Clinical leaders identify electronic, evidence-based order sets as tremendously effective at improving clinical decision support (CDS). Well-implemented order sets streamline workflow, drive computerized physician order entry (CPOE) and improve core measures right at the point of care.
Order sets are as good as the time and effort spent on them, says Dirk Stanley, MD, MPH, CMIO at Cooley Dickinson Hospital in Northhampton, Mass. “If you put a good process together and make sure you are incorporating safety, regulations and evidence-based best practices, you will love your order sets.”
He advises leaders to examine their current environment, practices, stakeholders and resources, and then develop a clear vision. A testing phase is critical, and providers need to determine if matching paper order sets are necessary in case systems go down.
“Always remember, order sets are a team sport,” Stanley says. “Physician engagement is important, but you can’t play a soccer match without your entire team.”
Lucio Martinez, MD, led the efforts to create standardized, electronic order sets at FHN Hospital in Freeport, Ill., and offers a four-step approach: (1) assess overall core measure performance to identify opportunities for improvement; (2) identify, analyze and revise order sets; (3) embed CDS functionality; and (4) measure outcomes.
At FHN Hospital, a governance structure was set up with assistance from an outside consultant to navigate culture issues, Martinez says. The governance structure consisted of five champions: an order set champion, a metrics champion, training and support champion, workflow and devices champion, and a communications champion. All reported to the PEHR committee.
Martinez served as the order set champion. “You need to have governance structure in place to fend off resistance,” he explains.
At the suggestion of a consultant, the physicians took on less of an advisory role and more of a driving role. “It would not have succeeded if it was IT driven,” he notes. “That was the key to our success but it was a multi-disciplinary approach.”
At the beginning, physician participation worked in part due to naivety. “It’s not as easy now, as their eyes are open. They’ve seen the impact of CPOE and the change to their practice style,” Martinez says.
This especially played out in the order set review process. “We used ProVation order sets and web review tools, and we needed to have physicians willing to be trained on how to do the web reviews.” The hospital experienced active participation during the initial development of order sets. But, during a review and updating effort last year, there was minimal turnout, cooperation and participation from the physicians. Martinez attributes that either to contentment with the order sets or apathy.
“We got 98 percent utilization when we went live. The paper order sheets have disappeared. We’ve been live in every unit of the hospital,” he says, noting initial challenges of physicians ignoring and skipping narratives and cues embedded in the electronic order set due to alert fatigue.
The order set committee was able to redesign the order sets and saw significant improvement in core measure compliance without alerts and pop ups. The hospital looked at core measures for venous thromboembolism (VTE) prophylaxis, urinary catheter removal post-op and beta blocker use in myocardial infarction, among others.
Metrics showed improvement in some of the Appropriate Care Measures and the Surgical Care Improvement Project measures—reduction in mortality and in length of stay for 11 months in 2011, Martinez reports.
Robert Budman, MD, MBA, CMIO at Yuma Regional Medical Center in Yuma, Ariz., has worked with four providers updating their order sets. Success rests on a sturdy governance structure when undertaking the process of moving order sets from paper to EHRs. “The real key is that it has to be interdisciplinary,” he says.
Budman says the process of standardizing order sets can be done internally, without a consultant, if the CMIO is willing to proactively lead the process and oversee governance activities. “The cost of maintenance is significantly higher the minute you add a content provider. Some [providers] spend millions of dollars,” he says. “We did it here fast and somewhat frugally.”
To get it done, governance must be put in place quickly, he says. With the CMIO ultimately in charge, an oversight committee of physicians from different departments should review order sets as developed by subject matter experts, he says. Once satisfactory protocol is established, the IT department takes the marching orders to build it into the systems and archive them on the EHR database. In the end, there is a global system of sets and also some that apply to specific departments only, Budman says.
He cautions against over defining and over perfecting. “More general tends to be better than complex,” he says. When order sets are over-engineered to drive too many decisions, workflows become onerous, Budman emphasizes.
At Yuma Regional, staff maintains about 200 order sets. Of those, 20 are for the emergency department (ED) and 180 are spread through different departments.
Budman knows providers with 400 to 500 order sets—and one place with as many as 1,600—that struggle to maintain that level of quantity and complexity. “You need general things to get patients into the hospital, and basic things in place for surgery or medical management,” he says.
“Your multiple use items like admit orders, diets, IVs, VTE prophylaxis, convenience meds and analgesics all should be fairly compact and uniform to cover the majority of patients and avoid differences to limit errors in CPOE over-utilization and/or incorrect ordering,” he says.
Regulatory requirements for accountability and Meaningful Use, related to antibiotics, blood cultures and pneumonia led to true improved outcomes. “Core measures are what really drove standardization of care,” he says. One initiative at Yuma Regional involved a hip fracture pathway established for patients admitted to the ED. The order sets include a specific protocol, including post-op and discharge instructions. It took four to six months to develop the whole pathway, and required coordination with the ED, hospitalist and orthopedic surgeon, he recalls.
“Literally within 60 to 90 days, we saw an improvement of baseline metrics. I’m talking tremendous improvements that really prove what order sets can do,” Budman says, noting reduced length of hospital stay, better cost containment and improved follow-up care.
While the journey from paper to electronic order sets is fraught with its share of challenges, clinical leaders agree that it is a critical stepping stone to improving outcomes through better standardization of care.