Computerized Physician Order Entry Helps Hospitals Improve Care

Ten years after the Institute of Medicine released its influential report "To Err Is Human" (www.iom.edu/en/Reports/1999/To-Err-is-Human-Building-A-Safer-Health-System.aspx), hospital care still has many safety problems, and the quality of care remains lower than it should be in many institutions.

Hospitals could improve both quality and patient safety by using health information technology to standardize the processes of care and to ensure that vital information is available to clinicians when they need it. However, electronic health record (EHR) systems are multi-faceted and challenging to implement in acute-care settings, and few health care facilities have complete EHRs. As an initial step toward the automation of patient care, about 10 percent of U.S. hospitals have implemented computerized physician order entry (CPOE), which includes medication orders and orders for laboratory and imaging tests, as well as the ability to view test results and medication lists. While this falls short of a complete EHR, which also incorporates clinical documentation, CPOE, when it is properly implemented and utilized, represents a giant step toward better patient care.

My organization, Summa Health System, includes six hospitals in the greater Akron, Ohio, area. Our two largest facilities, Akron City Hospital and St. Thomas Hospital, which together have 1,050 licensed beds, have been using CPOE since 2006. We selected Eclipsys as our software vendor in 2003, and we spent the next three years planning and configuring the system with the help of consultants provided by the vendor.

The reason it took so long to prepare for CPOE implementation is that the system touches everything that occurs in the course of inpatient care, from admission to discharge. Planning was very important, because we were redesigning the way that people worked every day. First, we had to observe and analyze the work processes in the paper-based environment. Then we had to figure out how to make the CPOE system fit into our workflow, and, when necessary, how to change that workflow. At the same time, we had to get buy-in from our "users," which included physicians, nurses, pharmacists, and every discipline that is involved in patient care—not an easy task, and one that is still ongoing.

What facilitated the entire effort was the flexibility of the CPOE system we chose. In fact, we picked it partly for that reason. We knew that we would have to customize the application to meet the needs of our organization, and the system we acquired has allowed us to do that.

As a result of our successful implementation of CPOE, Summa Health has significantly improved the process and outcomes of care. That advance can be attributed mostly to the use of standardized, electronic evidence-based order-sets. In simple terms, for those unfamiliar with order-sets, an order-set is a preselected set of orders for medications, lab work, etc., based on the patient's condition. Physicians pick and choose which orders to include rather than having to hand write the same information over and over again for the same set of conditions. This article will explain how we transformed care by building paper-based order-sets into an electronic environment and through the successful physician adoption of using this innovative technology for patient care.

Climbing the hill of CPOE

A big advantage in our transition to CPOE was that we had already developed many evidence-based order-sets in the paper environment. When developing these order-sets, a multidisciplinary team involved in patient care was convened. This included physicians, nurses, pharmacists, and laboratory staff. When formulating the order-sets, the group reviewed the most current research and best practices. The order-sets were then formally approved through key hospital governing committees including physician medical and surgical department committees and the pharmacy and therapeutics committee.

Those helped prepare us for the transition to CPOE, because our staff already understood the concept of using order-sets, an essential component of CPOE. The paper order-sets had some problems, however. First, if physicians didn’t like the preprinted orders, they would handwrite their own. Second, many doctors didn’t like the idea of following evidence-based guidelines, which they considered “cookie-cutter medicine.” And third, it was very difficult to keep the paper order-sets updated, and the latest versions coexisted with older order-sets that were still in circulation. As a result, many physicians got frustrated and decided not to use the order-sets. Before CPOE adoption, only about 30 percent of orders involved the use of order-sets.

To transform the paper order-sets to electronic order-sets for CPOE, we formed a clinical decision support team. This team included five clinical nurse informaticists and a database administrator who built and configured the order-sets into the electronic environment. A pharmacist, a clinical nurse informaticist, and a quality nurse analyst reviewed each order-set to ensure accuracy and optimal ease of use.

Because medical research is continually advancing, we update orders as needed to incorporate the latest evidence. In the case of stroke care, we’re creating two new order-sets and updating the existing ones, because American Stroke Association guidelines have changed. Every order-set is updated at least once every three years.

When we implemented CPOE, we banned paper orders. Every physician is expected to enter orders into the system directly—about 80 percent of orders are done this way—or through verbal or phone messages to nurses who input the orders. A physician can deviate from an order-set in response to a particular patient’s condition. As a result, physicians now use the order-sets for 94 percent of their orders.

