11(2): 100 – 103.