The request by physician leaders to develop more evidence-based order-sets is continually rising. We currently have more than 150 evidence-based order-sets available for CPOE.

Increasingly, the physicians are seeing that the use of the order-sets leads to better outcomes for their patients. Being scientists at heart, doctors believe in hard data, and when we showed it to them, most became convinced that order-sets could help them provide higher-quality care.

Decision-support tools and outcomes toolkits

The data we gave to our physicians was the result of much internal development work, coupled with the advice we received from other hospitals around the country that use the same CPOE system that we do.

The internal development project, which was part of our CPOE implementation, involved the creation of decision-support tools that help physicians improve safety and quality. For example, the evidence showed that administering the clot-busting medication tPA can help if it’s given within three hours after a stroke, but if it’s administered any later than that, it can be dangerous. So if a physician enters information into the CPOE system that the onset of symptoms occurred at 2 a.m., and the physician tries to order the drug at 7 a.m., the system is programmed “hard-stop” the prescription.

Our aggregate knowledge about best practices, population reporting, and outcomes assessments is collected in what we call “outcomes toolkits,” which are posted on the vendor’s web sites. These toolkits are now available for a dozen conditions, include stroke, acute myocardial infarction, deep-vein thrombosis, and diabetes. Every health care system that uses this CPOE software has access to the toolkits. All of us benefit from them, because they show us how to make better use of our information systems in a shorter amount of time than we would on our own. We also routinely share information we have gained during vendor-facilitated calls with other U.S. and Canadian hospitals that are using the same CPOE system. In return, they provide us with the fruits of their experiences.

Advances in Stroke Care

Since we had an outcomes toolkit for stroke and had reengineered our process for stroke care, we did an extensive analysis to show our physicians how CPOE had improved care. As the basis of our study, we used a number of Joint Commission performance measures and compared the results of these measures with and without the use of order-sets. In every performance category, compliance with the Joint Commission guidelines was higher (by an average of 40.5 percent) when physicians used the order-sets. When order-sets were used, 9.4 percent more patients went home directly from the hospital, and 16 percent fewer patients were discharged to a skilled nursing facility. In addition, 21 percent fewer patients went to a rehabilitation facility after discharge.

Readmissions within 31 days after discharge were 35.7 percent lower with the use of the order-sets, and the length of stay in the hospital was 7.5 percent shorter. Average direct costs dropped 11.4 percent, and indirect costs were 12.7 percent lower. All of this enhanced our hospitals’ bottom lines as well as patient outcomes.

When patients are discharged sooner from the hospital, but don’t go to a nursing facility or a rehab facility, that is a strong indication that they are recovering faster. The lower readmission rate also points to fewer complications as a result of hospitalization. What this indicates is that by hewing closer to the best practices recommended by experts, physicians are producing better outcomes. And what made the difference, the study shows, is their adherence to computerized order-sets.

Lessons Learned

Ultimately, CPOE is about the standardization of certain aspects of inpatient care. We have shown that this standardization can drive major improvements in the quality and safety of care. It is the collaborative effort of a highly dedicated and engaged patient-care team that has accomplished this successful implementation of technology. While a few physicians continue to resist any infringement on their autonomy, most of our doctors have seen the light.

What we tell our physicians is that they’re like the pilots of commercial jetliners. Those pilots have the knowledge and skills to fly their airplanes under almost any condition, but they rely on the gauges and tools built into their planes to help them avoid errors. Like the pilots, physicians are at the helm, but they can use the resources and the knowledge embedded in CPOE to deliver higher-quality care and prevent medical errors.

Ultimately, patients are the real winners. CPOE facilitates our ability to provide them with the most up-to-date, evidence-based treatment. In addition, CPOE makes the workflow more efficient, which means that lab results and medications are available sooner. No longer do paper orders for tests have to be hand-carried to the lab, or paper prescriptions picked up by the pharmacy. Lab techs and pharmacists receive the online orders instantly. Moreover, those orders are now completely legible 100 percent of the time, greatly reducing the chance that they will be misunderstood.

In short, CPOE is vital to the proper functioning of a modern hospital. When more institutions implement these systems, health care will be better, safer, and cheaper.

About The Author